Handbook of Polyblends
Handbook of Polyblends
Handbook of Polyblends
and Composites
Volume 1
Editors:
A.K. Kulshreshtha
and
C. Vasile
A catalogue record for this book is available from the British Library.
Editors:
A.K. Kulshreshtha
and
C. Vasile
A catalogue record for this book is available from the British Library.
Editors:
C. Vasile
and
A.K. Kulshreshtha
A catalogue record for this book is available from the British Library.
Every effort has been made to contact copyright holders of any material reproduced within
the text and the authors and publishers apologise if any have been overlooked.
Editors:
C. Vasile
and
A.K. Kulshreshtha
A catalogue record for this book is available from the British Library.
Every effort has been made to contact copyright holders of any material reproduced within
the text and the authors and publishers apologise if any have been overlooked.
Editors:
C. Vasile
and
A.K. Kulshreshtha
A catalogue record for this book is available from the British Library.
Editors:
C. Vasile
and
A.K. Kulshreshtha
A catalogue record for this book is available from the British Library.
1. They provide the ability to lower costs without sacrificing those properties that are
most desired;
2. They provide the ability to control or tailor properties without the invention or creation
of completely new polymers, extending engineering resin performance by diluting it
with a low cost polymer adjusting the composition of the blend to customer
specifications;
Polymer alloys and blends are one of the most dynamic sectors of polymer research and
industry, representing one of the fastest growing areas in polymer science, as it appears
from the large number of publications and patents. Review articles and books are very
frequently published due to the necessity for up-to-date information and specialised
international and national symposia are held at least once a year. Not only is the science
of polymer mixing steadily improving but the technology is also undergoing a rapid
evolution. Polymer blends have invaded the ever-growing market of automobile and
1
Handbook of Polymer Blends and Composites
electronic industries and much of the equipments that is in day-to-day use in our homes.
In particular the annual growth rate of engineering polymer blends and alloys is 13%
17% while that of plastics industry is only 2%-4 %. Clearly the plastics industry is
moving toward more complex systems.
This handbook is intended to provide an overview of the theory and practice of polymer
blends and composites. It is a collection of monographs on the subject of polymer blends
and composites but much remains to be done and understood. The subject is huge and the
number of pages is limited. The choice of subjects has changed over the years as science
and technology have progressed. The Handbook of Polymer Blends and Composites is in
four volumes. The first two volumes (of which this is the first) are concerned with the
state-of-the art of composites development, characteristics of particulate fillers and fibre
reinforcements and interface characteristics, main procedures of composites manufacture
and their applications. The other two volumes are dedicated to polymer blends. These
volumes start with general aspects including terminology, thermodynamics of polymer
mixing, peculiar behaviour of polymer blends (glass transition, crystallisation, ageing,
degradation, commingled waste treatment, etc.), and the progress recorded in investigation
and application of various classes of polymer blends (based on polyolefins, polyvinylchloride,
styrenic polymers, polyesters, polyamides, polyacrylates, polyvinylalcohol, polyurethanes,
etc.), ending with speciality polymer blends such as silicones, liquid crystalline polymers,
lignocellulosics and eco-friendly polymers blends.
Practical and theoretical investigations are presented, which are aimed at generating an
understanding of the fundamental nature of polymer mixtures and composites and
describing progress in the thermodynamics of mixing (both in solution and solid state)
of binary and multi-component systems.
It has been established theoretically that most pairs of chemically different polymers are
mutually immiscible and mixing usually results in materials, which are phase separated
and have weak polymer-polymer interfaces, and therefore have poor mechanical
properties. Recent studies are showing many more examples of existing polymer mixtures
exhibiting regions of miscibility than previously anticipated.
The mixing together of different polymers has been carried out for many years for various
purposes. Rubbers have been used to flexibilise rigid plastics. Rigid plastics have been
used to stiffen rubbers. Many adhesives, sealants, and coatings have been based on blends
of polymers. We have been dealing at all times with heterogeneous rather than
homogeneous systems. It even appears that many of the successful results of mixing
polymers are attributable to the properties obtained from two-phase systems.
Consideration of heterophase systems has been made not only for physical blends and
composites but also for copolymers and interpenetrating polymer networks.
2
Preface
The earliest blends were prepared by exploiting the natural miscibility of the components.
Good examples are blends of polyvinylchloride and acrylonitrile rubber or polyphenylene
ether (PPE) with high impact polystyrene (HIPS). The miscibility was the main criterion
for blending. For example the disadvantages of PPE (processability and impact strength)
are compensated for by the advantageous properties of either polyamides or HIPS. Earlier
Noryl was developed based on PPE/HIPS, then the second generation of such material
was formulated by using reactive methods of compatibilisation. During the next stage
miscibility still was the main criterion to develop new blends. New blends of immiscible
polymers were developed using graft or preferably block copolymer having one part
interacting with one polymer and another interacting with the other. The third stage of
chemical modification of immiscible polymers to ensure their miscibility is via for example,
ionic or donor-acceptor interactions. The latest developments are methods of reactive
processing, reactive compounding, reactive impact modification, etc. In recent years the
blends are increasingly used as a base for either foams or composites. A better control of
multi-component interfacial properties either via sophisticated chemistry or reactive
processing made this step possible. Foamed polymer blends are used for computer
housings, hospital or office furniture, automobile parts, etc. Reinforced grades of polymer
blends are also in commercial use.
Polymer blends and composites will certainly be continuing as lively research areas -
greater understanding will allow controlled manipulation and optimisation of properties.
The field of thermodynamics of polymer blends is too extensive to be covered in one
book. There are many questions that have not been considered or were considered only
briefly. New aspects are being explored. The phenomenology of chemical reactions
between polymers at interfaces is only now being examined in depth from both
experimental and theoretical standpoints.
Much less attention has been given in the published monograph and reviews on polymer
blends to the thermodynamic properties of complex materials like gels,
biomacromolecules, micelles, colloids, block copolymers and similar substances that are
often called soft materials. Such materials abound in nature and technology, application
of chemical thermodynamics to soft materials has been delayed because of experimental
difficulties and because, until recently, there were few theoretical models available for
describing assemblies of complicated molecules. New classes of multi-component polymer
systems not considered for a long time as polymer blends, are now included as formulations
for controlled drug delivery, hydrogels, membranes, multi-responsive systems containing
block and graft copolymers, etc. Progress in this area is presented in two chapters of the
handbook. Thanks to recent advances in statistical mechanics and molecular physics,
and thanks to increasingly fast computers, it is now possible to develop a hard science
for soft materials The creation and understanding of soft materials depend primarily
on experimental science. Now statistical mechanics is able to provide guidance toward
3
Handbook of Polymer Blends and Composites
Over 50 professionals from academia and industry (coming from Europe, America, Japan,
India) have contributed to this work. They were asked to provide a survey of their area
of expertise that would enable non-specialists to become aware of advances and future
trends in the area. Certainly, such a distinguished group of authors provides the needed
balance and perspective. However, including many authors can also lead to particular
complexities in a project of this type. Even with so many authors, all subjects relevant to
polymer blends and composites cannot be addressed. Compromises have been often
reached, and the end product is different from that originally envisioned. The production
of a such book is not easy, we open ourselves to the inevitable charges of incompleteness
and imbalance. The authors have attempted to cover most of the important areas but
there are bound to be some omissions and the balance depends on the readers knowledge
and interests. Nevertheless, the editors and authors hope that this book will lead to more
texts on the topic and will stimulate interest in this field of polymer materials. The polymer
blends and composites fields has always been wide open with opportunities, stimulation,
comparison and intellectual ideas.
The editors and the contributors are well aware that no book of such wide coverage can
be free of shortcomings, and all critical comments will be gratefully appreciated.
We offer our special thanks and we wish to express our sincere gratitude to Ms Frances
Powers of Rapra Technology, who initiated this book and who waited patiently for
result, for her understanding, guidance during production of this handbook, pertinent
English language corrections and scientific comments. Thanks are also due to Ms Claire
Griffiths (Editorial Assistant) and Mr Steve Barnfield for typesetting and cover design.
Cornelia Vasile
Petru Poni Institute of Macromolecular Chemistry, 41A Grigore Ghica Voda Alley,
Ro. 6600 IASI, Romania.
A.K. Kulshreshtha
Indian Petrochemical Corporation Ltd., Vadodara 391346, India.
4
Preface
1. They provide the ability to lower costs without sacrificing those properties that are
most desired;
2. They provide the ability to control or tailor properties without the invention or creation
of completely new polymers, extending engineering resin performance by diluting it
with a low cost polymer adjusting the composition of the blend to customer
specifications;
Polymer alloys and blends are one of the most dynamic sectors of polymer research and
industry, representing one of the fastest growing areas in polymer science, as it appears
from the large number of publications and patents. Review articles and books are very
frequently published due to the necessity for up-to-date information and specialised
international and national symposia are held at least once a year. Not only is the science
of polymer mixing steadily improving but the technology is also undergoing a rapid
evolution. Polymer blends have invaded the ever-growing market of automobile and
1
Handbook of Polymer Blends and Composites
electronic industries and much of the equipments that is in day-to-day use in our homes.
In particular the annual growth rate of engineering polymer blends and alloys is 13%
17% while that of plastics industry is only 2%-4 %. Clearly the plastics industry is
moving toward more complex systems.
This handbook is intended to provide an overview of the theory and practice of polymer
blends and composites. It is a collection of monographs on the subject of polymer blends
and composites but much remains to be done and understood. The subject is huge and the
number of pages is limited. The choice of subjects has changed over the years as science
and technology have progressed. The Handbook of Polymer Blends and Composites is in
four volumes. The first two volumes (of which this is the first) are concerned with the
state-of-the art of composites development, characteristics of particulate fillers and fibre
reinforcements and interface characteristics, main procedures of composites manufacture
and their applications. The other two volumes are dedicated to polymer blends. These
volumes start with general aspects including terminology, thermodynamics of polymer
mixing, peculiar behaviour of polymer blends (glass transition, crystallisation, ageing,
degradation, commingled waste treatment, etc.), and the progress recorded in investigation
and application of various classes of polymer blends (based on polyolefins, polyvinylchloride,
styrenic polymers, polyesters, polyamides, polyacrylates, polyvinylalcohol, polyurethanes,
etc.), ending with speciality polymer blends such as silicones, liquid crystalline polymers,
lignocellulosics and eco-friendly polymers blends.
Practical and theoretical investigations are presented, which are aimed at generating an
understanding of the fundamental nature of polymer mixtures and composites and
describing progress in the thermodynamics of mixing (both in solution and solid state)
of binary and multi-component systems.
It has been established theoretically that most pairs of chemically different polymers are
mutually immiscible and mixing usually results in materials, which are phase separated
and have weak polymer-polymer interfaces, and therefore have poor mechanical
properties. Recent studies are showing many more examples of existing polymer mixtures
exhibiting regions of miscibility than previously anticipated.
The mixing together of different polymers has been carried out for many years for various
purposes. Rubbers have been used to flexibilise rigid plastics. Rigid plastics have been
used to stiffen rubbers. Many adhesives, sealants, and coatings have been based on blends
of polymers. We have been dealing at all times with heterogeneous rather than
homogeneous systems. It even appears that many of the successful results of mixing
polymers are attributable to the properties obtained from two-phase systems.
Consideration of heterophase systems has been made not only for physical blends and
composites but also for copolymers and interpenetrating polymer networks.
2
Preface
The earliest blends were prepared by exploiting the natural miscibility of the components.
Good examples are blends of polyvinylchloride and acrylonitrile rubber or polyphenylene
ether (PPE) with high impact polystyrene (HIPS). The miscibility was the main criterion
for blending. For example the disadvantages of PPE (processability and impact strength)
are compensated for by the advantageous properties of either polyamides or HIPS. Earlier
Noryl was developed based on PPE/HIPS, then the second generation of such material
was formulated by using reactive methods of compatibilisation. During the next stage
miscibility still was the main criterion to develop new blends. New blends of immiscible
polymers were developed using graft or preferably block copolymer having one part
interacting with one polymer and another interacting with the other. The third stage of
chemical modification of immiscible polymers to ensure their miscibility is via for example,
ionic or donor-acceptor interactions. The latest developments are methods of reactive
processing, reactive compounding, reactive impact modification, etc. In recent years the
blends are increasingly used as a base for either foams or composites. A better control of
multi-component interfacial properties either via sophisticated chemistry or reactive
processing made this step possible. Foamed polymer blends are used for computer
housings, hospital or office furniture, automobile parts, etc. Reinforced grades of polymer
blends are also in commercial use.
Polymer blends and composites will certainly be continuing as lively research areas -
greater understanding will allow controlled manipulation and optimisation of properties.
The field of thermodynamics of polymer blends is too extensive to be covered in one
book. There are many questions that have not been considered or were considered only
briefly. New aspects are being explored. The phenomenology of chemical reactions
between polymers at interfaces is only now being examined in depth from both
experimental and theoretical standpoints.
Much less attention has been given in the published monograph and reviews on polymer
blends to the thermodynamic properties of complex materials like gels,
biomacromolecules, micelles, colloids, block copolymers and similar substances that are
often called soft materials. Such materials abound in nature and technology, application
of chemical thermodynamics to soft materials has been delayed because of experimental
difficulties and because, until recently, there were few theoretical models available for
describing assemblies of complicated molecules. New classes of multi-component polymer
systems not considered for a long time as polymer blends, are now included as formulations
for controlled drug delivery, hydrogels, membranes, multi-responsive systems containing
block and graft copolymers, etc. Progress in this area is presented in two chapters of the
handbook. Thanks to recent advances in statistical mechanics and molecular physics,
and thanks to increasingly fast computers, it is now possible to develop a hard science
for soft materials The creation and understanding of soft materials depend primarily
on experimental science. Now statistical mechanics is able to provide guidance toward
3
Handbook of Polymer Blends and Composites
Over 50 professionals from academia and industry (coming from Europe, America, Japan,
India) have contributed to this work. They were asked to provide a survey of their area
of expertise that would enable non-specialists to become aware of advances and future
trends in the area. Certainly, such a distinguished group of authors provides the needed
balance and perspective. However, including many authors can also lead to particular
complexities in a project of this type. Even with so many authors, all subjects relevant to
polymer blends and composites cannot be addressed. Compromises have been often
reached, and the end product is different from that originally envisioned. The production
of a such book is not easy, we open ourselves to the inevitable charges of incompleteness
and imbalance. The authors have attempted to cover most of the important areas but
there are bound to be some omissions and the balance depends on the readers knowledge
and interests. Nevertheless, the editors and authors hope that this book will lead to more
texts on the topic and will stimulate interest in this field of polymer materials. The polymer
blends and composites fields has always been wide open with opportunities, stimulation,
comparison and intellectual ideas.
The editors and the contributors are well aware that no book of such wide coverage can
be free of shortcomings, and all critical comments will be gratefully appreciated.
We offer our special thanks and we wish to express our sincere gratitude to Ms Frances
Powers of Rapra Technology, who initiated this book and who waited patiently for
result, for her understanding, guidance during production of this handbook, pertinent
English language corrections and scientific comments. Thanks are also due to Ms Claire
Griffiths (Editorial Assistant) and Mr Steve Barnfield for typesetting and cover design.
Cornelia Vasile
Petru Poni Institute of Macromolecular Chemistry, 41A Grigore Ghica Voda Alley,
Ro. 6600 IASI, Romania.
A.K. Kulshreshtha
Indian Petrochemical Corporation Ltd., Vadodara 391346, India.
4
Preface
Preface
There are not many things to add to the preface of the entire handbook presented in the
first volume for the preface of the third and fourth volumes. It is intended that these
volumes will provide an overview of the theory and practice of polymer blends, from
fundamental thermodynamics of mixing to polymer blends processing, from experimental
research to modelling developments, although many other very specialised and superb
reviews and monographs are frequently published in a relatively short period of time.
The choice of subjects has changed over the years as science and technology have
progressed. New aspects are constantly being explored. The polymer blends and alloys
had a rapid growth in importance during 1980s, although the initial interest in them was
much earlier. The prospect of producing new materials with technologically useful
properties by mixing different polymers, to form the equivalent of the metal alloys has
long been an attractive goal. It is a field characterised by practical needs, basic research,
advanced technological development, industrial involvement and legislation. The reasons
for blending are: economy, extending engineering resin performance by diluting it with a
low cost polymer, adjusting the composition of the blend to customer specifications and
polymer waste recycling. About 65% of polymer alloys and blends are produced by
resin manufacturers, about 25% by compounding companies and the remaining by
transformers who have their proprietary blend formulations for certain products.
There are over 40 chapters (each volume containing 15 chapters) coming from more
than 30 distinguished specialists. The 15 chapters of the third volume deal with general
aspects of polymer blend morphology, properties and behaviour in various conditions,
while the fourth volume is mainly concerned with the various chemical classes of polymer
blends.
Unfortunately many polymer blends, have poor mechanical properties because most pairs
of chemically different polymers are mutually immiscible and mixing usually results in
materials which are phase separated and have weak polymer-polymer interfaces. The
thermodynamics of polymer mixing have throughout the years seen many developments
and it is still a very fascinating domain because of its attractive and exciting theoretical
aspects in solving problems related to structure, conformation and configuration of
macromolecules and also due to the direct practical implications. Bulk demixing of binary
polymer mixtures has been studied extensively, both theoretically and experimentally,
during the last few decades. The presence of an interface, however, may lead to domain
xiii
Handbook of Polymer Blends and Composites
morphologies and domain growth kinetics, which differ significantly from those in the
bulk. Interface phenomena are still being studied, and a chapter is devoted to this. The
third volume includes the topics related to the criteria of selection of the components of
the blends, thermodynamic fundamentals of mixing, methods of compatibilisation, general
characteristics of the blends such as glass transition, crystallisation behaviour, ageing
and thermal degradation, etc. The surface treatment methods to improve some properties
such as polarity and biocompatibility, preparation and properties of the interpenetrating
polymer networks, etc., are also discussed. The selection of the blend components is the
most difficult task in the development of a material with a full set of the desired properties.
Material selection works on the principal that the main advantages of one of the
component will compensate for the deficiencies of the second and vice versa. The trial-
and-error approach used to obtain a good blend consists of: define the desired physical
and chemical properties of the blend, list the properties of several resins which may
provide some required characteristics, tabulate the advantages and disadvantages of
various candidates, determine the miscibility and method of compatibilisation, examine
the economics including cost of components and operations, define the morphology
which assures the optimal performance, select the characteristics of resins (mainly
rheological properties, molecular weight, concentration of ingredients, amount of
compatibilising agent, etc.), necessary to generate the desired morphology, determine the
method of stabilisation of morphology (controlled cooling rate, crystallisation, chemical
reaction, irradiation, etc., and select the optimum fabrication method which assures the
final morphology. All these aspects are considered in the third volume as well as the up-
to-date theoretical aspects regarding the selection of the components of a blend with
tailored properties. An enhanced optimisation of material performance through a better
understanding of theoretical basis and development in computer simulation of all
dependences between composition-miscibility-processing parameters and properties is
necessary. Studies show many more examples of polymer mixtures existing which exhibit
regions of miscibility than were previously anticipated. The science of complex materials
is still at an early stage but there is good reason to expect it to grow dramatically in the
near future thanks to the creative efforts of dedicated researchers.
In recent years, much effort has been devoted to the study of polymer blends and the
factors, which have an effect on their miscibility. The phenomenology of chemical reactions
between polymers at interfaces is only just now being examined in depth from both the
experimental and theoretical viewpoints, but much remains to be done and understood.
In Volume 3, there are chapters which highlight the problems related to the morphology
of the blends and its direct implications in glass transition and crystallisation phenomena.
Chapters 5 and 6 focus on water-soluble polymers, which were a very attractive subject
of research in the last decades because of their utilisation in medicine and pharmacy and
for environmental protection. Chapters 11-14 also feature recent advances on the effect
xiv
Preface
of radiation and ageing on polymer blends and also behaviour during degradation. Topics
of IPN, heterofibres and reactive blending are also included.
The fourth volume contains mainly the descriptive aspects of the various polymer blends
classified on their main component, although there is some overlap because sometimes
even the minor component plays an important role in blend properties. The following
classes are presented: polyolefin, poly (vinyl chloride), styrene polymers, ionomers,
polyamides, polyesters, poly (vinyl alcohol), polyacrylates, silicones, polyurethanes,
lignocellulosics, liquid crystalline polymers, thermostable polymers, environmentally
friendly polymer blends, rubber toughened epoxies/thermosets, etc.
Each chapter is based on a rich and up-to-date literature in the field and the personal
research of the authors.
We hope that the selected topics will be readily appreciated by the readers and that they
will acknowledge the quality of the contributions.
This handbook will offer to scientists and also to R&D engineers useful updated
information on the present developments in the polymer blends area of science and
technology.
The editor thanks to all contributors to the handbook because without their efforts this
publication should not be possible. Many thanks are due to the Rapra editorial staff,
Claire Griffiths (Editorial Assitant), Sandra Hall (Graphic Designer and Typesetter) and
in particular Frances Powers for her professionalism and wholehearted support during
entire working period.
Cornelia Vasile
Petru Poni Institute of Macromolecular Chemistry
Romania
Anand Kulshreshtha
Indian Petrochemical Company
India
2002
xv
Preface
Preface
There are not many things to add to the preface of the entire handbook presented in the
first volume for the preface of the third and fourth volumes. It is intended that these
volumes will provide an overview of the theory and practice of polymer blends, from
fundamental thermodynamics of mixing to polymer blends processing, from experimental
research to modelling developments, although many other very specialised and superb
reviews and monographs are frequently published in a relatively short period of time.
The choice of subjects has changed over the years as science and technology have
progressed. New aspects are constantly being explored. The polymer blends and alloys
had a rapid growth in importance during 1980s, although the initial interest in them was
much earlier. The prospect of producing new materials with technologically useful
properties by mixing different polymers, to form the equivalent of the metal alloys has
long been an attractive goal. It is a field characterised by practical needs, basic research,
advanced technological development, industrial involvement and legislation. The reasons
for blending are: economy, extending engineering resin performance by diluting it with a
low cost polymer, adjusting the composition of the blend to customer specifications and
polymer waste recycling. About 65% of polymer alloys and blends are produced by
resin manufacturers, about 25% by compounding companies and the remaining by
transformers who have their proprietary blend formulations for certain products.
There are over 40 chapters (each volume containing 15 chapters) coming from more
than 30 distinguished specialists. The 15 chapters of the third volume deal with general
aspects of polymer blend morphology, properties and behaviour in various conditions,
while the fourth volume is mainly concerned with the various chemical classes of polymer
blends.
Unfortunately many polymer blends, have poor mechanical properties because most pairs
of chemically different polymers are mutually immiscible and mixing usually results in
materials which are phase separated and have weak polymer-polymer interfaces. The
thermodynamics of polymer mixing have throughout the years seen many developments
and it is still a very fascinating domain because of its attractive and exciting theoretical
aspects in solving problems related to structure, conformation and configuration of
macromolecules and also due to the direct practical implications. Bulk demixing of binary
polymer mixtures has been studied extensively, both theoretically and experimentally,
during the last few decades. The presence of an interface, however, may lead to domain
xv
Handbook of Polymer Blends and Composites
morphologies and domain growth kinetics, which differ significantly from those in the
bulk. Interface phenomena are still being studied, and a chapter is devoted to this. The
third volume includes the topics related to the criteria of selection of the components of
the blends, thermodynamic fundamentals of mixing, methods of compatibilisation, general
characteristics of the blends such as glass transition, crystallisation behaviour, ageing
and thermal degradation, etc. The surface treatment methods to improve some properties
such as polarity and biocompatibility, preparation and properties of the interpenetrating
polymer networks, etc., are also discussed. The selection of the blend components is the
most difficult task in the development of a material with a full set of the desired properties.
Material selection works on the principal that the main advantages of one of the
component will compensate for the deficiencies of the second and vice versa. The trial-
and-error approach used to obtain a good blend consists of: define the desired physical
and chemical properties of the blend, list the properties of several resins which may
provide some required characteristics, tabulate the advantages and disadvantages of
various candidates, determine the miscibility and method of compatibilisation, examine
the economics including cost of components and operations, define the morphology
which assures the optimal performance, select the characteristics of resins (mainly
rheological properties, molecular weight, concentration of ingredients, amount of
compatibilising agent, etc.), necessary to generate the desired morphology, determine the
method of stabilisation of morphology (controlled cooling rate, crystallisation, chemical
reaction, irradiation, etc., and select the optimum fabrication method which assures the
final morphology. All these aspects are considered in the third volume as well as the up-
to-date theoretical aspects regarding the selection of the components of a blend with
tailored properties. An enhanced optimisation of material performance through a better
understanding of theoretical basis and development in computer simulation of all
dependences between composition-miscibility-processing parameters and properties is
necessary. Studies show many more examples of polymer mixtures existing which exhibit
regions of miscibility than were previously anticipated. The science of complex materials
is still at an early stage but there is good reason to expect it to grow dramatically in the
near future thanks to the creative efforts of dedicated researchers.
In recent years, much effort has been devoted to the study of polymer blends and the
factors, which have an effect on their miscibility. The phenomenology of chemical reactions
between polymers at interfaces is only just now being examined in depth from both the
experimental and theoretical viewpoints, but much remains to be done and understood.
In Volume 3, there are chapters which highlight the problems related to the morphology
of the blends and its direct implications in glass transition and crystallisation phenomena.
Chapters 5 and 6 focus on water-soluble polymers, which were a very attractive subject
of research in the last decades because of their utilisation in medicine and pharmacy and
for environmental protection. Chapters 11-14 also feature recent advances on the effect
xvi
Preface
of radiation and ageing on polymer blends and also behaviour during degradation. Topics
of IPN, heterofibres and reactive blending are also included.
The fourth volume contains mainly the descriptive aspects of the various polymer blends
classified on their main component, although there is some overlap because sometimes
even the minor component plays an important role in blend properties. The following
classes are presented: polyolefin, poly (vinyl chloride), styrene polymers, ionomers,
polyamides, polyesters, poly (vinyl alcohol), polyacrylates, silicones, polyurethanes,
lignocellulosics, liquid crystalline polymers, thermostable polymers, environmentally
friendly polymer blends, rubber toughened epoxies/thermosets, etc.
Each chapter is based on a rich and up-to-date literature in the field and the personal
research of the authors.
We hope that the selected topics will be readily appreciated by the readers and that they
will acknowledge the quality of the contributions.
This handbook will offer to scientists and also to R&D engineers useful updated
information on the present developments in the polymer blends area of science and
technology.
The editor thanks to all contributors to the handbook because without their efforts this
publication should not be possible. Many thanks are due to the Rapra editorial staff,
Claire Griffiths (Editorial Assitant), Sandra Hall (Graphic Designer and Typesetter) and
in particular Frances Powers for her professionalism and wholehearted support during
entire working period.
Cornelia Vasile
Petru Poni Institute of Macromolecular Chemistry
Romania
Anand Kulshreshtha
Indian Petrochemical Company
India
2002
xvii
Contents
Preface ................................................................................................................... 1
i
Handbook of Polymer Blends and Composites
3 Composites in Asia........................................................................................ 97
3.1 Composites in India ............................................................................. 97
3.1.1 Introduction ............................................................................. 97
3.1.2 The non-aerospace composites industry ................................... 97
3.1.3 Overview ................................................................................. 99
3.1.4 Integrated helmet for LCA ..................................................... 102
3.1.5 Existing opportunities in India ............................................... 103
3.1.6 Indias monopoly: mica as filler ............................................. 104
3.1.7 Preferred fabrication methods ................................................ 104
ii
Contents
iii
Handbook of Polymer Blends and Composites
iv
Contents
v
Handbook of Polymer Blends and Composites
vi
Contents
vii
Handbook of Polymer Blends and Composites
viii
Contents
11 Curing and Bonding of Composites using Electron Beam Processing .......... 459
11.1 Introduction .................................................................................... 459
ix
Handbook of Polymer Blends and Composites
x
Contents
12.7 Completely Recyclable Fabric Insert Moulding (for Auto Interiors) .. 506
12.8 Laminated Object Manufacturing (LOM) Fabrication of
Fibre Reinforced Plastic (FRP) Parts ............................................... 507
12.9 Vacuum Assisted Resin Transfer Moulding (VARTM)
Fabrication Process ......................................................................... 508
12.10 Monitoring End of Cure Using Fibre Optics and Ultrasonics ........ 509
12.11 GMT ............................................................................................... 509
xi
Handbook of Polymer Blends and Composites
xii
Contents
Preface ................................................................................................................... 1
i
Handbook of Polymer Blends and Composites
ii
Contents
iii
Handbook of Polymer Blends and Composites
iv
Contents
6.4 Fibre Reinforced Plastic Composites for Orthopaedic Applications ... 205
6.4.1 Carbon Fibre Reinforced Carbon Composites (CFRC) .......... 206
6.4.2 Carbon Fibre Reinforced Plastic Composites (CFRPC) .......... 208
6.4.3 Glass Fibre Reinforced Plastics Composites ........................... 211
6.4.4 Polyethylene Fibre Reinforced Plastic Composites ................. 211
6.4.5 Polyester and Other Fibre Reinforced Plastics ....................... 213
v
Handbook of Polymer Blends and Composites
6.5 Fibre Reinforced Plastic Composites for Dental Applications ............ 214
6.5.1 Carbon Fibre Reinforced Poly Methyl Methacrylate
Composites ............................................................................ 214
6.5.2 Glass Fibre Reinforced Acrylate Composites ......................... 216
6.5.3 Glass Fibre Reinforced Polycarbonate and Polyester
Composites ............................................................................ 218
6.5.4 Polyethylene Fibre Reinforced Plastic Composites ................. 219
6.6 Fibre Reinforced Plastic Composites for General Applications .......... 220
6.7 Conclusion ......................................................................................... 222
vi
Contents
vii
Handbook of Polymer Blends and Composites
viii
Contents
ix
Handbook of Polymer Blends and Composites
x
Contents
Contents
1 Terminology .................................................................................................... 1
1.1 Miscibility and Compatibility ................................................................ 1
1.2 Related Terms used for Polymer Blends ................................................. 7
i
Handbook of Polymer Blends and Composites
ii
Contents
iii
Handbook of Polymer Blends and Composites
iv
Contents
v
Handbook of Polymer Blends and Composites
vi
Contents
vii
Handbook of Polymer Blends and Composites
viii
Contents
Contents
1 Polyolefin Blends............................................................................................. 1
1.1 Introduction ........................................................................................... 1
1.2 Blends of Polyolefins .............................................................................. 1
1.3 Blends of Polyolefins with Other Polymers .......................................... 26
i
Handbook of Polymer Blends and Composites
ii
Contents
iii
Handbook of Polymer Blends and Composites
iv
Contents
v
Handbook of Polymer Blends and Composites
vi
Contents
vii
Handbook of Polymer Blends and Composites
viii
Contents
ix
Handbook of Polymer Blends and Composites
x
Contents
xi
Handbook of Polymer Blends and Composites
xii
Contents
xiii
Abbreviations and Acronyms
529
Handbook of Polymer Blends and Composites
530
Abbreviations and Acronyms
531
Handbook of Polymer Blends and Composites
532
Abbreviations and Acronyms
533
Handbook of Polymer Blends and Composites
534
Abbreviations and Acronyms
PA Phthalic anhydride
PADC Product Application Centre
PAI Polyamide imide
PAN Polyacrylonitrile
PBO Poly-p-phenylenebenzbisoxazole
PBT Polybutylene terephthalate
PBZT Poly-p-phenylene benzobisthiazole
PC Polycarbonate
PCB Printed circuit board
PCC Polymer cement composites
PDMS Polymethylsiloxane
PE Polyethylene
PEEK Polyetheretherketone
PEI Polyetherimides
PES Polyether sulphone(s)
PET Polyethylene terephthalate
PETP Polyethylene terephthalate
PF Phenol-formaldehyde
PGE Phentyl glycidyl ether
PGEAA Phenyl glycidyl ehter of acetoacetanilide
PGEHA 4-Hydroxy acetanilide
phr Parts per hundred rubber
PI Polyimide
PMC Polymer matrix composites
PMDA Pyrometlallic dianhydride
PMMA Polymethylmethacrylate
PO Polyolefin
POE Polyethylene oxide
POM Polyoxymethylene
PP Polypropylene
535
Handbook of Polymer Blends and Composites
536
Abbreviations and Acronyms
537
Handbook of Polymer Blends and Composites
UP Unsaturated polyesters
UV Ultraviolet
UV-VARTM Ultraviolet vacuum-assisted transfer moulding
VARTM Vacuum assisted resin transfer moulding
VCD Vinyl cyclohexene dioxide
VCDAA Vinyl cyclohexane dioxide of acetoacetanilide
VCDHA Vinyl cyclohexane dioxide of 4-hydroxy acetanilide
VCDRC Vinyl cyclohexane dioxide of resorcinol
VNC Vibrating needle curemeter
VSSC Vikram Sarabhai Space Centre
WLF Williams Landel Ferry
XMC Continuous fibres moulding compound
538
Abbreviations and Acronyms
407
Handbook of Polymer Blends and Composites
CF Carbon fibre
CFC Chlorinated flurocarbon
CFM Continuous filament glass mat
CFRC Carbon fibre reinforced carbon
CFRP Carbon fibre reinforced plastic
CFRPC Carbon fibre reinforced plastic composites
CGFRP Carbon/graphite fibre-reinforced plastics
CIRTM Co-injection resin transfer moulding
CNC Computer numerical control
CPE-AS Chlorinated polyethylene-acrylonitrile-styrene copolymer
CSM Chopped strand mat
CTBN Carboxy terminated butadiene acrylonitrile
DDM Diaminodiphenyl methane
DGEBPA Diglycidyl ether of bisphenol-A
DMC Dough moulding compound
DNV Det Norske Veritas
DSC Differential scanning calorimetry
EC Ethyl cellulose
E-glass Electrical glass
EMI Electromagnetic interference
ER Epoxy resin
EPA Environmental Protection Agency
EPDM Ethylene-propylene diene terpolymer
EVA Ethylene-vinyl acetate copolymer
FCC Face centred cubic structure
FIA Federation Internationale de lAutomobile
FPS Floating production unit with storage
FPSO Floating production unit without storage
FRP Fibre reinforced plastic(s)
FRPC Fibre-reinforced plastic composite
408
Abbreviations and Acronyms
409
Handbook of Polymer Blends and Composites
410
Abbreviations and Acronyms
411
Handbook of Polymer Blends and Composites
412
Abbreviations and Acronyms
413
Handbook of Polymer Blends and Composites
414
Abbreviations
Density
Tensile strength
Volume fraction
Volume fraction of amorphous phase
1 Volume fraction of PVA in the amorphous phase
12 Thermodynamic interaction parameter
b Elongation at break
-CD -Cyclodextrin
diff. Diffusion factor
e Electrical conductivity
g Swelling ratio
p Percolation threshold
(PP-MA)-g-PEO Maleated polypropylene grafted with polyethylene oxide
(R)SF (Regenerated) Silk Fibroin
sep Separation factor
sorp. Sorption factor
1, 2 EPB Epoxidated 1,2-polybutadiene
1, 4 EPB 1,4-Polybutadiene
12DA Dodecane diamine
2DA Ethylene diamine
3HV 3-Hydroxyvalerate
4AP 4-Amino phenol
4VP 4-Vinyl pyridine
6DA Hexamethylene diamine
6F Hexafluoroisopropylidene
6F-PC Bis(hydroxyphenyl)-hexafluoropropane - PC copolymer
721
Handbook of Polymer Blends and Composites
AA Acrylic acid
AAM Acrylamide
AAS Acrylonitrile-acrylate (ester)-styrene copolymer
ABR Acrylate (ester)-butadiene rubber
ABS Acrylonitrile-butadiene-styrene terpolymer
ACN Acrylonitrile
ADC Allyl diglycol carbonate
AEP 1-(2-Aminoethyl) piperazine
AFM Atomic force microscopy
AHL Acid hydrolysis lignin
AIBN 2,2-Azobis-isobutyronitrile
AMSMMA -Methylstyrene-methyl methacrylate
ANR Allyl novolac resin
aPHB Atactic PHB
aPMMA Atactic PMMA
aPP Atactic polypropylene
aPS Atactic polystyrene
APU Aqueous polyurethane
aPVA Atactic PVA
ASA Acrylate-styrene-acrylonitrile terpolymers
ASTM American Society for Testing and Materials
ATBN Amine Terminated Acrylonitrile Butadiene copolymer
BA Butylacrylate
BD 1,4-Butane diol
BEPD 2-Butyl-2-ethyl-1,3-propanediol
BES Bis(2-hydroxyethyl)-2-aminoethanesulfonic acid
BIE Benzoin isobutyl ether
BisA-PSF Bisphenol-A polysulfone
BMA Butyl methacrylate
BMI Bismaleinimide
BOD Biochemical oxygen demand
722
Abbreviations
723
Handbook of Polymer Blends and Composites
724
Abbreviations
DS Degree of substitution
DSC Differential scanning calorimetry
DSD Duales System Deutchsland
DSDA 3,3,4,4-Diphenylsulfone tetracarboxylic dianhydride
DTA Dynamic thermal analysis
DVB Divinyl benzene
E Ethylene
EA Ethyl acrylate
EAA Ethylene-acrylic acid
EAM Ethylene-maleic anhydride copolymer
EB Ethylene-butene rubbers
EBA Ethylene butyl acrylate
EC Ethyl cellulose
ECA Ethylene-carbonate
ECO Epichlorohydrin
EDMA Ethylene dimethacrylate
EDP Environmentally degradable polymers
EEA Ethylene-ethyl acrylate
EG Ethylene glycol
EGDMA Ethylene glycol dimethacrylate
EGMA Ethylene-glycidyl methacrylate copolymer
EHEM Ethylene-hydroxyethyl methacrylate
EM4VP Poly(ethyl methacrylate-co-4-vinyl-pyridine)
EMA Ethylene-methacrylic acid random copolymer
EMMA Ethylene methyl methacrylate copolymer
ENR Epoxidated natural rubber
EO Ethylene oxide
EOS Equations of state theories
EP Ethylene-propylene rubber
EPB Ethylene-propylene-butylene terpolymer
EPDM Ethylene-propylene terpolymer
725
Handbook of Polymer Blends and Composites
726
Abbreviations
727
Handbook of Polymer Blends and Composites
728
Abbreviations
M6NOMe (6-Methoxy-2-naphthylmethylene)aniline
M6Ome (4-Methoxybenzylidene)aniline
M6Q2 (2-Quinolinylmethylene)aniline
M6Q3 (3-Quinolinylmethylene)aniline
MA Maleic anhydride
MAA Methacrylic acid
MABS Methacrylate-acrylonitrile-butadiene-styrene
MA-MI Poly(methyl acrylate-co-maleimide)
MAO Methylaluminoxane
MAP MA grafted PP
MBC Metallocene-based catalysts
MBS Methacrylate-butadiene-styrene
MDA Methylene dianiline
MDEA N-methyldietanol amine
MDI 4,4-Methylenebis(phenyl isocyanate)
MDPE Medium density polyethylene
ME Entanglement molecular weight
MEK Methyl ether ketone
MFI Melt flow index
MFR Melt flow rate
MGE MMA-GMA-ethyl acrylate terpolymer
MI Melt index
MIT Massachusetts Instituet of Technology
MLL Metallocene low linear
MLLD LDPE prepared using a metallocene catalyst
mLLDPE Metallocene catalysed LDPE
MMA Methyl methacrylate
MMAS Methyl methacrylate-styrene
MMDI Modified 4,4 diphenylmethane diisocyanate
MMW Medium molecular weight
Mn Number average molecular weight
729
Handbook of Polymer Blends and Composites
MNDA 1, 8-Diamino-p-menthane
m-PE Metallocene polyethylene
MPS-5 5 mol% hydroxystyrene
MPW Municipal plastic waste
MRGT Multi catalyst reactor granule technology
MS Microphase separation
Mw Weight average molecular weight
MWD Molecular weight distribution
MWL Mechanical wood milling
MZCR Multi-one circulating reactor technology
N Nematic
NaPET Sulfonic natrium-neutralised
NBR Acrylonitrile butadiene rubber (nitrile rubber)
NC Native corn (72% amylopectin)
N-MAm N-methylolacrylamide
NMP N-methylpyrolidone
NMR Nuclear magnetic resonance
n-PrOH n-Propanol
NR Natural Rubber
OB Oxybenzoate
ODA 4,4-Oxy-dianiline
ODMS Oligodimethylsiloxane
OS Oligostyrene
OSL Organosolv lignin
OXA Oxazoline
P Polydispersity
P((B-co-S)-g-AN) Poly((butadiene-co-styrene)-g-AN)
P(-MSAN) Poly(-methyl styrene-co-acrylonitrile)
P(AN-AM-AcAc) Poly(acrylonitrile-acrylamide-acrylic acid)
P(CHMA-co-MMA) Poly(cyclohexyl methacrylate-co-methyl methacrylate)
P(DMS-DPhS-VyMS) Poly(dimethyl-diphenyl-vinylmethyl)siloxane
730
Abbreviations
731
Handbook of Polymer Blends and Composites
732
Abbreviations
733
Handbook of Polymer Blends and Composites
PES Polysiloxane
PEST Polyester (s)
PESU Polyethersulfone
PET Poly(ethylene terephthalate)
PET/I PET-isophthalate
PET12 Polyethylene terephthalate-12
PET-OB Polyethylene terephthalate-oxybenzoate
PEU Poly(ether-urethane)
PF Phenol formaldehyde
PGI Polyglutarimide
PGPTA Polyglycerylpropoxy triacrylate
PH Phenolic resin
Ph Phenyl
PhA Phthalic anhydride
PHA Poly (hydroxyalkanoates)
PHB Poly-hydroxybutyrate
PHBV Poly(hydroxybutyrate-co-hydroxy valerate)
PHEA Polyhydroxyethyl acrylate
PHEE Polyhydroxy ester ether
PHEMA Poly(2-hydroxyethyl methacrylate)
PHMS Polyhydromethylsiloxane
PHQ Phenyl hydroquinone
phr Parts per hundred rubber
PHS Poly(styrene-co-p-hydroxystyrene)
PHT Polyhexylene terephthalate
PHV Polyhydroxyvalerate
PHXMS Polyhexylmethylsiloxane
PI Polyimide (s)
PIB Polyisobutylene
PIBMA Poly(isobutyl methacrylate)
PICA Poly(itaconic acid)
734
Abbreviations
PIR Polyisoprene
PLA Poly(lactic acid)
PLLA Polylactide
PM6BA-OMe Poly 4{[6-(methacryloyloxy)hexyl]oxy}-N-(4- Methyl-
oxybenzilidene) aniline
PM6Cz Poly 4{[6-(methacryloyloxy)hexyl]oxy}-N-(carbazolylmethylene)
aniline
PM6Q2 Poly 4{[6-(methacryloyloxy)hexyl]oxy}-N-(2-quinolinyl-
methylene) aniline
PM6R Poly 4{[6-(methacryloyloxy)hexyl]oxy}-N-(4-R-benzilidene)
aniline
PM6SBNO2 Polymethacryloyloxyhexyloxynitrostilbene
PMA Polymethacrylates
PMAA Polymethacrylic acid
PMDA Pyromellitic dianhydride
PMMA Poly(methyl methacrylate)
PMMA-g-PS Methyl methacrylate-g-poly(styrene) graft-copolymer
PMMA-OH -Hydroxy PMMA
PMO Poly(2-methyl-2-oxazoline)
PMOC Poly(2-methyl-2-oxazoline) copolymer
PMPI Poly(methylene(phenylene isocyanate))
PnBMA Poly(n-butyl methacrylate)
PNI Polynaphthimidazole
PNVP Poly(N-vinylpyrrolidine)
PO Polyolefin(s)
POD Poly(p-phenylene-1,3,4-oxadiazole)
POE Polyolefin elastomers
POEOD Poly(oxyethylene oxide diane)
POHMS Polyhydroxymethyl-siloxane
POM Polyoxymethylene
PP Polypropylene(s)
PPB Polyphenyl boronic compounds
735
Handbook of Polymer Blends and Composites
736
Abbreviations
PSiMS Polysila--methylstyrene
PSiS Polysilastyrene
PSLG Poly(sodium L-glutamate
PSMA Polystyrene-co-maleic anhydride
PS-P(MMA-CO-MAA) Polystyrene-b-poly(methyl methacrylate-co-methacrylic acid)
PS-PEO Polystyrene-poly(ethylene-oxide)
PS-PU Poly(siloxane-urethane)
PSSA Poly(styrene sulfonic acid)
PSSNa Poly (sodium styrene sulfonate)
PSU Polysulfone
P-t-BA Poly-t-butyl acrylate
PTFE Polytetrafluorethylene
PTHF Polytetrahydrofuran
PTMA Polytetramethylene adipate
PTMG Poly (tetramethylene glycol)
PTMO Poly(tetramethylene oxide)
PTMPS Polytetramethylsil-phenylenesiloxane
PU Polyurethane
PVA Polyvinyl alcohol
PVA(B) Oxidised PVA
PVA(D) Acetalised PVA
PVA-F PVA crosslinked with p-formaldehyde
PVC Polyvinylchloride
PVC-PP Polyvinyl chloride/polypropylene
PVDC Polyvinylidene chloride
PVDF Poly(vinylidene fluoride)
PVF Polyvinyl fluoride
PVME Poly(vinyl methyl ether)
PVMK Poly(vinyl methylketone)
PVP Poly(vinyl pyrrolidone)
PVPh Poly(p-vinylphenol)
737
Handbook of Polymer Blends and Composites
738
Abbreviations
SC Soluble collagen
SCB Short chain branch
SCLCP Side-chain liquid crystalline polymers (s)
SCORIM Shear controlled orientation in injection moulding technique
S-co-TMI Styrene-m-isopropenyl-,-dimethylbenzyl isocyanate random
copolymer
SD Spinodal decomposition
SE Styrene-ethylene
SEBS Styrene-ethylene-butylene-styrene block copolymer
SEBS-g-MA Styrene-poly(ethylene-co-butylene) block, graft maleic acid
SEM Scanning electron microscopy
SEN3PB Three-point bend type tests
SEP Styrene-ethylene/propylene copolymer
SEVOH Starch/EVOH
SGMA Styrene-glycidyl methacrylate copolymer
SIN Simultaneous interpenetrating network(s)
S-IPN Semi-interpenetrating network
SIS Styrene-isoprene-styrene copolymer
SMA Poly(styrene-co-maleic anhydride)
SMAA Poly(styrene-co-methacrylic acid)
SMC Sheet moulding compound(s)
SMMA Polystyrene-co-methyl methacrylate
SNaMA Poly(styrene-co-sodium methacrylate)
SPAN Sulfonic acid ring-substituted polyaniline
sPMMA Syndiotactic PMMA
sPP Syndiotactic polypropylene
SPS Sulfonated PS
sPS Syndiotactic PS
sPVA Syndiotactic PVA
SRF Salicylic acidresorcinol-formaldehyde polymeric resin
SS Soft segment
739
Handbook of Polymer Blends and Composites
740
Abbreviations
741
Handbook of Polymer Blends and Composites
WM Waxy maize
Xc The weight fraction crystallinity for blends
XPS X-ray photoelectron spectroscopy
XRD X-ray diffraction
ZN Ziegler-Natta catalyst (s)
ZNLLD LLDPE prepared using a Ziegler-Natta catalyst
ZnPET Zinc-neutralised PET
ZnSPS Zinc sulfonated polystyrene ionomers
742
1
History of Composites
A. K. Kulshreshtha
1.1 Introduction
There are several reviews already available on the history of composites [1-5]. This chapter,
therefore attempts to exemplify that the Mother Nature used her wisdom for millions of
years to produce life and environment around us and our inspiration has been mainly
derived from its various laws. For example Indian villages even today reinforce mud
with straw in order to make their huts and coat its floor and walls with cow dung.
Trees provide homes for wildlife and birds but perhaps more importantly, they use sunlight
to convert carbon dioxide from the air and water as the basic building blocks to produce
their woody substance, bark, and leaves. A by-product of these photosynthesis reactions
is oxygen, which is emitted into the air. Wood is one of the oldest construction materials
in human use and continues to be an extremely valuable material to this day. Wood is
lightweight, strong, stable, easily worked and fastened, a good insulator, warm to the
touch, and pleasing in appearance, among many other attributes.
As trees grow, cellulose molecules are produced. These molecules are long strands of 6-
carbon ring structures attached end-to-end. They may be as long as 1500 units. Many
cellulose molecules are laid side-by-side to produce microfibrils. The microfibrils form
the basic structure of the wood fibres. Most fibres develop so that their strongest,
lengthwise direction is parallel to the vertical direction of the stem. A substance called
lignin is formed to reinforce and bond the fibres to one another. As the tree grows, these
layers of fibres develop. In normal wood cells, the microfibrils are organised into several
distinct layers (Figure 1.1).
As with all living things, trees have a finite lifespan. Only a few species will flourish for
more than 100 years, the majority passing into old age at 30 to 80 years.
5
Handbook of Polymer Blends and Composites
Figure 1.1 Diagram of a tracheid wall illustrating the layers and their microfibrillar
organisation. S refers to secondary wall; S3 is sometimes interpreted as a tertiary
wall layer [6].
This reminds us of the huge skeletons of dinosaurs, whales and other giant animals
which are dug up from time to time. Human skeleton is the backbone of our body. Its
strength originates from calcium. After 50 years the human body begins to lose calcium.
This leads to osteoporosis, due to which bones become brittle and fragile. This can cause
fractures. Carbon/carbon composites have become very useful, e.g., as hip bone implants,
as bone substitutes.
It is the worlds first preform (nest) maker (although it is not actually used in making
composites).
Jute or hessian is a natural product which is fibrous and can be easily spun and woven
into fabrics. Before the advent of plastics, jute had been used for centuries for making
twine, sacks and other packaging materials. India is a leading producer of jute and in
6
History of Composites
rural areas, jute twine is still used to weave fabrics in situ in cot and chair frames. Ropes
made of jute are still used for taking animals for grazing or tying them up and more
importantly, to draw water from the well.
Nature is so intelligent that all its polymers, fibres and composites can be recycled (with the
exception of bones). Plants and trees apart from being a source of chemicals and biochemicals,
are renewable and natures factories for producing and protecting our environment.
Perhaps, the most advanced macromolecular composites are found in plants and animals.
The remarkable properties of wood are well known. This material is quite ductile at cryogenic
temperatures due to its hierarchical composite structure. Certainly, the connective tissues
in mammals are among the most advanced macromolecular composites known to mankind.
The efficient use of collagen, a fibrous protein, in the design of both soft and hard connective
tissues are worthy of comment. Very crudely, in bone, collagen serves as a highly efficient
binder for the inorganic hydroxyapatite which stiffens the structure. The interactions
between the organic fibre of collagen and the inorganic material seems to occur at a collodial
level of organisation. Epitaxial crystallisation of the inorganic phase on the fibres has been
reported to give a highly anisotropic, stress responsive, structure.
7
Handbook of Polymer Blends and Composites
Soft connective tissues also have sophisticated oriented hierarchical structures. The
collagen fibres are glued together by a gel-like proteoglycan (highly hydrated) matrix.
One of the simplest structures of this type is the tendon which functions primarily in
uniaxial tension as a reinforced elastomeric cable between muscle and bone. In
Figure 1.2 six levels of organisation of collagen in a tendon are shown. Also, the
outstanding toughness or resistance to damage of this system is due to the composite
character of the collagen fibre. First, yielding occurs by a localised shear process
within the fibre which is followed by a selective micro-fracture process of the sub
and microfibres. Catastrophic failure occurs only after vast damage within the entire
composite system has taken place.
The first man-made composite based upon polymers appeared in about 5000 BC, in the
Middle East where pitch was used as a binder for reeds in boat-building (Table 1.1).
Pitch is still being used for this purpose in the UK, more specifically in Wales, by the
descendants of the Celts who themselves had a connection with the Middle East, as it has
been for perhaps 2000 or more years in the building of coracles for fishing. Incidentally, it
is of interest to note that the same material, pitch, is presently being assessed as a precursor
for one of the most important components of ultra-modern reinforced plastics, namely
carbon fibre. Laminated wood dating to about 1500 BC has been found at Thebes and
similar laminates based upon shellac resin have been known and used in India for at least
3000 years. In the first century AD, Pliny was describing the use of wood veneers laminated
with shellac instead of glue and egg which were used in Europe, Asia, and America [1]. For
well over a thousand years, natural resins were still used, and far-seeing intellectuals like
Leonardo da Vinci were quite clear as to their potential. On the filler side, it is of interest to
note that glass fibre was known to the Phoenicians and was used by them in bottle making,
but it was not until 1713 that it again gained momentary prominence when exhibited as a
scientific curiosity at the Paris Academy. Even then the scientific world soon lost interest
again until it was once more rediscovered in the 1930s, being quickly adopted for
incorporation in polymers to form composites, so emphasising the point made earlier, that
8
History of Composites
developments in one area can assist growth in another, this time to the mutual advantage
of both glass fibres and unsaturated polyester resins.
The proliferation of composites that has occurred during the last one hundred years may
be gleaned from a review of composites and material origins over the past several thousand
years. The expression composite first appeared early in the twentieth century, and is
now generally used to describe the union of two or more diverse materials to attain
synergistic or superior qualities to those exhibited by the individual members.
Early cultures found it advantageous to use two materials to attain a superior product
such as a weapon or tool. Thus, for example, the bone or ivory handle on a metal
dagger provided a more easily handled weapon. Thin films of hammered gold leaf
decorations, for wood or leather products, yielded more attractive articles. Wood,
leather, and bone are examples of early composites, as they are derivatives of natural
polymers, cellulose, a polysaccharide of glucose, is present in vegetable matter in
substantial quantities. Collagen is the major protein component of connective tissue in
mammals. As the ageing process occurs, it becomes molecularly crosslinked. By tanning
collagen, leather products are formed.
9
Handbook of Polymer Blends and Composites
After the development of bronze alloys in the second millennium BC, bronze artifacts
were in wide use in the Near East and in China because of their strength and processing
advantages. Iron and steel production and artifacts of these materials may be traced to
the early smelting of iron ore in ancient Turkey (Amenia and Anatolia).
Before the twentieth century, most organic polymers were of natural origin and when
exposed to adverse environments, such as moist climates and ultraviolet radiation, their
tendencies to degrade over long periods of time contributed to their deterioration. Wood,
as a living tree, has remarkable longevity over many centuries. Animals and plant matter
trapped in the frozen waters of arctic regions have also maintained their integrity. Few
natural polymer/metal composite examples have survived. Exceptions are those composites
maintained in a dry desert climate (as in Egypt) or by exclusion from sunlight and oxygen
(when buried in a moist bog, as in Denmark).
1.4 Reinforcements
In all probability, the original fibre glass was a creation of nature known as Peles Hair,
This material resulted when high winds exerted a force on the leading edges of waves of
vitreous volcanic dross during a lava flow. The lava was attenuated into a fine-fibred
mass which was carried by the wind and deposited in trees. Strangely indicative of things
of come, this fibrous mass was used by birds as reinforcement in construction of their
nests. The phenomenon has been observed many places in the world, but notably in
Hawaii from the Kilauea volcano, and from Mount Vesuvius in 1906.
The first glass artisans in ancient Syria, Greece, and Egypt learned to draw fibres from
a heated glass rod and to impress them while fluid onto the surface or stems of fabricated
10
History of Composites
articles. This technique was used even prior to invention of the blowpipe in
approximately 250 BC. These fibres, as laid down, were either left in situ or further
manipulated by reheating and tooling of scores to form ingenious, artful decorations
on the glassware of the day.
Similar and other related ways of incorporating fine-drawn glass fibres were used by
Venetian glass makers in the 16th and 17th centuries (fine decorated, thin-walled lace
glass stemware) and by the Romans (millifiori glass). English, German, and French
glass producers (1650-1720) devised methods for drawing fine fibres from small melts
or heated glass rods, and in 1832, glass threads from a hot glass blob onto a revolving
wheel. The material so produced remained only a curiosity. Some was woven into
fabric structures with other textile materials, and some was used for such decorative
applications as the tails of birds in Christmas tree ornaments.
Glass fibres drawn from heated rods were produced to make a dress and neckties and
exhibited at a Chicago fair in 1893 by entrepreneur Edward Drummond Libbey and
his technical genius, Michael J. Owens. A near-success by Libbey-Owens-Ford at a
commercial fibre glass product occurred in the 1930s when coloured glass fibres were
laminated between sheets of flat window glass to form a glare-reducing structural
panel. Unfortunately, these ventures were still classed in the novelty category, and the
real expansion of the glass fibre industry was still to come.
The research was carried out at Purdue University. The original Owens-Corning
production facility was located at Newark, Ohio.
Original products were white wool insulation, coarse fibred air-filters, and continuous
filament material which was used to weave a cloth subsequently impregnated with
varnish for electrical laminates. The development of polyester resins in 1935 gave a
natural impetus to the combination, and the reinforced plastics industry was at last
launched.
Growth of fillers and reinforcements has kept up pace with that of plastics during last
25 years (Figure 1.3). This can be attributed to the availability of glass fibres in many
forms and configurations.
11
Handbook of Polymer Blends and Composites
Figure 1.3 Comparison of use of fillers and reinforcements vs. growth of plastics
1.4.2 Chronology
1916 Kemp applied for a patent on fibre reinforced plastics (FRP). US Patent
1,393,541 was issued in 1921 and assigned to Westinghouse.
1930-1931 Owens-Illinois glass company, at its plant in Columbus, OH, made significant
improvements to the process of discontinuous glass fibre manufacture which
made it commercially viable. Owens-Illinois developed the steam or air
blowing technique using a platinum melting crucible. In 1930-1931 this
company placed on the market a coarse glass fibre air filter, and in 1934
insulation mats were produced in its new Newark OH, fiber glass plant.
1933 Carlton Ellis patented unsaturated polyester resins. The resins were first
developed around 1935.
12
History of Composites
1938 In November, Owens-Illinois and Corning Glass Works jointly formed the
Owens-Corning Fiberglass Corporation.
1940 The FRP industry started because of electronic needs of World War II.
Manufacture of radomes for military aircraft was the first important application.
1943 Fiberglass reinforced plastic was first conceived, developed and designed
for light air-frame structures by the Air Force-Wright Field Structures and
Air Force Materials Laboratory (AFML).
1944 Glass Fibers, Inc., a second fiberglass company, was formed. In 1954, it
merged with Libbey-Owens-Ford and in 1958 with Johns-Manville
Corporation.
1951 The first allylsilane glass sise was patented by Robert Steinman.
1951-1952 First overseas licences issued to St. Gobain in France and Pilkington in
England.
13
Handbook of Polymer Blends and Composites
1966 Introduction of glass belted bias tires by the Armstrong Tire and Rubber
Company.
1975 The use of glass fibre in making bituminous roofing felt and domestic roofing
(shingles) based on random chopped glass filament mat.
The outstanding properties of boron were first determined at Texaco Experiments, Inc.,
in 1959. The potential of such a material for advanced aircraft applications was recognised
by the AFML, which funded a large-scale research and development effort to advance
the filament from the laboratory stage to a production level.
During the early 1960s, boron was a laboratory curiosity priced at over $11,100/kg, but
subsequently, boron filament production has expanded to a level of over 20,411 kg/y,
primarily for use on the two advanced military aircraft, the F-14 and F-15. About 90%
of the filament produced is converted into a continuous boron/epoxy tape, which is the
raw material form used in fabricating the high stiffness stabiliser sections of the F-14 and
F-15. Boron/epoxy tape comprises about 65 weight percent boron and costs about $355/
kg). In addition to its use in fabricating the empennage (tail assembly) components on
the F-14 and F-15, boron/epoxy was used in the B-1 aircraft development programme
for fabricating the horizontal stabiliser and for reinforcing the steel longeron (horizontal
member). It represented a market for boron filament of over 45,359 kg during the early
1980s. Boron filament is also converted into boron/aluminum composite forms. One of
the most important applications of the material is in the tube members used to reinforce
the mid-fuselage structure of the Space Shuttle, Orbiter.
In 1973, boron was used in the sporting goods market, and the first large application
was the reinforcement of graphite golf shafts. Over 1,360 kg of filament was used in this
segment in both 1974 and 1975. This hybrid structure, i.e., the combination of boron
and graphite similar to its counterpart in aircraft components, utilises boron for selective
reinforcement of the highly stressed lower region of the graphite shaft.
14
History of Composites
In 1974 and 1975, tennis rackets became the predominant market for boron/aluminum.
In 1975 over 455 kg of the composite was produced for just one racket, the AMF/Head,
Arthur Ashe Competition 2, and this segment of the boron/aluminum market was a
factor of two greater than the rest of the military and aerospace market combined. Other
tennis rackets and cross-country skis were developed to take advantage of borons unique
strength and stiffness. The AMF/Head racket utilises a few tenths of an gram of boron/
aluminum composite in the throat section to enhance its performance.
When boron filaments high strength and stiffness properties were first demonstrated, it
was heralded as the greatest single material advance in 3000 years.
In November 1879, the first patent on carbon fibre development was filed by Edison
who used them as electrically conducting filaments in the incandescent light bulb. At
about the same time, an English scientist, Sir Joseph Wilson Swan, produced lamp filaments
by carbonising cotton threads that had been parchmentised in sulphuric acid. In 1883,
Swan was granted a British patent for a process to produce fine cellulose filaments,
which he had developed as a means towards producing smaller diameter carbon filaments.
Although tungsten largely displaced carbon in lamp filaments beginning around 1910,
carbon filaments were still used in US Navy ships as late as 1960, because they withstood
ship vibrations better than tungsten. During the latter half of the 1950s, carbon fibres
were reinvented in more versatile forms. Crystallographic structures of carbon fibre
were proposed as early as 1917 (Figure 1.4). Because of their military usefulness, nearly
every major development in carbon fibre technology was strongly supported by
government agencies in the US, UK, Japan, France and other countries. In the US, rayon-
polyacrylonitrile (PAN) and pitch-based carbon fibre processes were developed under
sponsorship of the AFML. In England, the Royal Aircraft Establishment (RAE) and the
Atomic Energy Research Establishment (AERE) carried out much of the early PAN-
based carbon fibre research and development as did the Government Industrial Research
Institute in Japan.
The credit for investigating the structural applications of carbon fibres goes to the research
team at Wright-Patterson Air Force Base in Ohio in the late 1950s. High strength carbon
fibres were prepared using viscose rayon and this process was developed commercially
by Union Carbide and in about 1964 Thornel 25 was placed on the market at a price in
excess of $1000 per kg.
On a commercial scale, carbon fibres are prepared from only three main precursors,
namely, cellulosic fibres, PAN and pitch. PAN-based carbon fibres have proven to be
15
Handbook of Polymer Blends and Composites
superior to rayon-based carbon fibres in several respects, notably in tensile strength and
have largely dominated the exposive growth of the industry since 1970. Pitch-based
carbon fibres, however are uniquely capable of achieving extremely high axial Youngs
modulus and thermal conductivity and, therefore, have an assured place in critical military
and space applications.
A team led by Akio Shindo of the Goverment Industrial Research Institute was the
first to make carbon fibres from PAN fibres in 1961. The team failed, however, to
exploit the commercial potential of their findings and merely patented the process
they had used. Quite unaware of Shindos contribution, Watt, Johnson & Phillips of
the RAE in Farnborough, UK, produced carbon fibres from Courtalds Courtelle
acrylic fibres, converted into a special acrylic fibre, and applied for a patent in 1964.
Roll Royce who had been working independently of RAE, successfully adopted the
RAE Process.
16
History of Composites
In the spring of 1966, AERE were invited by RAE to develop production of carbon fibres
on a large scale by the UK Ministry of Technology. Exploitation of RAE to develop
production of carbon fibres on a large scale by the Ministry of Technology. Exploitation
of RAE Patent was handled by the National Research Development Corporation. The
British carbon fibre technology, to which workers at Rolls Royce and the AERE
contributed significantly, was licensed to three manufacturers: (i) Morgan Crucible Co.,
who were experienced in high temperature technology, was the first, in 1967, to produce
PAN-based carbon fibres commercially (Modmor I) but has since dropped out of
business, (ii) Rolls Royce, who developed the jet engine fan blades which unfortunately
failed due to bird-strike impact and led to bankruptcy of the Rolls-Royce engine
manufacturing business has contributed its development of carbon fibre applications in
jet engines, but no longer produces its own fibre, (iii) Courtaulds, the producers of the
acrylic precursor fibre, extended their technology to include carbonising and is still a
major producer of carbon fibres.
In 1971, the production of carbon fibres (to the RAE Patent) commenced in USA by two
companies - Hercules, Inc. and the Whittaka Corporation.
However, once the key elements of the British processes were known, the most important
factor in the further advancement of PAN-based carbon fibres was the development of
superior precursors. In this endeavour, the Japanese PAN fibre industry excelled. Toray
Industries developed a precursor far superior to any previously available ones and offered
it to Union Carbide in the USA under a technical exchange agreement (1970) under
which Toray obtained Union Carbides carbonising technology. The resulting high strength,
moderate modulus (Type II) carbon fibres were produced briefly by Union Carbide in
1971-1972 and simultaneously by Toray in Japan where it became the dominant carbon
fibre sold throughout the world for the next dozen years. Similarly, Sumitomo offered
their precursors to Hercules, who had been producing carbon fibres under license from
Courtalds, and this partnership has led to several high strength carbon fibres. Another
successful US Japanese partnership has been that between Celanese (later BASF) and
Toho Belson.
Various structures have been proposed for carbon fibres using electron microscopy.
Carbon fibres have been made from mesophase pitch which is capable of orientation
through flow and shear. In 1970, Singer of Union Carbide was able to produce high
modulus carbon fibres from pitch by melt spinning and subsequent processing. Union
Carbide commercialised pitch-based carbon fibre technology in 1975 when Thornel P-
55 yarn was produced. Ultra-high modulus pitch-based fibres were commercially
produced in 1980-1982. Mesophase pitch fibres are raw stock which are converted to
carbon fibres, retaining 80-90% of their weight and easily develop high modulus merely
17
Handbook of Polymer Blends and Composites
by heat treatment, requiring no stretching. Other companies which have developed carbon
fibres from mesophase pitch include Exxon Enterprises (who sold their carbon fibre
technology to DuPont) and many Japanese companies.
1.4.5 Whiskers
Whiskers are fibrous reinforcements having mechanical strength equivalent to the binding
forces of adjacent atoms. They are strong because they are perfect crystals and their
extremely small diameter allow little room for defects that weaken larger crystals (see
Figure 1.5).
The first commercial source of whiskers was Thermokinetic Fibers, Inc. (TKF), a company
that was started in 1962 with the intention of producing whiskers as their only product.
From 1962 to 1968, TKF produced and offered for sale a wide variety of whisker products.
One of the major drawbacks in the development of whisker composite technology has
been the price and availability of quality whisker products. During TKFs 5 years of
major activity (1963-1968), they produced about 23 kg of whiskers per year.
World production of whiskers during the late 1960s was less than 46 kg. Over 90% of
those produced were silicon carbide whiskers.
The Explosives Research and Development Establishment, Waltham Abbey, Essex, UK,
developed a process for the production of beta silicon carbide. During 1970, this group
made about 13 kg of these whiskers. Their process has been licensed to Lonza Co.,
Switzerland, where about 10 kg of whiskers were produced during the 1970s, primarily
for internal research and development work.
Also, in the late 1960s, Carborundum Corporation introduced an alpha grade of silicon
carbide whisker which they were able to produce in 46 kg quantities and sell at $560/kg.
Unfortunately, the quality of alpha silicon carbide never approached that of beta silicon
carbide in strength and uniformity of product. Therefore, this alpha silicon carbide could
not compete with beta silicon carbide.
18
History of Composites
Figure 1.5 Comparing the strength and elongation of current reinforcing fibres to
those of whiskers
1.5 Honeycombs
Long before man was concerned with strength-to-weight ratios, honeybees entrusted
their young and their honey to a honeycomb made of beeswax. Though beeswax has
never been counted among the worlds best structural materials, the high strength- and
stiffness-to-weight ratios of the hexagonal honeycomb structure have served admirably
to protect developing workers and store honey. Even dragonfly wings, apparently no
more substantial than cigarette paper, use the honeycomb structure to carry the dragonfly
through aerial acrobatics that would give any stunt pilot nightmares.
As demonstrated many times in nature, materials not normally considered for load-
bearing, fatigue-resisting, or otherwise high-performance applications can be utilised in
this configuration, often at great savings in weight. Honeycomb can be made up of any
material that can be supplied in roll or web form and can be used in flat, simply curved,
or complex and tortuous surfaces.
The great majority of honeycomb is found in sandwich constructions, as the core between
two thin, usually high strength facings. These composite sandwiches are held together
19
Handbook of Polymer Blends and Composites
with adhesives strong enough to ensure that loads are transferred from the face, to the
core, to the second face, so that the composite acts as a single structure.
Relying primarily on DuPont Nomex aramid honeycomb surfaced with unwoven biaxial
or triaxial DuPont Kevlar aramid fibre, Avtek designers reduced the operating empty
weight of the Avtek 400 to less than half that of most similar aircraft. Some 72% of the
airframe is composed of 51 moulded honeycomb parts.
Other composite aircraft include Beechcraft Starship I and the Lear Fan 2100. Commercial
manufacturers were not alone in their pursuit of an all-composite aircraft. Both the
primary and secondary structures of the Bell Helicopter Textron D-292 and the Sikorsky
Aircraft S-75 were made from predominantly Kevlar epoxy, Kevlar/graphite epoxy hybrids,
and glassfibre epoxies.
Figure 1.6 shows the flexibility of honeycomb to assume complex curved shapes as well
as one of its various applications (a helicopter blade). Figure 1.7 depicts the corrugation
process for fabricating a honeycomb.
Fibrous proteins are crosslinked through the disulphide linkages of cystine; animal and
vegetable proteins can be crosslinked by the reaction of tannic acid, chromic acid, or
formaldehyde; and natural rubber can be vulcanised by heating with sulphur. However,
most thermosetting polymers are synthetic polymers. These include saturated polyesters
(Glyptals) synthesised by Smith in 1901, phenolic plastics produced by Baekeland in 1907,
20
History of Composites
21
Handbook of Polymer Blends and Composites
urea and melamine plastics produced by John and Henkel in 1918 and 1935, alkyds patented
by Kienle in 1933, glycol maleates patented by Ellis in 1937, and epoxy resins patented by
Schlack in 1933. Many of these developments were made prior to the advent of commercial
synthetic thermoplastics. However, because their fabrication is more labour intensive than
that of thermoplastics, thermosetting plastics now account for 20% of all plastics produced.
Thermoset castings were obtained by Leo Baekeland in the early 1900s by the catalytic
acidification of a resol liquid resin or prepolymer produced by the condensation of phenol
(C6H5OH) and formaldehyde (CH2O). This catalytic process, in which thermoplastic
prepolymers are converted to thermosets has been extended to urea-formaldehyde (UF),
melamine-formaldehyde (MF), unsaturated polyester, and epoxy resins. UF, MF, and
epoxy resins are cured by condensation reactions and the unsaturated polyesters are
cured by radical chain reactions.
The modern interest in composite materials can be traced to the development of Bakelite,
or phenolic resin, in 1906. Bakelite was a hard, brittle material that had few if any mechanical
applications on its own. However, the addition of a filler, the earlier applications used
short cellulose fibres, yielded Bakelite moulding compounds that were strong and tough
and found early applications in mass-produced automobile components. The wood flour
additive improved Bakelites processibility and properties, as well as reducing its cost.
Todays technology of epoxy resins had started only by late 1930s and early 1940s,
when a number of patents were applied simultaneously in the USA and Europe. Special
mention many be made of the work carried out by Castan of Switzerland and Greenlee
of the United States simultaneously around year 1938, who had patented the first epoxy
resin from a reaction between epichlorohydrin and bisphenol-A. Subsequently, in the
early 1940s, CIBA of Basle took the licence of the patent by Castan and almost
simultaneously Shell Chemicals of the USA took the licence of patent by Greenlee and
others. By the end of the 1939-1945 War, the commercial utilisation of epoxy resins
intensified, and a very rapid growth in production of epoxy resins took place in the USA
and the whole world. The tremendous rate of growth that has taken place can be observed
from the initial production of 9000 tonnes per year in the United States by 1954 to
1,63,000 tonnes by end of 1979.
The state-of-the-art in fibre glass reinforced polyesters given in Table 1.2 is for the period
1948 to 1972. There has been a steady improvement in the dry and wet flexural strengths
of the composite due to better interfacial bonding caused by improvements resulting in
better coupling agents over three generations. Retention of wet strength is also improved.
22
History of Composites
Straw pulp mixed with 15 g of gum shellac, a quantity of alcohol to cut the gum, 15 g of
glycerine, and a quantity of ammoniacal solution of copper to make the pulp moist was
used as a compound for making picture frames. The concept of thickened glass reinforced
polyester composites originated in an invention by Charles Fisk in 1953 [6]. In his patent,
he used metal oxides to thicken and dry the polyester resin before moulding. The concept
was refined by Bayer in the early 1960s to the form the currently used (polyester moulding
compound). Throughout the 1960s, Europe and Japan dominated SMC development
and use. US Rubber introduced the first sheet moulding compound (SMC) product into
US Markets in 1965. Vibring used a chopped-strand mat for reinforcement. The E.B.
Blue Company developed the compound machinery in the same year. Cal Resin produced
the first domestic commercial application in 1966 - overhead passenger service units in
DC-8 jetliners. Goodyear Aerospace introduced the use of chopped-strand from rovings
in place of mat in 1967. Finn and Fram followed this development in 1968 with the first
commercial SMC machine containing a glass roving chopper.
Several centuries ago, the Egyptians produced articles such as baskets and trays by placing
tortoise shells or animal horns on hot forms. In the nineteenth century, piano keys were
made by thermoforming celluloid sheets over wooden cores. In fact, in spite of many
advances in plastic fabrication, the name horner is still used to classify some workers in
the plastics industry.
23
Handbook of Polymer Blends and Composites
Figure 1.8 Injection moulding machine was patented by Hyatt Brothers in 1872.
Eichengrun (1921) designated its modern version
Massive structures such as helicopter and windmill blades were produced on filament
winding equipment. Pultrusion of FRP made its appearance in 1950s.
Plastic laminated sheets produced in 1913 led to the formation of the Formica Products
Company and the commercial introduction in 1931, of decorative laminates consisting
of a urea-formaldehyde surface on an unrefined (kraft) paper core impregnated with
phenolic resin and compressed and heated between polished steel platens. Since 1937,
the surface layer of most decorative laminates has been fabricated with melamine-
24
History of Composites
25
Handbook of Polymer Blends and Composites
formaldehyde, which can be prepared with mineral fillers, thus offering improved heat
and moisture resistance and allowing a wide range of decorative effects.
Plywood is a laminate consisting of thin sheet of wood arranged with the grain in
alternate sheets at right angles. Invented in the mid-nineteenth century using natural
based adhesives, the manufacture of plywood was greatly improved by the introduction
of phenolic resins. Indeed, plywood accounts for more than 60% of the total phenolic
resins production.
Intense development of RRIM began when it was discovered that the physical properties
of polyurethane could be improved by the addition of glass reinforcement. Although
compounds were highly viscous and abrasive, suppliers began experimenting on
improvised machines, and it became clear that the practical use of filled material
unmodified RIM systems was possible. Milled glass (1.59 mm long) has been the accepted
form of reinforcement but ICI have recently introduced 1.59 mm long chopped glass
which, it is claimed, offers a number of distinct advantages over milled glass.
1.10 Composites
The first laminated boat (4 m long) was fabricated from manilla paper and glue in 1867,
four of which were acquired by the US Naval Academy in 1868. Several fabrication
processes were developed and patented in the ensuing period. The decade of the 1940s
was marked by great industrial progress in the development of reinforced plastics. The
Society of Plastic Engineers (SPE) was founded in 1941.
Since the first FRP for the light airframe structures were conceived, developed and designed
by the Wright-Patterson Air Force, Structures and Materials Laboratory, Dayton, Ohio
in 1943, the FRP composites have been used at an accelerated pace by the defense,
aerospace, aircraft, and sporting goods industries.
26
History of Composites
Reinforced plastics showed a significant growth following the 1939-1945 War. They
came to be utilised in the manufacture of boats, cars, furniture, and other consumer and
industrial applications. However, their full potential as high performance structural
materials was not realised until a technological breakthrough occurred which resulted in
the development of a class of reinforced plastics known as advanced composites.
This breakthrough was the result of innovative work conducted at Wright-Patterson Air
Force Base, Ohio, in the early 1940s. In this work, a BT-15 composite aft fuselage was
designed, fabricated, static tested, and flight tested. This fuselage was a sandwich
construction which incorporated glass reinforced polyester face sheets and a balsa wood
core. On a specific strength basis, the composite structure was 50% stronger than the
baseline aluminum structure. In subsequent work at Wright-Patterson, an AT-6C glass
reinforced composite wing was designed, fabricated, and flight tested.
This early pioneering work with glass reinforced composites provided the confidence
and impetus for the subsequent work which expanded the advanced composite technology
base. This subsequent technology explosion spanned a multitude of disciplines, such as
design, processing, materials and testing. An ever increasing usage of advanced composites
mirrored this technology explosion.
What may be termed the modern era of advanced composites began about 1960s when
boron filaments were developed. In addition to high specific strengths, these reinforcing
filaments were capable of providing the high specific stiffness missing with glass
reinforcement. Further, the introduction of boron filaments was a significant catalyst to
advanced composite technology. The technology base that grew out of boron composites
demonstrated that these advanced materials had a significant role to play and set the
stage for an even further expansion of advanced composite technology.
In terms of a materials technology expansion, the boron filaments were soon followed
by the carbon/graphite fibres, aramid fibres, and recent specialty fibres. Concurrent with
this expansion of fibre technology was an expansion of resin technology. The initial use
of polyester resins was followed by the use of epoxies which eventually became the state-
of-the-art, baseline resins for the higher modulus composites. Numerous specialty matrix
resins such as high temperature polyimides were also developed for use in advanced
composites. Current trends in matrix resin development are addressing the needs for
improved toughness and improved environmental durability.
The chronology of the use of composite materials in cargo aircraft is given in Table 1.3
for the period (1946-1998).
The term advanced composites has been coined to cover those reinforced plastics
containing continuous strong, stiff fibres such as carbon (graphite), boron, aramid, or
27
Handbook of Polymer Blends and Composites
glass. Advanced composites have been developed primarily for the aerospace industry, in
which the demand for strong and stiff lightweight structures overcame the prohibitive
costs of early composite material systems. Currently, advanced composite materials, mainly
carbon-reinforced epoxy resins, are used widely in military aircraft and increasingly in
civil aircraft. The largest volume usage of these materials has, however, been in the
recreational area in applications such as tennis rackets and golf club shafts.
28
History of Composites
The passenger car developed through the nineteenth century, reaching a watershed in
about 1900, when the hallmarks of the modern car were established. The Oldsmobile
curved dash (named for its bold design of the curved front) signalled the end of the car
as merely a replacement of the horse-drawn carriage for high society, and the beginning
of a vehicle for the general population. Approximately 5,000 curved-dash vehicles were
sold each year from 1900 to 1905, which was high volume production for the time. This
volume demonstrated that there was a popular market for the car, and Henry Ford
produced the volume the market required by developing the assembly line concept. The
original lightweight cars, closely styled to the earlier horse-drawn carriages, evolved to
the muscle cars of the 1950s with power of more than 225 kW. By the 1980s cars
decreased in size in dimensions, weight, and horsepower. Composites became recognised
as an ideal material to meet the new requirements of the modern car.
Demonstration cars with all-composite bodies have been built since the early 1900s
beginning with Henry Fords coupe. The Rohm & Haas acrylic Explorer, the Borg-
Warner acrylonitrile-butadiene-styrene (ABS) Cycolac Research Vehicle (CRV), and
the Mobay polyurethane car were built in the late 1950s and early 1960s. The all-
composite-body Pontiac Phoenix Ventura, Chrysler PXL, and Ford Lightweight Vehicle
(LWV) represented demonstration cars of the 1970s, and Fords Probe V concept car
indicates the all-composite-body car of the 1990s. Commercial volumes of all-composite
Chevrolet Corvettes have been produced since the late 1950s; and Pontiac Fieros
represent the all-composite-body car of the 1980s. The principal difference between
these two cars is that the Corvette is a low production volume car with an annual
production range between 25,000 and 50,000 units, whereas the annual production
range for the Fiero was 100,000 units.
Automobile parts made of SMC account for a major portion of reinforced plastics used
in the land transportation industry.
It was a milestone for the composites industry, when the first Pontiac Fiero rolled off the
assembly line with an all fibre glass composite body (300 kg) in 1953. The newly designed
all glass reinforced plastic (GRP) truck Bronco was introduced in 1981 by Ford. When
Corvette leaf springs were introduced in 1980, it caused excitement in the industry. Ford
29
Handbook of Polymer Blends and Composites
Motor Company has introduced one of the key components, Air Springs, on their Lincoln
Continental and Mark VII vehicles in 1984. They are producing more than 250,000
individual air springs due to lower tooling cost. A one piece air intake manifold made of
glass fibre reinforced nylon was used in the 1994 Mercedes-Benz, 3.2 litre engine to
reduce weight, improve engine performance and reduce manufacturing cost.
1909 Baekeland was granted his Heat and Pressure patent for phenolic resins.
1941 Henry Ford swung an axe on an automobile RP body to demonstrate the feasibility
of RP for automobiles. RP used in World War II for cotton-phenolic ship bearings.
1943 Howald of Plaskon working with personnel at Wright Field made the first RP
honeycomb using large soda straws for the form. Fiberglass RP were first
conceived, developed, and designed for light airframe structures by the Air Force
- Wright Field Structures and Materials Laboratory. Reinforced plastic tooling
30
History of Composites
1945 Meyer of Western Products made first production RP honeycomb by using soda
straws to index corrugated sheets.
1946 First automatic injection moulding of RP. Filament winding process was under
development by Young and Kellogg for the production of rocket motor cases and
filament wound pipe. The first moulding of production fibre-glass parts in low-
cost matched metal dies. Patent application for first fibreglass spray technique.
1953 The Mobile Plastics Division of Carlyle Corporation introduced the first
preimpregnated roving. General Motors launched an exploratory program with
its Chevrolet Corvette all FRP body.
1960 The Dow Chemical Company started building the Windecker fibreglass-epoxy
RP wing for a monocoupe, low-wing airplane.
1962 S-glass introduced. The Polaris A-3 submarine launched ballistic missile, included
FRP filament wound motor cases, nozzles, and electrical connectors. Pontiac
had SMC front ends on five models.
1965 The United States astronauts walked in space with FRP helmets, a filament wound
altitude control gun, and a FRP survival kit.
31
Handbook of Polymer Blends and Composites
1967 A new fibre became available, S-2 fibre glass, a commercial grade of S glass.
Pontiac introduced SMC in their front end fender extender and lamp housing for
the first high volume automotive application for SMC.
1969 FRP went to the moon on the Apollo 11 Spacecraft with Neil Armstrong.
1970s First epoxy resin patent to Pierre Castan (Swiss), licensed to CIBA.
1971 Apollo 15 - First use of Lunar Roving Vehicle (FRP fenders). DuPont introduces
Aramid Fibre Kevlar, an aramid fibre which fills the gap between high-cost, high-
modulus carbon/graphite intermediate-modulus polymer fibre.
1978 Boeing adopts S-2 glass for aircraft flooring in all 700 series aircraft (727, 737,
747, 767, etc.). Helicopter blades in Boeing Vertol, were made with highly
engineered hybrid-titanium, S-glass prepreg, Nomex honeycomb, glass or kevlar
skin. Hull Corporation introduces BMC injection machine.
1979 RIM, polyurethane elastomers have become important, chiefly in the USA, for
soft front-end mouldings for automobiles, however, additional automotive
applications for RIM were limited because of low stiffness properties and lack of
dimensional stability. Intense development of RRIM began when it was discovered
that the physical properties of polyurethane could be improved by the additional
of glass reinforcement. Automotive manufacturers evaluated hang-on RRIM body
panels, e.g., boot lids, bonnets, door panels, and wings in many models of cars.
First attempts to use sandwich structures in aircraft were in the Vickers Viscount and
Vikings aeroplanes.
However, due to the unsupported spaces between successive crests, the panels were easily
damaged by sharp impact. To overcome these problems a similar construction was used but
with 0.91 mm aluminium skins instead of in plywood Boeing 707, Douglas DC 6 and DC 9.
In order to overcome the weak, unsupported skins, areas of the structure, the stringers
(type of internal brace), were replaced with end-grain balsa wood. Typical aircraft
incorporating this type of floor construction were the BAC 1-11, Boeing 707 and the
present Boeing 747. The Fokker F27 used the material as a fuselage structural member.
In 1950s stiletto heels were a major problem for lightweight aircraft flooring material.
The high local loadings caused puncturing of the floor structure with subsequent
32
History of Composites
delamination (a 145 kg person wearing 6.35 m square stilleto heels produces a local
stress 11 MPa with each step.
Mallite (aluminium sheet bonded on both sides to end-grain balsa wood) sandwich
panels were tested in the VC 10 and proved satisfactory after 4000 h.
The product, Satellite 47 (Fibrelam, USA), was tested in the Boeing 747 which is still in
operation today. Fibrelam has also undergone 55000 flying hours without crazing,
whereas balsa cores survive only 5000 h.
Fibrelam sandwich panels were adopted and continue to be used to the present day.
The use of fibrelam flooring in the 747 afforded a weight saving of 145 kg compared
with using aluminium skins/balsa core skins/PVC foam core (toilet and galley areas).
Composite pressure vessels began their development in the 1950s using rocket motor
case technology. They typically used glass fibre reinforcement and a rubber liner. While
these vessels offered lighter weight than steel vessels, they required higher safety factors
than steel due to stress rupture considerations, and there was a measurable permeation
of gaseous contents through the wall.
Metal liners came into use during the 1960s. This allowed the reinforcing fibre to be
either a full wrap (helical and hoop) or only a hoop wrap if the liner was sufficiently
thick. Composite vessels with metal liners provided significantly lower permeation rates
than rubber lined vessels, but the fatigue life of the liners was limited. Thinner liners
might develop a through crack and leak between 100 and 1000 cycles, while thicker
liners might leak between 10,000 and 30,000 cycles.
The primary use for composite pressure vessels in the 1950s and 1960s was for defense
and aerospace applications. These pressure vessels were used abroad military aircraft for
ejection systems, emergency power systems, and engine restart applications. They were
also used for Skylab oxygen tanks and as power sources for missile systems.
33
Handbook of Polymer Blends and Composites
Two new reinforcing fibres were introduced in the 1970s. Aramid fibre was lighter and
stiffer than glass fibre and had superior fatigue properties. It found almost immediate
acceptance in high performance applications, although its higher cost prevented it from
being a replacement for glass fibre in all aplications. Carbon fibre was used in some
pressure vessel applications in the 1970s, but it did not become the material of choice
until the 1980s, when it became lower cost, higher strength, and easier to handle.
Plastic liners have been developed for use in composite pressure vessels in the 1990s.
Plastic liners have very low permeation rates and virtually unlimited cyclic fatigue life.
They are relatively inexpensive and have a short lead time.
Plastic lined pressure vessels have been used as fuel containers for Natural Gas Vehicles
(NGV), and more recently for use in tensioning systems on off-shore oil platforms.
Shell has been a leader in deepwater activities in the Gulf beginning with installation of
the Cognac fixed platform in 312 m in 1978. In 1989 Shell pushed this technology about
to its limit with its Bullwinkle fixed platform in 415 m. Shells next breakthrough came
in 1994 with their Auger TLP (tension leg platform) in 875 m (see Figure 1.10). Augers
production records far surpassed even Shells expectations and fuelled a flurry of interest
in deepwater projects. In 1996, Shell and BP Exploration installed the Mars tension leg
platform (TLP) in 900 m. In May 1997, Shell, Amoco and Exxon began production
from the Ram-Powell TL in 980 m. Ram-Powell was nearly identical to the Mars TLP in
design. In July 1997, Shell set two world records with their Mensa Subsea Development
- World waerdepth record for production at 1625 m and world record of 110 km for
tieback distance to the HUB platform in West Delta 143.
But Shell is not alone in setting records and is aggressively pursuing deepwater plans in
the Gulf. In 1997, Kerr McGee (Oryx) and CNG began producing from the Neptune
Spar platform in 580 m. In 1998, Chevron along with partner Exxon and Fina brought
the Genesis Spar platform on line in 790 m. A third Spar platform is under construction
by Exxon for its Diana/Hoover project in 1460 m. In 1998, British-Borneo installed the
first mini-TLP in 520 m with its Morpeth project. Also in 1998, Amerada Hess installed
the Baldpate compliant tower in 495 m.
34
History of Composites
Figure 1.10 Application of carbon fibre reinforced tubes in off shore drilling. The light
weight of carbon fibre reinforced composites becomes most important with increasing
drilling depth
1.16 Summary
The material evolution which occurred millions of years ago has led to the development
of composites in the 20th century (Figure 1.11). Simultaneously with composite usage
as fuselage materials, aircraft engines have also used this technology (Figure 1.12).
Advanced Composite Materials now forms about 80% of structural weight in fighter
aircraft, adding to their speed & maneuverability (Figure 1.13). A helicopter rotor
blade was made from composites in 1990. An all composite aircraft Avtek 400 has
been made.
FRP are used not only in aircrafts, rockets, missiles and satellites, but have found common
usage in home appliances, thus benefiting the man on the street.
35
Handbook of Polymer Blends and Composites
36
History of Composites
Bibliography
37
2
Particulate Fillers and Fibre Reinforcements
2.1 Introduction
Important changes in the properties of plastics resulting from the incorporation of special
additives permit their use in applications where the polymer alone would have had small
chance to meet certain performance specifications. Fillers and reinforcements are solid
additives that differ from the plastic matrices with respect to their composition and structure.
The basic role of a filler is to fill, i.e., increase the bulk at low cost, thereby improving
economics while, by definition, the main function of a reinforcing filler is to improve the
physical and mechanical properties of the basic polymer. Out of these properties, stiffness
and strength are the most important among short-term properties in engineering
applications. Resistance to creep and fatigue failure are principal long-term properties.
The effect of heat on both these groups of properties is shown by the heat deflection
temperature (HDT) and dimensional stability. All these properties can be upgraded by
reinforcing fillers. Non-mechanical properties of the basic polymer, e.g., electrical
properties, abrasion resistance, flammability, may also be strongly modified by the presence
of reinforcing fillers. With fibrous fillers, the influence of the fibre aspect ratio and the
anisotropic effect of fibre orientation can further magnify the improvements. That is
why, the most effective reinforcing fillers are fibres of high modulus and strength.
Modern fillers can take on many of the functions of reinforcements. Usually, fibres and
lamina structure are counted as reinforcements, while the ball type additives are counted
as fillers. Inert fillers or extender fillers increase the bulk, solve some processing problems,
and lower the price; no improvement is seen in the mechanical or physical properties
compared with the unfilled polymer, although by a higher thermal conductivity, they
improve the production rates.
39
Handbook of Polymer Blends and Composites
essentially induce rigidity and thermal stability, but do not attain the stiffening effect of
fibre-type reinforcements. Average particle diameter, specific surface and surface energy
are decisive in the reinforcing effect of fillers in elastomers.
Enhancers and reinforcing fillers induce the following improvements in thermoplastics [1]:
increase in tensile strength and tensile stress at break, and in compressive and shear
strength,
increase of the modulus of elasticity, stiffness and hardness of the composite material,
improving creep behaviour and bend-creep modulus, as well as partial impact strength,
reducing the viscoelastic yield under load and lower shrinkage, giving good
dimensional stability.
The action of active fillers can be attributed to three causes [2-5], namely: (1) chemical
bond formation between filler and the material to be reinforced; (2) immobilisation of
polymer segments attached to the filler surface by secondary or primary valence bonds,
an interfacial layer with characteristic properties thus appearing; (3) when the polymer
molecules are subjected to stress with energy absorption, they can slide off the filler
surface; the impact energy is thus uniformly distributed and the impact strength increases.
Detailed theories on these aspects are presented in many reviews [5-13].
For their use in plastics the most important characteristics of fillers are: chemical
composition, particle shape, average diameter, grain distribution, specific surface and
value of surface energy, thermo-oxidative and UV-stability for outdoor applications and
moisture and water-soluble compounds content.
Fillers and reinforcements can be differentiated by the aspect ratio (AR) which is the
ratio between length (or length and width) to thickness (L:T) - as follows [14]:
(b) Enhancers such as short fibres, e.g., wood flour, milled or chopped glass fibres,
wollastonite, whiskers (a mineral fibre as a single crystal fibre), talc, etc., with an L:T
varying from 10 to >> 100.
40
Particulate Fillers and Fibre Reinforcements
(c) Reinforcements such as filaments, non-woven or woven textile products, L:T being
very large.
(d) Special active reinforcements used principally in elastomers, such as carbon black,
pyrogenic highly dispersed silica or precipitated ultrafine, carboxylated rubber coated
calcium carbonate.
For a controlled modification of the various properties of the composite materials, certain
characteristics of the filler/reinforcement are necessary.
The main requirements for the properties of fillers and reinforcements necessary to obtain
of composite materials that can be further used in specific applications are [1]:
Low moisture absorption and high bulk density; they should preserve their properties
during storage prior to compounding.
Filler particles should be as round as possible with a small specific surface, low surface
energy, and absorptivity, thus assuring a low viscosity during compounding.
A high compounding rate is obtained with fillers having low specific heat and high thermal
conductivity.
A composite with high tensile strength and high elongation is obtained when using a
filler/reinforcement having high strength in comparison with the matrix, high length/
diameter ratio, and good fibre/matrix adhesion, as well as a good distribution in the
matrix while, for high flexural strength, it is very important to obtain, additionally a
smooth surface in the finished article.
Fillers with low compressibility and small round particles are suitable for obtaining
composites with high compressive strength from crystalline polymers.
41
Handbook of Polymer Blends and Composites
There are high strand integrity products with good fibre feed and handling characteristics
in 3.17 and 4.76 mm chopped lengths. The high performance carbon/
polyetheretherketone (PEEK) types are pre-pregs in which continuous filament and
matrix have been combined by a form of pultrusion process. They require only placing
in position and heating to fuse the thermoplastic polypropylene (PP) matrix. The
interfacial bonding is improved. PP sheet moulding compounds and bulk moulding
compounds known as glass mat thermoplastics (GMT) are compounded in granules
for injection moulding and extrusion.
Good long-term behaviour and fatigue and weathering resistance of the composite
materials are achieved with filler/reinforcement with permanent polymer/matrix bonds,
good resistance to heat, light, water, chemicals, etc.
Also, a low cost results when low cost filler is used with low processing cost and
maximum possible degree of filling; the other requirements regarding the properties
being, of course, fulfilled.
It is well known that mineral fillers have a catalytic effect on resin cure and that the
effect is specific to each resin and each curing system. Thus, polyester resins cured with
a benzoyl peroxide initiator are less sensitive to the mineral surface than the same resin
cured with a cobalt-promoted ketone peroxide initiator. Glass fibres treated with a
chrome finish, retarded gelation of a polyester more than those treated with a silane
finish. Barium sulphate, calcium carbonate and zircon promoted a much faster cure
than an unfilled resin. With some materials such as clay, silica and talc, the inhibition
of cure is severe enough to limit their usefulness in highly filled systems. Most polymers
undergo shrinkage during cure. Therefore, many composites are translucent when
removed hot from the press, but they become opaque due to crazing as the composite
cools. Mixing finely divided particulate fillers with resin produces total expansion, yet
merely transfers interfacial stresses from a macro to a microscale. The result is that
particulate filled plastics generally have lower mechanical properties than the milled
filled plastics.
The value of filler depends on the difference between the coefficients of thermal
expansion that must have minimum values. Some thermal expansion coefficients of
the materials used in composites are given in Table 2.1.
The value of asbestos in high-temperature composites depends more on its close match
with the resins coefficient of thermal expansion than on fibres heat stability.
42
Particulate Fillers and Fibre Reinforcements
43
Handbook of Polymer Blends and Composites
The particulate fillers, which produce true reinforcing effects, are the fillers whose particles
have low aspect ratios (if needle- or rod-shaped) and which may be roughly spherical
(with various degrees of irregularity) or plate-like. The properties mainly improved by
particulate reinforcing fillers are stiffness, hardness and heat deflection temperature.
They also change the colour and the cost. Considerable research effort has been directed
to the use of silane coupling agents in composites [15].
The most common particulate fillers, which are used in one form or another in plastic
composites are listed in the next sections [1, 15-27]. With a few exceptions, these fillers have
been subjected to treatments with a silane coupling agent, in a variety of composites [5].
2.3.1.1 Silica
Synthetic fillers are wet process silica, fumed colloidal silica and silica aerogel. Their
specific surface may be very large from 50 800 m2/g.
2.3.1.2 Silicate
Kaolin is also known as porcelain earth and china clay, is lamina with an AR of 5 100,
s=500-600mJ/m2; c=0.91 W/m.K, cp= 0.92 J/kg K. It is used as an extender/pigment in
wire and cable, sheet moulding compound (SMC), bulk moulding compound (BMC)
and polyvinyl chloride (PVC) flooring, etc.
Talc Lamina has an AR of 5 - 100, s=60-70mJ/m2, c=2.1 W/m.K, cp= 0.86 J/kg K. It
can be used as a filler extender/reinforcement and improves stiffness, tensile strength and
resistance to creep.
44
Particulate Fillers and Fibre Reinforcements
Feldspar, nepheline, syenite are cuboid with an AR of 1. They are abrasive; speciality
fillers, easily wetted and dispersed, permitting transparency/translucency, are resistant
to chemicals and weathering.
Fibre has an AR of > 10, high strength, dimensional stability, heat resistance and chemical
resistance.
Limestone
Marble
Dolomite
Alumina
Magnesium
45
Handbook of Polymer Blends and Composites
These have good electrical and/or thermal conductivity or magnetic properties, they reduce
friction, and they are expensive.
2.3.1.6 Salts
Calcium phosphate
Calcium sulphate is an extender, which, increases impact, tensile and compressive
strength
Barium sulphate heavy spar is cuboid with an AR of 1.4-4. It is used as a filler and
a white pigment. It increases specific gravity, chemical resistance and frictional
resistance.
Potassium titanate (Fybex).
Silicon carbide
Carbon black and graphite are lamina with an AR of 5 100. They are used as
fillers, pigments and antistatic agents. They are electrically conductive.
Carbon fibres
Ceramic microspheres
Organic fillers are wood flour, nut shells, corncobs, rice, and peanut hulls. They are
used as extenders/fillers.
Polymeric fillers s=15-60 mJ/m2. These are used as reinforcements and are lightweight.
Cellulose fibres have an AR of > 10
Whiskers have an AR of > 10.
46
Particulate Fillers and Fibre Reinforcements
Calcium carbonates mainly occur naturally in the form of chalk, limestone, marble, etc.,
from which the filler is obtained by fine milling or synthesised by precipitation. Due to
its low price and to its improvement of polymer properties and ageing resistance, natural
calcium carbonate is the most important filler used in plastics. Natural and synthetic
calcium carbonate generally have a calcite structure, and may contain some aragonite of
metastable phase. Grinding of the calcite crystal is known to result in a aragonite structure.
Small amounts of finely dispersed calcium carbonate, silica, or various silicates reduce
sticking and improve the paper-like feel of polyethylene (PE) films.
The synthetic precipitated calcium carbonate fillers [28], are more expensive than ground
chalk. Due to their larger surface area, the shearing forces during processing are
appreciably higher, thus high filler addition is not possible. Ground calcium carbonate
takes the shape of sponge-like aggregates. Precipitated calcium carbonate having calcite
structure forms cube, spindle or rod shapes, while the aragonite structure exhibits a rod-
like shape. It has a higher absorptive effect on plasticisers, stabilisers, lubricants, etc.
The adhesion of the calcium carbonate-polymer matrix can be improved by a surface
treatment, commonly using stearates.
Ultrafine particles of 5-20 nm or 15-70 nm average diameter have been prepared by the
recarbonation process. The product has an aragonite structure and a good dispersability
in polymer matrix.
High quality carbonate fillers should have the following properties: high chemical purity,
low plasticiser absorption, high degree of whiteness being possible by partially substituting
the expensive white pigments, coloration in pastel shades, non-abrasive, non-toxic,
odourless, reduce shrinkage, increase stiffness and modulus of elasticity, improve the
surface quality of finished parts, improve stability and ageing resistance, reduce plate-
out effect, etc. The synthetic calcium carbonate has higher absorptive effects on plasticisers,
stabilisers and lubricants.
Dolomite is much more abrasive than calcium carbonate, it has a good wettability and
dispersability, but because of water solubility, weathering resistance is inferior to that of
calcium carbonate.
Calcium sulphate (gypsum) is mainly used in three different forms: dihydrate, a filler
which is used in foamed polymers, hemihydrate available in fibre form and anhydrite.
After treatment, the latter does not absorb moisture. Calcium sulphate offers a very
white acid resistant filler with low surface energy and low abrasiveness.
Barium sulphate has high density (4.3-4.6 g/cm3), Mohs hardness (2.5-3.5), and high
reflectivity. It is primarily used in foams, floor covering, and gives protection against
high energy radiation and sound deadening.
47
Handbook of Polymer Blends and Composites
Silanisation of the filler surface facilitates the incorporation of mica in polyolefins (PO),
aminosilanes being used. A metallised form is also found.
Feldspar and nepheline syenite are characterised by: density 2.6 g/cm3, Mohs hardness
6.0 - 6.5, refractive index of 1.53, pH 8 9, oil absorption 13 - 14 g/100 g powder. Due
to the low oil absorption, they are used in thermosetting and PVC.
Silica and silicates are both natural and synthetic. The natural ones (diatomaceous earth,
novawite, perlite, quartz, quartzite, sand and tripoli) differ in their particle size, degree
of crystallinity, and hardness. -Quartz is, however, the hardest of the common minerals.
Quartz has a density of 2.65 g.cm-3 and a Mohs hardness of 7.
Synthetic silicates are obtained by relatively complicated procedures because of the fine
spherical primary particles, which can form agglomerates and aggregates. Depending on
the manufacturing process, the surface may be very large, attaining values of 50 to 800
m2g-1. Apart from the chemical composition, not all-synthetic silicas are crystalline. All
silicas and silicates are supplied with various silane coatings. They have the following
functions: shrinkage reduction, reinforcement, improvement of dimensional stability,
electrical properties, increase hardness, thickening, rheological and thixotropic effects.
Metals and metallic powders. Metallic powders consisting of aluminium, bronze, copper,
and nickel are always used in thermoplastics, if products with very high thermal or
48
Particulate Fillers and Fibre Reinforcements
electrical conductivity are required. Additionally, heavy metal powders increase the
resistance to neutron and gamma rays [20].
Metals in fibre form and metallic oxides are also supplied as fillers and reinforcements
for selective changes such as aluminium oxide for improved electrical values, zinc oxide
for weathering stability, beryllium oxide to improve thermal conductivity, iron oxides
for magnetic effects, lead oxides for high density and magnesium oxide to increase stiffness
and thickening of unsaturated polyester resins, etc.
Kaolins, also known as porcelain earth and china clay, are hydrated aluminium silicates
possessing a clearly determinable crystal lattice with a plate-like, hexagonal structure.
Kaolin consists of primary and secondary kaolinites with the molecular formula, Al2O3
SiO2 2H2O containing Al2O3 (38.8%), SiO2 (45.4%) MgO (0.2%), K2O (0.97%) Na2O
(0.07%), FeO (variable). In their plate-like hexagonal lamellar structure, primary kaolin
has a length to thickness ratio of 10:1. Kaolins generally possesses a high degree of
whiteness and are electrically non-conductors; they are highly resistant to chemicals,
even to strong acids.
Kaolins used in polyolefin (PO) composites should have fine particles of 0.6 to 6 m,
Mohs hardness of 2.5, and a pH of 5.0 - 5.6.
The calcinated kaolin form is appreciably harder than natural kaolin; it considerably
improves the electrical properties of a polymer and reduces the water absorption (cable
insulation materials based PVC or ethylene-propylene-diene terpolymer; EPDM). Kaolin
is obtainable with various silane coatings that facilitate dispersion of the filler in the
plastics. Kaolin is the most important filler in rubber industry, after carbon black. The
rubber chemist distinguishes between hard (<2 m) and soft (coarse) kaolin, depending
on whether the finished product has a high or low modulus.
Carbon black is a special form of carbon obtained through partial combustion of liquid
or gaseous hydrocarbons. It can be used as a black pigment (particle dimensions of 15 to
20 m), as an improver of electric conductibility (particle dimensions of 17, 24, and 90
m), and as a filler/reinforcement material (particle dimensions of 23 to 28 m) [1, 12,
16]. Hollow carbon black spheres are used in foams.
49
Handbook of Polymer Blends and Composites
Wood flour is obtained by fine milling of soft or hardwood wastes, leading to many sorts
with the following characteristics, such as: particle size 60 to 80 m, bulk density 182 to
285 kg/m3, specific volume 3.5 to 5.49 m3/kg. It improves the surface gloss.
Glass spheres or glass beads. Both solid (ballotini) and hollow glass spheres (glass
microballons) are obtained from sodium borosilicates or silica. As fillers, and also as
flame-retardants, solid glass spheres with diameters ranging from 4 to 5000 m are used.
For plastics, the size typically used is 30 m in diameter and densities of 2.5 g.cm-3.
Plastichem Ltd offers six standard size ranges, all with the same spherical shape,
transparent and compression-resistant; they have controlled granulometry and high
thermal stability. The shear forces between polymer and filler are low. Moulded parts
filled with glass spheres exhibit isotropic behaviour, shrinkage in the reinforced material
being the same in all directions, so that sometimes the properties can be predicted
beforehand. Solid glass microspheres (< 50 m) improve modulus, compressive strength,
hardness, and surface smoothness. Favourable flow properties permit high filler content.
The action of glass spheres is decisively influenced by their wettability, silanes are therefore
often used as coupling agents.
Density of the hollow glass spheres ranges from 0.3 to 0.6 g cm-3, their diameter is
usually between 10 and 250 m (sometimes even 850 m).
The size distribution of solid glass spheres determines the glass spheres type. For example,
A-glass (soda lime) spheres have a mean size from 11-22 m, while the dimensions of E-
glass (borosilicate) spheres range between 11-36 m.
They decrease density, usually impair the mechanical properties and are mainly used in
foams. Hollow microspheres are not reinforcing fillers in the true sense. They displace
the same volume of resin as solid spheres, but are lighter (weight reduction can reach 20
- 25% compared with mineral filled polymers). They are colourless but in bulk they
appear as a buff-coloured, free flowing sand-like powder. They are available in two
densities and wall thickness ranges. Their main uses are in syntactic foams based on
thermosetting resins, and as fillers to reduce simultaneously the density and the dielectric
constant of epoxy resins used as encapsulants in electrical applications.
50
Particulate Fillers and Fibre Reinforcements
Glass spheres are used as fillers in both thermoplastics and thermosetting compounds.
The viscosity of the melt is increased to a much lesser extent by glass spheres than it is by
irregular particles, e.g. clay, powdered quartz, or glass fibres. This makes the production
of complex or thin-walled mouldings easier with spheres-filled than with glass-fibre-
filled thermoplastics. In comparison with the basic polymer the principal effects on
properties include reduction of shrinkage, improvements in abrasion resistance (up to
750% in acrylonitrile-butadiene-styrene (ABS) cups), compressive strength, hardness,
tensile strength, modulus and creep. A cost saving of ~ 13 - 16% could be attained [7].
Conductive glass microspheres coated with silver are also manufactured. The silver coating
lowers the volume resistivity of the moulding materials to 0.03 cm. They are used for
manufacturing of electrically conductive moulding compounds, which are inexpensive.
51
Handbook of Polymer Blends and Composites
52
Particulate Fillers and Fibre Reinforcements
Other mineral fillers are: carbides, silicon carbide being used to increase the abrasive
action of polymers, molybdenum sulphide reduces the friction, ZnS is used for white
coloration, barium titanate for electrical applications, and barium ferrite to produce
magnetisation of polymers (magnetic seals).
In the analysis and specification of fillers, several properties are of particular importance:
ASTM designation referring to specific gravity, D153-84 [32]. ASTM designation referring
to bulk density, D1513-99ae1 [33]; to pH-value, D1512-95 [34], or oil absorption, D281-
95 [35], D1483-95 [36].
Other tests include the determination of purity, freedom from coarse particles, etc., and
other properties, extensively treated in several books [1, 5, 10, 21, 24 - 26].
2.3.3 Uses and Problems and Examples of the Applications [6-28, 37-43]
53
Handbook of Polymer Blends and Composites
stress crack resistance and printability. The melt flow index, shrinkage, elongation, water
vapour and oxygen permeability decrease. Surface coating calcium carbonate use gives
high impact strength. The effects are much stronger in high density polyethylene (HDPE)
than in low density polyethylene (LDPE).
Molecular water diffuses readily even through the most water-resistant polymers. The
mechanical properties of mineral-filled polymers are most seriously affected when clusters
occur at the mineral-polymer interface.
2.3.3.1 Thermoplastics
Fillers and reinforcements suitable for polyolefins are [6-28, 37-40] mineral fillers such
as: natural and precipitated calcium carbonates, talc, mica, silica and silicates, metal
powders [37-39], kaolin, carbon black, aluminium trihydrate, wollastonite, wood flour,
asbestos, glass spheres, glass fibres and reinforcing fibres. Carbon fibres (graphite),
whiskers, etc., are also used.
Many non-black fillers such as talc, clay, silicates and carbonates continue to be used in
thermoplastics. Silanes, improve the dispersion of most mineral-filled thermoplastics and
provide protection under water immersion conditions. In many thermoplastics systems,
a relatively high load of low-cost filler can be tolerated with the addition of incremental
amounts of silane, coupling agents being used to restore the physical properties of the
composites to acceptable performance levels. A number of different fillers are used in PO
either as extenders or as physical properties modifiers. Clays are most used: in HDPE to
improve rigidity and tensile strength for automotive and pipe applications, while in
crosslinked PE to improve the dielectric constant. For very high voltage cables, surface
modified-clays are added.
Direct incorporation of fillers bypassing the preceding compounding stage is one of the
areas under investigation. Research activities are focused on the development of
inexpensive technological solutions, by combining cheap, spherical materials with more
expensive lamellar or fibrous reinforcements, and by minimising the interfacial tension
between fillers or reinforcements and the polymer matrix.
2.3.3.2 Vinyls
Fillers (extenders) are generally used in vinyls to lower the cost. The asbestos is used in
floor tile, to give dimensional stability, and processed clays are used especially to improve
the electrical properties in wire compounds. Silane-treated clays provide faster extrusion,
54
Particulate Fillers and Fibre Reinforcements
shorter Banbury mixing times and improve die wear and cables applications. For ethylene-
vinyl acetate copolymer (EVA) technical applications, the following fillers are used: calcium
carbonate for toys, heavy spar (barium sulphate) to increase resistance to X-rays and for
sound deadening, carbon black for electrically conductive articles, barium ferrite (even
up to 90 wt%) for magnetisable sealing strips, aluminium hydroxide as a flame retardant
and talc for stiffening.
The properties of PVC can be varied within large limits by additive use. Natural calcium
carbonates are by far the most important fillers for both plasticised and rigid PVC. In the
latter, only fillers with very small particles size that have been surface treated are used.
The quantity of incorporated filler can be increased in the presence of an impact modifier.
The results are: increased stiffness, lower shrinkage and reduced plate-out, notched impact
strength, enhanced ageing resistance and deflection temperature. Precipitated calcium
carbonate improves surface, shade, degree of whiteness, surface gloss and plate out.
Other fillers used in PVC articles are: talc, wood flour, silicates.
The high inherent stiffness of polyoxymethylene (POM) can be further increased by the
use of fillers. POM can be filled up to 80 wt% of glass spheres without any change in
processing conditions.
The glass spheres incorporated in polystyrene (PS) and styrene copolymers impart higher
stiffness, in particular at elevated temperature, along with higher compressive strength,
improved dimensional stability, higher HDT, but reduce impact strength. Other particulate
fillers used in PS are: calcium carbonate as extender, or nucleating agent, and fine, surface-
coated grades which serve as partial substitute of TiO2, while aluminium hydroxide
increases flame retardance. Up to 30 wt%, calcium carbonate is used in the production
of components for robots to improve mechanical properties of ABS, without appreciable
loss of impact strength.
In polyamides up to 50%, glass spheres may be incorporated. They improve tensile and
flexural strength, elongation at break, impact strength and HDT. Glass spheres permit
uniform shrinkage, close tolerance, good abrasion resistance and smooth surface of the
finished part. A notched impact strength enhancement can be obtained by the use of
surface coated glass spheres combined with glass fibres. Other improvements in
polyamides properties by using particulate fillers are: tensile strength, extrudability and
surface smoothing with silicon dioxide, stiffness, mechanical strength, hardness, thermal
and dimensional stability, surface quality with silicates (up to 40%); calcium carbonate
is a true reinforcement for polyamide (PA), increasing, besides mechanical properties,
the resistance to solvents; the amount of bronze can be up to 90 wt%, improving electrical
conductivity as aluminium, copper, lead, nickel and zinc also did. PA filled with up to
80% barium ferrite is used for small magnets while MoS2 improves slip properties.
55
Handbook of Polymer Blends and Composites
Reinforced polycarbonate (PC) is suitable for parts subjected to high static load and
which have to meet high requirements relative to dimensional accuracy, dimensional
stability and HDT.
The dispersability of TiO2 has improved by silane treatment both in PS and polymethyl
methacrylate (PMMA) as are metallic fibres in polytetrafluoroethylene (PTFE).
For most uses, epoxy resins are filled with particulate inorganic fillers. In addition to
decreasing cost, these fillers serve to increase hardness, act as heat sink for the exothermic
curing reaction, decrease shrinkage during curing add opacity and colour, and improve
other general processing and performance parameters. Silane (epoxysilane) treated fillers
(quartz, silicate, wollastonite) provide improvements in flexural strength and complete
retention of electrical properties after extensive water immersion.
The physical properties improvements result from the use of epoxy functional and
aminofunctional treated metals (aluminium needles and powder, iron powder), while
wollastonite in epoxy composites improves flexural strength with complete retention
after boiling test. Hydrated alumina is commonly added to cycloaliphatic epoxy resins
used for outdoor, high voltage insulation, largely as replacement for porcelain.
Phenolic and other condensation resins are typically used filled with alumina (abrasive
grit for grinding wheels), foundry sand (for shell moulding), glass spheres, silicon carbide
granules or treated with very low amounts of silane.
56
Particulate Fillers and Fibre Reinforcements
2.3.3.2 Elastomers
In addition to the use of some 1000 kilo tonnes of carbon black, the elastomers industry
consumes about 800 kilo tonnes of nonblack particulate fillers (~ 1300 kilo tonnes)
largely dominated by carbonates, clays, silica, titania and zinc oxide. The reinforcement
provided by many of these fillers is considerable, but their overall performance properties
are limited and, generally, are significantly lower than those obtained with carbon black
reinforcement. The use of silane coupling agents that are rubber-reactive and which will
through couple to the non-black filler surface, provides properties that are useful under
very severe end-use conditions. Unsaturated silanes in peroxide cured ethylene-propylene-
rubber (EPR) and EPDM wire and cable compounds give significant improvements to
high-voltage cable formulations. Silane-treated clays are used in considerable volume by
wire and cable and speciality goods as solid golf ball formulations, providing
improvements in resiliency, hysteresis and rebound. The use of mercaptofunctional and
aminofunctional silanes in sulphur-curing elastomers provides improvement in modulus,
compression and tensile strength. The use of these silanes in tyre compounds is also of
considerable value in improving heat build-up, abrasion resistance, etc. In many non-
black systems, combinations of TiO2 and mineral filler (as hydrous clay) are used. Specialty
elastomers such as polychloroprene, epichlorohydrin and nitrile are filled with silane
treated-silica. Modulus increases by 150-300%, compression is reduced by 30-40% and
abrasion resistance is improved by 55-70%.
SMC are thin sheets made of fibres precompounded with a thermoset resin, being used
primarily in compression moulding processes. The various types of SMC in current use are:
57
Handbook of Polymer Blends and Composites
58
Particulate Fillers and Fibre Reinforcements
59
Handbook of Polymer Blends and Composites
2.4 Reinforcements
In the field of reinforcements, the new developments are in long fibre and high performance
fibres (such as aramids, carbon or polyethylene fibres), for injection moulding products,
and in the improvement of surface treatments developing new coupling agents.
The cellulose granules derived from a woody growing ring or corncobs are introduced as
extenders in plastics. They are environmentally inert and biodegradable, non-dusting
and absorb a high quantity of water (95%) retaining the free-flowing characteristics.
They are compatible with a wide range of active agents.
60
Particulate Fillers and Fibre Reinforcements
Recently, the E-CR glass (corrosion-resistant E-glass) was developed, being particularly
designed for reinforcement of plastics submitted to acidic environment. Vetrotex
Certain Teed has developed Twintex, a commingled reinforcement of unidirectional
fibre roving and PP filaments. Glass reinforcement offering superior mechanical
properties in compounding PP, is M Star Stran from Schuller Mats and Reinforcements.
In addition to glass, the principal composite reinforcements are carbon (or graphite)
filaments, alumina whiskers, and filaments of high-strength metal such as boron.
These fibres will be less sensitive to water than glass is, because they are less
hygroscopic.
The interest in fillers and reinforcements has risen considerably since 1974, due to
the increase of the raw materials cost and the industry search for materials with
good mechanical properties and lower densities as those of metals.
The predominant fibres used for reinforcement are made of glass or carbon (graphite).
Polymeric and metal fibres are used for specialised circumstances. Mineral fibre usage
is low and declining because of health concerns and lower performance, although
these materials were once of considerable commercial importance. Natural fibres
have yet to demonstrate their performance level to justify their usage in any but non-
demanding applications. Although they provide improvement in stiffness and impact
resistance, their use is severely limited by their relatively low resistance strength loss
sets around 124 C, while thermal degradation begins around 163 C, so they impart
a dark coloration to the composite, tend to degrade quickly on exposure to sunlight
and microbial attack and absorb water and oils with diminished mechanical and
dielectric properties.
Inorganic fibres are of interest for advanced composite materials because of their thermal
resistance and their compatibility with metal and ceramic matrices. Inorganic fibres of
commercial interest are based on silica, alumina, boron, silicon carbide and high melting
temperature metal fibres (steel or alloys of molybdenum, nickel, titanium and tungsten, etc).
Table 2.4 presents the chemical composition of inorganic fibres and Tables 2.5 and 2.6
list the representative properties.
61
Handbook of Polymer Blends and Composites
62
Particulate Fillers and Fibre Reinforcements
63
Handbook of Polymer Blends and Composites
Glass fibres are the most common of all reinforcing fibres for polymeric (plastic) matrix
composites (PMC). The main advantages of glass fibres are: low cost, highly crosslinked
polymer, high tensile strength, high chemical resistance, and excellent insulating properties.
The disadvantages are: low tensile modulus, relatively high specific gravity among the
commercial fibres (2.62 g/cm3), sensitivity to abrasion with handling (which frequently
decreases in tensile strength), relatively low fatigue resistance, and high hardness (which
causes excessive wear on moulding dies and cutting tools).
After melt spinning, glass fibres are sized to protect their surface, to bind fibres together,
to improve interfacial bonding and antistatic properties. Necessary constituents of the
sizing are: a film-forming polymer to provide protection, e.g., polyvinyl acetate, a lubricant
and a coupling agent, e.g., organosilane.
The two types of glass fibres commonly used in the fibre-reinforced plastics (FRP) industry
are E-glass and S-glass.
E-glass is used in electronic boards because of its good electrical and weathering properties,
dimensional stability, good strength and stiffness, moisture resistance and lower cost
[51-53].
S-glass, originally developed for aircraft components and missile casings, has the highest
strength and modulus, high temperature resistance among all fibres in use [54, 55].
However, the compositional difference and higher manufacturing cost make it more
expensive than the E-glass. A lower cost version of S-glass, called S-2-glass, has become
available in recent years. S-glass is used in advanced composites where cost-performance
benefits can be justified.
Alkali-containing (eventually boron, too) grade A-glass is used in articles not excessively
subjected to stress and exposed neither to the action of weathering nor to moisture.
Grade C-glass is a chemical glass suitable for application requiring higher resistance to
acids but with a low strength, while dielectric grade D-glass is limited to electrical insulating
materials.
For special applications, AR, M and R grades are indicated for products subjected to
extremely high stress, for products with high strength and with good thermal and radiation
stability, respectively. AR-glass contains an increased addition of zirconium oxide and
is particularly alkali-proof. M-glass contains beryllium and is characterised by a high
modulus. R-glass is used in applications requiring radiation protection. These special
types are very expensive.
64
Particulate Fillers and Fibre Reinforcements
Oxides, such as B2O3 and Al2O3, are added to modify the network structures of SiO2, as
well as to improve its processability, while the Na2O and K2O content in E- and S-glass
fibres is quite low, which gives them a better corrosion resistance to water as well as
higher surface resistivity.
The basic commercial form of continuous glass fibres is a strand that is a collection of
parallel filaments, numbering 204 or more.
Chopped strands are produced by cutting continuous strands into short lengths from 3.2
to 12.7 mm.
Milled glass fibres are produced by grinding continuous strands in a hammer mill into
lengths ranging from 0.79 to 3.2 mm.
Glass fibres are also available in woven form, such as woven roving or woven cloth. All
these forms of glass fibres are suitable for hand lay-up mouldings.
The average tensile strength of freshly drawn glass fibres may exceed 3.45 GPa. However,
surface damage produced by abrasion, either by rubbing against each other or by contact
with the processing equipment, reduces it to a value that is in the range of 1.72 2.07
GPa. The tensile strength of glass fibres is reduced in the presence of water or under
sustained loads (static fatigue) due to existence of alkaline soluble oxides and other
alkaline impurities.
In recent years, numerous investigations have been devoted to the multiple aspects of
glass fibres utilisation in composites, as: obtaining fibres [56-65] and composites [66-
92]; investigation methods [93-112]; properties of reinforcing fibres and of the composites
[113-127]; economic aspects [128-134].
Silicon carbide (SiC) and aluminium oxide (Al2O3) fibres are examples of ceramic fibres
notable for their high-temperature applications in metal and ceramic matrix composites.
Their melting points are 2830 C and 2045 C, respectively. Silicon carbide retains its
strength well above 650 C, and aluminium oxide up to about 1370 C. Both fibres are
65
Handbook of Polymer Blends and Composites
suitable for reinforcing metal matrices in which carbon and boron fibres exhibit adverse
reactivities [46, 135-137].
Silicon carbide fibres are available in three different forms [47, 50, 51]:
3. Whiskers, which are 0.1-1 m in diameter and around 50 m in length, have been
initially produced from man-made rice hulls, which contains 10-20 wt% SiO2. The
silicon carbide whiskers contain 10 wt % SiO2 and up to 10 wt% Si3N4. The tensile
strength and modulus of these whiskers are reported as 13 and 700 GPa, respectively,
[143-145].
It is known that high thermal and mechanical resistance require strong chemical bonds. In
ceramic fibres, such a requirement is met through three-dimensional covalent bonds. Thus,
a perfect crystal, of a perfect geometry in a network, is represented by silicon carbide.
In the case of silicon carbide, the geometry of the chemical bonds is simple, as both the
carbon and the silica are tetracovalent elements, permitting regulated tetrahedral networks.
The oxide-type fibres have much more complicated networks, as a result of the elements
different valencies. The Nicalon fibre (of the silicon carbide-type) shows ultrafine -
silicon carbide crystals, while the Fibre FP alumina fibre has a polycrystalline structure
of the -Al2O3 type, with a particle size of about 0.5 m. The most representative properties
of SiC fibres are presented in Table 2.7.
Out of the most recent achievements recorded in the field of silicon carbide fibres and of
their corresponding composites, special mention should be made of:
66
Particulate Fillers and Fibre Reinforcements
new fibre made by UBE Industries with trade mark Tyranno [155, 156];
ceramic fibre tow reinforced metal matrix composite [157, 158];
microstructure control for optimising properties of ceramic composites[159-163];
honeycomb structure of thermostrucural composite material [164-170].
Continuous multifilament aluminium oxide yarn is available under the trade name Fibre
FP [45, 137, 171]. It is a high-purity (>99%) polycrystalline -Al2O3 fibre, dry spun
from a slurry mix of alumina and proprietary spinning additives. The filament diameter
is 20 m, and there are 210 filaments in the yarn. Fibre FP retains almost 100% of its
67
Handbook of Polymer Blends and Composites
room temperature tensile strength after 300 hours of exposure in air at 1000 C. Fibre
FP has remarkably high compressive strength estimated to be about 6.9 GPa.
Another ceramic fibre, containing approximately equal parts of Al2O3 and silica (SiO2),
is available as staple under the trade name Fibrefrax (3M, USA) [137, 172]. The fibre
diameter is 2-12 m and the fibre aspect ratio is greater than 200. It is manufactured
either by melt blowing or by melt spinning processes.
Some of the properties of the alumina fibres are listed in Table 2.8.
Boron fibres have an extremely high tensile modulus: 379-414 GPa and with their relatively
large diameter, they offer excellent resistance to buckling, which in its turn contributes
to high compressive strength for composites. The main disadvantage of boron is its high
cost, which is even higher than that of many forms of carbon fibres. For this reason, its
use is at present restricted to aerospace applications [49, 147, 157].
Boron fibres are manufactured by chemical vapour deposition (CVD) of boron onto a
heated substrate (either a tungsten wire or a carbon monofilament [137]). Boron vapour
is produced by the reaction of boron chloride with hydrogen:
68
Particulate Fillers and Fibre Reinforcements
Boron fibres are used in epoxy, aluminium or titanium matrices. The resulting composites
have high strength-to-weight ratio, and good compressive strength [162].
Some of the main properties of boron fibres are presented in Table 2.9.
By CVD of silicon carbide on boron fibres, borsic fibres are produced [137, 144]. Borsic
fibre loses 50% of its room-temperature tensile strength after only one hour of exposure
to air at 500 C [137].
The commercial term asbestos includes a large number of fibrous hydrated magnesium
and sodium silicate minerals of different composition, crystalline state, strength, flexibility
and length. There are six kinds of asbestos from the chrysolite and amphibole group,
which are either fibres or prismatic crystals. The colour varies from white, green and
brown; Mohs hardness 2.5 to 6; fibre diameter of 20 to 90 nm and length up to 500 nm,
mean strength; service temperature limit of 1510 C.
The long-fibre types are used as fibres and yarns, while the short-fibre ones are used as
fillers, extenders, and viscosity adjusting aids.
Asbestos fibres differ from glass fibres in that they have an unusually large surface, the
stiffness is twice as high at a similar level of tensile strength; also they are more compactly
bundled, have stronger interfibrillary bonds, and exhibit a higher degree of felting. All these
characteristics induce a significant improvement in the isotropy of the reinforcing effect.
69
Handbook of Polymer Blends and Composites
Asbestos fibres have been used in the past, but their use has ceased following discoveries
of the health hazard they may induce.
For the manufacture of composites with lower technical performances classical fibres,
characterised by normal elasticity modulus and tenacity values, are still used.
Nevertheless, in several cases, attempts have been made to replace high modulus (HM)
fibres, known to be expensive and less accessible, by classical fibres, to obtain composites
with comparable qualitative characteristics. To this end, both natural and chemical fibres
are used.
Suitable natural fibres, e.g., cellulosic fibres, bonded together by means of modified binders
containing preservatives, are used to reinforce lightweight components used in the
construction of walls, ceilings and other structural items [50, 173, 174-195].
Natural fibres are undergoing a revolution that could see them replace composite materials,
plastics reinforced with glass or carbon fibres, in the future. Daimler Benz is a world leader
in using natural fibres in its vehicles and has been using flax fibres in the door panels of its
Mercedes G-Class cars since 1995. The real challenge though is to eliminate the oil derived
plastic and replace it with a renewable version [175, 176].
German car manufacturers are researching the use of naturally occurring materials to
replace synthetics. Mercedes-Benz is now using animal hair and fibres made from coconut,
cotton, flax and sisal in its cars and want to progress their use further. Currently, they are
investigating the potential to replace glass fibres in reinforced plastics with cellulosic fibre.
Current problems are processing and the poor heat resistance of natural fibres [51].
Originally, high performance composites were defined as being: carbon, aramid, ceramic
fibres (silica carbide and Al2O3), high molecular PE, aromatic polyester fibres (liquid crystal
polyester; LCP) and special glass fibres from R- and S-glass. However, over the last ten
years, natural fibres such as flax and hemp have been used in low performance composites
and the development in this field has lead to the assumption that natural fibres are set to
replace E-glass fibres in high performance composites [54, 177, 178].
Cotton fibres [176, 177, 191, 194], flax or hemp fibres [174, 177, 184, 185, 190, 192,
193], sisal fibres [175, 178, 179, 186-188], pineapple-leaf fibres [183], palm tree leaves
[173, 183, 195], bamboo fibres [181, 183] and jute fibres [182, 187, 188] have all been
used as reinforcing fibres.
70
Particulate Fillers and Fibre Reinforcements
Several carbocatenary and heterocatenary chemical fibres have been used as reinforcing
agents for composites [47, 50, 51]. Thus, special mention should be made here of both
carbocatenary and heterocatenary chemical fibres [196, 197].
Out of the carbocatenary fibres the following have been used: acrylic fibres [198-200],
Vinylon (PVA) fibres [201-202] and classical polyolefinic fibres [203-208].
Among the heterocatenary ones, the most representative are: polyester fibres [209-211]
utilised especially for the fabrication of nonwovens; polyamide fibres [212-214], mainly
employed in reinforcing car types; nylon based composites are claimed to reduce component
cost and weight when used instead of cast metal; elastic or polyurethanic fibres [215, 216].
Also, fibres based on both regenerated and modified cellulose have been successfully
used [217].
Synthetic fibre reinforced plastics have significantly lower moduli than glass reinforced
plastics in the direction of the fibre, but there is a higher resistance to damage by
distortion.
Carbon fibres are produced by heat treatment of organic precursors such as rayon [218,
219-224], polyacrylonitrile (PAN) and pitch.
PAN-based carbon fibres can be obtained by the following method: stabilisation at 200-
300 C in oxygen, depolymerisation and aromatisation, converting thermoplastic PAN
to a non-plastic cyclic or ladder compound; carbonisation at 1000-1500 C in an inert
atmosphere to get rid of non-carbon elements; stretch and graphitisation at > 1800 C.
Pitch-based carbon fibres may be obtained by the following method: heating at > 350 C,
condensation reaction, formation of mesophase (liquid crystal; LC); melt spinning into
pitch fibres; conversion into graphite fibres at 2000 C.
71
Handbook of Polymer Blends and Composites
Carbon fibres have the highest modulus and strength of all reinforcing fibres at both
room and high temperatures, with a tensile modulus ranging from 207 GPa on the low
side to 1035 GPa on the high side. Low-modulus fibres have lower specific mass, lower
cost, higher tensile and compressive strength and higher tensile strain-to-failure than
high-modulus fibres [218].
Table 2.10 shows the typical mechanical properties of commercially available carbon fibres.
72
Particulate Fillers and Fibre Reinforcements
The density is low, however, the performance:cost ratio limits their use to high
performance applications such as aerospace, where stiffness:weight ratio is the primary
concern [219-226].
Carbon fibres are electrically conductive, which require special care when using them
around electric motors and electronic controls [227].
Since carbon fibres are difficult to wet with resins, surface treatments are used to increase
the number of active chemical groups and harden the fibre surface [228].
Structurally, carbon fibres contain a blend of amorphous and graphitic carbon. Their
high tensile modulus results from the graphitic form, in which carbon atoms are arranged
in crystallographically parallel planes of regular hexagons [47, 218]. The distance between
the planes (3.4 ) is larger than that between the adjacent atoms in a plane (1.42 ). The
planes are held together by weak van de Waals-type forces, and a strong covalent bond
exists between the carbon atoms in a plane. This results in highly anisotropic physical
and mechanical properties for the fibre [233-236].
There are different types of carbon fibres reinforcements and recent achievements have
come from the three main types of composite materials:
carbon fibre reinforcing-polymer composites (CFRP) [229-232, 241-254];
carbon fibre reinforcing-metal composites (CFRM) [228, 235, 255-263];
carbon fibre reinforcing-carbon composites (CFRC) [234, 240, 264-276].
Among the aspects to which the literature of the field has been predominantly devoted,
one should note:
obtaining carbon fibre-based composites;
investigation of composite fibres properties considered as a whole and in their diversity,
including fibre degradation and compatibilisation;
applications based on strength and stiffness, thermophysical properties, chemical
inertness and corrosion resistance, specific rigidity and good damping, biological
inertness and x-ray permeability, electrical conductivity, fatigue resistance, etc.
73
Handbook of Polymer Blends and Composites
Aramids are a group of highly crystalline aromatic polyamide fibres that have the lowest
specific gravity (1.47 g/cm3), the relatively high thermal and chemical resistance and the
highest tensile strength-to-weight ratio.
Some imperfections among crystals may be noticed, along with imperfections in crystals
packing and deviations from perfect orientation. These fibres have a sheath-core structure
with a porous skin region with arrays of needle-shaped voids aligned approximately
parallel to the fibre axis.
There are several versions of Kevlar used in textile structural composites, as shown in
Table 2.11.
The thermal treatment under tension improves structural perfection, less ordered forms,
with a smaller modulus, e.g., Kevlar 29, being present along with more ordered forms,
with a higher modulus, e.g., Kevlar 49 and 149.
74
Particulate Fillers and Fibre Reinforcements
Nomex, the meta-aramidic fibre, has a partially oriented structure, once the shape of
the poly-meta-phenylene terephthalamide prevents the formation of liquid crystals.
The structure of the Nomex fibre is closer to that of the synthetic polyamidic and
polyesteric fibres.
Other HM fibres made from liquid crystals are the Technora aramidic fibres, the Vectra(n)
aromatic polyesteric fibres and some new polymeric fibres with highly oriented structures,
which make them highly crystalline, such as the poly-p-phenylenebenzbisoxazole (PBO),
poly-2,5,(6)-benzoxazole (ABPBO), poly-p-phenylene benzobisthiozole (PBZT). Like
carbon fibres, they also have a negative coefficient of thermal expansion in the longitudinal
direction, which is utilised in designing low thermal expansion composites for printed
circuit boards. The major disadvantages of aramidic fibre for reinforced composites are
their low compressive strength, tendency to fibrillate, poor transverse properties,
susceptibility to abrasion and difficulty in cutting or machining them [47-48, 50-51].
Absorbed moisture seems to have very little effect on the tensile properties of Kevlar 49
fibres. However, at high moisture content, they tend to crack internally at the pre-existing
microvoids and produce longitudinal splitting.
A second-generation Kevlar fibre is Kevlar 149, which has the highest tensile modulus of
all commercially available aramid fibres. The tensile modulus of Kevlar 149 is 40%
higher than that of Kevlar 49; however, its strain-to-failure rate is lower. Kevlar 149 has
an equilibrium moisture content of 1.2% at 65% relative humidity and at 22 C, which
is nearly 70% lower than of the Kevlar 49 under similar conditions. Kevlar 149 also has
a lower creep rate than Kevlar 49.
Kevlar 149 fibres are commercially available as untwisted textured yarns with 134, 267,
768 and 1,000 filaments, roving with 3,072 and 5000 filaments, and woven and non-
woven fabrics [283, 284].
75
Handbook of Polymer Blends and Composites
76
Particulate Fillers and Fibre Reinforcements
Extended chain PE fibres, commercially available under the trade name Spectra, are
produced by solution spinning of a high molecular-weight PE. Solution spinning yields a
highly oriented fibrous structure with exceptionally high crystallinity (95-99%) [47, 48,
50, 286, 287].
Spectra PE fibres from the Allied Company have the highest strength-to-weight ratio of
all commercial fibres available to date. Two other outstanding features of Spectra fibres
are their low moisture absorption (1% compared to 5-6% for Kevlar 49) and high abrasion
resistance.
The melting point of Spectra fibres is low (147 C), and since they exhibit a high level of
creep above 100 C, their application temperature is limited to 80-90 C. Their poor
adhesion with resin matrices, can be partially improved by their surface modification
with gas-plasma treatment.
Spectra fibres provide high impact resistance for composite laminates even at low
temperatures and are finding increasing application in ballistic composites. However,
their use in high-performance aerospace composites is limited, unless they are used in
conjunction with stiffer carbon fibres to produce hybrid laminates with improved impact
damage [288-291].
Some of the physico-mechanical properties of the high performance PE fibres are presented
in Table 2.13.
The structure and deformation behaviour of UHMWPE fibres and composites based on
them are explained, and advances in the development of self-reinforced composites are
predicted [295-300].
77
Handbook of Polymer Blends and Composites
The literature gives a survey on the development of reinforcing fibres used for polymeric
composites along with the description of the improvement of the properties of carbon
fibres. In addition, the status quo of the development of inorganic fibres for metallic and
ceramic matrices is presented. A cost rating of the different fibres is performed on the
basis of their production processes. Also, the market development for glass, carbon and
aramide fibres is described geographically and by application.
FRP continue to be used in new and innovative designs. Sales of advanced composite
components were exceeding 20,000 metric tons in 1995, yet, FRP users/suppliers need
to demonstrate their recyclability to maintain this market acceptance.
Most of the composites in existence today were originally developed for aerospace
applications. Although the results have been very successful, the cost factor has hindered
spreading of composites into other application areas and industries. Other factors that
prevent the faster growth of composite usage include capital investment, inspectability,
part consistency, lack of well-established manufacturing techniques and industry-wide
specifications.
With the end of the cold war, the composite research in aerospace and defence areas is
decreasing, which leaves transportation and construction as the biggest potential areas
for composite growth. As a result, due to different and probably less demanding
applications, new low cost composites need to be developed, for the utilisation in
mechanical, civil and structural engineering or in automotives and electronic applications.
The emphasis laid on smart or intelligent composites and materials has increased recently.
A smart or intelligent material may be defined as a material that possesses intrinsic
sensing, controlling and actuating capabilities. The growing use of composites in highly
sophisticated systems such as aerospace, may necessitate the development of smart or
intelligent composites. This, in turn, may require new developments in polymers, fibres
and other textile structures. Although some intelligent materials can be found in naturally
occurring biosystems, the development of synthetic intelligent materials is still in the
conceptual stage.
The future of advanced composites is reported, with the long-term status of the market
predicted for the next 20 years. The future looks good even if prospects may not look so
attractive in the short term. Four major sectors are identified as providing the bulk of the
future growth: infrastructure and constructions; transportation and alternate fuels for
transportation; marine structures; and industrial machinery [301].
78
Particulate Fillers and Fibre Reinforcements
The 21st century will usher in the need for repair retrofit and replacement of facilities,
which are deteriorating at a rapid rate. Composites will provide the material to resist
corrosion and maintain the safety of structures.
References
7. P. D. Ritchie, Plasticisers, Stabilisers and Fillers, Illife Book Ltd., London, UK,
1977.
10. L. Mascia, The Role of Additives in Plastics, Edward Arnold, London, UK, 1974.
12. H. S. Karz and J. V. Milewski, Handbook of Fillers for Plastics, Van Nostrand
Reinhold, New York, USA, 1987.
14. H. Schtling, International Plastics Handbook for the Technologist, Engineer and
User, 2nd Edition, Hanser Publishers, Munich, Germany, 1987.
79
Handbook of Polymer Blends and Composites
18. Handbook of Fillers for Plastics, Eds., J. V. Milewski and H. S. Katz, Van
Nostrand Reinhold, New York, USA, 1987.
23. J. Weiss and C. Bord, Les Materiaux Composites, Volume 1, CEP Edition, Paris,
France, 1983.
27. C. Vasile and M. Rusu in Handbook of Polyolefins, 2nd Edition, Ed., C. Vasile,
Marcel Dekker Inc., New York, USA, 2000.
29. ASTM D1366-86, Standard Practice for Reporting Particle Size Characteristics of
Pigments, 1997.
80
Particulate Fillers and Fibre Reinforcements
30. ASTM C92-95, Standard Test Methods for Sieve Analysis and Water Content of
Refractory Materials, 1999.
31. ASTM D1510-99, Standard Test Method for Carbon Black-Iodine Adsorption
Number, 1999.
32. ASTM D153-84 (1996) e1, Standard Test Methods for Specific Gravity of
Pigments, 1996.
33. ASTM D1513-99ae1, Standard Test Method for Carbon Black, Pelleted-Pour
Density, 1999.
34. ASTM D1512-95, Standard Test Methods for Carbon Black - pH Value, 2000
36. ASTM D1483-95, Standard Test Method for Oil Absorption of Pigments by
Gardner-Coleman Method, 1995.
41. R. F. Jones, Guide to Short Fibre Reinforced Plastics, Hanser Publishers, Munich,
Germany, 1998.
42. P. Bataille, L. Richard and S. Sapieha, Polymer Composites, 1987, 10, 2, 103.
81
Handbook of Polymer Blends and Composites
52. Seong Sik Cheon, Jin Ho Choi and Dai Gil Lee, Composite Structures, 1995, 32,
1-4, 491.
53. G. O. Shonaike and T. Maksuo, Composite Structures, 1995, 32, 1-4, 445.
54. M. Akay and D. F. ORegan, Plastics, Rubber and Composites Processing &
Applications, 1995, 24, 2, 97.
59. C. R. Key and Y. J. Shih, inventors; National Starch & Chemical Investment
Holding Corporation, assignee; US 5,491,182, 1994.
82
Particulate Fillers and Fibre Reinforcements
65. C. R. Key and Y. J. Shih, inventors; National Starch & Chemical Investment
Holding Corporation, assignee; US 5,665,470, 1997.
73. G. J. Cook and A. Hughes, inventors; United Wire Ltd., assignee; E 0 697 921,
1996.
75. P. Davoine and J-C. Berthier, inventors; Sarl ADN, assignee; EP 0703307, 1996.
79. H. Kimura, M. Maeda and T. Hirai, inventors; UBE - Nitto Kasei Co., assignee;
US 5,540,986, 1996.
80. H. T. Van de Grampel and Y. Hou, inventors; Azdel Inc., assignee; US 5,500,433,
1996.
83
Handbook of Polymer Blends and Composites
82. S. Kichmeyer, inventor; Bayer AG, Bayer Antwerpen NV, assignee; E 0751100,
1997.
85. F. E. Penrose, inventor; Smith and Nephew plc, assignee; EP 0773800, 1997.
87. K. D. Beer, B. Anderson and R. A. Davis, inventors; PPG Industries Inc., assignee;
US 5,672,641, 1997.
89. M. Ushida and M. Ogata, inventors; Shin-Kobe Electric Machinery Co., assignee;
US 5,662,776, 1997.
94. R. J. Brooks and G. J. Turvey, Composite Structures, 1995, 32, 1-4, 203.
97. Y. Katahara and T. Katayama, Composite Structures, 1995, 32, 1-4, 567.
98. M. R. Wisnom and M. I. Jones, Composite Structures, 1995, 32, 1-4, 615.
84
Particulate Fillers and Fibre Reinforcements
100. Z-F. Li, D. T. Grubb and S. L. Phoenix, Composites Science & Technology, 1995,
54, 3, 251.
101. M. Pinfold and G. Calvert, Composite Structures, 1995, 32, 1-4, 549.
103. T. E. Tay, H. G. Ang and V. P. W. Shim, Composite Structures, 1995, 33, 4, 201.
105. X. Yuanming and W. Xing, Composites Science & Technology, 1996, 56, 2, 155.
106. M. L. Benzeggagh and M. Kenane, Composites Science & Technology, 1996, 56,
4, 439.
107. H. Leber and J. M. Lifshitz, Composites Science & Technology, 1996, 56, 4, 391.
110. M. Salvia and L. Vincent, Composites Science & Technology, 1996, 56, 7, 797.
111. E. V. Pisanova and S. F. Zhandarov, Composites Science & Technology, 1997, 57,
8, 937.
112. R. Park and J. Yang, Composites Science & Technology, 1998, 58, 6, 979.
113. D. C. Lee, J. J. Lee and S. J. Yun, Composite Structures, 1995, 32, 1-4, 39.
115. B. Bompard and V. Veauville, TUT Textiles Usages Techniques, 1997, 23, 32.
85
Handbook of Polymer Blends and Composites
123. J. P. Tancrez, J. Pabiot and F. Rietsch, Composites Science & Technology, 1996,
56, 7, 725.
86
Particulate Fillers and Fibre Reinforcements
143. S. Kumar and R. N. Singh, Composites Science & Technology, 1996, 56, 11,
1271.
144. D. B. Gundel and F. E. Wawner, Composites Science & Technology, 1997, 57, 4,
471.
145. L. Zheng, Y. Jin and P. Li, Composites Science & Technology, 1997, 57, 4, 463.
146. U. Cocen, K. Onel and I. Ozdemir, Composites Science & Technology, 1997, 57,
7, 810.
148. Hua-Tay Liu and P. F. Becher in Thermal and Mechanical Test Methods and
Behaviour of Continuous-Fibre Ceramic Composites, Ed., M. G. Jenkins, Marcel
Dekker, New York, 1997, 128.
149. S. Lee and P. V. Mosher, inventors; SGL Carbon Composites Inc., assignee; US
5,759,688, 1998.
150. B. Guichet and J. C. Sangleboeuf, Composites Science & Technology, 1998, 58,
5, 665.
153. A. Thionnet and J. Renard, Journal of the European Ceramic Society, 1995, 15,
12, 1235.
87
Handbook of Polymer Blends and Composites
157. J. Pan, J. H. Li and H. Fukunaga, Composites Science & Technology, 1997, 57,
3, 319.
160. P. Colomban and N. Lapous, Composites Science & Technology, 1996, 56, 7, 739.
167. C. G. Cofer, J. Economy and Y. Xu, Composites Science & Technology, 1996, 56,
8, 967.
169. R. Kahraman, Composites Science & Technology, 1996, 56, 12, 1453.
88
Particulate Fillers and Fibre Reinforcements
171. S. Q. Wu, H. Z. Wang and S. Tjong, Composites Science & Technology, 1996,
56, 11, 1261.
173. S. K. Al-Oraimi and A. C. Seibi, Composite Structures, 1995, 32, 1-4, 165.
174. H. Burger, A. Koine, R. Maron and K. P. Miek, GAK Gummi Fasern Kunststoffe,
1995, 48, 7, 475.
177. C. Joly, R. Gauthier and B. Chabert, Composites Science & Technology, 1996,
56, 7, 761.
178. K. Joseph, S.Thomas and C. Pavithran, Polymer, 1996, 37, 23, 5139.
181. N. Akiyama and S. Irie, inventors; ASK Corp., assignee; GB 2,202,152 A, 1997.
186. A. Paul, K. Joseph and S. Thomas, Composites Science & Technology, 1996, 56,
1, 25.
187. J. Gassan and T. Bledzki, GAK Gummi Fasern Kunststoffe, 1996, 50, 5, 360.
89
Handbook of Polymer Blends and Composites
192. K. P. Mieck, T. Reussmann and Ch. Hauspurg, GAK Gummi Fasern Kunststoffe,
1998, 52, 9, 740.
194. J. Gassan, I. Mildner and V. Bledzki, Technical Textiles, 1998, 41, 2, 74.
200. G.M. Lorenz and W. Fester, inventors; Hoechst AG, assignee; US 5,591,509, 1997.
204. E. K. Tschegg and N. Finger, Composites Science & Technology, 1996, 56, 8, 947.
206. T. Stern, A. Teishev and G. Morom, Composites Science & Technology, 1997, 57,
8, 1009.
90
Particulate Fillers and Fibre Reinforcements
213. T. Masuda, Y. Horii and T. Iwama, inventor; Toray Industries Ltd., assignee; GB
2,292,385, 1996.
215. R. Gilat and J. Aboudi, Composite Structures, 1995, 32, 1-4, 81.
217. G. Leumer and E. Gebauer, TUT Textiles Usages Techniques, 1996, 20, 59.
218. J-B. Donnet and R. C. Bansal, Carbon Fibres, Third Edition, Marcel Dekker Inc,
New York, USA, 1998.
219. J. Longhlan and M. Ata, Composite Structures, 1995, 32, 1-4, 13.
220. J. Eiblmeier and J. Longhlan, Composite Structures, 1995, 32, 1-4, 97.
221. M. S. Found and M. J. Friend, Composite Structures, 1995, 32, 1-4, 115.
223. M. S. Found and C. I. Howard, Composite Structures, 1995, 32, 1-4, 159.
224. L. V. Smith and S. R. Swason, Composite Structures, 1995, 32, 1-4, 177.
226. K. S. Jeong, D. G. Lee and S. H. Oh, Composite Structures, 1995, 32, 1-4, 557.
227. Y. G. Kim, K. S. Jeong and D. G. Lee, Composite Structures, 1995, 32, 1-4, 575.
91
Handbook of Polymer Blends and Composites
229. S. Singh and I, TUT Textiles a Usages Techniques, 1996, 19, 13. K. Partridge,
Composites Sciences & Technology, 1995, 55, 4, 319.
231. H. Hamada and S. Ramakrishna, Composites Sciences & Technology, 1995, 55,
3, 211.
232. J. A. Barnes and C. Newton, Composites Sciences & Technology, 1995, 55, 4, 287.
233. S. S. Cheon, J. H. Choi, D. G. Lee, Composite Structures, 1995, 32, 1-4, 491.
236. B. Revzin, D. Fuks and J. Pelleg, Composites Sciences & Technology, 1996, 56, 1, 3.
237. L. V. Smith and S. R. Swanson, Composites Sciences & Technology, 1996, 56, 3, 359.
243. L. Caramaro and G. Vignie, TUT Textiles Usages Techniques, 1995, 16, 29.
244. L. Ye, K. Friedrich, J. Kastel and Y-W. Mai, Composites Sciences & Technology,
1995, 55, 4, 349.
246. E. Leblanc, C. Piuna, P. Sigety and I. Vincon, Composites Sciences & Technology,
1996, 56, 7, 855.
92
Particulate Fillers and Fibre Reinforcements
251. K. P. San, T. K. Chaki and D. Khastgir, Polymer, 1998, 39, 25, 6461.
253. K. E. Atkinson and C. Kiely, Composites Science & Technology, 1998, 58, 12,
1917.
254. T. Nojima and T. Kusaka, JSME International Journal, Series A: Mechanics and
Material Engineering, 1998, 41, 2, 225.
255. K. Gamble, M. Pilling and A. Wilson, Composite Structures, 1995, 32, 1-4, 265.
258. H. Kishi and N. Odagiri, inventors; Toray Industries Inc, assignee; US 5,557,831,
1996.
260. J. Raghovan and M. Meshii, Composites Science & Technology, 1997, 57, 4,
375.
263. J-M. Ting and M. L. Lake, inventors; Applied Sciences Inc, assignee; US
5,814,408, 1998.
93
Handbook of Polymer Blends and Composites
266. S. C. Khatri and M. J. Koczak, Composites Science & Technology, 1996, 56, 6,
181.
270. S. C. Khatri and M. J. Koczak, Composites Science & Technology, 1996, 56, 4,
473.
272. D. Leveque, M. H. Auvray, Composites Science & Technology, 1996, 56, 7, 749.
275. B.Y. Park, S. C. Kim and B. Jung, Polymers for Advanced Technologies, 1997, 8,
6, 371.
276. V. Bianchi, P. Goursat and W. Sinkler, Journal European Ceramic Society, 1997,
17, 12, 1485.
280. H. van der Werff and M. H. Hoffman, Chemical Fibres International, 1996, 46,
12, 435.
283. R. Neuert and H. Jung, inventors; Hoechst AG, assignee; EP 0688743, 1995.
94
Particulate Fillers and Fibre Reinforcements
288. S. W. Shalaby and M. Deng, inventors; Smith & Nephew Richards, Inc., assignee;
EP 0714460, 1996.
293. S. Shalom, H. Harel and G. Marom, Composites Science & Technology, 1997,
57, 9-10, 1423.
295. R. H. Zee and C. Y. Hsieh, Materials Science and Engineering A, 1998, 246, 1-2,
161.
296. B. J. Field and R. Soar, inventors; Pacific Safety Products Inc., assignee; US
5,796,028, 1998.
297. E. Devaux and C. Caze, Composites Science & Technology, 1999, 59, 6, 879.
95
3
Composites in Asia
A. K. Kulshreshtha
3.1.1 Introduction
Use of composites is not yet as popular in India [1-11] as compared to their extensive
and multipurpose use worldwide. The composites industry in India which could be
broadly grouped under the aerospace and defence and non-aerospace segments is
about three decades old. The aerospace and defence establishments in the country
most of which are under the Government of India are the technology developers and
main users of composites for their major design and development programmes like
the Light Combat Aircraft. These programmes have spear-headed the design, analysis,
fabrication, testing and certification techniques for composites. The technologies and
expertise on composites developed for these programmes offer potential applications
in the non-aerospace industrial sectors as well. Figure 3.1 shows India to be the
poorest consumers.
The lack of sustained technological back-up and continuous research and development
(R&D) efforts are responsible for the industry not penetrating into high-cost and high-
volume applications.
97
Handbook of Polymer Blends and Composites
A careful analysis of the status of the fibre reinforced plastics (FRP) industry in India and
its specific needs reveals the following:
Most of the FRP manufacturers are small-scale units whose infrastructure and
technology levels are far from satisfactory. These units require upgrading in order to
enter into high-value and high-volume production. Technology information, design
and production development are some of the important services required by the
industry to grow further and penetrate the national and international markets.
The spin-offs of various technologies and expertise that have already been developed
for the aerospace and defence industries in the country also have good market potential
for civil applications. However, proper technology absorption and transfer mechanisms
are required. With the Governments new policy of economic liberalisation the industry
is fast undergoing a metamorphosis and, by 2000-2001, there is likely to be an upward
surge in the usage of composites in the country. In order to accelerate the growth and
development of the industry, there is a need for concerted and coordinated efforts
from all concerned - the industry, Government, R&D institutions, etc.
98
Composites in Asia
3.1.3 Overview
The Indian market for composites is limited to the electronics and aerospace sectors. The
current domestic market size has been estimated to be 13,000 tons for thermoset
composites and about 3,500 tons for thermoplastic composites with an approximate
turnover of about 75 million US$. Average annual growth rate in the last decade has
been about 15 per cent per year. The share for thermoset composites by sector is presented
in Table 3.1.
The major market for thermoplastic composites has been automotive and engineering
industries. With the boom in two wheeler production in the country, the usage of thermoplastic
composites has increased in the automotive sector. Also, passenger car manufacturers with
an export thrust would gradually switch over to using composites from the conventional
materials due to their energy savings potential. The Automobile Association of India (AAI)
has devised a programme of substituting steel springs (weight 25 kg) by ones made of
composites (weight 5 kg). Bajaj Auto is also actively pursuing a programme for improved
usage of composites for the components namely gears, bearings, etc.
Demand for composites in the domestic sectors was 10,000 to 15,000 tons in the year
2000. While the bulk of it was in low cost components, medium cost composites were
used for leaf springs, propeller shafts, bumpers, etc. With the increased manufacture of
light commercial vehicles (LCV) and heavy commercial vehicles (HCV), demand for
medium composites was around 1,000 tons for four wheelers and around 2,000 tons for
two wheelers.
Fabrication of chemical process equipment has been an established user segment for
composites in India. The process equipment handling corrosive fluids are potential
application areas for FRP.
99
Handbook of Polymer Blends and Composites
These include storage tanks, bodies for water treatment plants and cooling towers,
scrubbers, ducting, piping, exhaust stacks, pump casings etc. With the introduction of
resin transfer moulding (RTM) processes for fabrication, chemical equipment would be
more cost effective and would adhere to the manufacturing standards. The usage of
composites in the chemical industry grew at 10 per cent per year to reach the level of
about 11,000 tons by the year 2000.
Composites can be used for the fabrication of boats or as lining for the wooden hulls in
conventional boats. There exist about 30-40 small boat builders mainly around the
coastline. The consumption of composites by this segment is mainly restricted to glass
fibre reinforced polyesters and the total usage of composite is estimated to be about 750
to 1,000 tonnes per year. With the development potential for marine fishing, the boat
building industry is expected to grow at 12% per year. This would certainly have bearing
on the consumption of FRP for nonstructural components for bigger vessels as a measure
of weight reduction and tackling corrosion problems.
Glass epoxy laminate has been used for two end uses, (i) insulation laminates for the
electrical industry and for (ii) printed circuit boards (PCB). SSB Industries, Ltd., under
the trade name Xtruglas make these.
The usage volume in this segment was about 5,000 TPA in 2000. Hindustan Aeronautics
Ltd. (HAL) is the only aircraft manufacturer catering for defence requirements in India.
The present production programme of HAL is indicated in Table 3.2.
100
Composites in Asia
Kevlar and a carbon fibre tail section. The target of production of AHL for civilian and
foreign markets beyond 1997, is 300. With such ambitious projections taking shape in
the future, the consumption of composites in aerospace sector rose to 100 TPA by 2000.
Composites have found extensive usage in the Indian Space Programme as well as missiles.
Vikram Sarabhai Space Centre (VSSC), Trivandrum, has been using around 2 TPA of
carbon phenolic composites in the rocket nozzle. The future requirement of VSSC is
around 15 tons over a period of 4 years for their Polar Satellite Launch Vehicle (PSLV)
and Augmented Satellite Launch Vehicle (ASLV) projects (Figure 3.2). On successful
completion of these projects, the Indian Space Programme may require around 5 TPA of
composites based on carbon fibre. The missile programme under Defence Research and
Development Organisation (DRDO) for testing and production of the missiles would
require around 5 TPA of carbon fibre reinforced composites (Figure 3.3).
101
Handbook of Polymer Blends and Composites
FRP have been taking the world by storm. Lamentably, in terms of mass production, they are
still in their infancy in India. However, India still has a long way to go. Indias composite
consumption measures a meager 20,000 tonnes as compared to Chinas 1,80,000 tonnes and
order of higher magnitudes achieved by the developed countries in the world.
While the rest of the world have comfortably adapted to composites, seeing in them a safe and
durable substitute to such conventional material like wood and metal, Indian users are still
lagging behind. However, the launch of INSAT-2C on 7 December 1995 was successful and the
roll out of the LCA on 17 November 1995 was indeed an epoch making event, particularly for
the composites community. The LCA uses composites extensively. Wings, Fin, Rudder, Elevons,
Air Brakes and Landing Gear Doors in this aircraft are all built with composites.
The Defence Bio-Engineering and Electro-Medical Laboratory (DEBEL) Bangalore has developed
an improved version of light-weight integrated helmet for use in high G aircraft like LCA. This
helmet is fabricated out of a Kevlar-FRP combination which can withstand high shock attenuation
and penetration resistance. It is provided with ear pads with communications and mask
102
Composites in Asia
connections. A dual visor system consisting of clear and tinted polycarbonate visors is fitted for
windblast and antiglare protection. The inner visor is provided with an adjustable lock of
greater security under blast conditions. Attachments for oxygen mask microphone/radio
telephone connector, and night vision goggles are also mounted inside the helmet.
Vast investment potential exists in the processing sector (see Tables 3.3 and 3.4).
103
Handbook of Polymer Blends and Composites
India holds the world monopoly in the production and export of mica. Muscovite or
potassium or ruby mica (H2KAl3 (SiO)3) is a first class insulator of electricity. The function
of an electrical insulator is to strongly resist the passage of electrical current. Mica has
proved itself as an invaluable and an ideal insulator for electrical and electronic equipment
operating at variable temperatures and other conditions. This is ascribed to its volume
constancy, i.e., freedom from contraction and expansion.
Next to India, the output comes from Brazil and other countries reporting small production
individually are Argentia, Colombia, Guatemala, Mexico, Mozambique, Sri Lanka, Sudan,
Tanzania and USSR.
In India, the most commonly used technique is hand lay-up, accounting for 80 per cent of the
total volume produced, followed by filament winding (12%), press moulding (6%) and
others (2%). Most of the composite processing units are in small scale sectors, using mainly
contact moulding process which cannot be used for high volume production. A few bigger
units with press facilities using the sheet moulding compound (SMC) process are there but
they have problems of size and mould availability. Indian composite industry presently is not
in a position to meet the processing of thermoplastic composites for high volume productions.
Some companies have the facilities for pultrusion to cater to the needs of the electrical industry.
Experts are of the view that RTM and Reinforced Reaction Injection Moulding (RRIM) are
the two processes suitable to be adopted and developed for the domestic environment.
Contact moulding has been the predominant method of glass reinforced plastic (GRP)
processing in India to the extent of nearly 70%. Filament winding, compression moulding,
pultrusion and injection moulding of reinforced thermoplastics account for the balance
30%. The Glass Fibre Technology Centre (GFTC), FGP Limited, India has been in the
forefront (since 1981) in process, product and applications development in the field of
GRP composites. The GFTC is devoted to the propagation and development of GRP
composites with the intention of broadening the market base for glass fibres.
Realising the need to break away from the traditional labour intensive contact moulding
technique, the GFTC ushered in the concept of automated RTM technique in India in 1994.
The approach consisted of developing suitable grades of glass fibre reinforcements having
delayed styrene solubility (since continuous strand mat is not available in India), RTM moulds
for doors, passenger car bumpers, modular panel water tanks, chair shells, etc.
104
Composites in Asia
Realising the importance of composite materials and their potential benefits, particularly
in industry, the Technology Information Forecasting and Assessment Council (TIFAC),
Department of Science and Technology, launched a programme in 1990 with the active
collaboration with the National Aerospace Laboratories (NAL), Bangalore, for creating
and updating a technology database for dissemination of information on composites for
the benefit of various industries in India. This programme has led to an increased awareness
about composites and their applications and to a greater interaction between the industry
and the R&D institutions.
In order to further synergise the development efforts in R&D institutions with those in the
industry, and to promote the increased applications of composites in various areas, TIFAC
has set up a Technology Project in Mission Mode on Advanced Composites (TPMM-ACM)
in 1994. Under this mission, TIFAC has planned to set up Composites Design Centres in
various parts of the country to provide the necessary links between R&D/academic institutions
and the FRP industry to ensure the rapid growth of the technology and its applications.
The first composites design centre to be established in the country is the TIFAC composites
design centre in Bangalore.
Organising HRD and training programmes for the composites industry at various levels,
particularly covering students, moulders and entrepreneurs.
The composites design centre will focus mainly on the design and development of selected
products having high value as well as high volume requirements in the country. The range
of products identified for immediate attention is:
105
Handbook of Polymer Blends and Composites
Energy Sector: windmill blades, compressed natural gas bottles and other corrosion
resistant applications;
Sports and Leisure Goods Sector: sports goods and musical instruments.
The Composite Structures Laboratory has made very impressive progress: it successfully
developed a carbon fibre composite rudder and fin for the LCA, using the novel co-cure,
co-bonding technology, and delivered spars and other components to LCA. In addition,
NAL also successfully developed and made operational, its 3 m x 7 m computer-controlled
autoclave, and went on to undertake the development of the countrys largest autoclave
(4 m x 8 m) for HAL.
1992 saw the development of carbon fibre composite wings for the LCA. It was a
challenging task to design, fabricate and test the first ever wing of its kind. The LCA has
integrated with carbon fibre composite (CFC) wings, rudder, fin and landing gear doors.
In spite of a sluggish start, the Indian GRP composites industry has been enjoying a
healthy uninterrupted average annual growth rate of 15% since 1987. This trend is
expected to continue which bears ample testimony to its excellent growth prospects.
106
Composites in Asia
The annual growth rate of 15% of GRP composites in India is much higher than the
average growth of 5-6% worldwide and is typical of the good growth prospects of GRP
composites in the Asia Pacific region in general, and India, in particular. However, though
India and the Peoples Republic of China (PRC) have been contemporaries in the glass
fibre industry (the glass fibre industry in both countries took off in the mid-1960s), the
current glass fibre consumption in China is nearly five times that of India.
The current installed capacity of the two companies in India presently involved in glass
fibre manufacture, (FGP and UP Twiga Fibreglass), is in the region of 14,000 tonnes.
One more company (Binani Glass Fibre) which has an installed capacity of 6,700 tonnes
started commercial production in July/August 1996. The global leader, Owens Corning,
has already drawn up ambitious plans of setting up a greenfield project in India with an
installed capacity of 30,000 tonnes the plant was commissioned in late 1998-1999.
The growth rate of the GRP industry in India has been remarkable in the last decade.
The annual growth rate has been in the range of 15% plus during the last 4-5 years and
has taken a quantum jump (in the range of 25%) by 1999-2000. Some of the reasons for
the healthy growth rate have been:
A greater awareness amongst end users of the advantages of using glass fibre
composites. This has mainly come about by the development of several new
applications by Indian glass fibre manufacturers and also through a systematic process
of educating the end users on the long-term advantages of GRP products, e.g., reduced
maintenance, lighter weight, etc., in spite of higher prime cost.
107
Handbook of Polymer Blends and Composites
Right from its inception in the late sixties, the chemical and marine industries have been
considered to be the prime domain of GRP in India apart from its high usage in translucent
rooflight sheets by the building industry. Even after three decades, the chemical sector
still continues to be the main outlet (23%) for the use of GRP composites for corrosion
resistant applications in India. The significant difference today is that high volume
applications, e.g., underground gasoline storage tanks, large diameter pipes for water
transport, sewage, etc., have been the new areas where GRP has made significant inroads
especially in the last three years. Following closely is the building/construction sector,
which accounts for 21% of glass fibre consumption and which is expected to have a
boom for the next 5-7 years. GRP translucent roofing sheets continue to be the mainstay
in building applications by taking advantage of Indias tropical conditions, where the
ambient temperature for the major part of the year exceeds 30 C which results in the
need for diffused lighting in the interior of buildings. The use of GRP in cooling towers
has also seen a significant increase (5%) over the years. It is expected to increase still
further as cooling tower manufacturers are not seriously contemplating replacement of
even timber/concrete sections of GRP pultruded profiles. A heartening feature in the use
of GRP for building/construction has been its increasing use in recent years in structural
applications such as dome light formwork/shuttering, doors with frames, window sections,
etc. It is expected that GRP pultruded profiles would be used to quite an extent from the
year 1998-1999 onwards for building applications.
The automotive industry in India is poised for a tremendous growth in the next decade.
The present growth rate is 25-26% which is expected to increase significantly in the next
two years with most of the global passenger car manufacturers, (for example, Daewoo,
General Motors, Mercedes Benz, Rover, Volkswagon, Peugeot) having decided to set up in
India by tying up with leading Indian automobile manufacturers. Hitherto use of GRP for
automotive applications in India was restricted mainly to jeep body components, engine
hoods, radiator fans blades for trucks and several components for bus bodies. However,
108
Composites in Asia
this trend is expected to extend into extensive use of GRP for passenger car components in
the light of their tremendous success in USA and Europe. SMC and reinforced thermoplastics
have had a major application in several of the new passenger cars rolling out on the Indian
roads from 1997. For the Indian GRP industry, the task has been made simple since most
of the leading car manufacturers mentioned above have all had the successful experience
of GRP components in cars in the US and European markets.
In view of the geographical location, India is an ideal country to harness wind power.
Based on exhaustive wind mapping studies it has been estimated that India has the potential
to generate approximately 20,000 MW wind power. It is for this reason that the
Government of India has been encouraging projects, relating to setting up of wind farms
in the country at strategic locations. More than 45 wind farms have been successfully set
up in the southern, western and eastern parts of India and these have been in operation
for more than 5 years. To date, Indias installed capacity has been 500 MW and 2000
MW was the target for the year 2000. In terms of installed capacity, India stands fourth
in the world with the United States topping the list with 1400 MW. The wind turbine
blades are all made by GRP and several manufacturers have already started manufacturing
the blades in GRP in India. This application involves use of glass fibre reinforcements
with specialised orientation of the fibres and contributes significantly to the quantum
jump (of 25%) forecast for glass fibre consumption in the year 1999-2000. This is one of
the largest outlets for GRP in India in the years to come.
Indcarf was made in 1990s in a 75 million US$, 12 tonnes/year facility (Figure 3.5) at the
Research Centre of Indian Petrochemicals Corporation, Ltd., (IPCL) at Vadodara. The
facility was designed for processing polyacrylonitrile (PAN) based special acrylic fibre
(SAF) from Courtaulds Grafil, UK and has the capability to produce oxidised acrylic
320K tow and Type II grades (INDICARF 25 and INDCARF 30) of carbon fibres of 3K,
6K and 12K filament tows. These tows are available with size finishes applied to them
either for weaving or prepreging. Carbon fibre applications are used in defence and
other sectors.
The steps involved in the manufacture of Indcarf carbon fibre from PAN-based SAF are
oxidation precarbonisation, carbonisation and graphitisation (Figure 3.5).
109
Handbook of Polymer Blends and Composites
110
Composites in Asia
The carbon fibres and products from PADC are expected to find use in addition to the
strategic area of defence and aerospace, in the areas of automobiles, medical field, sports
goods, aircraft industries and other industrial applications. The PADC facilities were
created to have capacity to size/desize carbon fibre tows and to produce chopped fibres,
carbon fibre tapes, carbon fibre fabrics, thermoplastic granules reinforced with short
carbon fibres, prepregs, pultruded sections, injection moulded reinforced engineering
thermoplastic materials and to produce thermoset as well as thermoplastic composites
including large scale manufacturing of composite products.
Details of PADC production are given in Tables 3.5 and 3.6 and shown schematically in
Figures 3.6 to 3.10.
111
Handbook of Polymer Blends and Composites
112
Composites in Asia
113
Handbook of Polymer Blends and Composites
114
Composites in Asia
Figure 3.9 Processing of carbon fibres from raw material to final product
115
Handbook of Polymer Blends and Composites
load from fibre to resin and again to fibre and should therefore improve compression
strength. For polymeric materials a possible figure of ~5 GPa should be achievable as
against the current resins with shear modulus of about 2 GPa. As far as hydrothermal
degradation is considered, newer systems based on cyanate ester look very promising
and some of these have already been used. Another route being investigated is the use of
thermoplastic resins and their blends. Polyether-ether-ketone (PEEK) looks promising,
but the industry needs to resolve the problems associated with high temperature (>350
C) processing of the material. Current approaches to new resins appear to be directed
towards producing polymeric systems which can be processed in the way composites
industry is used to (such as autoclave curing up to 180 C). Two promising classes of
materials under development are:
The other area of advances in resins is that of low-loss resins specially for radomes
which use high-performance advanced radars. Varieties of low-loss polyesters and cynate
esters are being explored.
The establishment of satellite technology has successfully taken place in India. The first
Indian satellite Aryabhata was designed, fabricated and launched from USSR
Cosmodrome in 1975.
116
Composites in Asia
The follow-on indigenous second generation INSAT, namely INSAT-II, system has also
been designed.
The Indian Remote Sensing Satellite (IRS), the first semi-operational remote sensing
satellite weighing 950 kg, a highly complex three axis stabilised polar orbiting satellite
capable of providing state-of-the-art images with a resolution of 30-50 metres has been
designed and fabricated. IRS-1A and IRS-1B were launched successfully in 1988 and
1992, respectively, and are now providing vital inputs for monitoring and management
of Indias natural resources.
ASLV launch vehicle programme capable of launching 150 kg satellite into a low earth
orbit and the PSLV launch vehicle capable of launching 1000 kg IRS class of satellites
into a polar orbit were fabricated. Cryogenic engine technology development has been
initiated which will lead to the establishment of GSLV (Geo-Synchronous Launch Vehicle)
capable of launching of 2-tonne class of satellites into a geostationary orbit. R&D activities
in the area of composite materials and structures were established with the primary
objective of developing composite technology for aerospace applications.
India is now entering the golden era of FRP and composites. After many difficult years
the Indian composite industry has now attained a stable equilibrium and is poised for
take-off. Significant progress has been achieved in several technology areas of advanced
composites and many milestones have been reached in composite product development
117
Handbook of Polymer Blends and Composites
for civil and military applications during 2000. This was evident when the indigenously
built ALH, all composite aircraft, HANSA-21, airborne early warning rotodomed aircraft
first flew.
With regards to advanced composites, still a dominating aerospace and defence sector,
has marked an unparallel success with production of carbon composite structures such
as fin, rudder and wing for an Advanced Fighter Aircraft (AFA), using state of the art co-
cure and co-bonding technology.
The technologies which were considered to be very complex and much advanced for the
Indian technologists to achieve, were successfully developed and demonstrated. One such
technology developed at National Aerospace Laboratories (NAL), namely derivative
tooling technique, was used for producing a complex main landing gear door for the
LCA. These developments have paved the way to even more challenging tasks for
producing carbon composite fuselage parts for the AFA, in a one shot operation. Perhaps,
the next logical step is to attempt production of carbon composite wing for the LCA,
using similar technology.
Technological breakthroughs are not new to NAL especially in the field of composites, in
which several noteworthy advances have been made. The latest in this string of success is the
co-cured fin (Figure 3.11) for a High Speed Fighter Aircraft (HSFA) developed by the composite
structures laboratory. The technology is first of its kind in India, it utilises advanced composite
materials, state-of-the-art tooling and fabrication processes, quality control and inspection
and test methods for indigenous development of a primary aircraft structure.
The technology developed for a high performance aircraft has promising potential considering
the wide ranging use of composites in non-aerospace applications to which it can be put.
The composite materials and technology gained momentum at NAL in the early 1990,
with launching of a major project called Development of Carbon Composite Fin for
HSFA, sponsored by Aeronautical Development Agency (ADA). The project envisaged
development and supply of CFRP composite rudder and fin. The project demanded
expertise in advanced tooling and fabrication technology including setting up of a modern
processing and testing facility. The team has achieved the break through in technology,
and successfully developed the CFRP rudder and fin, satisfying the stringent specifications.
118
Composites in Asia
Figure 3.11 Schematics of (a) a CFRP co-cured fin and (b) a pressure pad mandrel for
the fabrication of a torsional box
The fin has a multi spar-rib skin construction, comprising of a CFRP leading edge torsional
box, closing ribs, and metallic fittings, through which it is attached to the fuselage. The
torsional box is the single largest composite component in the fin. This is also a classic
example of a multi rib-skin co-cured construction.
After the rollout, NAL scientists got a shock. Weighing 888 kg (including the weight
of two pilots and the fuel), Hansa had exceeded the weight-limit of small-trainer
aircraft by 138 kg. If NAL could not reduce the weight, the aircraft would have had
to go through the more rigorous and expensive certification process meant for larger
aircraft. The problem generated a fierce debate among NAL scientists. A large number
of them felt that it was impossible to reduce the weight. An equal number felt that it
was possible, and worth trying. After a year and a half of effort, Hansa 3 rolled out,
weighing only 750 kg, and was formally certified by the Director-General of Civil
Aviation.
The 50-odd scientists who worked on the Hansa consider the weight reduction as
one of their greatest achievements. The pilot and the fuel together weigh 225 kg
without compromising on the strength of the aircraft. How they achieved this is a
story in itself. Behind it is a bigger story of the development of state-of-the-art
composite materials.
119
Handbook of Polymer Blends and Composites
Hansa is now widely recognised as being as good as the best aircraft in its category, the
Diamong Katana. Priced at US$85,000, it costs about US$17,000 less than the latter. In
addition it has at least one feature which the Katana lacks: lightning protection. If marketed
well, the Hansa could sell in large numbers across the world. In fact, enquiries have
already come from firms in the US, Canada and Australia.
The Centre for Air Borne Systems (CABS), a laboratory under DRDO, has taken up a
project for the development of air borne surveillance platform. The project needed a
radar transparent rotodome (rotating radar dome) to carry the primary and secondary
antennae. The rotodome with two antennae would be mounted on the aircraft fuselage
with the help of two pylons. The rotodome rotates to scan 360 C in the azimuth.
The rotodome is made up of two composite radome shells, two end caps and one
aluminium metal centre section. The radome shells and end caps are built with a sandwich
structure using glass prepreg and Nomex (polyamide; DuPont) honeycomb cores. The
broad specifications are given under technical data. The exploded view of rotodome is
shown in Figure 3.12.
CABS with the assistance of HAL, Bangalore and MBB (of Germany) developed the first
rotodome. The preliminary design was carried out by the CABS design team, and technical
assistance was provided by Indian Institute of Science for wind tunnel testing and NAL
for design. It was also agreed to transfer the technology with necessary toolings and
drawings to CABS.
CABS was in need of a second rotodome for its second aircraft. Importing another
rotodome (only the composite parts) would have cost the exchequer millions of US Dollars.
120
Composites in Asia
To save this enormous amount of foreign exchange and to gain expertise in the
development of a large size sandwich radome, it was proposed to develop the rotodome
indigenously. However, the tools and the technical experience gained during the
development of first rotodome were alone not sufficient for the development. A major
infrastructure required was the autoclave, to cure the fabricated large size radome.
Autoclaves available in India were not able to meet the size requirements for this project
(4 m x 8 m). However, a (4 m x 8 m) oven was under installation at HAL LCA-complex
for the LCA project. A decision was taken to cure the radomes in oven, after ascertaining
that the cure cycle meets the quality requirements.
The radomes and end caps were developed in the following stages:
Tool modification
Tool rework/preparation
121
Handbook of Polymer Blends and Composites
122
Composites in Asia
of glass fibre reinforced plastic (GFRP) material whereas the fuselage is a combination
of aluminium and GFRP. The aircraft has an endurance of 90 minutes while carrying
a mission payload of 20 kg, and can achieve a speed of 180 km/h in flight.
The use of GRP in automotive applications which was restricted to jeep bodies a
decade ago, has gained rapid momentum over the last couple of years and is now
being extensively used in bus body construction, LCV bodies, truck drivers cabin,
etc. The key factors that favour GRP as a material of construction in automotive
applications are reduction in weight of the vehicle leading to lower fuel consumption,
less wear on tyres, reduced maintenance cost, etc.
The Indian automobile industry has been witnessing some hectic activity over the
past few years. With the advent of the economic reforms, barriers for the entry of
multinational car manufacturers have been broken and as many as 17 foreign car
manufacturers announced their projects either individually or as joint ventures with
a regional partner (Table 3.7). Tables 3.8 and 3.9 provide additional information
relating to this.
3.1.24 Railways
India owns the largest railway network in the world and the total production of
coaches is about 2,000 per year. The usage of composites has recently been initiated
by Indian railways.
Based on the components that have been developed, the requirement of composites
for coaches would be around 150 kg per coach giving rise to a total use of around
300 TPA of composites by the railways. If the advanced fabrication processes namely
RTM are adopted and concerted efforts are undertaken towards design development,
usage of plastic composites and light metals may make a significant contribution to
railway coaches. This may eventually raise the volume to about 10,000 tons of
composites (for an average 2-3 tonnes per coach and a production level of about
3,000 coaches). This sector will consume low cost or to a limited extent of medium
cost composites.
123
Handbook of Polymer Blends and Composites
124
Composites in Asia
Work on the development of FRP-based gear case for traction motor (TM 4906 AZ) of
broad gauge diesel locomotives of 2600 hp for use in Indian Railways was carried out by
Regional Research Laboratory, Bhopal.
In one locomotive six gear cases are required, which constitutes one set. Each gear case is
mounted on a railway traction motor with the help of two bosses. The heavy weight of the
steel impact, is very common to locomotives. The light weight of the FRP gear case is also
expected to prevent such occurrences. The specially developed FRP material is used in moulding
two halves of FRP gear case. Thirty four steel fixtures were fixed on the FRP gear case during
assembly. The improved gross weight efficiency of FRP gear case per locomotive per year is
estimated to be 108,624 tonnes per kilometre. Additional income to the railways by adding
capacity of five percent shall be in the order of US$20,000 per locomotive per year. The
saving in weight will not only assist in reducing fuel consumption but also provide an
opportunity to the locomotive designers to add additional features in the present model.
125
Handbook of Polymer Blends and Composites
GRP has made significant inroads in this field in India in the following applications:
Following extensive field trials for over three years, the Department of Non-Conventional
Energy Sources (DNES) has now approved GRP as a material of construction for biogas
126
Composites in Asia
holders. GRP is also an accepted material for solar cooker bodies. Field trials are now
progressing GRP blades in windmills intended for minor irrigation.
China and Brazil have also done a considerable amount of work on biogas plants. In fact,
Brazil has developed the entire biogas plant (consisting of gas holder and digester) in GRP.
Another development in the UK has been the use of GRP anaerobic digesters for generation
of biogas from livestock solid waste dumped from slaughter houses.
3.1.26 Defence
The main advantages of GRP for defence items are its light weight and shrapnel-proof
properties. A few such applications are:
Bullet-proof helmets
Ammunition boxes
Bridge sections/ladders using GRP pultruded sections that can be easily assembled
and installed in a short time with minimum manpower.
Command posts
Pickets for fencing using pultruded sections
Mobile shelters in modular form
Pistol grips
Single-piece moulded drivers cabin for military trucks
Mine casings.
The defence sector offers a good potential of numerous medium cost composites along
with the following products:
Bullet resistant composite armours have already been developed indigenously for defence
applications. The reinforcements used include glass, Kevlar and ceramics depending on
specific protection needs.
127
Handbook of Polymer Blends and Composites
The FRP modular container is a mobile container to house various kinds of equipment
(electronic, electrical or mechanical) and to protect them from the effects of environmental
changes in various sectors of operations. The Research and Development Establishment
(Engineers)/RDEE, Pune, has developed such a container which can also be mounted on
a vehicle to facilitate mobility of equipment and instruments for in situ usage. Further, it
also provides for electromagnetic interface (EMI)/electromagnetic conduction (EMC)
shielding requirement of sensitive electronic equipment.
The salient features of the FRP modular container developed by RDEE are:
The FRP container weighs 1,400 kg and 3,500 kg when completely laden with equipment
on the vehicle. There exists a margin for incorporating an additional facility with a
load capacity of 500 kg. The other technological features of the container include:
fresh air sucking louvres, hot air outlets, exhaust fans for proper ventilation. AC and
DC power supply for proper lighting, and anti-vibration. AC and DC power supply
for proper lighting, and anti-vibration mountings for the equipment to absorb shocks
up to 4 g level.
3.1.28 Miscellaneous
Bicycles are a potential application area for composites. While the Indian bicycle industry
today produces 8 million per year, the usage of composites has not caught on so far.
Given a 10 per cent penetration, the total production of bicycles in India can generate a
business volume of about 1000 TPA for composites. The sector also enjoys excellent
export potential.
While the domestic consumption of composites for sports and leisure goods is almost
negligible, the segment presents excellent export potential. The global consumption of
carbon fibre for sports and leisure goods has been estimated at about 1,000 TPA at
128
Composites in Asia
present. The production value of the present market is about US$ 100 million excluding
the branded products. With 10 per cent share in the global market by India, the demand
rose to 250 TPA of composites by 1995.
3.1.29 Constraints
The availability of raw materials for composites in India is fraught with some typical
economic problems. The following factors need to be considered:
Prices for fibre glass available in India are very high. This attributed to the higher
energy cost and smaller plant size.
Government policies have imposed high excise and import duties on raw materials
and finished products.
For category II PMC raw materials are mostly imported and high import duties have
direct bearing on the raw material prices.
While the Indian industry of GFRC is quite well established, the industry lacks
standardisation and proper testing facilities [10]. The test facilities located at a few
educational institutions and defence laboratories, etc., are not accessible by the small
scale industries. No serious efforts have so far been undertaken towards design
development, standardisation and quality assurance programmes for composite products
in India.
3.1.30 Future
In future, use of composites in India will be very significant for both domestic and export
segments. The processing capability is now restricted to few organisations, namely HAL,
NAL, Indian Space Research Organisation etc. However, it is not difficult to develop a
technology/production base. In fact, the existing facilities and expertise available in these
organisations should be fully exploited. The current problem in commercial production
is the low volume of demand for composites in a particular segment of application. With
increasing awareness of the entrepreneurs and the users, composites have established
themselves also in the domestic material market. Consolidation of demand in minimum
varieties, moderate costs, use of proper material specifications, development of standards
etc. would further lead to improvement in the commercial viability of the composites.
129
Handbook of Polymer Blends and Composites
3.1.31 Conclusion
The rapid turnaround in the fortunes of the glass fibre industry in India in the last 10
years has been mainly due to concentration of R&D efforts in the critical areas of glass
fibre manufacture and development of new applications. The advent of automated
processing techniques in recent years such as RTM, pultrusion, etc., and the awareness
of a greater sense of quality consciousness on the part of processors have also contributed
to the growth. With a wider range of unsaturated polyester and phenolic resins now
available in the Indian market, leading resin manufacturers have already made plans for
enhancing their production capabilities to match the increased glass fibre demand. The
thrust being given to composites in India is evident from the setting up (in 1993) of a
Technology Mission Mode for Advanced Composites by the Ministry of Science and
Technology, for more widespread propagation of composites in the country in the coming
years. The Indian Governments emphasis on indigenisation of carbon fibre technology
coupled with new standards on glass fibres and allied products, along with the
development of new applications have all resulted in a synergistic effect on the Indian
composites scenario which is presently growing from strength to strength.
It is up to the industry now to take advantage of the design and fabrication facilities of
the Composites Design Centre to innovate their product design and development. But,
can the industry come out of the traditional brush and bucket methods and enter into a
modern, state-of-the-art technology? This is indeed a moot question and a challenge.
The user-industries too have a responsibility to encourage the manufacturers to venture
into producing cost-effective and good quality products and accept them in preference
to, and in place of, products made out of wood and metals. A good partnership between
the manufacturers, the users and the composites design centre is essential, and if this
happens, it can, no doubt, revolutionise the Indian composite industry.
130
Composites in Asia
Byung-Sun Kim
3.2.1 Introduction
The advanced composite materials and technology in Korea have been used from the middle
of the 1970s in some government institutes, universities, defense and aerospace industries.
Initially, the high performance ablative composite materials were studied and products
were developed for solid rocket engines. Later structural composite materials began to get
attention due to the wide and important applications to aircraft and aerospace industries.
In the beginning of the 1980s the carbon fibre prepregs were available and carbon fibres
were produced from industries in the middle of the 1980s. The development and production
of carbon fibre prepregs was an important landmark in the advanced composite materials
in Korea, since the delivery and storage of prepregs are very important factors.
Government research institutions, universities and industries have carried out research
and development independently or jointly. Parts are developed for robots, aircraft, land
transportation, civil engineering applications, sport and leisure applications, bullet-proof
helmets and other defense products, electric transformers, domestic applications, etc.
Research is carried out on design and stress analysis, process development, various fibre
and resin development, heat resisting composites for rocket propulsion units, sheet
moulding compound (SMC) composites for high impact energy absorption, bulk moulding
compound (BMC) and carbon/carbon (C/C) composites, thermoplastic composites, etc.
The aims of the research and development are for the practical application of the
composites and its efficient productivity.
The fibres, the resins and their prepregs are either manufactured by local industries or
imported. Fibres are glass, carbon and Kevlar while resins are epoxy and polyester.
However, those for aerospace applications are still in basic research and development
stage and thus rely on imports.
In this chapter, a brief review of recent research and development and the anticipated
applications of local advanced polymeric composite materials are discussed. This particular
chapter is the modified version of the overview published in 1988 [12].
Government research institutions, universities and local industries carry out research
and development on advanced polymeric composite materials. The research organisations,
their areas of specialty and products are discussed as follows.
131
Handbook of Polymer Blends and Composites
There are five main government research institutions involved in working with composite
materials. They are Korea Institute of Machinery and Materials (KIMM) [13], Korea
Institute of Science and Technology (KIST) [14], Korea Research Institute of Chemical
Technology (KRICT) [15], Agency for Defense Development (ADD), and Korea Aerospace
Research Institute (KARI) [16]. KIST concentrates on the development of raw materials
such as fibres and resins, KRICT concentrates on the research and development of fibre/
matrix interface and intermediate materials such as prepregs, honeycomb cores, etc.
KIMM is focusing on process technologies and product developments while ADD put
more emphasis on the development of defense products.
The Airframe Structural Laboratory of KARI has interests in the design and stress analysis
of composite structures, material characterisation, and performance evaluation of
structures. The development of composite parts for 4-8 passenger, all composite planes
and 100 seat mid-size planes is its major on-going project. Figure 3.14 is the composite
material mock-up model. The institution is also planning to develop the composite
antenna, solar panel, and ultra-light structures for Korean Multipurpose Satellite (KOSAT)
and composite pressure vessel for scientific rockets.
In the polymer processing laboratory of KRICT, the raw materials for composites such
as fibres, resin matrices, core materials for sandwich, and adhesives have been researched
for last decade. Currently the high performance resins for advanced composites are under
development in cooperation with local companies.
The research topics of polymeric materials research group of KRICT cover the areas of
advanced polymeric materials such as heat-resistant, high modulus, conductive,
photosensitive, biocompatible polymers, resins for selective separation of heavy/precious
132
Composites in Asia
metals, adhesives and solid polymer electrolysis. The research group has produced many
successful applications such as conducting polymers for electrolytic capacitors, less toxic
flame retardant polystyrene for electric appliances, insulin patchs for transthermal drug
delivery system, anti-slipping agents for stacking boxes and technology for high-quality
carbon fibre production.
One of the most active research areas is the development of basic technology for highly
functionalised polyimides, which have outstanding heat resistance, excellent mechanical
and electrical properties and high optical transparency. These polyimides can be used as
lightweight structural materials for automobiles, aerospace and marine vessels, low
dielectric materials for electric and microelectronics and as functional materials for
information and communications.
The research group is now expanding this research topic to new application fields such
as gas separation polymers, etc. Novel preimidised polyimides with good solubility,
outstanding heat resistance, desirable electrical/optical performance have been studied
for use in microelectronics.
The group has also found new technology for the good storage-stable copolyimides
precursor with alternating repeating units in polymer backbone. It is expected that the
technology can be applied as a low dielectric thin polyimide film.
Their research include the development of high performance polymer hybrid materials,
structural hybrid materials for transportation and civil engineering applications, new
fabrication technology to promote the maximum materials properties of the materials,
intelligent polymer hybrid materials, etc.
133
Handbook of Polymer Blends and Composites
resin transfer moulding (RTM), filament winding, tape wrapping, braided pultrusion,
vacuum bag moulding and textile technology, while developing parts such as a bicycle
frame (Figure 3.15) and other sports leisure products, pressure vessels for a compressed
natural gas (CNG) vehicle (Figure 3.16), robot arm, transmission shaft, excavator arm
for heavy machinery, composite rotor blade for helicopters, trough to contain high
voltage cables (Figure 3.17) and parabolic antenna used as domestic satellite dishes
(Figure 3.18).
Figure 3.15 First composite bicycle frame developed, reproduced with permission from
Samchuly Bicycle Co., Corex Co. and Composite Materials Laboratory of KIMM, 1994
Figure 3.16 Tested composite pressure vessel for CNG vehicle, reproduced with
permission from Daewoo Automotive Co. and Composite Materials Laboratory
of KIMM
134
Composites in Asia
The composite bicycle developed by the silicone rubber bladder moulding process is the
first one developed in Korea and the companies collaborating in its development Samchuly
Bicycle Co., and Corex Co. Corex Co., have been exporting their own models to Europe.
The composite pressure vessel for the CNG vehicle was developed by filament winding
technology for Daewoo Automotive Co. The vessel developed met part of North American
Department of Transportation regulations and other tests are to be performed to satisfy
the rest of the regulations. These vessels are to be used for CNG vehicles in early 21st
century. Carbon/epoxy robot arms produced by filament winding reduced its weight by
135
Handbook of Polymer Blends and Composites
40% improving the arms performance greatly. A composite trough was fabricated by
Pull-Form technology in collaboration with Dong Yang Industries and the trough was
reinforced with glass fibre. Each trough is four metres long and buried underground
while containing high voltage electrical cables. Carbon/epoxy parabolic antenna developed
by autoclave for the Hyundai Electronics Company is used as a domestic satellite dish
with which the wave reception is lot more stable in all weather conditions.
KIMM also leads the RTM process study group, publishes the Composite Newsletter
bimonthly and maintains the Korea Chapter of Society for the Advancement of Materials
and Process Engineering (SAMPE). Figure 3.19 shows the mould and carbon/epoxy
artificial leg shell for medical application fabricated by RTM. Figure 3.20 shows carbon/
epoxy composite coil spring of 5 mm-diameter wire also fabricated by RTM. Such
composite coil springs could possibly be used for aerospace applications.
Figure 3.19 The mould and artificial leg shell produced by RTM, reproduced with
permission from Composite Materials Laboratory of KIMM
136
Composites in Asia
Figure 3.20 Fabricated composite coil spring and RTM mould, reproduced with
permission from Composite Materials Laboratory of KIMM
Research on advanced composite materials have also been essential in the defense
sector. Some research laboratories in ADD are involved in developing heat resistant
composite materials and lightweight airframe parts. System and part design are their
major roles and the intensive cooperative work with industry have given them
motivation to improve composite materials technology. They require more strict
specifications that lead to the improved quality of composites parts. The filament
wound structures, autoclave and tape-wrapping products, and sandwich panels are
being developed for some specific applications.
137
Handbook of Polymer Blends and Composites
Among the universities, SNU and KAIST are the most active. At SNU, the study of
core process parameters is carried out for various process technologies such as
autoclave, RTM, filament winding, pultrusion and thermoplastics. At KAIST, research
is being carried out on material characterisation, fibre/matrix interface, some process
technologies of thermoset and thermoplastic composites, stress analysis of composite
structures, delamination and interlaminar fracture, buckling and post buckling
problems, fatigue and fracture of composites and nondestructive evaluation, etc.,
are carried out. In addition, they are offering some specific courses on composite
materials such as Introduction to Composite Materials, Composite Materials
Manufacturing, Design with Composite Materials, Mechanics of Composite
Materials at the department of mechanical engineering and aerospace engineering.
Research areas for the rest of the universities are the material characterisation, carbon
fibre of C-shape, carbon/carbon composites, thermoplastic composites, stress analysis
of composite structures, fatigue and fracture of composites and nondestructive
evaluations.
Recently, the Structural Safety Institute of Kookmin University has developed a glass
fibre reinforced composite bridge deck ( Floor Deck) with KookminCI Inc. A
demonstration bridge will be built by SSangyong Construction Co., in 2001. In the
construction, the composite bridge deck, from Floor Deck and the DuraSpan Floor
deck from Martin Marietta Composites of USA will be used. In manufacturing
Floor Deck, the parts with a triangular cross-section were fabricated by filament
winding technology and the part with a rectangular cross-section was fabricated by
the pultrusion process. Then the parts were bonded. DuraSpan Floor deck was also
manufactured by a pultrusion process. The decks fabrication processes are shown in
Figures 3.21 and 3.22.
Compared to the concrete bridge deck, the composite deck has a life span of 4~5 times
longer and due to the reduction of the decks weight by about 80%, the construction is
expected to be lot easier and quicker. The bridge decks have been examined and tested
by the Korea Construction Technology Institute. This verified the safe use of the bridge
deck. The composite bridge decks are shown in Figure 3.23.
138
Composites in Asia
Figure 3.21a Glass fibres are wound around a cement block with a cross-sectional area
that is a delta shape
Figure 3.21b As Figure 3.21a except it is viewed from the side of the cement block
139
Handbook of Polymer Blends and Composites
Figure 3.23 Composite Bridge Decks, D-DECK Floor Deck, DuraSpan Floor Deck,
Assembled Composite Bridge Deck
140
Composites in Asia
3.2.2.3 Industries
Local industries are involved in developing raw materials and product developments.
Glass fibres are produced by Han Kuk Fibre Glass Co., Hankuk Betrotex Co. and Lucky
Fibreglass Co. Taekwang Co. is the only carbon fibre manufacturer while Han Kuk
Fibre Glass Co. and Sunkyung Industry Co. produce carbon fibre prepregs. Their raw
material developments are summarised in Table 3.10.
Local consumption of carbon fibres in 1991 was 567 tons and gradually increased to
about 865 tons by 1998. About 95,000 tons of polyester, 5,500 tons of epoxy were
consumed annually over the past four years while 95,000 tons of polyester and 1,500
tons of epoxy were produced each year. Resin manufacturing companies are National
Chemical, Miwon, Aekyoung, Korea Chemical, Kukdo Chemical, Korea IPCO, etc.
About 80% of carbon fibres are used for production of sporting goods. Annual
consumption of carbon fibre is estimated to be about 1,000 tons. Most of the glass fibres
are consumed for commercial applications. Some of the parts produced by major local
industries are summarised in Table 3.11.
141
Handbook of Polymer Blends and Composites
fuselage is 1.8 m (W) x 10.95 m (L), and that of main wing is 2.5 m (length) x 0.7 m
(maximum thickness along length) x 13.9 m (span) [15]. The master model and lay-up
mould, laminating, vacuum bagging, autoclave curing, and finishing is carried out by the
Han Kuk Fibre Glass Company under subcontract by Samsung Aerospace Industry. The
first flight test was successfully performed in the middle of 1996. Korean Air has the
experience of developing a solar panel for satellite. Carbon/carbon composites are partially
produced in prototype scale by Daewoo Heavy Industries (DHI) by the cooperative
research with NII Grafit in Russia (Figure 3.24). The C/C brake discs are being considered
for use on aircraft as well as on high speed trains.
Figure 3.24 Carbon/carbon composite brake disc, reproduced with permission from
Daewoo Heavy Ind. Ltd.
142
Composites in Asia
Figure 3.25 is the composite Aileron for Turbo-Prop Airplane developed by Daewoo Heavy
Ind. Compared with the metallic aileron, the weight was reduced by 18% and number of
parts were reduced by 35%. As a result, the cost of assembly was reduced by about 25%.
Figure 3.25 Composite aileron for turbo-prop airplane, reproduced with permission
from Daewoo Heavy Ind.)
One of the commercial applications is seen in Figure 3.26. This vertical axis type windmill
system has composites blades fabricated in Han Kuk Fibre Glass Co. The blade is 2 m
wide and 30 m long.
Figure 3.26 Vertical axis type windmill system with composite blades
(courtesy of Han Kuk Fibre Glass Co.)
143
Handbook of Polymer Blends and Composites
The spar material is carbon/epoxy and the skin has glass/epoxy honeycomb structure
cured by autoclave. The Ministry of Commerce and Industry supported this programme.
For the local industries, the FRP Association was organised in early 1980. The association
now has more than 200 companies as members. These small to mid-sized companies
produce large tanks, containers, pipes as well as sporting goods (tennis rackets, fishing
rods, golf shafts, wheel support for roller skate, bicycle shoe insole, etc.) KSCM was
founded with the support of industry, universities and Government research institutions
in 1986. Table 3.11 summarises typical products developed by local industries.
The research funds from the government support two areas basic and applied research.
The government supports much more in applied research since this R&D can lead to a
product development. Basic research funds are supported by Ministry of Science and
Technology (MOST, www.most.go.kr), Korea Science and Engineering Foundation
(KOSEF, www.kosef.re.kr/), Ministry of Education and Agency for Defense Development
(ADD). However, applied research funds are supported by Ministry of Commerce and
Industry (Korea Academy of Industrial Technology, www.mci.go.kr), Agency for Defense
Development (ADD).
Demand for carbon fibre in 1987 is about 250 tons and its market size was about five
million dollars. In 1991 all composites market including carbon and glass fibre composites
grew up to 0.4 billion dollars. A detail of production and market share is shown in Table
3.12. Data of glass fibre reinforced plastics (GFRP) are taken from reference [19] and
those of carbon fibre reinforced plastics (CFRP) and Kevlar fibre reinforced plastics
(KFRP) are from 1994.
Simply considering the fact that the number of composites parts/components used in
Korea is only about 80 when compared with more than 1,200 parts in USA, it can easily
be expected to have great number of composites to be applied in future.
Other than some aircraft components which are being developed, the applications in the
civil engineering area have just begun recently by two companies in Korea and are expected
144
Composites in Asia
to expand. At the same time, environmental problems are also an issue. KIMM and KRICT,
have studied this problem for a few years. The recent recycling method utilised by most of
the local industries is landfill or burn up/landfill. However, some of the industries began to
grind the waste parts and use them to strengthen the cement, use them as a filler for a resin
system, asphalt, or use them as a portion of BMC ingredient. However, inexpensive grinding
methods are still to be studied for cost effectiveness. In the mean time, recyclable composites
such as low cost thermoplastic parts are being developed.
145
Handbook of Polymer Blends and Composites
The BMC market is also expected to grow rapidly in the areas of automotive, electrical
components and housewares.
3.2.5 Conclusions
Designing structures with composite materials seems to be mid-level. The design of shapes,
stress analysis, failure analysis, joint analysis, fractures and vibrations are studied in
many universities and institutes. Some companies appear to have the capability to design
and fabricate the secondary composites structures for aircraft.
The research and industrial history of advanced composite materials is not so long as the
developed nations. However, the production and applications are growing rapidly. The
nations momentum is moving towards aerospace industry and high-value-added products.
Thus, the future of advanced composite materials in Korea is further expanding its
capability both in quantity and quality. Many experts anticipate that the composite
industries will catch up with that of the developed nations and the technology level
would be the similar level in several years.
Further, expected applications and R&D areas extend to aerospace, land transportation,
offshore and civil engineering, and industrial applications. The defense industry is not so
large but expected to grow consistently.
146
Composites in Asia
3.3.1 Introduction
The research on carbon fibre reinforced carbon matrix composite (C/C) in Japan was
started by Yamada in 1971. Yamada, the pioneer of glassy carbon (GC) and
polyacrylonitrile (PAN)-based carbon fibre (CF), has tried to increase the strength of
glassy carbon by applying the high strength CF to GC maintaining the non-gas
permeability. However, he just obtained a different microstructure after high temperature
heat treatment. The authors were asked to investigate the microstructural change of the
composite. This is just the start of the research of C/C composite in Japan. During the
last 25 years, the research of C/C composite progressed steadily and the number of
applications for these materials has increased.
H-II Orbiting Plane program (National Space Development Agency of Japan, Tsukuba,
Japan).
Plasma facing materials for experimental fusion reactor JT-60U - Japan Taurus 60m3
Upgrade (Japan Atomic Energy Research Institute, Ibaragi, Japan).
Automotive ceramic gas turbine project (Petroleum Energy Center, Tokyo, Japan)
Basically graphite crystal shows a strong anisotropy, i.e., elastic modulus, thermal
expansion coefficient, thermal conductivity, etc., which are derived from the bonding
147
Handbook of Polymer Blends and Composites
nature of carbon atom in sp2 (hybrid orbital) as covalent bonding in plane and van der
Waals bonding between the planes. In addition to the crystal anisotropy, fibre
reinforcement of the composite enhances the anisotropy more and more. The main way
to control the properties of C/C composite will be to control the matrix anisotropy. In
this section, the research of C/C composite in Japan using anisotropy control will be
reviewed.
The weakest point of the pitch-based carbon fibre is the compression strength instead of
the high modulus. Tojima and others [21] have analysed the process to increase the
compression strength of the pitch-based carbon fibre. They prepared a pitch with low
molecular weight part insoluble in benzene. The pitch gave a strong preferred orientation
of the pitch molecule, but it gave smaller domain size and crystallite size after
carbonisation. As can be seen in Figure 3.27, the compression strength of the pitch based
carbon fibre follows the line of PAN-based carbon fibre. They succeeded in reaching a
compression strength higher than 900 MPa, with a Youngs modulus of 500 GPa. This
technology can open up the application field of the pitch-based carbon fibre.
The matrix microstructure has a strong relationship with the shear strength of the C/C
composite, and the first way to control the bonding nature of the carbon fibre with the
matrix is by applying 3% ozone to the carbon fibre at 110 C over 8 hours [22]. The
results showed that the strong bonding by the introduction of much amount of functional
group onto the fibre increased the shear strength as shown in Figure 3.28.
However, we can see two stages of the increment of shear strength. We can imagine two
factors controlling the shear strength as strong bonding of the fibre/matrix interface and
making the fine matrix microstructure. The initial increase of the shear strength shown
in Figure 3.28 will depend on the interface control. The increment during the long period
of ozone treatment will come from the chemical reaction between the matrix furan resin
and the functional group introduced on the fibre which changes the chemical composition
of the matrix resins.
Hirano and others [23] applied hot-pressing technique to obtain high density two-
dimensional C/C composite. After applying the pressure of 10 MPa at 600 C, they
applied 5 MPa up to 2000 C, and obtained the density above 2.00 g/cm3 without fibre
damage. They succeeded in higher tensile strength (>550 MPa) after post heat treatment
at 2000 C.
148
Composites in Asia
Figure 3.27 Compression strength of pitch based CF in comparison with the PAN CF
Figure 3.28 Shear strength of C/C composite as a function of ozone treatment period of CF
149
Handbook of Polymer Blends and Composites
Takayasu and others [24] developed an economical process to produce high density
pitch-based C/C composite by only one cycle process. They used a slurry mixture of raw
pitch coke, phenolic resin and small amount of solvent as a matrix precursor. The density
of the produced composite was higher than 1.73 g/cm3.
Yasuda and others [25] are now developing a novel process to fabricate high density and
fine grained matrix C/C composite by only one cycle process. They apply iodine vapour
to pitch matrix composite at 80 C. After iodine treatment the coal tar pitch (softening
point of 90 C, char yield; 55%) changed non-flow material to high char yield (>90%),
and fine grained (a few m) carbon.
All the previously mentioned processes are effective for obtaining excellent mechanical
properties such as high tensile strength and shear strength.
3.3.3 Properties
To compare the data reported from many organisations, standardisation of the testing
methods is most important. Testing methods for continuous fibre reinforced ceramic
matrix composite are not yet established by the Japanese Industrial Standard (JIS) and
ASTM, however the ASTM is now working on this, and in Japan some organisations
such as the R&D Institute of Metals and Composites for Future Industries (RIMCOF)
have devised their own standards, for tensile strength, bending strength, compression
strength and shear strength. The other testing methods used, are ones established for
FRP, JIS K7078 [26], ASTM D2344 [27] and CEN ENJ 658-6 [28].
Youngs modulus and the tensile strength of the C/C composite increase with the increasing
of the orientation of the graphite layer in the direction of applied stress, density and heat
treatment temperature (HTT, Figure 3.29) because of the large shear deformation and
appearance of the ductility and development of the graphite crystallite.
However there were few data for the compression strength because of the difficulty
of measurement. Hirano and others [29] showed that the compression strength of C/
C composite decreases with increasing HTT. The specimen they used was coal tar
pitch-based carbon fibre carbon composite with 0/90 two-dimensional crossply. As
can be seen in Figure 3.29, the compression strength decreases continuously from
2100 C to 3000 C, instead of the continuous increase of the Youngs modulus. This
was thought to be caused by the development of the graphite crystal which is easily
buckled [30].
150
Composites in Asia
The most exciting property of the C/C composite is its excellent high temperature mechanical
properties such as Youngs modulus and strength. Generally it is believed that these properties
increase with increase the temperature. Until now the increase of the Youngs modulus is
explained by Mrozovski model, which explains the temperature dependence of mechanical
properties by the microcracking depending on the large difference of the thermal shrinkage
between c-axis and the a-axis of the graphite crystal, during cooling of the graphitisation
process. However, Kohyama [31] reported different evidence from the Mrozovski model. He
measured the high temperature Youngs modulus for the C/C composite with different HTT.
The materials they used were unidirectionally reinforced composite with mesophase pitch-
based carbon fibre, and a mixture of green coke and phenolic resin composite. Youngs
modulus was measured by resonant frequency method of the rectangular specimen of 1 x 5
x 80 mm. The results are shown in Figure 3.30, in which the change of the Youngs modulus
normalised to that at room temperature are plotted as a function of elevated temperature.
When the material was heat treated below 1800 C, Youngs modulus increased with increase
of temperature. On the other hand, the material which was heat treated at 2500 C showed
negative temperature dependency. The temperature dependency of that with HTT of 2000 C
was positive but slight dependency and turned over to negative dependency above 925 C.
Kohyama [31] explained that the positive dependency will be the typical feature of elasticity
controlled by entropy like polymer or rubber behaviour. The negative dependency could
be explained by the elasticity controlled by energy.
151
Handbook of Polymer Blends and Composites
The shear strength is the weakest property in C/C composite. The short beam test does
not express the shear strength because the specimen fails by compression at the knife
edge, and reflects the mixed mode stress of shear and compression. There are some data
measured by double notch beam compression test (ASTM D3846) [32].
As mentioned in the fabrication process, the shear strength is strongly affected by the
development of graphite crystallite and the preferred orientation, as can be seen in Figure
3.31 [22].
The shear strength of UD C/C composite derived from thermosetting resin char matrix
decreases with increasing HTT, which is directly related the development of graphite
crystallite. The absolute value of UD C/C composite is between 20 to 30 MPa. On the
other hand, the shear strength of two-dimensional C/C composite is relatively low (<10
MPa) [29]. It will depend on the thermal mismatch between fibre direction, parallel and/
152
Composites in Asia
Figure 3.31 Balanced midterm target and present situation of the mechanical
properties of C/C composite
or perpendicular. In this case the control of the matrix microstructure is important factor
to increase the shear strength of two-dimensional C/C composite.
The other topics to be considered are the report of rising R-curve behaviour and the
target of the high-performance material for severe environments project. Sakai and
others [34] cleared the bridging stress of 900 MPa and the distribution to reach 4 mm as
shown in Figure 3.32.
153
Handbook of Polymer Blends and Composites
Figure 3.32 Change of tensile and compression strength and modulus of pitch based
UD C/C composite in various HTT
The high-performance material for severe environments project, set the mid-term target
of making the C/C composite a well balanced material. In Figure 3.33 tensile modulus
and tensile strength has already cleared up the target, but the present situation of inter
laminar shear strength (ILSS) is far from the target [29]. As mentioned previously, the
key technology to reach the target depend on how the microstructure be controlled.
154
Composites in Asia
free path length is shortened by the vibration of the lattice, then the thermal conductivity
decreases. Then to increase the thermal conductivity, it is useful to increase the crystallite
size of the graphite by high temperature heat treatment. The maximum value of the
graphite is ideally estimated as 5000 W/m.K at -173 C, however, the reported value was
usually lower than that of copper (400 W/m.K). The value of 700 W/m.K was reported
by Mutoh and others [36], recently. The material was hot isotactic pressed (HIP) pitch
fibre/pitch-based UD C/C composite heat treated at 3000 C as shown in Figure 3.34
and Figure 3.35. On the other hand, the conductivity perpendicular to the fibre axis is
fairly lower by two orders of magnitude than that in the parallel.
155
Handbook of Polymer Blends and Composites
Kohgo [39] developed a new process as liquid phase conversion as shown in Figure 3.36.
This process is economical and applicable to large scale material. C/C composite is coated
by Si powder resin slurry and converted to SiC at 1500 C, but some part of Si reacted
with the substrate C/C composite. To avoid the reaction with the substrate, they introduced
carbon powder and SiC powder to the slurry (SiC/C/Si/polymer process). The density of
the coating layer by SiC/C/Si/polymer process was dense for oxidation protection as well
as CVD process. However when SiC coated C/C composite was oxidised at above 1800 C,
bubbling phenomena was observed at the SiC/SiO2 interface.
The target temperature of the high performance materials for severe environments
program is 2000 C.
The researchers are now looking for the suitable materials. One candidate they believe, is
iridium. Iridium is inert against carbon and can act as a barrier against oxygen diffusion, but
is very weak against Si because iridium and Si make an intermetallic compound of low
156
Composites in Asia
melting temperature. Then iridium cannot be applied to the SiC coating system. Sekigawa
and others [40] measured the oxidation behaviour of the iridium coated carbon by CVD
method using organo-metallic compounds at 2000 C. However, oxygen penetrates from the
defect area of the iridium coating and the reaction products of carbon to utilise the iridium
diffusion barrier, many efforts are made for multi-coating. Kosaka and others [41] tried to
use Al2O3 and SrZrO3, and Echigoya [42] applied radio frequency (RF) sputter coating of
Al2O3 on iridium, but the oxide layer on iridium included some pin holes. Sugawara and
others [43] proposed a new concept of multi-layer as shown in Figure 3.37. The reaction of
HfO2 and Al2O3 or SiO2 will occur but not so much. The fresh result is expected this year.
Some interesting reports for oxidation protection are by Kobayashi and others [44] and
by Park and others [45]. The former described the effectiveness of lanthanum on carbon
fibre, and the latter the effectiveness of tantalum metal in thermosetting resin derived
carbon matrix composite. However, both of these are effective only for selected systems,
then the continuous research is expected.
157
Handbook of Polymer Blends and Composites
Figure 3.36 Schematic diagram for SiC/C/Si/polymer process for oxidation protection
Key: SiC SiC by conversion process; SiC (CVD) SiC by CVD process with high
density; Ir Ir metal coatuing to protect the oxygen diffusion; HfO2 HfO2 by CVD
process with high density; Al2O3SiO2 - HfO2 (sol gel) Oxide layer by Sol-Gel process
158
Composites in Asia
The phrase carbon alloys originated in 1992 at a meeting of a grant-in-aid for scientific
research. Kobayashi of Nagasaki University (at that time) and Yoshida of National
Research Institute, Kyushu, proposed the word graphite alloys to systematise
polycrystalline graphite-related materials. One of the authors (EY) and some Japanese
researchers discussed the possibility of systematising materials in much wider scope
including all carbon-related ones, and proposed the word of carbon alloys should be
adopted to the name of the new project. In CSJ, an organisation for carbon alloys
research meetings has been established. After deep discussion of the terminology of carbon
alloys the definition has been tentatively decided as: carbon alloys are materials mainly
composed of carbon atoms in multi component systems, in which each component has
physical and/or chemical interaction with each other. Here, carbons with different hybrid
orbital act as different components.
3.3.4.1 The Organisation and the Outline of the Carbon Alloys Project
The project of carbon alloys was accepted as a three-year project, from April 1997 to
March 2000, by the Ministry of Education, Science, Sports and Culture of Japan. The
aim of this project is to do systematic research on carbon alloys.
Space control at atomic and/or crystal levels, Professor Endo, Shinshu University.
Space control at the interface/boundary area and texture, Professor Yasuda, Tokyo
Institute of Technology.
To obtain visible/good results in three years, the project focused on four individual
materials: graphite intercalation related compounds (GIC; Professor Akuzawa), C/C
composites (Professor Kyotani), porous materials and new hybrid systems (Professor
Terai). The major and most important objective is to analyse the carbon-related materials
from the viewpoint of alloys among carbon-atoms and/or hetero-atoms. Another objective
is to discover new carbon-related materials as CBxN, carbine, etc., and new functions of
the alloys. The groups have focused on many types of materials: spl-related, graphite-
related, diamond-related; amorphous, crystalline, molecular and ceramic-based materials.
159
Handbook of Polymer Blends and Composites
3.3.5 Summary
The graphite crystal shows a strong anisotropy, especially for elastic modulus, thermal
expansion, thermal conductivity, etc. In this section the research on C/C composite in
Japan using anisotropy control was reviewed. To realise excellent properties such as
high modulus, high thermal conductivity,etc., in a single crystal of graphite, crystal
orientation is the major factor to be controlled. The important things to consider are:
how can the preferred orientation of the c-axis be controlled and how can the structure
of the fibre and matrix in the perpendicular direction be controlled precisely. The other
is how a suitable mineraliser or catalyser for oxidation prevention in carbon alloy
technology can be developed.
References
5. S. Wadhwa and S. Biswas, Chemical and Engineering World, 1992, 27, 12, 115.
160
Composites in Asia
21. H.Tojima, T.Naito, K.Murakami and T.Hino, Proceedings of the Annual Meeting
of the Carbon Society of Japan, Numazu, Japan, 1993, 94.
22. E.Yasuda, and J.Ariyoshi, Proceedings of the Annual Meeting of the Japanese
Society of Composite Materials, Tokyo, Japan, 1994, 49.
23. K. Hirano, N. Narita, T. Kohno and M.Sato, Proceedings of the 4th Symposium on
High-Performance Materials for Severe Environments, Nagoya, Japan, 1993, 124.
24. J. Takayasu and H. Tsushima, Proceedings of the Annual Meeting of the Carbon
Society of Japan, Numazu, Japan, 1993, 106.
25. E.Yasuda, H.Kajiura and A.Kaiho, Proceedings of the Annual Meeting of the
Carbon Society of Japan, Tokyo, Japan, 1994, 56.
26. JIS K 7078, Testing Method for Apparent Interlaminar Shear Strength of Carbon
Fiber Reinforced Plastics by Three Point Loading Method, 1991.
29. K.Hirano, N.Narita, T.Kohno and M.Sato, Proceedings of the 5th Symposium on
High-Performance Materials for Severe Environments, Nagoyo, Japan, 1994, 241.
161
Handbook of Polymer Blends and Composites
32. D3846-94 Standard Test Method for In-Plane Shear Strength of Reinforced
Plastics, 1994
33. K.Hamada, S.Satoh and A.Kohyama, Proceedings of the Annual Meeting of the
Carbon Society of Japan, Tokyo, Japan, 1994, 142.
34. K.Soh and M.Sakai, Proceedings of the Annual Meeting of the Carbon Society of
Japan, Numazu, Japan, 1993, 270.
39. Y. Kohogo, H. Hatta, Y. Goto, Y. Sawada and T.Yarii, Proceedings of the 19th
Annual Meeting of the Carbon Society of Japan, Tokyo, Japan, 1994,
41. K.Kosaka, H.Okamoto and K.Kishi, Proceedings of the 5th Symposium on High-
Performance Materials for Severe Environments, 1994, 291.
45. S.Park, E.Yasuda, T.Akatsu and Y.Tanabe, Proceedings of the Annual Meeting of
the Carbon Society of Japan, Numazu, Japan, 1993, 190.
162
4
Advances in Wood-based Composites in China
Y. Hua
4.1 Introduction
Wood-based composites are artificial materials that are manufactured from wood elements
blended with resin through forming and hot pressing. The wood elements consist of
lumber, wood strip, veneer, particle, fibre and so on, and the composites normally take
the form of multi-layered symmetric structure, such as plywood, particleboard, fibreboard
and their derivative products. Table 4.1 presents the production capacity of wood-based
composites in China from 1951 to 2010 [1], Table 4.2 gives the main wood-based
composites producing areas and capacities of 1995 and 1996 [2], and Table 4.3 shows
the usage ratio of the composites in different industries.
4.1.1 Classification
There is no unified classification method because the numbers of different types of wood-
based composites. One classification method based on kinds of wood (bamboo) elements
is described as follows [3, 4, 5, 24-30]:
Functional plywood: decorative plywood, plywood for specific use, plywood for
general use.
Structural glulam: glulam, scrimber, parallel veneer strip lumber. Glulam is a glued
laminated timber, glued either straight or curved with the grain of all the pieces
essentially parallel to the longitudinal axis. Scrimber is a kind of wood composite
made from small diameter logs.
163
164
Table 4.2 Main wood-based composite producing areas in China (unit: 1.0 x 104m3)
Area Total Plywood Fibreboard Particleboard Other board
1995 1996 1995 1996 1995 1996 1995 1996 1995 1996
Whole Country 1684.6 1203.26 759.26 490.32 216.4 205.5 435.1 338.28 273.84 169.16
Beijing 5.51 6.46 1.08 0.76 2.40 2.89 1.95 2.0 0.08 0.81
Tianjin 21.76 24.46 9.78 10.61 3.60 3.01 3.75 4.29 4.63 6.55
Hebei 131.57 161.11 79.44 95.16 12.20 13.81 27.41 30.36 12.52 21.78
Shanxi 6.97 6.58 3.11 2.61 0.99 1.39 2.44 2.08 0.43 0.50
Neimenggu 20.0 21.31 3.84 5.31 3.25 3.03 11.35 11.30 1.56 1.67
Liaoning 44.12 23.56 11.35 5.50 11.11 12.09 6.77 4.18 14.89 1.79
Jilin 48.74 60.39 10.44 22.56 6.30 8.14 20.8 22.84 11.20 6.85
Helongjiang 93.19 66.54 39.80 14.94 12.32 10.35 27.93 28.35 13.14 12.90
Table 4.2 Main wood-based composite producing areas in China (unit: 1.0 x 104m3) continued
Area Total Plywood Fibreboard Particleboard Other board
1995 1996 1995 1996 1995 1996 1995 1996 1995 1996
Shanghai 17.33 14.54 10.49 7.86 4.86 4.34 0.66 0.74 1.32 1.60
Jiangsu 180.44 133.54 76.85 62.23 10.49 12.77 33.89 13.22 59.21 45.32
Zhejiang 59.28 40.44 30.51 17.76 13.84 16.84 6.54 5.66 8.39 0.18
Anhui 81.19 36.01 56.62 14.81 2.67 2.85 12.91 10.51 8.99 7.84
Fujian 104.44 73.20 76.32 42.60 13.49 15.86 10.42 9.58 4.21 5.16
Jiangxi 38.81 45.57 20.56 20.27 5.34 5.59 6.98 8.68 5.93 11.03
Shandong 340.75 144.0 158.89 54.28 23.98 12.63 115.62 64.56 42.26 12.53
Henan 121.42 42.89 42.83 9.68 13.81 5.45 25.23 22.85 39.55 4.91
Hubei 101.84 42.67 50.71 12.12 12.06 9.14 21.78 12.67 17.29 8.74
Hunan 37.72 42.60 16.84 12.91 6.88 6.50 12.33 17.74 1.67 5.45
Note: The production of whole country does not include Taiwan province.
Handbook of Polymer Blends and Composites
Common particleboard:
Species: wood, bamboo, and other lignocellulosic fibre.
Binder: organic, inorganic.
Structure: single layer, graduated structure, and multi-layered.
Density: low, medium, and high density.
Functional particleboard
Structural particleboard: oriented strandboard (OSB), wafer board, and oriented
wafer board.
166
Advances in Wood-based Composites in China
Common fibreboard:
Manufacturing process: dry-process, and wet-process fibreboard.
Raw material: wood fibreboard, and non-wood fibreboard.
Density: low, medium and high-density fibreboard.
Binder: organic and inorganic bonded fibreboard.
Overlaid composite: veneer overlaid, impregnated paper overlaid, other material overlaid.
Decorative composite: paint, pigment.
Wood and other material composite: carbon fibre, metal wood composite.
Composite structural lumber: I-beam, composite board and lumber.
The properties of wood-based composites are determined by the final usage of the
products. For instance, the composites are required to have not only good mechanical
properties but also excellent durability if they are to be used as exterior structural material.
They must have a beautiful wood grain if they are to be used decoratively. Decorative or
other functional properties, as well as good fire-retardant performance, are necessary
for the composites used in high buildings.
The general properties are divided into two kinds: appearance and internal properties.
4.1.2.1 Appearance
The main indices to measure the appearance are as follows: dimensions and its tolerance,
warp, characteristic defects (tight knot, dead knot, rot, distortion), manufacturing defects
(overlap, open joint, blister, delamination, imprint), edge straightness, squareness and so
on. The detailed requirements are given in product standards.
167
Handbook of Polymer Blends and Composites
Physical properties
These are moisture content, density, water absorption, thickness swelling, free
formaldehyde emission. The main specifications of internal quality for composites is
shown in Table 4.4 and Table 4.5 [6, 7, 8, 9].
Mechanical properties
168
Advances in Wood-based Composites in China
Durability
The wood element and binder will become deformed as a result of weather changes. For
example, moisture adsorption and desorption, and extremes of heat and cold are experienced
during the service period of wood-based composites. The two elements do not deform
independently of each other, the force of binder, binding them together, results in appearance
of internal stress and damages the glueline and makes it weaker. This finally leads to a
reduction in quality of the physical and mechanical properties and destroys the composite
structure. Moreover, the ageing of the glue curing from the irradiation with ultraviolet rays
169
Handbook of Polymer Blends and Composites
and sunlight will reduce the properties of the composites. The whole process takes place
slowly, and so it takes a long period to measure the relationship between the drop in
quality and the service time, thus many accelerated ageing testing methods have been
developed according to the key weather variables which have an important influence on
the long term service properties of the composites. The methods involved adopt high-low
temperature cycles and high-low moisture content cycles to determine the aging resistance
capability of the composites. The accelerated ageing test used and the remaining properties
specification after the test in China is presented in Table 4.6 [3, 10-12].
Functional performance
Wood-based composites for certain uses have to meet the special requirements, such as being
fire resistant, insect proof or have low formaldehyde emissions, as shown in Table 4.7 [13-16].
Table 4.6 Accelerated ageing test and the remaining property requirement for
wood-based composites
Ply wood Birch
Humid Bamboo
Properties plywood
class class for aircraft OSB resistant mat
I II MDF plywood
use
Bond strength (MPa) after
treatment: boiling 4 h,
drying 20 h at 63 C, 0.7
boiling 4 h
Bond strength (MPa) after
treatment: immersion in
0.7
water for 3 h
Bond strength (MPa) after
1.6
boiling for 1 h
MOR (MPa)
Internal bond strength 1.2
0.10
(MPa) after boiling in water 0.12
for 2 h
MOR (MPa) after
treatment: boiling 3 h,
frozen under 20 C below 50
zero for 24 h, drying at 103
C for 3 h
170
Advances in Wood-based Composites in China
171
Handbook of Polymer Blends and Composites
4.2.1.5 Blockboard
Depending on the core type, blockboard can be divided into two types, solid and hollow
core blockboard. The core of the solid blockboard is made of wood strip jointed in
longitudinal and perpendicular directions, and its two surfaces can be overlaid with two
layers of veneers. The product is used as material for furniture, sewing machines and
vehicles. The core of hollow blockboard is constructed with wood frame and veneer strip,
or paper cellular board. Veneers or thin plywood are overlaid on each surface of the core,
and the board has good sound and thermal insulation properties, and is used widely.
In order to get lumber with high strength in one direction, veneers are primarily laminated
along the grain of the member hence LVL. The material can be substituted for sawn wood
in building applications such as roof and purlin (a framework in a roof to bear loads).
The main categories of wood plastic are manufactured, plastic plywood, wood laminated
plastic, and impregnated compression wood.
Common low density wood is impregnated with resin, then the wood is compressed
and the resin hardened under pressure and heat, and the final product is a kind of
solid wood plastic which is called resin-impregnated compression wood, it is not as
good quality as wood laminated plastic, but it is cheaper.
Composite plywood is a kind of board that uses oriented strandboard, particleboard and
fibreboard as a baseboard, and veneer or bamboo-based material as surface material, and
172
Advances in Wood-based Composites in China
bonds them together. This kind of composite can save high quality wood and raise the recovery
of wood and bamboo up to 80%, and the final product can take place of thick plywood.
4.2.1.9 Glulam
The characteristic defects are removed from dried plank, and then it is jointed into long
planks by finger joint, and spread with PF resin that is cured under ambient temperature,
and bonded together. Glulam can be used as wood beams, especially in the construction
of gymnasiums.
Characteristic defects are removed from the dried plank, and then its veneer mat is
pressed into formed plywood in curved mould by one step. For instance, the back, leg
and face of the chair, and other part of assembled furniture. Also two-layered plywood
can be rolled up into tubes and circles. This not only saves raw material, but also produces
a beautiful moulded product.
The thicker glued plates, which are assembled into a three or multiple layer mat according
to the plywood construction principle, are then pressed to a certain thickness through
high pressure at hot temperature and in this way the product is produced. It is normally
used as vehicle deck.
The product is constructed with dry bamboo strips, which are impregnated with PF
resin, assembled together in a frame and pressed using the cold, hot and cold pressing
method. It is similar to parallel strand lumber.
To manufacture bamboo mat plywood, the first step is to weave a bamboo mat with thin
strips and then glue is spread on it, then assemble a board mat is assembled according to
requirements, finally the board mat is bonded through hot pressing.
173
Handbook of Polymer Blends and Composites
When bamboo curtain is woven with bamboo strip, and laid-up in a common plywood
such as three-layered, five-layered mat, then the mat is bonded under high temperature
and high pressure, a product named bamboo curtain plywood is manufactured.
Strips processed from bamboo are glued together, then as a whole its cut into floor
blocks with tenon tongues and gullets.
4.2.1.16 Strip Ply Bamboo for Bottom Boards of Trucks and Buses
Plywood made of bamboo strips with adhesives, which is widely used as the bottom
boards in trucks and buses on account of its considerable hardness and superior wear-
resistant properties.
Bamboo strips or mats are bonded with resin adhesives and hot-pressed into thick plywood
which mainly acts as models to form concrete boards.
Wood veneers laid up parallel to the grain are glued together with phenol formaldehyde
resin or other bonding agents and hot pressed into plywood specially for textile industry,
e.g., shuttles.
Wood species must be specially selected in avoidance with pungent odors, and
corresponding bonding agents shouldnt do harm to mans health.
Veneers are treated with fire-retardants and then manufactured into plywood which is
characterised by its difficult flammability.
174
Advances in Wood-based Composites in China
The two major parts of plywood manufacturing are veneer manufacturing and bonding.
Veneer manufacturing
Veneer manufacturing can also be divided into two types according to the veneer thickness,
i.e., thin veneer and thick veneer.
The process of thin veneer manufacturing involves, log grading, storing, cross-cutting,
heat treatment, debarking, bolt centering, peeling, sealing and rolling, storing, drying,
clipping and sorting, repairing, jointing and splicing.
The manufacturing difference between thick and thin veneer is that the wet veneer should
be clipped before drying, and the touch drying method is better to ensure that the veneer
is flat and smooth.
Categories of bonding
Two categories of bonding are thin plywood manufacture (three-layered plywood) and
thick plywood manufacture.
The bonding procedure of thin plywood is as follows: preparation of face veneer, back
veneer and core veneer, glue regulating, glue spreading, laying up, prepressing and
repairing, hot pressing, cooling, trimming, sanding, grading, repairing and packing. Two
methods are used to eliminate overlap and open joint (the gap which appears if two
adjacent pieces fail to adhere properly). One is given as follows: laying up the mat after
core veneer spreading, use a knife to cut open the back of the prepressed mat, and remove
the overlap of the core or place a veneer strip in the open joint, and finally hot pressing.
The other method is: first, the core is layered up, after glue is spread on a single side, and
then it is prepressed and repaired, next the glue is spread once more on the single side
and layering up with a face veneer into a three-layered mat, the final procedure is pre-
pressing and hot pressing.
The manufacturing of thick plywood: both whole core or narrow veneer strips are suitable
to act as the core board. If the core board is a whole sheet, then glue spreading, layering
up, prepressing and hot pressing. In the case that the core is a narrow veneer strip, then
the glue spreaded veneer should open assembly layering up, prepressing and hot pressing.
The other procedure is the same as that of thin plywood manufacturing.
175
Handbook of Polymer Blends and Composites
Figure 4.1 to Figure 4.5, taken in a plywood mill in China, show the main procedures of
plywood manufacturing.
176
Advances in Wood-based Composites in China
177
Handbook of Polymer Blends and Composites
Aviation plywood:
A thin dried veneer is layered up with PF resin film, hot pressed, trimmed, graded,
packed and stored.
Marine plywood
A 1-1.5 mm dried veneer is spread with glue or impregnated with PF resin, dried,
layered, hot pressed, trimmed, graded, packed and stored.
A thin dried veneer is impregnated with PF resin, dried, layered up, hot pressed,
graded and inspected, packed and stored.
Solid blockboard
Dried plank is sawn into wood strip, cross-cut, spread with glue on the side, jointing
into the solid core, planed, spread with glue again, layered up with dried veneer, hot
pressed, cooled, trimmed, graded, packed, and stored.
178
Advances in Wood-based Composites in China
The outer layer of bamboo is cut away to make a thin strip, a mat is woven, dried,
resin impregnated glue is spread on to it, the plywood is assembled, hot pressed,
trimmed and then graded.
Bamboo floor
The outer layer of bamboo is cut away and shortened, split into strips, spread with
glue, assembled, layed up, hot pressed, trimmed, tenon processed, graded, packed
and then stored.
The outer layer of bamboo is cut away and split into thin strips, impregnating with
glue, strip dryed, layed up, hot pressed, trimmed, graded, packed and stored.
Wood veneer is dryed, spread with phenol formaldehyde resin, dried, assembled,
layed up, hot pressed, trimmed graded, packed and stored.
Flammable plywood
Dried veneer is impregnated with fire retardant and then dried. Adhesive is mixed
and blended with the fire retardant, this is applied to the dried veneer, spread with
glue and then the plywood is assembled. The plywood is then layed up, hot pressed,
trimmed, graded, packed and then stored.
Physical properties of several types of plywood are illustrated in Table 4.8 [6, 7, 10].
179
Handbook of Polymer Blends and Composites
The specification for mechanical properties is different for different categories of plywood.
Here are several examples: plywood for general use (Table 4.9) [6], plywood for concrete-
form (Table 4.10) [17], plywood for aircraft use (Table 4.11) [10], carrige plywood with
sliced veneer (Table 4.12) [18], solid core blockboard (Table 4.13) [19], laminated wood
plastic (Table 4.14) [7], bamboo plywood (Table 4.15), bamboo mat plywood (Table
4.16) [12], plywood for containers (Table 4.17) [20].
Table 4.9 Bond strength of plywood for general use (unit: MPa)
Items Bond strength of each sample
Species class I, class II class III, class IV
Linden, poplar, China-bells 0.70
Ash, schima, sweet-gum, maple, oak 0.80
0.70
Birch 1.00
Masson pine, Yunnan pine, larch, spruce 0.80
180
Advances in Wood-based Composites in China
Table 4.11 Mechanical properties of Birch plywood for aircraft use (unit: MPa)
Plywood Tensile strength parallel to Shear strength of glue
thickness the grain () line ()
mm Grade 1 Grade 2 Dry Wet
1.0 78.4 66 .2 - 2.0
1.5-4.0 78.4 66.2 2.2 1.8
5.0 73.5 63.7 2.5
6.0 68.6 58.8 2.5 1.6
8.0-12.0 63.7 49 2.7
Note: boiled in boiling water for 1 h.
181
Handbook of Polymer Blends and Composites
182
Advances in Wood-based Composites in China
Table 4.16 The physical and mechanical properties of bamboo mat plywood
Thickness Property Type I Type II
Grade Grade Grade Grade Grade Grade
1 2 3 1 2 3
MC (%) 12 15 12 15
Thin MOR (MPa) 90.0 80.0 60.0 80.0 70.0 50.0
board Remain strength a
60.0 50.0 40.0 50.0 40.0 30.0
2~6mm (MPa)
Free formaldehyde 50
(mg/100 g)
MC (%) 12 15 12 15
Thick MOR (MPa) 90.0 80.0 70.0 90.0 70.0 60.0
board MOE (MPa) 6.0 5.0 1.0 6.0 5.0 1.0
>7mm x103 x103 x103 x103 x103 x103
Remain strength 70.0 60.0 50.0 60.0 50.0 40.0
(MPa)
Impact Toughness 60 50 40 60 50 40
(kJ/m2)
Free formaldehyde 80
(mg/100g)
Remark: a: After boiling-frozen-drying
MC: Moisture content
183
Handbook of Polymer Blends and Composites
The board is generally made from particles, and used for furniture manufacturing.
Common particleboard can be classified into 3 different types as follows:
184
Advances in Wood-based Composites in China
Three-layer particleboard
When formed, the fine particle forms on the two surface layers of the board, and the
rough particle in the core layer, it is clear that there are three layers from the cross
section of the board.
A particleboard with fine particle on the surface and rough particle in the core, and
largest particle forms the centre layer. There is not an evident layer-by-layer structure
in the cross section.
A special particleboard manufactured from long flat strands in which the strand is
orientationally formed, which means the longitudinal direction of the strand in the same
layer is approximately the same. There are three different types of OSB.
An OSB of which the entire strand along the cross section is oriented according to
one direction.
An OSB of which structure is similar to that of three layer plywood, which means
there are three layers obvious across cross section of the board and the oriented
direction of the strands in the same layer are similar, while that of adjacent layers is
perpendicular.
An OSB with fine particle formed on the both surface to make its surface smooth.
185
Handbook of Polymer Blends and Composites
A small diameter log is chipped, put into a wet chip bin for refining, then the particles
produced are dried and sorted into surface or core particles (depending on their size).
Overlarge particles are first further refined into surface or core particles.
186
Advances in Wood-based Composites in China
Surface and core particles are blended individually with resin. Both types of particles are
then mixed, formed, prepressed, hot pressed, trimmed, sanded, graded and packed.
A small diameter log is chipped, put into a wet chip bin for refining, then the particles
produced are dryed and screened into surface or core particles (depending on their size).
Overlarge particles are first futher refined into surface particles or core particles.
Surface particles are blended with resin to form a surface layer and core particles are
blended with resin to form a core layer. Both types of particles are mat prepressed
together, hot pressed, trimmed, cooled, sanded, graded and then packed.
Figure 4.6 and Figure 4.7 show the spreader and forming and hot pressing line of common
particleboard industry, respectively.
187
Handbook of Polymer Blends and Composites
A small diameter log is stored, debarked and then flaked, to produce wet flake. Core
strand and surface strand are dried and then blended with the wet flake.
After orientation of the strands, the mat produced is prepressed, hot pressed, cooled,
trimmed, graded and then packed.
Figure 4.8 and Figure 4.9 illustrate the oriented forming and single-open presser for
producing OSB in one factory in China.
Small diameter logs and wood residue are flaked and refined. The particles are then
screened and sorted into overlarge particles, surface particles (face particles) and core
particles (core particles). The overlarge particle types are further refined until they become
surface or core particles.
Cement and additives are mixed with both types of particles individually, then formed
and cross-cut. Both types of particle are then stacking together to form mats, are cold
pressed, heated, allowed to stand, removed from the mould, trimmed, left to stand at
ambient temperature, treated for resistance to humidity, inspected and then stored.
188
Advances in Wood-based Composites in China
189
Handbook of Polymer Blends and Composites
Particles are prepared by mixing with chemical additives, gypsum and water and
then formed and pressed. After drying and cutting the particleboard is inspected and
then stored.
The marrow is removed from the Bagasse, it is screened and then dried. The particles
obtained are classified into core particles or fine particles for the surface. Overlarge
particles are further refined. Both types of particles are then blending with resin,
formed, hot pressed, cooled, trimmed, sanded, graded and packed.
Broken cotton stalk is transported to the mill where it is stored and then cut and
sorted into flake or waste. The flake is separated, dryed, then screened and classified
into fine particle or core particle. Both types of particle are refined, blended, formed,
hot pressed, cooled, cross-cut, and then stored until ready for use.
Wheat straw is cross-cut and the particles prepared by screening into surface
particles, core particles or dust waste. The surface and the core particles are spread
with glue, formed, prepressed, hot pressed, cut, cooled and then the final product
is ready for use.
Table 4.18 shows the physical and mechanical properties of high-class particleboard
of Grade A according to the China national standard GB4897-92 [8], and Table
4.19 and 4.20 give physical and mechanical properties and sound adsorption
coefficient, respectively, of OSB. Properties of gypsum and cement bonded
190
Advances in Wood-based Composites in China
191
Handbook of Polymer Blends and Composites
192
Advances in Wood-based Composites in China
193
Handbook of Polymer Blends and Composites
A fibreboard is manufactured from fibre by glue spreading, drying, air forming and
hot pressing.
A fibreboard is manufactured from wet fibre that is not dried by forming and hot pressing
or drying. The capacity of the product in China becomes smaller and smaller due to
difficulty and high cost of the treatment of wastewater.
A fibreboard is manufactured by forming and hot pressing, with a density greater than
0.8 g/cm3. If the product made from a wet process or a dry process, then the board is
called wet-process high density fibreboard or dry-process fibreboard, respectively.
A fibreboard of a density between 0.5 and 0.8 g/cm3. Depending on the manufacturing
process, it also has two categories, wet-process medium density fibreboard and dry-
process fibreboard.
A fibreboard with a density of less than 0.4g/cm3. It is commonly called low weight
fibreboard, it can manufactured from both wet-process and dry-process.
194
Advances in Wood-based Composites in China
A fibreboard manufactured from non-wood fibre material, normally this means cotton
stalk, bagasse and other agricultural plant straw.
A fibreboard does not burn but only become carbon of which chip, fibre or mat are
treated with chemicals, or the fibre mat impregnated or spreaded with chemical additives
during manufacturing.
A fibreboard with gypsum as binder and fibre as reinforcement through mixing and
pressing procedure, and it is fire retardant and used for building material.
A fibreboard bonded with PF resin, or treated with oils such as tung oil (oil from Vernicicia
fordii Airyshaw).
An MDF which cannot withstand high humidity and also cannot be impregnated in cool
water for short time.
An MDF that can be impregnated in cool water for short time or service in high humidity
environment.
An MDF that can withstand aging due to the weather, water immersion and the humid
heat action of water vapour in a ventilated environment.
195
Handbook of Polymer Blends and Composites
Four main procedures of dry-process MDF production in one mill in China are present
from Figure 4.10 to Figure 4.13.
196
Advances in Wood-based Composites in China
197
Handbook of Polymer Blends and Composites
198
Table 4.25 Physical and mechanical properties of MDF for interior use
Nominal thickness (mm)
Propeties Unit
1.8-2.5 >2.5-4.0 >4-6 >6-9 >9-12 >12-19 >19-30 >30-45 >45
Special class 0.65 0.65 0.65 0.65 0.60 0.55 0.55 0.50 0.50
Internal bond
Class 1 MPa 0.60 0.60 0.60 0.60 0.55 0.50 0.50 0.45 0.45
strength
Qualified class 0.55 0.55 0.55 0.55 0.50 0.45 0.45 0.45 0.45
MOR MPa 23 23 23 23 22 20 18 17 15
MOE MPa - - 2700 2700 2500 2200 2100 1900 1700
0.70
In order to improve the surface water
resistance, surface wear resistance, and surface
staining of the wood-based composites,
numerous surface decoration processes are
expansion
0.30-0.40
0.35-0.50
0.30-0.38
Linear
0.30
0.30
adopted to treat the surface of the composites,
(%)
of wood-based composites.
swelling (%)
Thickness
6-10
7-8
4-6
9
14-17
22-26
16-20
20
12
0.50-0.60
0.60-0.65
0.45-0.55
0.60-0.65
strength
(MPa)
30-40
32-36
26-30
30-35
25
18
9
9
9
9
Composites
Bamboo
Bagasse
Cotton
Stalk
Reed
200
Advances in Wood-based Composites in China
Overlay: the base composites are prepared by spreading glue, laying up the veneer
and base composites, hot pressing, trimming and sanding.
Overlay: the base composites are prepared by spreading glue, laying up the wet veneer
and composites, hot pressing, mending and sanding.
201
Handbook of Polymer Blends and Composites
Paper is impregnated with resin, dried, clipped, paper and composites are layed up, hot
pressed or cold-hot-cold cycle pressed and then trimmed.
Surface paper, decorative paper and overlay paper are impregnated with melamine, dried
and clipped. The base paper is impregnated with PF resin, dried, and clipped. All the
paper layers are layed up with cold-hot-cold cycle pressing, sanded and the back of the
laminate is roughened. After inspection and grading the laminate is ready for use.
Physical and mechanical properties of decorative melamine laminate are listed in Table
4.27, and physical and mechanical properties of wood-based composites overlaid with
PVC are shown as Table 4.28 [23].
202
Table 4.27 Physical and mechanical properties of decorative melamine laminate
Type of HPLa
Test items Properties Unit
HW HG VG CL P F SE
Resistance to boiling water Appearance Grade 2 2 3 3 3 2 2
Appearance gloss 3 3 - - 3 3 3
Dry heat resistance Grade
Others 2 2 - - 2 2 2
Boll height cm 100 100 100 100 100 100 100
Impact resistance
Hollow diameter Mm 10 10 10 10 10 10 10
Resistance to surface wear Weariness R 1000 350 150 - 350 350 350
Resistance to surface staining Appearance Grade 2 2 2 3 3 2 2
Resistance to cigarette burns Appearance Grade 2 2 - - 2 2 2
T<2mm 55 45
Boll height: a standard method to measure the impact resistance property of materials
Handbook of Polymer Blends and Composites
Acknowledgements
The author would like to thank colleagues from Nanjing Forestry University, Mr.
Dongrong Shi in the Foreign Affairs Office for checking the English manuscripts and
Professor Yonglan Xu in College of Wood Science and Technology for offering the valuable
photographs. Special acknowledgments are due to publishing office of China Forest
Product Industry Journal and China Wood Industry Journal, China Forest Publishing
House and China National Standardisation Institution of Wood-based Panel for providing
abundant references and information.
References
204
Advances in Wood-based Composites in China
6. Plywood for General Use, National Standard of the Peoples Republic of China,
GB9846-88.
10. Birch Plywood for Aircraft Use, Industry Standard of the Peoples Republic of
China, ZB B 70005-87.
13. Code for Fire Prevention in Design of Interior Decoration of Buildings, National
Standard of Peoples Republic of China, GB 50222-95.
14. Test Method for Fire Resistance of Building Materials, National Standard of
Peoples Republic of China, GB 8625-88.
16. Test Methods of Evaluating the Properties of Wood-Based Panels and Surface
Decorated Wood-Based Panels, National Standard of Peoples Republic of China,
GB/T 17657-1999.
17. The Plywood for Concrete-Form, Industry Standard of the Peoples Republic of
China, ZB B 70006-86.
18. Carriage Plywood with Sliced Veneer, Industry Standard of the Peoples Republic
of China, ZB B 70001-85.
205
Handbook of Polymer Blends and Composites
19. Blockboard, National Standard of the Peoples Republic of China, GB/T 5849-
1999.
20. Plywood for Containers, National Standard of the Peoples Republic of China,
GB 1938-91.
22. Wood-based Panel Decoration, Ed., L. Longhe, China Forest Publishing House,
Beijing, 1991.
23. Decorative Wood-Based Panel with Polyvinyl Chloride Film, National Standard
of the Peoples Republic of China, GB/T 1279-1998.
25. Plywood for Tea Packing Box, Department Standard of the Peoples Republic of
China, LY/T 1170-95,.
26. Wood Laminated Board for Textile Industry, Department Standard of the
Peoples Republic of China, LY/T 1416-1999,.
28. Strip Plybamboo for Bottom Boards of Trucks and Buses, Department Standard
of the Peoples Republic of China, LY/T 1575-2000,.
29. Bamboo Floor, Department Standard of the Peoples Republic of China, LY/
T1573-2000.
30. Composite of Cement and Wood Particle, Department Standard of the Peoples
Republic of China, JC 411-91.
206
5
Overview of the Use of Composites Worldwide
The reinforced plastics industry essentially began in 1940 with the introduction of
glass fibres for reinforecement. Early attempts to fabricate glass-fibre reinforced
phenolics and melamines by high pressure moulding did not show much commercial
promise. Glass-fibre reinforced silicones prepared in 1941 by Hyde at the Corning
Glass Works were outstanding as heat resistant electrical insulation, but were too
costly for structural applications. The first glass-fibre-reinforced polyester resins were
prepared with an allylglycol carbonate resin by King in February 1941 at the Columbia
Chemicals Division of the Pittsburg Plate Glass Company. By 1942 low-pressure
curing polyester resins derived from maleic anhydride were available, and by 1944
they were used to turn out millions of helmet liners and thousands of parts for aircraft
and boats for the armed forces in World War II. The appearance of epoxy resins in
1939 and their application in the early 1950s gave a major boost to performance of
glass-reinforced plastics.
In 1951 Bradt [2] found that glass fibre reinforced polystyrene (PS) and others material
specifications for military applications and a company, Fibrefil Inc., was formed as the
first supplier for reinforced thermoplastics. At about the same time (1960) Nylon 66 was
used as a moulding material in the USA and was the first engineering plastic. In 1960 ICI
in the UK, Bayer in Germany and the LNP Corporation in the United States developed
new grades of GR (short glass fibre) nylon and Union Carbide developed carbon fibres
207
Handbook of Polymer Blends and Composites
By 1970 practically all structural plastics were available with reinforcement and sheet moulding
compounds (SMC) and bulk moulding compounds (BMC) have been also introduced.
The chronology of reinforced plastics composites has been reviewed by Rosato [3],
Seymour [4, 5] and others [5-13].
In 1994 and 1997, at the European Days of Composite Materials at Paris (Journes
Europennes des Composites), the manufacturers of five continents exposed their products
and developments in technologies. An estimation of the composites production and
consumption was made mainly by Vetrotex and the Society for the Advancement of
Materials and Process Engineering (SAMPE) [14, 15]. Each continent has its own
characteristics regarding materials, technologies and markets. The worldwide market of
polymer composites (thermosets and thermoplastics) was estimated as 3.24 million tonnes/
1997 and about 5.5 milllion tonnes/1998 and reached 6 million tonnes in 2000, with the
following distribution: 48% USA and 2% for South America, 27% Europe and 23%
Asia with an increase of 3.7% compared to 1996 and a global increase between 1993
1997 of 7.2% and by continent 5.7% North America, 11% South America, 11.6% Asia
(Japan has a smaller increase) and 5.8% for Europe. The worldwide increase is 45% in
1998 with a higher increase for thermoplastics (USA: 4.5 kg/head > Europe: 3.5 kg/head
> Asia: 0.3 kg/head) than thermosets. The European production of composites is 1.7
million tonnes from which 65% thermosets, with 32% Germany and Austria, 18% Italy
and 16% France.
The glass fibres (GF) are mainly produced (80%) by Vetrotex, Owens Corning and PPG.
A small decrease of GF production occurred in 1996 and 1997 and then production
increased again. The composites consumption at European level is 910,000 tonnes
(210,000 by France) with an increase of 12% in 1996-1998. Although their average
increase was only 2.5%, the thermosets are the materials with the largest volume utilised.
The increase of thermoplastics use is 8% and 3.1% for textiles. In USA, the consumption
of composites was of the order of 1.5 million tonnes in 1997 with increases of 6.1% in
1996, 3% in 1997 and 2% in 1998.
It is expected that there will be an important increase in 2001 for thermoplastic composites
of 79% while that of thermosets will be 5% per year. Between 19911998, the
thermoplastic composites exceeded 2.5 times the production of thermoset composites,
which was an especially remarkable increase in the automotive industry in the last years.
208
Overview of the Use of Composites Worldwide
The European market for SMC/BMC (26%) in 1998 was estimated at 256,000 tonnes
(26% in automobiles and resin transfer moulding resin injection moulding (RTM
RIM) (11%) from 800,000 tonnes worldwide market. SMC and BMC recorded an
increase between 1991-1998 in the USA especially in automotive industry. In Europe
this sector did not record progress during this period, but a significant development was
recorded in the electrical industry, the sanitary sector and the industrial sectors.
The major market for US composites is currently transportation and a 20% rise to over
108,000 tonnes in the use of SMC in automotive applications was projected for 1996 an
increase of 63% by the end of the year 2000 [16].
The major suppliers of reinforced thermoplastics are located in USA, Europe and Japan.
The ability of the plastic industry to buck the trend and outperform the economy in
good and bad times has always been due largely to its ability to substitute for alternative
materials and process technology. In this respect the use of additives to enhance polymer
properties for a wide variety of applications has played an increasing role.
The US funds for composite research programmes are very high to create new materials.
In Japan, government funds are directed to universities and national institutions primarily
for basic research. The Ministry of Education supports university research and the Science
and Technology Agency research budgets go to material research institutes. The Ministry
of International Trade and Industry funds both national research institutes and private
sector firms and associations for developing composite materials.
Governments within the EC are mobilising R&D efforts to meet US and Japanese competition.
EC-sponsored programmes such as Brite, Eureka and Euram including that for advanced
polymer composites (APC). To be considered for a variety of industrial applications, improved
performance must be accompanied by lower cost compared to competitive alternatives. Low
costs can be obtained by three ways: automated fabrication techniques, lower cost fibres and
thermoplastic resins. Significant R&D efforts are particularly made in US to develop high-
temperature thermoplastic resins for advanced polymer composites. These together with
rapid fabrication techniques such as stamping will take the leading role in the development
of composites in the USA, while in Japan and Western Europe, non-military government
agencies play a much more significant role in funding APC research.
In the US, Amoco and DuPont offer prepregs and finished shapes. Many European
companies have integrated via acquisition of US firms. ICI purchased Fiberite (US) and
209
Handbook of Polymer Blends and Composites
In Asia the composites consumption was 775,000 tonnes in 1997. This consumption
was represented by 47% thermosets, 22% thermoplastics and 31% textiles. Electronic
applications represent 31% with an important export to Europe and USA. The Asiatic
market is dominated by Japan (51%) with marine applications of 55% then Korea (11%),
Taiwan (18%), China (10%) and Asian countries (Singapore, Malaysia, Thailand and
so on). An important increase is expected in the next years in Asian countries. By
application [14, 15] the following order can be written: (46%) automotive > railway
transportation > marine > electrical/electronics (21%). Out of the 450,000 tonnes used
in automotive industry 42% are thermoplastics while engineering plastics with 59% PA.
Polypropylene (PP), (polybutylene terephthalate/polyethylene terephthalate (PBT/PET)
and polycarbonate (PC) also have a good position. The PPG company who manufacture
glass reinforced plastics (GRP) estimate that the production of GRP is 5.3 kg per person
in the USA and 3.3 kg per person in Europe [15, 17, 18].
Fast cure systems and plant automation are being re-evaluated to get greater production
without significant investment. That is why moulders such as, The Budd Company (USA),
Bailey Transportation Products (USA) and Columbus Plastics (part of Core Materials,
USA) expanded the capacity for closed moulding (SMC) [19].
SMC is being replaced by TSR in manufacturing of Volvo and Mercedes. TSR known
many technologies from their discovery such as Hybrid Thermoplastic Composites (HTPC)
or Plastic Omnium dExperss. HTPC consists of a combination of a tissue of GF and
plastic. They are developed for PP and now new HTPC-based PA and PP are developing.
Plastic Omnium considers that complementary techniques are necessary to increase the
performance. Appryl Composites have discovered the advantage of the long fibres. The
PA with 30% long GF exhibit equivalent properties and a density 10% lower than PA
reinforced with 40% short GF. Appryl Composites estimates that long fibres reinforced
thermoplastics use have an increase of 1520% in Europe, they will commercialise
210
Overview of the Use of Composites Worldwide
aesthetic long PP-fibres for use in automotive interiors and have also made progress in
long fibre reinforced polyethylene and PA/PP blends.
The industry buzz word at the moment is Composite Technology Transfer. For example,
until now composite bicycle wheels were constructed using aerospace technologies.
Vacuum bagging, hand lay-up of prepreg materials and a high temperature oven is
generally utilised [20].
The engineering polymer industry goes for globalisation. GR compounds have been
estimated to account for around half of all thermoplastic usage and showing higher
growth rates than unfilled material. Improved glass coupling technology and long fibre
reinforcement (ICI Vitron technology) will increase the performance levels [19, 21]. Factors
responsible for the continuous increase include continuing economic growth created by
high consumer confidence and a diversification into non-traditional applications. Many
of the current resins suppliers are nearing the limits of their capacity, resulting in expansion
and consolidation plans by key resin suppliers. Transportation accounts for about one-
third of all composites shipped. This was largely fuelled by the continuing consumer
demand for cars and trucks (light and heavy) [19].
The SCRIMP process is applied to polyesters, vinyl esters and modified acrylic resins.
Among the three technologies identified by the National Science Council of Canada as
promising for developmental work: composites, dosage barrier and membrane technology,
the composites rank the first [22].
Thermoplastics are of particular interest because of their toughness, suitability for high
volume production, automated production techniques, ease of storage and low quality
control costs but they should also have dimensional stability, high temperature resistance
up to 220250 C and improved chemical resistance. New engineering thermoplastic prepreg
forms have been developed for easier processing. These are available as yarn, cloth, felt,
tapes and paper. Fibres also used include polymer fibres such as: polyphenylene sulphide
(PPS), PA, polyimide (PI) and polyetherether ketone (PEEK). Application of heat and pressure
causes the thermoplastic to melt and coat the reinforcing fibre. Thermoset-based prepregs
211
Handbook of Polymer Blends and Composites
Increased automation is necessary to bring advanced composite costs down for large and
complex part fabrication. Computerised filament winding, tape placement and pulforming,
high-speed pin winding and process hybridisation are promising techniques in high volume
production. The greatest market potential for advanced composites exists where their
light weight, strength and part consolidation are key characteristics. The automotive
industry is a giant market, also proving useful in bridge construction (Bayer AG, US
Navy) and aerospace applications (Boeing).
The total plastic production in Japan in 1998 had a 78% reduction compared with
1997 due to the financial troubles in Asian countries. The demand for high temperature
engineering plastics such as: PPS, liquid crystalline polymer (LCP), polysulphone (PSF),
polyether sulphone (PES) and PEEK is growing for electronic components like
semiconductor and manufacturing tools. LCP increases its usage in computer or other
components. Sumitomo Chemical will expand its PES production and also PPS, PEEK
and fluoropolymers production.
Investment and capital participation by Japanese companies in Europe have risen very
significantly in recent years because of increased business opportunities, as a logical step
in the globalisation of multinational businesses of potential protectionist trade policies
under the unified European Community.
212
Overview of the Use of Composites Worldwide
All these activities have had a significant influence on the demand for polymeric products.
Initial investment was primarily by moulders specialised in the requirements of the
electronics sector such as Sanko Gosei. The purchase of one of the largest UK moulding
companies, Birkbys Ltd in 1990 by Marubeni, established major bridgeheads in the
non-Japanese automotive and business machinery markets.
Eastern Europe could well prove to be the region with the worlds highest growth rates
for plastics, although several rigid elements of the system will continue to exist in the
short term. Plastic demand for the East European auto industry, consumer electronics,
domestic appliances, packaging and building products will also increase.
US companies (Schulman, Ferro, Ampacet, LNP, Wilson Fibrefill, Cabot, Akzo, Ashland
Plastics, Cookson) are involved directly in the European compounding industry since
the Second World War, also mainly in the UK.
BASF can sustain a position in high temperature thermoplastics while ICI and Phillips
have been forced to withdraw. By the same argument the acquisition of Akzos engineering
plastics business including Wilson Fibrefill by DSM should place DSM in a better position
to compete for business in this market with Bayer, DuPont and GE.
Vinyl esters have been used in structural formulations because they provide a strong heat
corrosion resistant matrix but they are expensive and are difficult to thicken. Alpha
Owens Corning claims that a polyester/vinyl ester hybrid with a saturated polyester
thermoplastic additive can be chemically thickened and retain its mechanical properties
after heat aging.
In the recent years the Society of the Plastics Industry (SPI) Composites Institute developed
premanufactured composite carbon tubes (Carbon Sgell System) which are either fully
or partially filled with concrete (depending on the strength, stiffness and stability required)
for short and medium span bridges. The carbon shells have a dual function of stay-in-
place framework and reinforcement for standard elements.
213
Handbook of Polymer Blends and Composites
Wooden railway sleepers can also be strengthened and protected by glass fibres reinforced
composites based on epoxy and resorcinol formaldehyde resins (project carried out by West
Virginia University, PPG industry, Indspec Chemical Corp and Lockheed Martin Aeronautics)
Advanced polymer composites (APC) have moved from the development phase to
commercialisation for many applications in the aircraft/aerospace, industrial and
recreational markets [23]. APCs exhibit two key components, matrix and continuous
fibre reinforcement. Composites with chopped fibre reinforcements are not generally
considered advanced. Carbon, S-2 type glass and aramid fibres are the most common
reinforcements. Specialised fibres include boron, ceramics, metals and extended chain
PE while for the matrix, thermosets (epoxy, maleimide and polyimides) and thermoplastics
(polyamide-imide; PEEK; polyetherimides (PEI), PES, PPS) are used.
The international market from APC is huge and growing rapidly $1.8 billion in 1986
with an 11% annual growth rate. It is a dynamic and highly competitive industry, see
Figure 5.1 and Table 5.1.
214
Overview of the Use of Composites Worldwide
to the pursuit of technological supremacy. Even though demand from the Military fell,
there has been a slow but steady rise in industrial applications, especially in the automotive
and sports good sectors [24].
Some 30 polymers have been used as base polymers in commercial reinforced thermoplastic
materials mainly classified as thermoplastics (PE, PP, PS, polyvinyl chloride (PVC), PA,
PBT/PET, polyphenylene ether (PPE), polyurethane (PU) and others and thermosets
(phenolics, epoxy, unsaturated polyesters, others).
5.2.3.1 Thermoplastics
Thermoplastic composites are generally considered for low cost applications in various
material configurations. PA 6, PA 66, PA 12 and thermoplastic polyurethanes (PU) are
available with long fibre carbon reinforcement, as injection moulding materials, or as
advanced material product forms such as tapes and tow. Long fibre reinforced
thermoplastic materials offer unprecedented metal replacement opportunities. Their
mechanical properties such as, improved tensile creep resistance, flexural strength and
modulus at elevated temperature, impact strength and dimensional stability have been
used in many proven applications. The registered mark of Polymer Composites, Inc. (a
subsidiary of Hoechst Celanese Corporation, USA), are Celstran R and Fiberod R for a
215
Handbook of Polymer Blends and Composites
number of engineering resins as matrices and for various reinforcing fibres as long carbon.
Long carbon fibre reinforced engineering polymer (11 mm long) are a material class that
provide easy processing at high volume, utilising injection moulding equipment.
The market for particulate (mineral) additives (fillers) and glass fibre reinforced
thermoplastics of all types in moulding applications is steadily developing particularly in
automotive and domestic applications. The market is dominated by polypropylene and
polyamides (more than 90%)
The trend is the development of hybrid grades incorporationg both particulate and
fibres as suppliers seek to more closely match the performance requirements in key
applications [25, 26].
Polyamides
At present, polyamides have a leading role among the thermoplastic use in composite
materials. The mineral reinforced PA is continually expended by new metal replacement
applications in the automotive industry such as wheels for cars.
The main producers of PA in Europe are Allied Signal, BASF, Bayer, DSM, DuPont, EMS
and Nyltech. Hls (Germany) developed 15% carbon fibre reinforced PA-6 and PA-12
materials [27]. It is a tendency to concentrate the production. Atochem is leader for PA-
12 and PA-11 and Nyltech is the next highest worldwide producer for PA 66. The
worldwide market exceeded 1 million tonnes of engineering PA and reached 4.25 million
tonnes of PA fibres. PA 66 and PA 6 are mainly used as composites loaded with 30 - 50%
GF replacing the metals. They present a good recyclability [28].
PolyolefinsPolypropylene
The area of thermoplastic matrix composites is one of the most promising and challenging
fields for future applications of PP. The main driving factor is the substantial strength
and stiffness improvement achieved by a glass-fibre reinforcement which elevates the
material properties above those of other common thermoplastics into the region occupied
by metals as the price seems to be competitive (see Figure 5.2) [29].
Environmental factors are a strong additional driving force, recyclability being seen as
the issue of prime importance. PP-based materials are in a unique position in this respect,
the polymer itself is readily recyclable and used especially in automotive industry.
In Europe there is a tendency to restrict the use of polymers in cars to either one material or to
a limited range of recyclable polymers. Recyclability is beginning to become a strong negative
factor for thermosetting composites, for which non-economic solutions of recyclung are found.
216
Overview of the Use of Composites Worldwide
PP is the ideal material of choice for a single material design concept because it can be
made to fulfil all varying mechanical requirements of the structure. Although PP was the
last plastic to find a niche in the engine environment, today it is the most widely used as
a front end support replacing metal, chair structure, body panels, car dashboard. Whether
a rigid, load bearing support is required, a softer impact cushioning material or even the
smoothly texture, tactile outer skin PP could be a material of choice.
A higher- tech derivative of the glass mat thermoplastics (GMT), the new class of material
is referred to as structural thermoplastic composite or STC. Exxon was one of the first
companies to commercialise the technology STC. The ingredients are water-based slurry
particles containing PP particles or other thermoplastic and glass fibres (or other fillers).
Over the last few years reinforced PP has substituted for acetal in jug kettles, for PBT in
toaster components, for polyamides in power tool housing, for ABS in a variety of
applications including vacuum cleaner housings.
Critical to many of these applications is the degree of gloss and surface finish which can be
obtained with mineral reinforced compounds. The developments in this area are pursued
by all suppliers. The greatest potential for white goods lies with laundry (about 70 parts)
and dishwashers representing the most interesting areas for development. Most of these
217
Handbook of Polymer Blends and Composites
PP is the polymer in the best position to become the dominant material in the market
[29, 30].
About 30 kg of PP is used in each vehicle this means that 25 - 40 wt% of material in each
vehicle is plastic, so in ten years PP became the most useful plastic in the automotive industry.
The real consumption varies from one country to another, as in France and Italy it is of 35
kg, while in Germany and the Netherlands it is 25 kg. It is mainly used in interiors of vehicles
but also in exterior parts and is the base polymer for 30% of electrical/electronic components
and for 10% of the vehicle chassis. At the european level the consumption varies between
450,0000520,000 tonnes plus 80,000 tonnes for battery case.
PP has replaced ABS and PVC as the material of choice for interior vehicles therefore the
consumption of ABS and PVC decreased to 14% and 19%, respectively. In order to
assure impact resistance, thermal resistance, etc, propylene copolymer filled talc have
been used, which are competitive with high crystallinity PP (HCPP). HCPP with a degree
of crystallinity of 70% exhibits a similar rigidity with PP filled with 20% talc, offering
also better quality and lower density being interesting from point of view of weight
reduction. The replacement of other polymers with PP or copolymers low-loaded or
modified with elastomers such as ethylene-propylene-diene monomer is a new research
development in automotive industry because a monocomponent is easily recyclable. Filled
(10-20 wt%) HCPP assures a good dimensional stability.
A PP-grade is also being considered for vehicle wings as a replacement for steel assuring
also much closer tolerance in dimensions (3-4 mm). However, the PP consumption in
cars seems to reach a plateau, due to the competition with PA and SMC especially related
to thermal content (see Figure 5.3) [29].
The increase in use of PP in the next few years will be of 6-7% per year. PP composites as the
GMT and STC are used in synthetic paper manufacture and automotive industry. Composites
reinforced with ultra strong PE fibres are more resistant and delaminate owing to the materials
superior energy absorption properties when used in conjunction with other fibres such as
glass or carbon properties of the composites will be suitable for more demanding application
[31]. High density polyethylene (HDPE) rods are effective reinforcement in concrete. Compared
with steel, they are much lighter and do not rust [32, 33].
218
Overview of the Use of Composites Worldwide
Areas where the progress is necessary are thermoplastic pultrusion and filament winding
for sectional and pipe-like parts; low pressure lamination technology for large area
mouldings; the stamping of GMT [32, 33], etc.
In the USA, the use of the commodity plastics as PS, PE and quasi-engineering plastics ABS,
styrene acrylonitrile (SAN) and PP in reinforced form grew rapidly in importance due to
their utilisation in the automotive industry. In Europe these materials did not get the same
importance, but Nylon in reinforced form was widely exploited and, on a much smaller
volume level. PC became the next most useful reinforced plastics. In the USA, with the
exception of the economic recession of 1970, the use of these materials have been growing at
the rate of over 20% per year. The polyolefins become the largest tonnage group. The
automotive industry is of prime importance accounting for over 50% of the total reinforced
thermoplastic materials used. Domestic appliances and business machines rank second.
PEEK
PEEK is used where a very high temperature resistance is required [34]. Traditionally,
the aerospace composite market dominated the composite industry with new thermoplastic
matrix technology development such as PEEK. These high temperature matrices are
generally costly ($200/kg), and thus have limited use for non-aerospace applications
and limited market acceptance.
219
Handbook of Polymer Blends and Composites
Carbon fibre reinforced PEEK was developed for biomedical implants by ICI Advanced
Materials. Light weight, corrosion resistant polymer composites have potential for in
vivo use, particularly in structural applications such as fracture fixation plates and hip-
joint replacements.
Thermoplastic matrices are considered more inert in the human body than thermosets. PEEK-
based composites are attractive because they are biocompatible and corrosion resistant [35].
PEEK is also produced as non-reinforced and glass and carbon fibre reinforced grades.
The ultra-pure grades for contact with ultrapure water are also of interest.
5.2.3.2 Thermosets
The thermosets are generally used as reinforced glass fibres and their use are growing. In
1992, the consumption was 348,342 tonnes, while after only two years the consumption
reached 1.16 million tonnes in Europe with 196,000 tonnes in France [36, 37].
Phenolic Resins
The history of thermosets began in 1909 with the first synthesis of a phenoplast by
Baeckeland. Then, the phenoplasts have found applications in electrical, telephone, radio,
automobile, structured materials, etc. The competition with other plastics was great, but
in the present day they regained their position (with a small increase of 1% in 1996) due
to their properties. Phenolic materials excel in fire performance because of their low
ignitability, spread and flame, heat release, low smoke and toxic gas emissions and an
excellent performance:price ratio. Among their applications are: aeronautics,
transportation, etc. The potential new markets will be open in southern Europe [38].
They are used to develop parts for road and air transport, building and safety equipment.
Some developments of phenolic thermosets are given in Table 5.2.
Aminoplasts
Aminoplasts were discovered in 1918 and they were applied very soon for electric items,
communication being competitive with phenolics. They have a good surface purity, are
bright and easily cleaned particularly appreciated. Worldwide consumption is 2.5 million
tonnes with 200,000 in France.
220
Overview of the Use of Composites Worldwide
221
Handbook of Polymer Blends and Composites
Epoxy Resins
Epoxy resins were introduced in Europe in 1946 by Ciba. Their main applications are,
paints, adhesives, moulded and structural parts for mechanical, electrical and electronic
industry and if reinforced with aramid, carbon or boron fibres can be used in automobiles,
aeronautics and aerospace. They are crosslinkable at 120 C. Epoxy resins are in the
forefront of development of advanced composites for aircraft and aerospace applications
to replace metal with an increase in use of 7% annually. A Boeing project is expected to
consume ~ 21,000 kg carbon fibre reinforced epoxy [38].
In advanced composites, Dow occupied a good place especially with epoxy-based products.
Galt composites, UK, extended its manufacturing capacity in GRP (glass reinforced
plastics) sheets and moulding [39]. The worldwide market of epoxy resins is higher than
200 thousands of tonnes, both in USA and Europe. Germany has the highest consumption
of epoxy resin in Europe [36].
Increase of consumption during 1988/1993 was 3.2% for phenolics and 4% for epoxy
resins and the average increase was 5% in 1996.
Unsaturated Polyesters
Unsaturated polyesters were discovered in 1894 and have been used in paints, then were
used in shipping and now are used as important element in automotive industry. They
are well-known as SMC (impregnated resins) and BMC, etc. They have a huge
consumption.
Evolution of East Europe consumption of polyester resins in automobiles as SMC and BMC is:
Alpha/Owens Corning developed a lot of grades of low density foam from unsaturated
polyesters (UP) for applications in boat construction and also for the automotive market
for a Ford truck engine and under-the-bonnet applications [40, 41].
Another new series of resins offers the flexibility required for vertical panels which need
good impact properties such as wings or bumper beams.
222
Overview of the Use of Composites Worldwide
UP production is a core business for Reichhold which has plans to expand in Europe and
South America (Mexico and Venezuela) with resin and gel-coat plants. Restrictive
environmental legislation related to toxicity of styrene created some problems.
Polyurethanes
Polyurethanes were developed by Bayer for thermal and acoustic insulating materials. In
1950, Bayer presented the first elastomeric PU and in 1959, France had the first industrial
unit. The production of RIM was 3.1 million tonnes in 1991 and 3.7 million tonnes in
1996, an increase of 3.3%.
Polyimides
Polyimides are also used due to their long-term high temperature resistance.
Silicones
Silicones were discovered in 1899 and the industrial production began in 1943 by
Corning Glass Work (USA) and Dow Chemicals. Today, the worldwide market is
dominated by Dow Corning, General Electric and Rhone-Poulenc. In 1991 the
worldwide market was 650,000 tonnes and in 1996, 900,000 tonnes with an increase
of 7.3% per year; the increase is explained by their penetration in new sectors such as:
electric, electronics, informatics, the automotive industry and recently in the medical
sector. The excellent thermal resistance, non-toxicity, anti-adherent properties are the
reasons for these variable applications and their competitive application with
thermoplastic elastomers.
5.2.4.1 Reinforcements
The consumption of fillers and reinforcements rose by 5.8% per year while for reinforcing
fibres it is expected to be between 7 and 9% per year for the last ten years.
223
Handbook of Polymer Blends and Composites
Many suppliers of fillers and reinforcements have added capacity in the past years to
meet growing worldwide demand and they are facing challenges to meet higher
performance requirements.
Vetrotex introduced a process for commingling continuous roving and fabric and two
reinforcement products for the growing pultrusion market [42]. New facilities have been
opened. All capacities for technical fibres will increase [1]. About 30 companies produced
reinforced fibres, the main ones are: Vetrotex (St Gobain), DuPont for aramid fibres,
GF-Owens Corning, Toho Rayon and Hercules for carbon fibres, Enka, Hoechst, Bayer
and Rhodia and 40 other companies produced structural composites [43, 44].
PPG Industrie (USA) introduced several new fibre glass products including MatVantage-
P, continuous strand mat designed specifically for use in pultrusion moulding. The
materials softness and drapeability make it suitable for complex pultruded shapes.
Hyban 2788 direct draw rowing is used for low flame, smoke parts of reinforced
phenolics by filament winding or pultrusion and Hyban 715 for pultrusion with epoxy
resin. F-type composites (long fibre) show superior surface properties. Applications
are under body panels, engine noise shields, heat shields, battery trays, and seat
components, and so on [45].
The European market for GF as reinforcements was 500,000 tonnes in 1996. A leader in
their production is Owens Corning which has 30 plants and is expected to expand in the
Asia/Pacific region. The glass fibre reinforced composites represent 50% of total composite
production of East Europe [46]. The major problem encountered in PP composites is the
high viscosity of polymers. This limits the quantity of glass used in short fibre injection,
surface finish is damaged when the glass content is increased, productivity is hindered,
impregnation of direct roving is difficult and the flow of the GMT prepreg is impeded.
The viscosity could be reduced by using a processing aid such as partly hydrogenated
polyphenyl (3%), that improves the flow and produces a better surface gloss [47].
GMT thermoplastics look set for substantial growth in Europe [47]. The Volkswagen
Golf front end alone will account for some 5,000 tonnes per year of material and reached
20,000 tonnes in 1995. GE Plastics and EKT (Worms-Germany) increased their capacities,
the latter company having 40% of European market for GMT. Symalit (Shell International
Chemical Switzerland) accounts for another 40% of the market. EKT-GMT is an
elastopreg (PP) matrix reinforced with 2045% glass mat as long or endless fibres. The
fibres can be either completely random if isotropic properties are required, or oriented
enabling anisotropic material properties.
224
Overview of the Use of Composites Worldwide
Europes first completely automated production line for GMT (Worms, August 1990)
produces the front seat shells for the BMW 3 Series with different designs (driver and
passenger seats). It has high volume production of painted GMT, PP parts. The frontal
structure of the Ford Taunus/Mercury Sable is manufactured by SMC containing 30%
GF. Some applications of GMT are given in Table 5.3.
GMT is lighter than metal and offers the opportunity for parts consolidation. GMT offers
a designer considerable freeedom permiting significant integration of functions. Head lights,
radiator and fan are held in place by a single part avoiding costly fastening and assembly
steps. The Volkswagen Golfs composite front end is more than 35% lighter than its metal
equivalent. GMT improved productivity compared with thermoset composites such as
SMC, the cycle time is 3060 seconds for large complex parts. The highly automated
processing is practised. GMT are relatively easy to be recycled (GMT-R) by minimising the
number of plastics in the car in the favour of thermoplastics.
EKT developed two techniques for GMT recycling by heating up and remoulding. This
can be done three or four times without any significant reduction in properties. The
225
Handbook of Polymer Blends and Composites
preferred method consists in shredding used GMT parts and reuse in mixture (25%)
with virgin material.The second generation of GMT sheets has better adhesion between
the fibres and matrix and therefore has improved modulus and strength. This improvement
was obtained by new tailor-made sizing and matrix modification.
Continuous fibre roving impregnated with thermoplastics (COFIT) are obtained mainly
by different techniques such as melt impregnation (or pultrusion) process and by the
powder impregnation process and also by fibre entanglement, solvent impregnation and
powder/water impregnation. The latter is limited in its use by cost/efficiency and so is
applied mainly to expensive polymers.
Carbon Fibres
The main production capacities are Japan (51%) and USA (34%), Europe (13%) and
South-East Asia 2%. The main producers being Torray/Soficar and Akzo/Toho.
Applications are in aeronautics, sport (tennis, golf, fishing) and industry. It is estimated
that there will be an increase in use of 2.5 times by 2010 [17].
The worldwide market was 7,5009,300 tonnes in 1996 from which the USA 3,500
3,800 tonnes per year, Europe 5001,700 tonnes per year and other 3,5003,800 tonnes
per year. The aerospace consumption is 3035%; industrial use is 3540% and sports
goods is 15%. At the European level the consumption is 600 tonnes per year in aeronautics
and in sports 300 tonnes per year. The future of this fibre is for aeronautics (helicopters
and small planes) for the military and the civilian sectors.
Aramid Fibres
DuPont developed Kevlar aramid with a new plant in Maydown (Northern Ireland).
Aramid fibre-reinforced resins will be used in a new generation of supersonic aircraft.
Potentially uses of aramid composites are found in vehicle components, ballistic laminate
fairings for aeroplanes, helicopters and industrial applications [48]. The Sikorsky S-75
helicopter is 24% lighter than others with aluminium airframes, thanks to the extensive
use of graphite/epoxy and Kevlar/epoxy composite materials. Extremely high tensile
strength, low density, high energy absorption and favourable laminar breaking behaviour,
all contribute to the multiplicity of composite uses for aramid fibres.
226
Overview of the Use of Composites Worldwide
The body of the Peugeot 205 Turbo 16 is constructed from a composite of DuPont
high strength aramid (80%) and carbon fibre (20%). Each component offers specific
advantages and by their combination a lightweight structure is obtained, which is very
resistant to impact.
Akzo Nobel produces fibres based p-phenylenediamine and terephthalic acid; worldwide
production is 25,000 tonnes per year and for Europe 12,00013,000 tonnes per year
from which 10% are used for composites. The composites containing aramid fibre have
superior impact and chemical (acids) resistance and thus are used in aerospace/aeronautics
and transportation and refridgerated trucks. Another interesting application is sports
goods which may contain both carbon and aramidic fibres.
Special Fibres
GF as a reinforcement for the polymer matrix is gradually being left behind [49, 50].
The strengthening of the plastics panels can be also performed with hemp, flax or even
sisal, so rendering them even lighter and cheaper. They can boost the local agricultural
economy so contributing to the conservation of energy resources. A fully functional
recycling concept can be created. Miraflex is a bicomponent fibre with PP, PET/PBT,
PC/PBT, GF/Kevlar.
LCP
227
Handbook of Polymer Blends and Composites
5.2.4.2 Fillers
Some interesting developments are realised by surface treatments and new preparation
procedures, new types, etc. [52].
Total demand for fillers and extenders is expected to grow about 4% per year.
ECC International and Huber Corporation (USA) has a new calcium carbonate filler,
Hubercorn and Optifill that minimises wear on extruders for PVC processing. Microcel
Technol, Inc., introduced hollow ceramic spheres, as energy absorbing fillers for PO
used in interiors at Ford, BMW, etc.
From material substitution to new product development many interesting examples can
be cited: a highly successful product in packaging has been introduced as pearlised films
to substitute for speciality grades of paper in biscuit wrapping and the chocolate industry.
The pearl effect and modified mechanical properties are the result of a micro-void structure
achieved using calcium carbonate loaded masterbatches.
Calcined clays are used to control optical properties, such as infrared reflectivity in thermal
films used in horticultural glazing.
The market for surface-modified pigments, fillers and fibres is on the threshold of major
growth and change. Currently, the global market for silanes alone is estimated at over
$400 million and growing at 710% per year.
However, in some cases, the additional costs of adding a coupling agent and compounding
process cost can reduce savings or even increase unit volume costs over the virgin polymer.
The largest use of coupling agents is for treating continuous glass fibre. The largest use
of the surface modifiers is in treatment of calcium carbonate. The fastest growing
application are in treating precipitated silica, especially for replacement of carbon black
in tyres and in surface coating, adhesives and inks.
228
Overview of the Use of Composites Worldwide
Mixtures of hydrophobic coupling agents with more reactive silanes may also improve
the water resistance of appropriate polymer composites.
Silane coupled silicas are increasingly used to reduce rolling resistance in green tyres
while maintaining acceptable wet traction and improving abrasion resistance. Without
inorganic surface treatments, the good dispersing, more durable and multi-purpose
TiO2 grades would not be possible. Nearly all the wollastonite produced in US destined
for the plastics market is surface treated [53].
Areas to be developed are: new applications for coupling agents and surface treated
fillers; replacement of carbon black with silica in green tyres etc.
229
Handbook of Polymer Blends and Composites
230
Overview of the Use of Composites Worldwide
Aerospace and defence applications are the key fields for composites consumption. The
market for composites in 1984 increased by 38% followed by 66% increase in 1995 but
then the cold war ceased, reducing this increase [23].
The aircraft/aerospace industry was the largest end-use market for advanced polymer
composites, firstly in military applications. Once proven successful in this area, they
moved into the commercial market and then into industrial and recreation markets.
Although Japan leads the world carbon fibre technology, the US is major consumer of
each end-use market; the greatest consumer in military aircraft/aerospace applications
(79%). In the next generation of military aircraft, the amount will increase from 10
30% to 4080%.
The increase of use of APC in Europe is due to the airbus industry. Collective European
efforts led to Airbus Industrie, a consortium of European companies which uses the
composite in primary and secondary aircraft structures with the carbon/epoxy empenage
[56]. The Advanced Tactical Fighter designer is experimenting with thermoplastic resin
matrices. Composites could make up to 65% of the operating weight of commercial
aircraft by the year 2000. In Finland aeronautics consumes the most composites [56].
Japan has the largest market for composites in the recreation industry.
The main high performance fibres used in aircraft industry are CF, S-type GF, E-type GF,
AF and speciality fibres with an increased consumption from year to year [57].
Over the past decade, the recreational boat market became a major end-user, because
composites have better product durability, lower production and operational costs,
enhanced surface finishes and general product aesthetics. The Viper 330 a high
performance luxury boat contains many parts made from composites. There are Kevlar-
glass hybrid cloth manufactured by Chomarat, used in the construction of other yachts
and boats [58].
231
Handbook of Polymer Blends and Composites
will make it possible for marine manufactures to construct ultra-high performance boats
with high-priced competitive racing circuits.
Sophisticated resins such as vinylester resins and blends are now extensively used in the
most high-performance laminates. This development has been primarily carried out in
the US.
The development of new families of strong and versatile urethane and acrylic-based methyl
methacrylate (MMA) structural adhesives is revolutionising the boat building industry.
From the worldwide composite production of 4 million tonnes per year, the automotive
plastics market is 37% in 2000 which means about 445,000 tonnes.
Plastics offer the vehicle manufacturers two benefits, weight saving and style and also less
ecological damage [59]. The use of plastics, composites and thermosets has been imposed
in automotive construction because of their low density, reduced production costs, easy
utilisation, excellent resistance to acoustic vibrations and satisfactory impact resistance
[60]. Plastics help to solve the problems of engineers concerning reduction of gasoline
consumption and consequently reduce environment pollution, they also enable easy change
in the design of automobiles and recycling of materials. Another aspect is related to the
costly and pollutant operation of car painting [61]. The plastics can be coloured in bulk so
eliminating some operations and toxicity. Advantages of thermoplastic (TP) compared to
thermosets (TS) are: simple formulations, unlimited conservation, rapid moulding,
recyclability and reduction of pollution, no chemical reaction and toxics emission (vapours
or volatiles), good impact resistance and good UV and thermal resistance. From the style
point of view features can be added to a vehicle more economically.
About 10% of motor vehicles in use are estimated to be disposed of annually [62], i.e.,
nearly 14 million units. Association of Plastics Manufacturers in Europe (APME) showed
that in 1997, the automotive industry in Western Europe used 1.7 million tonnes of
plastics that means 6% of plastics consumption. 100 kg of plastics in a car replace
200300 kg of materials [46, 56, 63]. The resultant weight saving represents a saving
in fuel consumption of 750 litres over a life span of 150,000 km, therefore a reduction
of oil consumption of 12 million tonnes and carbon dioxide emission by 30 million
232
Overview of the Use of Composites Worldwide
tonnes in Western Europe. In the recent years growing safety and comfort levels have
led to an increase in the weight of an average car from 1,015 kg in 1990 to 1,132 kg in
1998. If plastics had not replaced traditional materials these cars would have been at
least 200 kg heavier.
In the USA the automotive industrys consumption also amounts to about one million
tonnes annually. The US motor car industry consumes about 200,000 tonnes of
polyurethane, over 150,000 tonnes of polypropylene and about 100,000 tonnes of PVC
(mainly in upholstery) per year. The industry also consumes over 80,000 tonnes of ABS,
90,000 tonnes of SMC (unsaturated polyester), 60,000 tonnes of Nylon, etc. Nylon in
all its forms is probably the most widely used material for performance critical components
in a car and its range of applications stretch from small to critical component heavy
weight parts. Nylon is used in under the bonnet, inlet manifold as GFR-PA, vibration
welding, air ducting and cooling system components, body panels. DuPont claims 60-
80% of US-made vehicles will use Nylon intake mainfold within five years. There are
predicted developments of plastics products for transmission turbines and stators, cooling
system components, radiator. In a new concept, Nylon 66 strips are laminated to form
channels for cooling fluids. A Nylon radiator was tested on military all-terrain vehicles
as well as marine craft [49]. The US consumed in 1994 about 92,000 tonnes of SMC/
BMC. Twenty-eight car automotive manufacturers introduced more than 300 parts from
these materials in 115 car models ranging from sports cars to sedans, from station wagons
to commercial cars. A 14% increase in the SMC/BMC consumption in 1994 corresponds
to the estimated growth of the automotive production in the US for the same year. The
price factor is very favourable for SMC/BMC, as their use permits an economy by weight
compared to the cost of steel. They replace a large number of parts, and material is
recycled more easily. Integration in a single part decreases the assembly price of the
manufacturing process [50]. SMC Automotive Alliance is an association of 31 societies
that looks for new products of good quality which will also satisfy the requirements of
Clean Air Act of the USA.
It is beneficial to use a cocktail (GF and mineral fillers) of fillers tailored to get the best
stiffness and temperature resistance with the minimum of warpage needed for cooling
system. Advances in three-dimensional blow moulding have enabled the development of
complex tubing systems made as single components including flexible sections to replace
the assembly of steel tubing, rubber bellows and hose clips. European automobile production
use as their main materials PA/GF 80,000 tonnes (long fibre and short fibre; 4.5 mm), PA
6, PA 6/6, PA 4/6, PA 6/10, PA 11 and PA 12. Recent developments have been made in the
following PA/long glass fibre for injection and PP/GF granules for injection (20,000 tonnes),
TSR moulded by compression (15,000 tonnes), PET/PBT/GF (12,000 tonnes), PPO and
PC/GF 2,000 tonnes) each, PVC, ABS/GF, PU/GF, recently LCP/GF, PEEK, PEI, PES, PPS,
and as the newest development with a TP reinforcement [46].
233
Handbook of Polymer Blends and Composites
The known European vehicles with a high proportion of plastics are: Renault Alpine,
Treser 1, BMW Z1 and in USA, Fierosports, Saturn and GM 200 luxury [45]. The average
content of plastics in various parts of the European vehicles are exterior body panels
BMW chassis 6%, engine 4%, trim 57%, body 21%, electrical 12%). Are used both
low tech and high tech or advanced composites for secondary and primary stressed
components, respectively.
The material is supplied pre-coloured in a metallic grey to eliminate the painting process
and cut down the emission of volatile organic compounds. Solvay makes the first thin
wall moulded-in colour metallic fascia application. Volkwagens requirement for low
temperature impact strength durability of paint finishes and low coefficient of thermal
expansion is also achieved. Sequel-1733 grade materials also replaced a more expensive
ASA/PC blend cutting the cost by 10% for some parts of the Chevrolet Lumina and
Monte Carlo. Fiero sports car from Pontiac has a body made from SMC, PU and glass
flake reinforced PO [60].
Future utilisation of liquid hydrogen as fuel will complement the use of composites to
reduce the weight and to assure a high speed engine. Use of the semi-conductor polymers
for electrical vehicle, battery must be very lighter also will require composite reinforced
with glass or carbon fibres (thermoplastic and epoxy) and polyaniline, polypyrrole, etc.
The field of competition polyolefins and polyurethane is environmental because the first
exhibit a lower environmental impact and are easily recyclable, so they are preferred
[64, 65].
Use of reinforced composites continues to grow, the shipments recording an 11.6% rise
in 1998 over the previous year. Data in Table 5.4 give an indication of the plastics use in
automobiles in Western Europe [66-68].
The use of lightweight materials particularly plastic composites will permit cars to triple
their fuel efficiency within 10 years [49, 50, 65]. If it takes into account that one-third of
US air pollution is generated by cars and other ground vehicles, such modifications will
lead also to the reduction of pollution.
The 1.8 million tonnes of plastics used in European automobile production in 1995 was
made up from a multitude of common polymer types. Leading the field was PP (35%),
followed by PU (20%) and PVC (11%). The remainder is distributed between PE, PE,
styrene copolymers and others.
234
Overview of the Use of Composites Worldwide
The new processing procedures such as BMC, RTM, SMC, SRIM, TSR offered the
potential for the use of composites in various parts of vehicles.
Propylene glass mat thermoplastics (GMT) replace metal in highly loaded complex
structural parts and are also challenging thermoset SMC in many non-appearance parts.
The planned production will consume 5,000 tonnes [47].
PPG developed the Azdel range of technopolymer structures (TPS) in Shelby, North
Carolina, Dalton, Massachusetts and Europe (Netherlands).
TPS replace metals and thermoset composites with an equivalent strength, lower weight,
faster cycling per part and recyclability. TPS can be formed into complex geometry with
variable wall sections (326 mm). Compared with SMC, TPS have greater impact
resistance, are more ductile and electrically insulating. Azdel in the USA is producing ~
20,000 tonnes per year of various grade of TPS in non-fracturing and heat resistant
qualities suitable for automotive applications as engine noise shields in trucks and the
bumpers beams for the Honda Accord, the General Motors APV minivan, Mercedes S
class saloon cars, sun roof for Ford. The roof frame is completely recyclable. Azloy
composites are also applicable for underengine noise shields, battery trays, seat backs
and flooring systems.
Azfab are high performance composites which will allow the production of assemblies
and products (cross member and suspension components) with strength, stiffness, impact
scratch resistance, light weight and low cost tooling.
Azfab sheet covered in a PC film can be used for the outer body panel of motor vehicle.
235
Handbook of Polymer Blends and Composites
The long fibre reinforced PA are used in automobile jacks supplied by LNP Engineering
Plastics as well as stabiliser links, tilt head assemblies, breaking support and steering column
components, while reinforced PP are used for the construction of wooden patio decks [69].
Such compounds are processed by injection moulding, which permits high volume
production on conventional equipment with only minor modifications. 25% GF reinforced
Xenoy PC/PBT alloy supplied by GE Plastic is used for automobile parts. It provides a
design that is very difficult to produce in one piece of sheet metal. The rail road industry
uses concrete from Celstran and 33% GFR-PA to replace costly aluminium parts.
About one-fifth of all plastics consumed in Europe are used in the building field with an
application rate which reached 59%, if only PVC is considered. The most part is for
structural components 90% being used as: cavity, insulators and profiles. Consumption
seems to remain constant in Western Europe from 1997 but there has been an increase of
7-10% in Slovakia, Czech Republic and Poland [72].
The market for portable electric tools is growing at a rate of 7% per year both for do-it-
yourself activities and also tools for agriculture, horticulture and forestry sector [74]. In
the last few years plastics have replaced the metals in such tools, for example glass fibre
reinforced PA 6 and PP count for 70% and 20%, respectively, because of good impact
resistance, long service life, resistance to abrasion and chemical agents, excellent surface
aspect, good electrical insulation and thermal content. For GFRPP low cost is also an
advantage. The disadvantage of GFRPA 6 is its water absorption tendency. Absorbed
water reduces electrical insulating properties.
The low weight of white PP garden furniture is a market which grew explosively; the production
doubling from one year to the next. These applications alone represent a demand for over 50
kt of mineral reinforcement such as talc or calcium carbonate and other additives [54].
Carbon fibre reinforced epoxy resin with a PVC core is used for radar antennae.
236
Overview of the Use of Composites Worldwide
Some recent developments in composite production are given in Tables 5.6 to 5.9.
237
Handbook of Polymer Blends and Composites
238
Overview of the Use of Composites Worldwide
239
Handbook of Polymer Blends and Composites
240
Overview of the Use of Composites Worldwide
241
Handbook of Polymer Blends and Composites
242
Overview of the Use of Composites Worldwide
While the USA is the technology leader for most plastics materials, Europe appears to be
the leader in recycling of composites. So, why is the European environment better for
recycling? Europe is the continent with the highest population per square mile. This fact
243
Handbook of Polymer Blends and Composites
alone sets the stage for better utilisation and preservation of resources. There are fewer
places to put rubbish in Western Europe, than in any other continent.
New technology and applications research for GRP recycled may be developed as result
of automative recycling efforts (see Figure 5.6) [62].
Worn-out Products
Sisa Reduction
Shredded Composite
Granulation
Fine Grinding
Figure 5.6 A possible scheme for composite waste conservation and recycling [62]
With the goal of reducing the cost of waste GRP disposal, recycling is preferable to
landfill, the costs of which have been increased and because the favourable properties of
plastics are not utilised if landfill is practised [62, 78, 79].
Driven by market forces, recycling products alternatives are started to hit market. Ashland
Chemical and AOC use 25% recyclates for press moulding of unsaturated polyesters,
244
Overview of the Use of Composites Worldwide
Phoenix Fibreglass of Canada offers fillers as recycling alternatives for BMC, glycolysis
is also applied to convert thermosets in polyols [34].
Cannon have a new technology for recycling RRIM products and Moulding Compounds
in Italy have made sample products made from recycled BMC and SMC parts used on
FIAT cars. GE Plastics Italy emphasised the comparative ease of recycling thermoplastic
composites rather than thermosets and showed a range of automotive parts.
The recyclate would be used in the same equipment as is being used today to make
standard SMC and BMC [77]. The best example of open loop recycling is the use of
sheet moulding recyclate in the marine industry.
Increase the productivity in the processing of composites, plastics and rubber by use
of industrial robots and manipulators [10, 72, 75].
Develop new fibres from aluminium, silicon carbide, boron nitride, PIA, high tenacity
PO, graphite, boron and aramids.
Give innovation in thermosets for lower density, greater flexibility and lower viscosity.
Develop low pressure moulding compounds (LPMC) and new conventional sheet
moulding compounds (SMC).
Increases in composite production are also expected to be fuelled through the replacement
of steel and aluminium by composites.
The technology will be directed to high technology products in sport and leisure.
245
Handbook of Polymer Blends and Composites
References
8. R. Burns, Polyester Moulding Compounds, Marcel Dekker, Inc., New York, 1982.
14. J. Myers and J. J. Innace, Modern Plastics International, 1994, 24, 6, 42.
246
Overview of the Use of Composites Worldwide
21. A. Smith, J. Whitehead and J. Baker-Counsell, Plastics & Rubber Weekly, 1989,
1273, 9.
28. J. Gossot, Revue Generale des Caoutchoucs & Plastiques, 1998, 75, 767, 26.
29. J-P. Forest, Revue Generale des Caoutchoucs & Plastiques, 1998, 75, 767, 58.
32. A. G. Gibson, Progress in Rubber and Plastics Technology, 1997, 13, 2, 125.
36. S. Lavabre, Revue Generale des Caoutchoues et Plastiques, 1994, 731, 49/51.
247
Handbook of Polymer Blends and Composites
38. Skeist Lab. Inc., Chemical Marketing Reporter, 1984, 225, 22, 4.
47. W. Piret, A. Weaver and M. Pandit, Reinforced Plastics, 1992, 36, 3, 17.
48. M. Ward, European Chemical News, Plastic Supplement, 1986, July, 34.
56. P. Martille and J. Joveru, Composites Plastiques Renforces Fibre de Verre Textile,
1987, 27, 2, 9.
248
Overview of the Use of Composites Worldwide
60. S. Lavabre, Revue Generals des Caoutchoues et Plastiques, 1995, 745, 32.
61. M. Reyne, Revue Generals des Caoutchoues et Plastiques, 1993, 70, 728, 41.
73. P. Dufton, Polymers in Building and Construction, RAPRA, Shawbury, UK, 1997.
77. J. P. Simmons, Composites Plastiques Renforces Fibres de Verre Textile, 1997, 21, 70.
249
6
The Interface in Polymer Composites
The interface between the filler or reinforcing material and the polymeric matrix is essential
in polymeric composites [1-3]. The interface adjusts the composites mechanical resistance:
here, concentration of mechanical stress occurs, as a result of the differences between the
matrixs coefficient of thermal expansion and the coefficient of the filler or reinforcing
agent, due to contraction at crosslinking (in thermoreactive matrices) and to crystallisation
(in thermoplastic matrices). At the same time, it acts as a site for the production of
chemical reactions.
The force acting on a polymeric composites matrix is transmitted to the filler or reinforcing
material through the interface. Its resistance and durability are determined by several
factors which govern the matrixs adhesion to the surface of the included material.
With a view to increasing the interfacial resistance between the included material and the
matrix, the surface of the filling or reinforcing material is usually treated by special techniques,
so that to create the conditions in which the interface should assure shifting without any
discontinuities from the properties of the matrix to those of the inclusion material.
The adhesion force between the solid polymers and other substances, along with other
parameters, such as the contact surface and the diffusion distance, are strongly influenced
by the type and magnitude of the intermolecular forces manifested between the adhesion
parameters [4, 5]. The adhesion forces may be characterised by the application of three
main types of methods, as follows: direct measurement of the force, spectral determination
of the chemical composition of the solid surfaces and physico-chemical determination of
the energetic interactions.
The theories discussing the increase of the composites mechanical properties are based
on the hypothetical existence of a perfect adhesion between phases. Thus, Lee [6] has
obtained an equation describing the state of adhesion, expressing the contact angle of
the liquid wetting a solid, in terms of the free surface energy of the liquid. When
investigating the adhesion characteristics of the polymer-filler systems, the properties of
the free surface energy of both components are determined separately [7, 8].
251
Handbook of Polymer Blends and Composites
In polymer mixtures and composites, the interface occurs between the two phases as a
result of thermodynamic incompatibility. The inorganic reinforcing fibres, as well as the
particulate fillers, have a high surface energy; that is why, they cannot be wetted by the
polymer, an adequate surface treatment being necessary [3].
Surface modifications are indispensable, too, when utilising carbon or aramidic fibres. In
most cases, a primary surface treatment is applied, a low molecular compound with two
functional groups being utilised. One of the reactive groups should react at the surface of
the reinforcing or filling material, preferably through covalent chemical bonds, while the
other with the polymeric matrix. The chemical reaction of the adhesion agent with the
matrix does not necessarily lead to the attainment of some optimum properties for the
composite. Such a situation may be explained by an insufficient number of bonding centres,
on the side, and by some rigid connections, on the other. The different coefficients of
thermal expansion, along with the impact stresses, cause the occurrence of some cracks in
the materials mass, starting from the interface. The realisation of a partial interpenetrating
of the coupling agent with the polymer was also demonstrated as necessary [9].
In the case of composites with disperse material fillers, one should necessarily know,
besides the distribution of particles dimensions, the characteristics of their surface [10].
The contact surfaces cannot be determined directly, but only approximated as satisfactorily
as possible, by the application of various methods.
If adhesion at the interface between the matrix and charge or fibre is null, the properties
of the material will be highly similar to those of the matrix with internal voids, once known
that fibres contribute only to a low extent to global resistance, acting even as centres of
stresses concentration; consequently, high adhesion is necessary, for structures consolidation.
Under certain circumstances, the fibre/polymer adhesion should be good, but not very
good, an optimum interfacial adhesion being therefore desired. Characterisation of this
optimum value represents an aspect that should consider a multitude of factors, such as:
the elastic moduli and Poisson coefficients of the two phases, the dimension and geometry
of the two phases, the rheological properties of the polymer, and the aging effects. On
considering that, up to now, no general theory has been put forth, understanding and
characterisation of adhesion should be based on the already existing data in other domains,
in which the role played by the resistance to interface is much more obvious.
252
The Interface in Polymer Composites
The idea was stated that resistance to the interface consists of a mechanical penetration
of the adhesive in the rough surface of the substrate. The pores and the asperities represent
a field on which the adhesive may be set through simple geometric effects (Figure 6.1).
By its nature, the model can be applied exclusively to materials having a high roughness,
such as: wood, paper or textiles. If the adhesive does not correspondingly wet the substrate,
interpenetrating is weak. The air retained between the asperities hinders a good substrate/
adhesive contact, and, consequently, the non-wetted regions may represent zones inducing
crack occurrence.
253
Handbook of Polymer Blends and Composites
Deriagin and Krotova [11] proposed a theory based on the electrical effects observed
during tests of separation through the detachment of the reinforcing material off the
matrix. Electrical discharges are produced, the observation being made that the separation
energy is 10-1,000 times higher than the energy necessary for breaking the intermolecular
links. The explanation lies in a dissipation of electrical charges: the necessary energy
depends on the separation rate of the two phases and, by a reduction in the density of the
charges, adhesion decreases with temperature. The system may be assimilated to a
condenser, the adhesion energy being equivalent to the mechanical work necessary for
the separation of the two sides of a flat condenser.
Figure 6.2 The electrodeposition of a 30% zinc and 70% copper alloy at the surface of
the rubber previously crosslinked with sulphur
Such a type of adhesion is quite resistant to temperature as well as to the action of water,
while its formation requires well-controlled fabrication conditions.
Coupling with silanes represents another example of the formation of primary bonds at
the interface. Their efficiency has been demonstrated by Ahagon and Gent [12], even if the
254
The Interface in Polymer Composites
part played by silanes in increasing adhesion has been disputed by some authors [13]. The
theory of chemical adhesion, with some modifications, seems to be the most valid one as to
the successful treatment of the glass fibres utilised in reinforcing thermoplastics.
Initially, the theory known as self-adhesion, has been proposed by Voyutski [14], with
a view to explaining the adhesion between two identical polymers.
When two polymeric materials are in intimate contact and are thermodynamically
compatible, the molecular motion (at not too low temperatures) makes possible
interdiffusion of the two species. A mutual interdiffusion zone, or an interphase, which
is the only one responsible for adhesion, thus results. It has been determined that adhesion
of polyisobutylene with another polymer results from an interdiffusion that depends on
the contact time, temperature and pressure, the development of such an interdiffusion
phenomenon being suggested graphically in Figure 6.3.
Taking into consideration the separation of two surfaces glued one to the other, Bikerman
[15] drew attention to the fact that, possibly, breaking should not be developed strictly
at the interface, being instead deviated and following a route quite near the interface, yet
in the mass of one or another of the components. Consequently, breaking is cohesive,
occurring in a region in which mechanical resistance is low. Further on, he proposed
seven types of low cohesion interfacial layers, quite simply to be classified (Figure 6.4).
255
Handbook of Polymer Blends and Composites
For the first class, a simple propagation of breaking occurs inside the air bags from the
interface, where the contact of the solids is not assured. Considering the specific dynamic
properties of the interface, a more or less pronounced tendency of migration of the
impurities towards the surface of one another of the components in contact is manifested.
In the case of polymers, they may be even macromolecular chains of low molecular
weight, which induce lowering of the local resistance of the network. Such an explanation,
corresponding to the second and third classes, may be equally attributed to a
supracrosslinking effect in the vicinity of the interface, when the second material also
acts as a catalyst. The existence of a supracrosslinking tends to increase the brittleness of
the polymer layer occurring near the interface.
The fourth up to the seventh classes take into consideration the possible reaction between
the air and the matrix, between the two components and, finally, between the air and the
two components. It goes without saying that, when the composite material is immersed
into a liquid, it may replace the air. The liquid may show chemical affinity toward a
polymer, nevertheless physical interactions may be sufficient as such [16].
Although Bikermans model has been criticised in the past, nowadays, the importance of
the low-cohesion interfacial layers in certain breaking processes, considered as cohesive,
is acknowledged.
In the situation in which adhesion occurs at the interface between two materials, one or
even both of them may be in liquid state. Formation of chemical or physical bonds
requires the existence of a very good intimate contact, which involves, in the case of a
liquid applied on a solid, a suitable wetting of the latter; in other words, regardless of the
nature of the final bonds, the necessary condition for a good adhesion is intimate wetting.
256
The Interface in Polymer Composites
W0 = A + B - AB (6.1)
S = SL + L cos (6.2)
This model was initially developed for explaining the fact that adhesion measured during
the detachment of an elastomer off a rigid substrate is much more important than the
thermodynamically measured one. In spite of the fact that the measured adhesion energy,
W, is enormous comparatively with the Dupre energy, W0, a proportionality factor does
exist between them, with the exception of certain situations. This breaking energy varies
considerably with the separation rate and temperature, which may be explained through
the dissipation of energy, occurring as an hysteresis, during the detachment of the two
materials from one another.
A certain proportionality does exist between dissipation and intrinsic energy; nevertheless,
contrary to its dependence on W, dissipation is higher by several orders of magnitude. It
responds to the superposition equation of Williams, Landel and Ferry [18], which explains
the equivalence of the effects induced by the deformation rate and temperature on the glass
transition temperature of the elastomer. More recently, to explain the high adhesion energies
at very low breaking rates, processes of molecular dissipation have been mentioned.
257
Handbook of Polymer Blends and Composites
One of possible treatments with calcium carbonate involves the utilisation of fatty acids,
which are fixed, through their carboxylic function, to the particle, while the long organic
radical assures a decrease in its superficial energy, which is dependent on the amount of
fixed acid [19, 20]. The observation was made that the stearic acid reduces the surface
free energy, mainly through the decrease of the polar component. Its minimum value is
recorded for an amount of 1% (weight) stearic acid, which corresponds to a complete
surface covering, according to which the observed increase is due to the presence of the
free carboxylic groups from the acid, at the surface.
Surface modification of calcium carbonate may also be realised by means of the silane
coupling agents. In spite of the numerous studies devoted to such topics [21], it has been
observed that the powder does not always respond well to the treatment with silanes
[22-24] or with other coupling agents, regardless of the presence, at the surface, of OH
groups coming from the adsorbed water. Improvement of their action may be attained
through the introduction of some polar groups in silanes structure, or through the
utilisation of polymeric silanes, e.g., poly(oxyethylenyl propyl trimethoxysilane), vinyl
diethyl trimethoxysilane and vinyltrimethoxysilane-acid and maleic anhyride.
258
The Interface in Polymer Composites
Avnimelech [26] confirmed the formation at the surface of the calcium carbonate, of the
Ca3(HCO3)PO4 complex, in the presence of Na2HPO4.
The phosphoric and metaphosphoric acids are recommended too, as modifiers of the
calcium carbonate surface. Some phosphorus-containing compounds are utilised in
surface treatment of the glass fibres or of some mineral filling materials. Wollastonite,
treated with phosphoric acid prior to its introduction into a polyester matrix containing
tritolyl phosphate, induces a significant increase of the composites bending strength
[27]. The triethylphosphate may be successfully utilised in the treatment of glass fibres,
see Figure 6.5.
Other agents utilised in the treatment of the surface of glass, metal or mineral fillers,
although with quite restricted applications, are the carboxylic acids, titanates, amines
and the chromic compounds; an excellent survey of this has been done by Cassidy and
Yager [28].
Undoubtedly, silanes represent one of the coupling agents classes most frequently used in
the treatment of fillers and mineral reinforcing materials.
The silane-glass adhesion involves crosslinking of silanols on the glass surface, substitution
of the water retained at the surface, physical adsorption to the glass surface, formation
of hydrogen bonds with glass, as well as glass protection, elimination of microcracks,
prevention of structural defects generation, prevention of corrosion and of the access of
water to the surface.
259
Handbook of Polymer Blends and Composites
The silane-polymeric matrix adhesion involves increase of its wetting capacity, increase
of the roughness of the surface. Formation of a layer which assures the transfer of
the stresses towards the reinforcing material, barrier for water, increase of
compatibility and formation of a diffusional layer, along with the formation of
covalent bonds.
It has been established that the formation of covalent bonds between the organo-
functional group from the silane and the matrix resin represents the controlling factor
in the development and maintenance of good adhesion [29, 30]. Figure 6.6 shows
the pattern of the action of silanes as a compatibilising agent [31] between a mineral
inclusion (particle or fibre) and an elastomer (a thermoplastic matrix).
The manner in which silanes hydrolysed and condensed to silane oligomers, may
attain compatibility is quite varied; the layer of them may be compatible with the
liquid resin, a case in which a real polymer is being formed during the treatment; or,
siloxanole oligomers may show partial compatibility with the matrix polymer and
may form an interpenetrated network; or, the polymeric siloxane segments may
interdifuse in the polymer mass, without crosslinking.
Table 6.1 lists some types of commercial silanes utilised in surface treatments of
certain mineral charges and reinforcing materials [28-31].
260
The Interface in Polymer Composites
261
Handbook of Polymer Blends and Composites
Treatment with ozone shows a high oxidation rate, the induction period being absent.
The oxidation degree increases with the duration of treatment, higher and higher
262
The Interface in Polymer Composites
Prior to the synthesis of the composite, the PP particles have been functionalised
with maleic anhydride, showing subsequently, that during processing, grafted
copolymers may be formed between the reactive functions of the filler and the matrix
[42, 43].
Grafting of polymers at the surface of ultrafine inorganic particles is very well known
already, as a method assuring an increased dispersion degree of them in either solvents or
polymeric matrices. An example is the polymerisation of acrylamide at the surface of
very fine silica particles, using a cerium salt as photochemical initiator, or polymerisation
of vinyl monomers to mineral surface, functionalised with labile organic groups [44,
45]. The reaction scheme (Figure 6.7) presents a possible introduction of some peroxydic
groups to the surface of the colloidal silica.
263
Handbook of Polymer Blends and Composites
Figure 6.7 The reaction scheme of introduction of peroxide groups to the silica surface
In a similar manner, one may first functionnalise and then graft carbon black particles,
the degree of grafting attained depending on the type of peroxide initiator fixed to the
surface of the filler [46]. For a subsequent polymerisation of some monomers, silica may
be functionalised at the surface by various methods [47]. Thus, the ultrafine particles of
silica or titanium oxide have been grafted photochemically [48], after the introduction,
on their surface, of azo groups, followed by subsequent photopolymerisation of monomers
such as, methyl methacrylate (MMA), styrene and N-vinylcarbazole (NVC).
For the introduction of the azo groups, the particles have been first treated with 2,4-
toluenediisocyanate (TDI), after which the product has been reacted with 4,4-azo-bis(4-
cyanopentanoic)acid (ACPA), according to the reaction shown in Figure 6.8. [49-52].
Figure 6.8 Introduction of azo groups to the silica or titanium oxide surface
Glass fibres can be grafted with vinyl polymers according to the same principle,
functionalisation through the introduction of azo groups, followed by polymerisation of
the vinyl monomers initiated through their thermal decomposition [53].
Table 6.2 lists the results obtained by grafting MMA and styrene at the surface of titanium
oxide containing surface azo groups.
264
The Interface in Polymer Composites
Figure 6.9 Grafting of MMA, styrene and NVC on the particle surface
Some recent investigations discuss the possible introduction of azo groups at the surface of
some barium sulphate, of carbon black and of carbon microbeads. Barium sulphate [54]
represents a mineral filler used in elastomer and resin-based composites. At its surface
there have been previously introduced 2-hydroxystearate groups which, further on, have
been condensed with the COOH groups of ACPA, using N,N-dicyclohexyl carbodiimide
(DCC) as a condensation agent. The reaction developed is shown in the Figure 6.10.
Figure 6.10 Introduction of azo groups on the surface of barium sulphate, carbon
black or carbon microbeads
265
Handbook of Polymer Blends and Composites
Further on, grafting is realised through thermal decomposition of the azo groups.
Introduction of the azo groups at the surface of the carbon black [55], on which hydroxylic
functions may be found, can be realised in two ways, either through the ACPA reaction
with the isocyanate groups previously introduced through the treatment with TDI (which
is similar to the treatment of silica), or through direct condensation of ACPA with the
surface OH groups, on using DCC (Figure 6.11).
Figure 6.11 The introduction of azo groups on the surface of mineral powder by direct
condensation of ACPA with the surface hydroxyl groups, using DCC
Under the influence of temperature, the functionalised intermediates lead to the formation
of radical active species capable of subsequently initiating polymerization of MMA,
styrene, N,N-diethylacrylamide. In a similar manner, one may polymerise MMA at the
surface of the carbon microspheres [56] functionalised through treatment with 2,2-azo-
bis(2-hydroxymethyl) propyonitrile.
The literature [57-58] also reports other methods of modification with organic polymers
and with polar or polar-nonpolar copolymers of the precipitated silica surface. These
methods include four distinct stages (Figure 6.12):
adsorption of a surfactant
solubilisation of a monomer in the surfactant layer
in situ polymerisation of the monomer in the surfactant layer
partial removal of the surfactant
According to the described procedure, grafting of silica with isoprene copolymers and
1,3-butadiene with vinylacetate, acrylonitrile, 4-methoxystyrene, 4-chlorostyrene, MMA,
styrene, polybutadiene and copolymers, has been realised.
Tsubokawa and others [59] have realised grafting of silica, ferrite and titanium oxide after
a preliminary treatment with 4-trimethoxysilyltetrahydrophthalic anhydride, followed by
anionic copolymerisation, accompanied by ring opening of some epoxies, initiated by the
carboxylate groups previously introduced to the surface of the mineral particles.
266
The Interface in Polymer Composites
The model of the calcium carbonate particle, treated superficially with a poly(acrylic
acid-b-polybutadiene) copolymer, is presented in Figure 6.14.
267
Handbook of Polymer Blends and Composites
Figure 6.14 The model of the calcium carbonate particle superficially treated with a
poly(acrylic acid-co-butadiene) copolymer
268
The Interface in Polymer Composites
Cotton-type cellulose shows high resistance, durability and thermal stability, high
hydrophilicity and the highest molecular weight among all the other cellulose
types [63].
Physical (UV treatment and ionising radiation, thermal treatment, grinding in ball mills)
as well as chemical methods have been applied to synthesise various grafted copolymers
of cellulose and of its derivatives, known to have unique and very useful properties.
The literature mentions the polymerisation of MMA [64] on the surface of some cellulose
derivatives, such as methylcellulose, containing 24.7% OCH3 groups. The process is
conducted in an inert atmosphere, in the presence of ethanol or methanol solutions, in the
absence of light, at 85 C, for 4 to 9 hours, using MMA and vinyldiene chloride as monomers.
Grafting is realised through the formation of an active complex among the monomer, the
OCH3 groups of methylcellulose and the methanol used as solvent (Figure 6.15).
Figure 6.15 The formation of an active complex among the monomer, the methanol
and the OCH3 groups of methylcellulose
Under certain reaction conditions, the decomposition of the complex favours the
attachment of the molecule of monomer to the methyl cellulose chains, as well as its
subsequent polymerisation.
Canadian researchers [65] have studied grafting of styrene on partially xanthaned cellulose
pulp, in the presence of redox systems, which involve potassium persulphate. This or
oxygenated water may initiate polymerisation through thermal decomposition, or in
combination with ferrous ions at much reduced temperatures. Utilisation of divinylbenzene
besides styrene, in the process of grafting of cellulose, leads to a substantial increase of
the degree of grafting.
Cellulose grafting may be also realised with mixtures of vinyl monomers (vinyl acetate,
methyl acrylate), the initiation method used making use of redox systems created by the
269
Handbook of Polymer Blends and Composites
ceric ion [66]. Grafting may be also initiated through physical methods. The cotton has
also been grafted with difunctional monomers (divinylbenzene) mixed with styrene [67],
with uranylnitrate and benzoinethyl as photoinitiators.
The polystyrene (PS) pearls may be modified, on the surface, through grafting, in the
presence of benzoyl peroxide, of several monomers, such as, styrene, MMA, 2-
hydroxypropyl methacrylate, propyleneglycol methacrylate, used either alone or in
mixtures. Such reactions lead, on one side, to grafted copolymers, i.e., polymer fixed
to the substrate and, on the other side, to free polymer. The grafted copolymerfree
polymer combination leads to the formation of core-shell particles, presented
schematically in Figure 6.16.
270
The Interface in Polymer Composites
6.3.2.1 Polymer Adsorption at the Surface of the Filler and Reinforcing Materials
A polymers physical interactions with inorganic substances are determined by two major
contributions: the van der Waals forces and the acid-base interaction [68, 69]. Complete
coating of the filling or reinforcing material is quite improbable, which is due both to the
energy distribution of the active centres on the surface of the filler, and to the conformation
of the adsorbed chains. The estimation was made that, for a layer with a thickness of
about 5 nm of absorbed polymer, the degree of coating of the particle represents 10
20% of the whole surface [70]. The arrangement of the chains on the surface of the
filling particles is illustrated in Figure 6.17.
Studies on adhesion through adsorption of the polymers at the surface of the inclusion
thermodynamically approach the phenomenon [71], starting from the concept of selectivity
of the interaction of compounds at the surface. Such a perspective requires knowledge of
the intensity of the adhesion forces that may be characterised by various methods, such
as, direct measurement, spectral determination of the chemical composition of the surfaces,
physico-chemical measurements of the energy of interaction [72]. Most of the studies are
devoted to elastomers filled with carbon black [73-75], although some are extended to
composites with thermoplastic matrices filled with silica or with other mineral powders,
too [76] or with organic powders (lignin) [5].
271
Handbook of Polymer Blends and Composites
develop, at the surface, strong interactions of the acid-base type, on their deposition
from solutions of basic (aluminium oxide) or acid (hydroxylated glass) substrates. IR-
RAS spectroscopy is very efficient in the determination of the nature and free energy of
the adduct formed at the interface of the polymer/metal systems.
A correlation between the results obtained by IR-RAS spectroscopy and adhesion tests
permits the assumption that both the acid-base type interactions and the density of the
interaction interfacial centres lead together to an efficient adhesion at the interface.
Adsorption of an oligomer or polymer at the surface may be applied to both organic and
inorganic filling and reinforcing materials. Petraru and Popa [77, 78] have reported on
the possible treatment of the calcium carbonate powder with a high density polyethylene
oligomer, which permits the subsequent inclusion of the charge into a polyethylene matrix,
in ratios up to 100% (calcium carbonate related to the polymer); such a treatment
significantly influences the flow properties of the, during processing.
Fowkes and Mostafa [79] tried to explain in detail the nature of the interactions between
the calcium carbonate particles and the polymer adsorbed at its surface, from organic
solvents. They showed that such an adsorption is dominated by interactions of acid-
basic nature, occurring among solvents, polymer and the surface of the filler. It was
demonstrated that chlorinated poly(vinychloride) is adsorbed at the calcium carbonate
surface, much more intensely in the presence of less polar solvents.
McCarthy and others [80] have obtained composites with polysiloxanic matrix and
particles of the silica- or titanium dioxide-type, precipitated in situ, when the matrix
polymer is strongly adsorbed at the particles surface. A good adsorption of polyethylene
or of the polyethylene/PS mixtures at the surface of talc particles has been reported by
Karrad and others [81].
The former case involves mechanical mixing of silica with the elastomer, on a Brabender
mixer, while the latter is based on the elastomer adsorption from solution, for various
polymer/silica ratios, at a constant silica/solvent ratio.
Trying to avoid the utilisation of coupling agents for increasing adhesion between glass
fibres and the polyimide matrices, some researchers [83, 84] attempted to include them
in polyimide matrices containing a siloxane [poly(dimethylsiloxane)] (PDMS).
It was demonstrated that breaking of the two layers is cohesive, taking place in the
polyimide, in the vicinity of the interface realised.
272
The Interface in Polymer Composites
The glass fibre, previously covered with the PP modified with maleic anhydride (0.5%),
has been used in PP reinforcement [88], the observation being made that the coupling
agent induces transcrystallinity, while the rate of transcrystallinity increase depends
on temperature.
A frequently used method, at industrial scale, for improving the adhesion between
the carbon fibres and poly(vinyl esters) is coating the fibres with a thermoplastic
polymer compatible with the fibre and miscible with a matrix [85]. The polymers
utilised for surface treatment were: poly(vinyl pyrrolidone), poly(hydroxyether) and
poly(ethyloxazoline), compatibilisation with the matrix being assured by the formation
of hydrogen bonds at the interface, as well as by the reciprocal interpenetrating of
the two polymers.
Coverage of the carbon fibres with polyamide 66, through an in situ polymerisation
of hexamethylene diamine with adipoyl chloride [89], permitted their utilisation in
the reinforcement of epoxydic matrices, the observation being made that the treatment
does not influence the tensile strength of the composite, although it increases its
impact strength. Starting from these results, the authors obtained hybrid composites
[90] with epoxy matrix, carbon and Kevlar fibres, superficially covered with polyamide
66, being used for reinforcement, by the same technique.
The intense increase of adhesion, at the interface, is mainly possible through the chemical
bonding of the polymeric coupling agent, as a result of the reaction of the functional
groups of the two components. Calcium carbonate may be surface treated with a
copolymer based on ethylene oxide and propylene oxide (PEO/PPO), or even with
poly(ethylene oxide) functionalised with -COOH groups [91].
The functionalised PEO is dissolved in the water in which the calcium carbonate had
been suspended, after a 23 hour stirring period, the surface-treated calcium carbonate
powder being obtained. Further on, this is suspended in the water in which a latex of
carboxylated SBR, and hydrochloric acid as a coagulation agent, had been added. The
rubber latex is fixed, by means of the COOH groups, to the surface of calcium
carbonate. R defines the charge weight/dry SBR weight ratio (see Table 6.4).
Table 6.4 lists the differences observed when treating the same charge with poly(ethylene
oxide) functionalised with a POE/PP copolymer and SBR latex.
273
Handbook of Polymer Blends and Composites
The treatment with functionalised PEO is seen as more efficient than the one with PPO/
PEO copolymer, which is undoubtedly due to the presence of COOH groups on the ends
of the macromolecular chains.
Nakagaki and Ninomyia [92] discussed the effect of the poly(sodium acrylate) aqueous
solution on the aggregation and sedimentation of calcium carbonate particles, while
Kawalewski and others [93] studied the effect of the introduction of calcium carbonate,
treated with poly(ethyleneoxide) oligomers, in a PP matrix.
Another method for treating the surface of calcium carbonate consists of the modification
with PP, which has rests of maleic acid on the ends of the macromolecular chains. The
treated material is well dispersed, becoming compatible with the PP matrix in which it is
subsequently introduced [3].
The literature [95] describes an interesting grafting method of silica (previously modified
with amine groups), with living poly(izobutylvinylether), obtained through anionic
polymerisation in the presence of a CF3COOH/ZnCl2 system. The reactions developed
are shown in Figure 6.18.
Figure 6.18 The grafting of a living poly(izobutylvinylether) on the silica surface in the
presence of a CF3COOH/ZnCl2 system
274
The Interface in Polymer Composites
Another method applied for the modification of the surface of the particles for their
hydrophobisation, is described by Le Deore and others [96]. In the first stage, partial
hydrosylilation with silanic coupling agents occurring, followed by the reaction of the
functionalised particle with 1,2-polybutadiene oligomers, in the presence of
hexachloroplatinic acid. A second method is represented by hydrosylilation of
polybutadiene, which is subsequently linked, to the surface of unmodified silica. The
reactions developed are presented in Figure 6.19.
Figure 6.19 The two methods used for grafting of polybutadiene on the silica surface
Obviously, the ratio of grafted polymer increases with the content of surface anhydridic
groups, the highest degree of grafting being obtained with silica and the lowest with ferrite.
The strong increase of adhesion at the interface is realised by the stimulation of direct
chemical reactions between the matrix and the filling or reinforcing polymer. Recently,
Fakirov and others [98, 99] have introduced the concept of microfibrillar composites,
which refers to a polymeric matrix reinforced with microfibrils, which are also polymeric.
Realisation of the composite is possible through a process of fibrillation induced through
stretching, followed by melting of the component with a lower melting point. The high
275
Handbook of Polymer Blends and Composites
Figure 6.20 The oligomer condensation on the silica or titanium oxide surface by
means of previously introduced anhydric groups
The extensive experimental and theoretical evidence accumulated at the present time
indicate that the physical structure of interfaces differs from that of the bulk [100-
102]. The interface formation can be established by a deviation of one or other physical
properties of the filled system from values calculated according to the additivity rules.
The structural model of filled polymers involving interfaces allows to explain the
character of variations in absolute values of density, viscosity, heat capacity, and glass
transition temperature. At the same time, the changes of these properties provide
information about the structural organisation of the interfaces.
With increasing filler content, the density of the polymer part, calculated according to
the additivity scheme, at first diminishes, but beginning with a certain filler concentration
remains constant [103, 104]. This value corresponds to the state in which whole polymer
transforms into the interfaces. The interface thickness in these conditions can be
determined by equation:
d
= ( a 1 / 3 1) (6.3)
2
276
The Interface in Polymer Composites
where is the diameter of the filler particles and a is a volume fraction of filler. It was
shown that the interface thickness depends on the molecular weight, Mw, of polymers.
For PS the corresponding equation was found to be [105]:
The relationship in Equation (6.4) was obtained on the basis of determination of the
heat interaction of CH2Cl2 with the filled polymers.
Tg = Tg 0 + T (6.5)
where Tg0 is the glass transition temperature of unfilled polymer, is the polymer fraction
in interfaces, T is the maximum addition of the glass transition temperature for the
given polymer and filler. The maximum T corresponds to =1 when the polymer
transforms into the interface state. The values are calculated by equation:
=1 C / C0 (6.6)
where C and C0 are the heat capacity jumps at the glass transition temperature for
filled and unfilled samples, respectively. The value increases with increasing of the
specific surface s of the filler and the surface energy Es. Generally,
where is the parameter of macromolecular rigidity and Ec is the cohesion energy. The
transfer of macromolecules into interfaces must be facilitated with decreasing of
intermolecular interactions and increasing of the polymer-filler interaction energy, that
is with increasing of the Es/Ec ratio. At a compromise between Es, Ec and , the Tg increase
is most evident. The increase observed of the value with is due to the deterioration of
the polymer chain packing conditions in interfaces. This concept is in good agreement
with the assumption that the more friable packing of the rigid-chain polymers in interfaces
is caused by the entropy effects [107].
The viscosity measurements of the filled polymer melts also provides the means of the
interface thickness estimation [108, 109]. The dependence of for liquids on the solid
phase content is usually described by Einsteins equation:
277
Handbook of Polymer Blends and Composites
However, by the example of filled polyester [108], it was shown that the value grows
with the increase considerably more than is predicted by Equation (6.8). Such behaviour
of viscosity testifies that the layers of adsorbed macromolecules are formed on the filler
particle surface. The fraction of the polymer available in interfaces can be evaluated
assuming that represents a combined fraction of the solid phase with the adsorbed
layer that is
= 0 + (6.9)
where 0 is the filler true fraction. The interface thickness determined from the values
at 393K is 12-14 nm. It should be noted that a comparison of the interface thickness
values defined from various physical properties of the filled polymers show their limits
of 40-150 nm [103].
Several surface analytical techniques may be utilised to study the interface in composite
materials, Auger Electron Spectroscopy, X-Ray Photoelectron Spectroscopy and Secondary
Ion Mass Spectroscopy; they give different and complementary information [3].
Small-angle x-ray diffraction can be used to measure the interfacial thickness in polymer
blends, especially [115-116], with possibility to be applied to study the interface in polymer
composites.
Another way of studying interfaces can be found in the fundamental property of chemical
reactions in solid polymers related to the dependence of their kinetics on the molecular
organisation of media [117]. To understand the kinetics of chemical reactions it is necessary
to know how the polymer is structured. The opposite problem implies that knowing the
kinetic features and mechanism of the reactions, one can attempt to reveal the structural
particularities of polymers. Investigations of the thermal and thermo-oxidative degradation
of filled polymers [118] indicates a significant role of fillers in these processes. However,
in the case of high-temperature oxidation processes, the reaction kinetics can also be
affected by, besides the structural-physical factors, by the catalytic properties of the filler
278
The Interface in Polymer Composites
surface. From this standpoint, the use of radical reactions, going at low, ordinary or
moderate temperatures, has a certain advantage because their kinetics reflects special
features in molecular organisation of the interfaces, virtually in the pure form not
complicated by interaction of radicals with the filler surface.
The low-temperature reactions of the triplet carbenes in filled polymers are suitable for
conducting interface studies. The distinctive feature of these reactions is a formation of
short-living intermediate radical pairs resulting from the hydrogen atom transfer from
macromolecules to carbenes [119]:
where kav is the average effective constant of the CHC decay rate and n is the parameter
characterising the distribution width of carbenes in their reactivity. The kav values decrease
moderately as the filler content increases in the range of distances between the aerosil
particles l exceeding 20 nm. If l < 20 nm, kav sharply decreases by more than ten times.
This critical distance may be considered as an effective thickness of the interface layers in
filled PMMA and CTA.
On the basis of kinetic data [121], the following conclusions can be reached about the
nature of the matrix structure changes caused by filling. As the width distribution in the
carbene reactivity, determined by the n parameter, does not change with filling of the
polymers, the structural-physical modification does not influence distribution of the rate
constant of the primary process of the hydrogen atom transfer from macromolecules to
carbenes in Equation (6.10). This process is the limiting stage of the carbene decay and
determines the kinetic inequivalence of carbenes by the dispersion of distances between
the carbene centre and the C-H bonds of the matrix. In this connection, one can conclude
that structural changes affect the efficiency of the intermediate radical pair recombination,
defined by the spin state populations. As the filled polymers are prepared from solution,
the formation of more brittle packing of macromolecules is possible in interfaces because
of the difficulties with the relaxation processes during shaping of filled samples. Such
structural realignment can change the relative population of singlet and triplet states of
radical pairs as a result of their geometrical configuration changes. The larger the
279
Handbook of Polymer Blends and Composites
declination of the non-pair electron axes from the right-angled orientations, the higher
the population of the triplet level of radical pairs and, consequently, the smaller the rate
of carbene decay.
As shown in [122], some fraction of diphenylcarbenes (up to 20%) gives rise to clusters,
owing to the non-uniform distribution of the carbene precursors (diazocompounds) in
polymers. These cluster carbenes are capable of giving biradicals even at 77 K, by the
following reaction:
hv
[ ]
2Ph 2C = N 2 Ph 2C : Ph 2 : BR
kT
(6.12)
The character of the W0 relationship is out of proportion to the filler volume fraction. Its
increase at 20% of aerosol content is conditioned by polymer transition into interfaces
in which the biradical decay efficiency is high. The interface thickness ( 12 nm) is fairly
close to that determined from the filler effect on the kinetics of the matrix insulated
carbene conversion.
280
The Interface in Polymer Composites
From the data in the Table 6.6 it is obvious that the distribution width is not changed
with the PMMA filling. A minor narrowing of this distribution was observed with an
increase in temperature. It is remarkable that the rate of decay sharply increases by a
factor of 15-60 in samples containing upwards of 50% aerosil. The results obtained
agree with the interface concept. Evaluation of the interface thickness substantiates the
relative significance of this value, which is variable, depending on the chosen investigation
techniques and experimental conditions. So, this value changes in PMMA from 5.5 nm
at 170 K to 8 nm at 200 K.
Unlike PMMA, in which the filler influences the kinetics of the radical decay at T > 77 K,
the filler effect in PVP has come to light at the photoinitiation stage at 77 K [123]. The
composition of radicals generated by sample photolysis depends on the aerosil content.
The ESR spectra show the presence of two radical species: CH2C (N(CH2)3CO)CH2
( I ) and CH2CH (+N (CH2)3CO)CH2 (II). The radical (I) formation can be outlined
taking into account the photoinitiation mechanism caused by transition metal ions, as:
281
Handbook of Polymer Blends and Composites
II I + H +
kT
(6.14)
The filler effect on the efficiency of the radical (I) formation is defined by the fraction of
cation-radicals (II) stabilised at 77 K during photolysis of the samples. As few as 10% of
cation-radicals (II) are registered in high-filled ( 50% aerosil) samples, whereas this
fraction amounts to 70% in unfilled PVP [123]. A non-linear dependence of the radical
composition on the aerosol content is shown by the PVP structural modification having
the most pronounced influence on the conversion of (II) by reaction (6.14) as the filler
particles approach each other for a distance of 25 nm. This distance may be considered
as the effective thickness of the filled PVP interface.
The studies show that the friable molecular packing in interfaces is the reason for the
observed effect. The decrease of the packing density can be favourable to the decrease of
the energy barrier of the carbon atom rehybridisation in the macroradical (I). In this
case, the rate of the reaction (6.14) increases even at 77 K. A lower density of molecular
packing can also assist in the realignment of local surroundings to create the favourable
mutual orientations of PMMA macroradicals at their cage recombination.
Future research into the field of the interface structure was aimed at searching for
elementary reactions whose kinetics and mechanism reflect the morphology of the filled
polymer systems at conventional and elevated temperatures. The model elementary
reaction of thermal dissociation of the dimer of arylindandionyl radicals
(diphthaloylethane; DPE) is most useful for these purposes [124]. In the liquid phase, the
DPE free radicals system is in the equilibrium state:
DPE 2R (6.15)
The DPE thermal dissociation in the CTA filled composition obeys quite different laws
[125]. Within the entire temperature region studied, the stationary concentration of stable
radicals is achieved during a rather long time period (10-20 minutes), and the stationary
concentrations do not undergo reversible changes with temperature. The DPE
transformations in CTA proceed according to the following formal schemes. At
comparatively low temperatures (321-334 K):
[ ]
DPE R R 2R (6.16)
282
The Interface in Polymer Composites
The DPE thermal decomposition generates radical pairs. A rigid cage environment hinders
the diffusion separation of radicals. Therefore, a fraction of radical pairs decays due to
cage recombination. Radicals, which appeared in the bulk, may be involved in
recombination to yield the starting DPE as a result of diffusion contacts. As the temperature
is increased to 334-364, the linear decay of radicals in the bulk begins to predominate,
and the scheme of the process takes the following form:
At low temperatures, the stationary concentration of radicals in the filled polymer sample
exceeds the corresponding values for the unfilled CTA. However, in the region of high
temperature, a reverse dependence is observed due to the changes in the reactivity of
radicals. In a highly filled CTA (60% of aerosil), the temperature coefficient for a change
in the stationary concentration becomes negative at T > 344 K. On the contrary, at
relatively low temperatures, when the square-low decay mechanism prevails, the yield of
radicals in the filled sample significantly increases. These facts allow one to conclude
that structural modification caused by filling largely affects the stage of radical decay. As
a result, the inversion of temperature dependence is observed near the region where the
square mechanism of radical decay changes to a linear one. Thus, the temperature
dependence of the stationary concentration of radicals can be regarded as a characteristic
of the extent of structural changes in the polymer upon filling. For the highly filled CTA
containing 60% aerosil, the average distance between the filler particles was found to be
10 nm. The distances of this order correspond to the limiting value of l, at which the
interfaces begin to overlap and the entire polymer transfers to another structural state.
Thus, the kinetic regularities of radical reactions considered in filled polymers point
clearly to their dependence on the structural organisation specified by the filling degree.
Along with the investigation methods using different physical properties, such a kinetic
approach permits characterisation of the structural peculiarities of filled polymers. In
addition, the kinetic studies allow elucidation the features of the mechanism of chemical
processes in polymer composites.
The mechanical stresses acting upon a polymer composite are transmitted from the matrix
towards the filling or reinforcing material by means of the interface. The assertion may
be made that the interface regulates the mechanical properties of polymeric composites.
Thorough studies [126-133] have been developed aimed at correlating the mechanical
properties with the resistance of the bonds formed at the filling or reinforcing material/
283
Handbook of Polymer Blends and Composites
polymeric matrix interface. The mechanical properties were found as maximum when,
at the interface, the strength of the links among phases had been optimised. However, in
the absence of a quantitative approach, the nature of interfacial links cannot be wholly
understood; with a view to quantifying bonds resistance at the interface, it was assumed
that the mechanical properties of some polymer composites as filling material can be
correlated with the adhesion force, which represents the stresses necessary for separation
of components and breaking of such links.
In some silica-filled polymer systems, intermolecular forces occur between the silanolic
groups of the filler and the carbonyl groups of the polymeric matrix [134, 135]. In this
way the mechanical properties of some composites have been studied with poly(ethylene-
co-vinyl acetate) matrix filled with silica, as a function of the filling content, its previous
treatment with trimethylchlorosilane, as well as of the type of silica used. It was observed
that the utilisation of a filler with finer granulation induces higher mechanical properties,
as a result of the superior specific surface of the charge and, consequently, of a higher
number of intermolecular forces acting at the interface.
284
The Interface in Polymer Composites
All these copolymers provoke significant increases in the values of the mechanical
properties, improving at the same time the crosslinking time of the matrix of the composite.
Several types of silane coupling agents have been used for the formation of ionic
bonds at the interface: 1,2-bis(trimethoxysilyl)ethane, for increasing the degree of
crosslinking of siloxane, phenyl trimethoxysilane and N-[2-(vinylbenzylamino)-ethyl]-
3-aminopropyltrimethoxysilane (Silane H) for increasing thermal stability, silanes
functionalised with carboxyl groups as well as carboxylated polymers and their zinc
salts. Mention should be made of the fact that Silane H induces a significant
improvement of the bending strength, in the case of composites with a polyamide
matrix reinforced with treated glass fibre, compared with the untreated silica,
especially after two hours of boiling in water [136].
Table 6.8 shows that the ionic bonds improve the mechanical properties more than
in the case of utilising Silane H, for composites with a PP matrix reinforced with
surface treated glass balls.
Table 6.8 Effect of the treatment with silanes on the mechanical properties of PP
Silane system % difference versus control
Tensile Flexural Izod Charpy
strength strength
No silane control control control control
Silane H 29.9 7.3 13.1 15.1
2+
Carboxysilane/Zn /carboxy PP 40.8 9.7 -5.6 21.1
Carboxysilane/carboxy PP 20.7 1.5 13.8 29.3
Silane Da / carboxy PP 44.7 16.7 -1.3 29.3
a
1% silan, 10% carboxy PP; silan CH2=C(CH3)COO(CH2)3Si(OCH3)3
285
Handbook of Polymer Blends and Composites
Table 6.9 shows the influence of superficial modifications on the mechanical properties
of a butadiene-styrene rubber composite filled with calcium carbonate.
Carboxylic acids and their salts (lauryl sodium sulphate), as well as preformed
polymers [poly(oxyethylenylpropyltrimethoxysilane), vinyltrimethoxy-silane-
diethylfumarate copolymer, vinyltrimethoxysilane-maleic acid or anhydride
copolymer represent the most efficient agents in treatment of the mineral surfaces
of the fillers (Table 6.10). When using polybutadiene as such or polybutadiene
with final siloxanic groups, significant increase of Izod impact strength may be
noticed.
286
The Interface in Polymer Composites
Table 6.11 presents the influence exercised by various derivatives of fatty acids, utilised as
coating agents for the mineral surface, on some mechanical properties of the PP-based composites.
Most efficient were the carboxylated salts of the aliphatic compounds, containing at least
10 carbon atoms in the molecule, which, at a ratio of 7%, gives a monolayer coverage of
the filling material.
The carbon fibres may be treated through an in situ polyamidation technique, which leads
to carbon fibres coated with polyamide 66, after which the fibres are introduced into epoxy
matrices [89]. Study of the surface effect of the polyamide content on some mechanical
properties has shown that such a parameter favourably influences impact strength, has no
influence on traction strength and affects negatively, when taking high values, bending strength.
They increase with the increase of the MAP content and especially by the previous
modification of the surface of Kenaf with silanes. Increase of impact strength in this last
case is remarkable, due to the fact that the long siloxanic chains form a ductile interface,
concomitantly with the increase of adhesion between matrix and fibres.
287
Handbook of Polymer Blends and Composites
288
The Interface in Polymer Composites
Table 6.12 Results of the mechanical tests performed for various PP/Kenaf
composite mixtures
Material Kenaf Tensile Flexural Izod
fibre impact
(%) Strength, Modulus, Strength, Modulus, strength,
MPa GPa MPa GPa J/m
PP only 0 28.4 1.2 34.8 1.3 42.1
PP without 20 26.9 2.7 43.1 2.3 43.8
MAPP 40 27.1 2.8 44.9 2.7 39.9
60 27.4 3.0 47.2 3.2 33.4
PP, 20 32.7 2.9 46.3 3.0 47.6
2% MAPP 40 41.3 3.4 54.6 3.9 41.4
60 53.8 4.1 63.2 4.4 38.7
PP, 20 38.1 3.2 52.1 3.8 50.1
5% MAPP 40 49.4 4.3 59.8 4.1 43.9
60 61.2 5.1 67.3 4.6 39.2
Silylated 20 42.5 3.3 57.7 4.0 54.6
Kenaf
MAPP: Maleic anhydride grafted polypropylene
References
289
Handbook of Polymer Blends and Composites
11. B. V. Deryagin and N. A. Krotova, Doklad Akademii Nauk SSSR, 1948, 61, 849.
12. A. Ahagon and N. Gent, Journal of Polymer Science, Polymer Physics, 1975, 13,
7, 1285.
15. J. J. Bikerman, The Science of Adhesive Joints, Academic Press, New York, 1961.
17. L. H. Sharpe and H. Schonhorne, Chemical Engineering News, 1963, 15, 67.
21. T. Nakatsuka, Journal of Colloid and Interface Science, 1981, 82, 298.
23. E. Fekete, B. Pukanszki, A. Tot and, I. Beroti, Journal of Colloid and Interface
Science, 1990, 35, 200.
290
The Interface in Polymer Composites
30. W. Zhuo, J. H. Dong, K. Y. Qiu and Y. Wei, Journal of Polymer Science, Part A:
Polymer Chemistry, 1998, 36, 10, 1607.
31. J. Weis and C. Bord, Les Matriaux Composites, Volume I, CEP Publishing
House, Paris, 1983.
291
Handbook of Polymer Blends and Composites
41. H. Li, T. Chiba, T. Nigashida, Y. Yang and T. Inoue, Polymer, 1997, 38, 15, 3921.
43. A. G. Montiel, H. Keskkula and D. R. Paul, Polymer, 1997, 36, 24, 4587.
45. N. Tsubokawa, Y. Shirai and K. Hashimoto, Colloid and Polymer Science, 1995,
11, 273, 1049.
54. N. Tsubokawa and K. Seno, JMS - Pure Applied Chemistry, 1996, A33, 1, p.23.
55. N. Tsubokawa and Masaki Koschiba, JMS - Pure Applied Chemistry, 1997, A34,
12, 2509.
56. N. Tsubokawa and Shinji Hayaschi, JMS - Pure Applied Chemistry, 1995, A32,
3, 525.
292
The Interface in Polymer Composites
60. M. Popa, C. V. Oprea and G. Riess, European Polymer Journal, 1995, 31, 11,
1043.
63. N. David and S. Hon, Journal of Polymer Science, Part B: Polymer Physics,
1980, 18, 1957.
293
Handbook of Polymer Blends and Composites
81. S. Karrad, J. M. Lopez Cuesta and A. Crespy, Journal of Materials Science, 1998,
33, 2, 453.
82. H. Haidar, F. Belraoui, V. Deborde and A, Vidal, Presented at IRC 98, Paris,
France, 1998, p.73.
83. K. Cho and D. Lee, Presented at Antec 95, Boston, MA, 1995, Volume 2, p.2889.
86. R. H. Shuster, International Polymer Science and Technology, 1996, 23, 11, 9.
294
The Interface in Polymer Composites
94. K. Chen, S. Chen and X. Xu, JMS - Pure and Applied Chemistry, 1994, A29, 1, 55.
96. C. Le Deore, A. Revillon, T. Hamaide and A. Guyot, Polymer, 1993, 34, 14,
3048.
100. Yu. S. Lipatov, Physical Chemistry of Filled Polymers, Khimiya, Moscow, 1977.
101. A. Silberberg, Faraday Discussions of the Chemical Society, 1975, 59, 203.
295
Handbook of Polymer Blends and Composites
109. R. Varoqui and P. Dejardin, Journal of Chemical Physics, 1977, 66, 4395.
112. M. Desaeger and I. Verpoest, Composites Science and Technology, 1993, 48, 1-4,
215.
113. T-H. Cheng, F. R. Jones and D. Wang, Composites Science and Technology, 1993,
48, 1-4, 89.
116. H. Yoon and C. C. Han, Polymer Engineering Science, 1995, 35, 18, 1476.
296
The Interface in Polymer Composites
120. E. Ya. Davydov, A. P. Vorotnikov and D. Ya. Toptygin, Izvestiya Akademii Nauk
SSSR, Seriya Khimicheskaya, 1989, 2453.
127. K-J. M. Soderholm and S. W. Shang, Journal of Dental Research, 1993, 72, 1050.
129. J. Jancar and J. Kucera, Polymer Engineering and Science, 1990, 30, 12, 707.
132. Y. Long and R. A. Shanks, Journal of Applied Polymer Science, 1996, 61, 11,
1877.
297
Handbook of Polymer Blends and Composites
138. P. R. Hornsby and C. L. Watson, Journal of Materials Science, 1995, 30, 21,
5347.
298
7
Novel Multifunctional Epoxy Resins
7.1 Introduction
Epoxy resins have been commercially available for more than four decades and are used
in wide range of applications in the modern world [1-4].
Epoxy resins are characterised by an epoxide group consisting of an oxygen atom attached
to two connected carbon atoms:
This three membered ring is also known as oxirane or ethoxyline or epoxy group. Epoxy resins
may contain aromatic, aliphatic or cycloaliphatic back bones. The earliest member of this
family is a conventional difunctional resin: a reaction product of bisphenol-A and epichlorohydrin
called as diglycidyl ether of bisphenol-A (DGEBA) (Figure 7.1) having the following structure.
Commercially available low molecular weight DGEBA resin is a viscous liquid and generally
stable up to 200 C. The most outstanding property of this resin is its excellent adhesion,
which is due in part to secondary hydroxyl group located along the molecular chain. In
addition, cured epoxy resins have good mechanical and electrical properties, superior
dimensional stability and good resistance to heat and chemical attack. Because of this diversity,
they are used in both thermal and ambient cure applications in industries such as aircraft and
aerospace, construction, automotive, chemical, electrical and electronic, leisure, light
engineering, marine, textile, composite structures for artificial limbs, etc. Even though the
difunctional epoxy resin is suitable for all these applications, for high performance applications,
299
Handbook of Polymer Blends and Composites
resin with higher functionality are required. Due to the presence of multifunctionality, the
degree of crosslinking is higher than that of the standard difunctional resins and consequently
they possess good adhesive characteristics. Their heat distortion temperature is also more
elevated. Multifuctionality arises as a result of N-glycidyl, O-glycidyl and N- and O-glycidyl
groups in the resin structure. These are generally triglycidyl, tetraglycidyl and epoxy novalac
resins. Some multifunctional epoxy resins, which have been noted by several researchers
because of their interesting characteristics, are described in this chapter.
Epoxy resins based on tris(hydroxy phenyl)methane (Figure 7.3) are one of the important
epoxy resins used in high performance applications. This resin shows excellent physical,
electrical properties, moisture resistance, formulation stability, reactivity and retention
of properties at elevated temperature compared to those of novalac epoxy resins [8, 9].
The semisolid epoxy resin, XD 7324, OOL (Dow Chemical Company)is useful for high
performance composites and adhesives where requirements include toughness, hot/wet
strength and long-term high temperature oxidative resistance. The solid epoxy resin, XD
9053, OOL (Dow Chemical Company), is used in the semiconductor moulding powder
industries where requirements include purity, formulation stability, fast reactivity and
electrical properties over a broad temperature range.
300
Novel Multifunctional Epoxy Resins
Triglycidyl isocyanurate (TGIC) (Figure 7.4) [10] is also marketed by CIBA-GEIGY under
the trade name PT 810. It is used as curing agent for carboxylated polyesters and used in
outdoor powder coating because of its good weatherability.
301
Handbook of Polymer Blends and Composites
Nippon Kayaku Co Ltd., Japan [12] has reported the epoxy resin (Figure 7.6) which on
crosslinking gives a product having a thermal deformation temperature of 161 C and
water absorption of 1.2 wt%.
302
Novel Multifunctional Epoxy Resins
First step:
Second step:
Various modifications have been made to the structure of TGDDM (R1 & R2 = H) in
order to get new resin with improved properties. Reports are available where methylene
hydrogens have been replaced by substituting phenyl rings and/or alkyl group in the
structure of TGDDM. The data are given in Table 7.1 [13] and Table 7.2 [14].
The above resins contain on an average, less than four epoxide groups per molecule. The
functionality of methyl phenyl substituted resin is 2.68 while the others have average
functionalities between 3.04 and 3.38 per formula weight. The commercially unsubstituted
TGDDM resin has an average functionality of 3.28.
In a novel method (Figure 7.9, Method-2) [14, 15] N-glycidyl amine of higher functionality
has been prepared by reacting aniline dichlorohydrin with formaldehyde using an acid
catalyst. Table 7.2 shows the characteristics of certain resins synthesised by this route.
The process for synthesising TGDDM by method-1 is very time consuming [13]. But in
method -2 the reaction time has been reduced from 65-67 hours to 11 to 12 hours by
modifying the process and by using the easily available raw materials.
The specification of this resin synthesised by both the methods is given in the Table 7.3 [14, 16].
Patel and others have reported a modified TGDDM (Figure 7.10) structure [17, 18].
303
Handbook of Polymer Blends and Composites
304
Novel Multifunctional Epoxy Resins
H2SO4 / HCHO
Innovation and improvement in the chemical design of newer epoxy resins is still continued
by several researchers by introducing various linkages between the aromatic rings, by
increasing the number of aromatic rings, by introducing various substitutions in the aromatic
rings to improve the strength of the resin in various respects or to impart flame retardancy
by introducing halogen and/or phosphorus atoms in the aromatic ring. Some typical resins
are described next.
These resins (Figure 7.11) show high heat distortion temperature, high bending strength, impact
and tensile strength and excellent moisture resistance with good dimensional stability [19].
305
Handbook of Polymer Blends and Composites
Figure 7.11
These resins (Figure 7.12) have good water resistance properties [20].
When Y = O, R = H; 1,4-Bis(4-N,N-diglycidylaminophenoxy)benzene.
Y = O, R = CH3; 1,4-Bis(4-N,N-methyldiglycidylaminophenoxy)benzene.
Y = S, R = H; 1,4-Bis(4-N,N-diglycidylaminothiophenyl)benzene.
Y = S, R = CH3; 1,4-Bis(4-N,N-methyldiglycidylaminothiophenyl)benzene.
Figure 7.12
The epoxy compound in Figure 7.13 is used to prepare epoxy resin compositions for
prepreg. The resultant compounds have good tensile strength and moisture resistance
after curing with diaminodiphenylsulphone [21].
Figure 7.13
306
Novel Multifunctional Epoxy Resins
Recent contributions on multifuntional epoxy resin by Patel and others include resins having
the following general structure shown in Figure 7.14. These resins show very good thermal
stability, mechanical properties, chemical resistance and electrical properties [22-25].
Figure 7.14
Cured castings and composites of tetra epoxides (Figure 7.15) are found [27] to be more
moisture resistant than other epoxy resins with comparable thermal stability, modulus,
strength, processibility and toughness.
307
Handbook of Polymer Blends and Composites
Some newer tetrafunctional resins have been reported (Figure 7.16). Curing behaviour
of these resins with aromatic amines and thermal stability of the cured resins have been
reported [28, 29].
The majority of the epoxy resin formulations are considered to have flammability
characteristics ranging from slow burning to self-extinguishing. Flame retardance of
epoxy resin can be achieved mainly through the addition of halogenated compounds
or phosphorus compounds or a mixture of both classes of compounds [31-34]. The
halogen or phosphorus may be present in the resin, in the curing agent or modifier or
in an additive specially designed for epoxy resins. Direct incorporation of a flame
retarding substance into the polymer backbone during the synthesis of epoxy resin
imparts permanence fire retardance, since in that case there will be complete dispersal
of the retarding species throughout the matrix. Two such multifunctional flame
retardant resins containing phosphorus (Figures 7.17 and 7.18) have been reported
[35-36].
The flame retardancy of cured resin samples can be determined mainly from the
values of limiting oxygen index (LOI). The oxygen index (OI)m is taken as the
minimum percentage of oxygen required in a nitrogen-oxygen atmosphere,
surrounding the sample to maintain its combustion for at least 30 seconds after
ignition. A material must be considered flammable as long as the LOI (% oxygen)
value is 26 or below [37]. The oxygen index value depends upon the extent of
crosslinking, thermal stability and phosphorus content of the matrix system. LOI of
some epoxy resin composition are given in Table 7.4.
308
Novel Multifunctional Epoxy Resins
In Table 7.4 DDS cured epoxy resin system has enhanced LOI value, hence can be ranked
as one of the best flame retardant compositions [16].
Table 7.4 LOI of glass fibre reinforced phosphorus containing epoxy resin composites
Resin Systems Proportion Curing agent LOI (% oxygen)
DGEBA HTAPPO 80:20 DDM 29.4
TGBATC HTAPPO 80:20 DDM 36.5
TGBATC HTAPPO 80:20 DDS 40.3
DDM: Diaminodiphenyl methane DDS: Diaminodiphenyl sulphone
309
Handbook of Polymer Blends and Composites
A high fire resistant phosphorus containing epoxy resins (Figure 7.19) with increased
compressive strength is described in literature [38].
The novalac epoxy resins are polyglycidyl ether of novalac resin (Figure 7.21). They
have increased crosslink densities and improved thermal and chemical resistance
properties [40].
310
Novel Multifunctional Epoxy Resins
Epoxy resins having the structure in Figure 7.22 is used in the formulation of sealant for
semiconductor devices. They have improved fluidity, low expansion co-efficient, high
glass transition temperature, adhesion and low moisture absorption [42].
Figure 7.22
311
Handbook of Polymer Blends and Composites
Multifunctional epoxy resins (Figure 7.23) (n = 1-6) is also used together with certain
phenolic resin curing agent in presence of inorganic fillers and curing accelerator as
potting composition for semiconductors [43].
Figure 7.23
The composition with the above resin has low softening point, good processibility and
give cured product with good heat resistance and low water absorptivity.
Naphthol-based pentafunctional epoxy resin [45] with the structure shown in Figure
7.25 have low thermal expansion co-efficient, good adhesion properties and generally
used in the composition for sealing semiconductors.
312
Novel Multifunctional Epoxy Resins
A tetrafunctional O-glycidyl resin (Figure 7.26) has very high glass transition
temperature, low thermal expansion co-efficient and low water and methylene dichloride
adsorption [46].
This resin has epoxy equivalent weight 166 g.eq1, percentage hydrolysable chlorine
content 0.43 and number average molecular weight 772 as determined by gel permeation
chromatographic technique. The same compound has also been used for powder coating
composition; flame blistering and solder resistance epoxy resins compositions [47, 48].
Another tetrafunctional epoxy resin used for epoxy resin potting compositions for
semiconductors [49] has the structure shown in Figure 7.27.
313
Handbook of Polymer Blends and Composites
Figure 7.27
Uncured liquid epoxy resins are generally characterised by epoxy content, hydroxyl
equivalent, density, viscosity, hydrolysable chlorine content, volatile content, gel
permeation chromatography (GPC), HPLC, nuclear magnetic resonance spectroscopy
(NMR) and infra-red (IR) spectroscopic study.
The epoxy content of epoxy resin is generally expressed in terms of epoxy equivalent
weight. The epoxy equivalent is the weight of resin in grams containing one gram
equivalent of epoxy group. There are many other techniques but it is normally
determined by reacting a known quantity of the resin with hydrochloric acid and then
estimating the unreacted acid [1]. Hydrochloric acid reacts with the epoxy group
according to the equation:
Likewise, the hydroxyl equivalent is the weight in gram of the resin containing one gram
equivalent of hydroxyl group and is normally determined by reacting the resin with
acetylchloride.
Viscosity of the liquid resins can be determined with a Cannon-Fenske capillary viscometer
at 25 C or a Brookfield viscometer.
314
Novel Multifunctional Epoxy Resins
Hydrolysable chlorine content of liquid and solid epoxy resins can be determined by
dehydrohalogenation with potassium hydroxide solution at higher refluxing temperature
and titrating with standard acid solution using phenolphthalein indicator.
Melting point of the solid resins can be measured either by ring and ball technique or by
Durrans mercury method [50].
GPC is used to determine molecular weight and molecular weight distribution of the
epoxies while differences in chemical composition of resin samples can be obtained from
HPLC. This process can be used to check the batch-to-batch reproducibility of neat
epoxy resins. Separation of epoxy resins has been done in both normal phase HPLC [51]
and in reverse phase HPLC [52, 53]. This technique is used for studies relating to kinetics,
purity evaluation, etc., in the normal phase. In reverse phase, separation of molecules is
based on a combination of solubility parameters, partition coefficients and polarity with
the most water soluble component (most polar) eluting first.
C13 NMR study of some tetrafunctional epoxy resins has been studied on a JEOL FX-
90Q FT-NMR spectrophotometer [5]. Chemical shift to () were recorded in parts per
million (ppm) with tetramethyl silane as the internal standard and chloroform-deuterium
as solvent.
Most of the epoxies are not considered useful if its molecules are not crosslinked by the
process known as curing. Curing is a process in which the linear resins in presence of
proper hardener or curing agent are converted into a three-dimensional thermoset network.
In this process, resin and curing agents are mixed together, once this mixing has occurred,
a point of no return has been reached, cure begins and proceeds at a rate dependent upon
factors such as temperature and the reactivities of the resin and curing agents towards
each other. Treatment with curing agents gives intractable and insoluble thermoset
polymers. Some curing agents can react with the resin at room temperature or below,
while others require heat to effect the polymerisation. The curing agent used in the
formulation plays a major role in determining the handling properties, curing schedule
and optimum performance properties.
The curing agent, which becomes an integral part of the cured epoxy resin, has a marked
effect on its thermal stability. The proper choice of resin and curing agent depends on the
315
Handbook of Polymer Blends and Composites
application and on their characteristics such as viscosity, gel time, pot life, curing time
and temperature, mechanical, thermal, electrical properties, toxicological and
environmental limitations and cost.
The major classes of epoxy curing agents are aliphatic amines, aromatic amines, acid
anhydrides, latent curing agents and flexibilising curing agents. Among those used for
aerospace applications are mostly aromatic amines namely DDS, DDM and
metaphenylenediamine, while the aliphatic amines like diethylenetriamine (DETA) and
triethylenetetramine (TETA) are used for the formulations to be cured at room
temperature.
The aliphatic amines, aromatic amines and acid anhydrides as curing agents are briefly
discussed in the next section.
The primary aliphatic amines are used extensively as curing agents for the epoxy resins
[15, 17-28] These are the hardeners used in adhesive systems capable of curing at normal
or slightly elevated temperatures. The most widely used aliphatic curing agents are
diethylenetriamine (DETA), triethylenetetramine (TETA) and diethylaminopropylamine
(DEAPA). The curing agents which contain both primary and secondary amino groups
and also amido groups are the polyamides known as the versamides. Versamid resins
are viscous, brown liquids termed polyamides. Since the versamids are themselves curing
agents, the composition need not contain additional amine. The structure of Versamid
160 is as shown in Figure 7.28.
Incorporation of versamid curing agent into certain tetrafunctional epoxy system increases
the flexibility due to a long carbon chain between the functional groups (Table 7.6).
316
Novel Multifunctional Epoxy Resins
Versamids having a long carbon chain between functional groups, are non-hygroscopic
and less irritant, and increase the flexibility and peel strength of adhesives.
In aromatic amines, amine nitrogens are attached directly to an aromatic ring. They
are generally solid at room temperature, and used in stoichiometric amount with epoxy
resins. Most common aromatic amine curing agents like MPD, DDM [1, 2] and DDS
[15, 16] are used to cure many multifunctional epoxy resins.
Benzidine diaminodiphenylether (DDE) and diamino diphenyl sulphide are also used
as curing agents. Benzidine cured epoxies, impart quite good high-temperature properties
and extremely good radiation resistance but rarely used due to its toxicity. DDE and
diaminodiphenyl sulphide have been reported to be of reduced toxicity and increased
solubility in the resins [14].
The reaction of epoxy groups with aromatic amines is slower than that of the aliphatic
amines, owing to lower basicity, steric factors of the aromatic ring [16].
A new generation of aromatic amines are developed for high temperature resistance
and for high performance applications. Aromatic diamines such as 2,2-bis-[4-(p-
aminophenoxy)phenyl]propane (BAPP) and 1,1-bis-[4-(p-aminophenoxy)phenyl]
cyclohexane (BAPC) have been synthesised and studied extensively [54].
317
Handbook of Polymer Blends and Composites
This are also used for curing of multifunctional epoxy resins [55]. Most commonly used
anhydrides are phthalic anhydride (PA), dodecenyl succinic anhydride (DDSA) pyromellitic
dianhydride (PMDA), trimellitic anhydride (TMA) and the like. Addition of accelerator
is essential for anhydride curing.
In general, the anhydrides provide long pot lives, measured in days and months. Reactions
of the anhydride with epoxies are complex and depend on number of factors like gel
time and temperature, post cure time and temperature, presence and absence of accelerator,
types of accelerator etc. A variety of accelerators are suggested for anhydride cure. Among
them tertiary amine like benzyldimethyl and tris(dimethylaminomethyl)phenol are most
commonly used [14].
Certain anhydride curing agents are used to impart flame retardant characteristics. Thus
(i) chlorendic anhydride, (ii) tetrabromophthalic anhydride, (iii) tetrachlorophthalic
anhydride may be used in the formulations with multifunctional epoxy resins to obtain
flame retardancy [16]:
The simple combination of resin-curing agent very rarely provides a material with all the
properties required for end use applications. Due to this many other materials like diluents,
fortifiers, fillers, toughening agents are used to modify the properties of the cured product.
The final properties and eventual performance of a system depends on the correct choice
of the type and amount of different components of an epoxy resin formulation.
7.5.1 Diluents
These are generally used to reduce the viscosity of the epoxy resin systems. Diluents
may be reactive and non reactive. A reactive diluent is a low viscosity liquid that is
318
Novel Multifunctional Epoxy Resins
capable of combining chemically with resin-curing agent system. The reactive groups
are usually epoxide. Many of the reactive diluents are glycidyl ethers. Allyl and butyl
glycidyl ethers, octylene oxide, styrene oxides are evaluated as high efficiency diluents
[54]. Other reactive diluents are 1,4-butane diol diglycidyl ether, vinyl cyclohexene
dioxide, monovinyl cyclohexene oxide, epoxy pentyl ether, etc.
Plasticisers such as dibutyl phthalate, dioctyl phthalate are non-reactive diluents. Dibutyl
phthalate in particular, is widely used for liquid resins.
7.5.2 Fortifiers
These are liquids or low melting point solids which are incorporated into the resin-
curing agent mixture, prior to curing, at loading of 10-40 parts per hundred (phr)of
resin. These fortifiers are generally the reaction product of substituted aromatic amines
or amides and mono or diepoxide. Different types of fortifiers are available in literature
[56, 57]. The fortifier phenyl glycidyl ether of 4-hydroxy acetanilide (PGEHA) was
prepared [56, 57] by reacting phenyl glycidyl ether (PGE; 1 mole) and 4-hydroxy
acetanilide (HA; 1mole) using diethylamine hydrochloride (0.05 wt%) as catalyst at
110 oC for an hour. The product obtained was semi solid in nature.
Besides PGEHA, the reaction of phenyl glycidyl ether with acetoacetanilide (AA)
gives PGEAA; reactions of cresyl glycidyl ether with 4-hydroxyacetanilide to give
CGEHA are known. Vinyl cyclohexene dioxide (VCD) when reacted with resorcinol,
HA and AA produce VCDRC [58], VCDHA, and VCDAA, respectively [59].
The mechanism of the cure reaction with fortifier is rather complex because many
chemical and physical changes occur during the cure cycle [57]. It may be the fact
that all the fortifiers are more or less reactive and due to this, possibilities of side
reactions can not be ruled out.
Addition of fortifiers to the epoxy resin-curing agent systems lower down the curing
temperature at the initiation and completion with increasing the curing rate. They
also increase the strength and modulus of cured resin products. Table 7.7 shows
incorporation of fortifier enhance the mechanical properties of multifunctional epoxy
resin system.
Above improvements in the properties are gained with little increase in cost with few
processing difficulties and only a slight decrease in glass transition temperature
(Tg) [57].
319
Handbook of Polymer Blends and Composites
Table 7.7 Mechanical properties and glass transition temperature of thick films
of TGDDM with different fortifiers [58]
Epoxy System Composition Cure Cycle Tensile Flexural Glass
(C/h) Strength strength transition
kg/cm2 kg/cm2 temperature
(C)
TGDDM:
100 100 412 870 155
DETA
TGDDM-DETA:
80:20 120 470 990 137
PGEHA
TGDDM-DETA:
80:20 160 520 1120 147
VCDRC
TGDDM:DDM 100 100 978 1500 205
TGDDM-DDM:
80:20 140 1249 1680 175
PGEHA
TGDDM-DDM:
80:20 180 1271 1820 190
VCDRC
TGDDM:DDS 100 120 1125 1850 253
TGDDM-DDS:
80:20 150 1275 2125 208
PGEHA
TGDDM-DDS: 244
80:20 200 1379 1999
VCDRC
7.5.3 Fillers
Fillers play a very important role in epoxy mouldings. Fillers for epoxy resins are more
or less finely divided inorganic powders, organic or inorganic hollow beads or
continuous or chopped organic and inorganic fibres. Principally they are used for the
control of viscosity, for reducing shrinkage and coefficient of thermal expansion, for
effecting cost reduction.
Inorganic powdered fillers like metallic powders and carbon black may be used to convert
the non-conductive epoxy formulation into the electrically conductive system [60]. They
also result in a significant increase in arc resistance.
Colloidal fillers such as colloidal silica has the property of imparting thixotropy. Antimony
trioxide appears to be a particularly satisfactory filler for high-temperature adhesives,
perhaps because of a catalytic effect and to impart flame retadency [61].
320
Novel Multifunctional Epoxy Resins
Effect of filler depends on the amount than the types. Loading of filler may affect the
viscosity of the formulation. The viscosity of the system can be modified by diluents or
increasing temperature.
The heat of reaction liberated during curing is a serious limitation of the epoxies in bulk
castings. Fillers absorb a part of the heat evolved and reduce the overall shrinkage during
curing. The thermal expansion of epoxy resins can be considerably lowered by
incorporating fillers.
Various thermal analysis techniques used for analysis of epoxy resins are differential
scanning calorimetry (DSC), differential thermal analysis (DTA), thermo-mechanical
analysis, dynamic mechanical analysis, thermal gravimetric analysis (TGA) and tortional
braid analysis (TBA).
DSC is the simplest and most widely used technique for quality control and to study the
mechanisms and kinetics of curing [62, 63]. It has proved to be a valuable technique for
determining curing behaviour of epoxy resin-curing agent systems. Different curing
characteristics like the temperature at which curing reaction is initiated (Ti), peaked (TP),
completed (Tf), cure range, cure time, degree of cure, heat of reaction of exothermic
transition can be evaluated. Curing characteristics of various N-glycidyl epoxy resins
using amine curing agent have been studied by Patel and others [63].
According to the Arrhenius law, the rate of reaction depends on the temperature of the
system. The temperature dependence is measured by the activation energy, E, which is
influenced by the type of chemical reaction.
Studies on the thermal degradation of polymers are of extreme importance from a practical
point of view. It gives an idea of behaviour of polymers under high temperature service
321
Handbook of Polymer Blends and Composites
and for selecting the right kind of materials for a specific high temperature application.
TGA is used extensively [14, 23, 25] to study the degradation behaviour of polymer over
various temperature ranges. The thermograms obtained in TGA are generally analysed
in terms of percentage weight loss as a function of temperature. These thermograms are
further analysed to obtain thermal characteristics like initial decomposition temperature
(T 0), temperature for 10% weight loss (T10), temperature for maximum rate of
decomposition (Tmax), etc. To get the overall nature of the thermogram, integral procedure
decomposition temperature (IPDT) [64] and activation energy [65] are also estimated.
Thermal characteristics of TGDDM cured with various amine curing agents are given in
the Table 7.8 [14].
The structure of curing agent system has an important role in the degradation process. In
Table 7.8, DDS cured systems have maximum thermal stability.
7.7 Applications
Among the various types of the matrix materials epoxy resins are gaining much interest
because of their excellent adhesion to a variety of fillers and reinforcing agents, low
322
Novel Multifunctional Epoxy Resins
shrinkage during cure, excellent moisture and chemical resistance, good electrical
insulation properties and thermal stability and high mechanical properties. Epoxy
polymers are sometimes too viscous for use in many applications. But this can be reduced
by adding solvents reactive diluents, plasticisers, modifying agents etc.
Proper selection of the polymer functionality and its crosslinking agent, types and
amount of reinforcing agents, curing conditions allows tailoring of properties for the
crosslinked products. A wide variety of reinforcing agents are used. They may be
particles, flakes, whiskers, short fibres, continuous filaments and fibres, sheet materials
and many others [66].
Most reinforcements used in composites have a fibrous form because they are stiffer and
stronger in this form. Jute and spider silk are the natural fibres while carbon fibres,
aramid fibres are organic in nature. Synthetic inorganic fibres like glass, silicon carbide,
silicon nitride, alumina are also used.
The properties of the composites depend on the interface between fibre and matrix because
interfacial bonding play very important role in this purpose. In case of glass fibre reinforced
composite this bonding can be improved by adding certain adhesion promoters also
known as coupling agent. Coupling agents contain organic functional groups capable of
co-reacting with the matrix. These groups can exert a strong physical-chemical interaction
with the hydroxyl groups of the glass surface. The surface of the most reinforcing fibres
is hydrophilic. Therefore the water present on the surface prevent physical or chemical
adsorption of the matrix molecules. Coupling agents disperse the water which may
accumulate at the interface and destroy the bond. -Aminopropyl triethoxysilane
(NH2(CH2)3 Si(OEt)3) is one of the extensively used coupling agent for epoxides [67].
The orientation of the reinforcing fibres is of equal importance with respect to the structure
of the composites. To achieve maximum strength in a specific direction it is necessary to
orient the fibre in the same direction. When all the fibres are arranged in parallel way
known as unidirectional. Here maximum strength is obtained in one direction. Generally
this type of fibres are used for making rods and bars. In bidirectional fibres half of the
strands are laid at right angle to the other half and the strength in any one direction is
less than the unidirectional fibre. These fibres are used in structural applications.
Multidirectional fibres give the lowest strength [66, 68].
323
Handbook of Polymer Blends and Composites
For making polymer matrix composites, glass fibres are most commonly used reinforcing
agent. Different compositions of mineral glasses are used for making fibres. The basic
ingredient of glass is silica (SiO2). Proportion of silica varies from 52-65% in different
glasses. Oxides of sodium, calcium, boron, aluminium, iron etc. are also present. Thus
three important types of glass are E-glass, S-glass and C-glass. E-glass has good strength,
stiffness, electrical and weather resistant properties. For fibres reinforced composites
this type is used. S-glass has higher Youngs Modulus. It is expensive compared to E-
glass. C-glass has good chemical resistant properties. At room temperature in bulk state,
glasses are hard, brittle and rigid. But when drawn in the form of fibre are flexible,
strong, defect free having coherent structure. Glass fibres are used as reinforcement for
epoxy, polyester and phenolic resins. The strength of glass fibres depends on the processing
conditions and the environment at which it is tested.
Carbon fibres are another family of fibres used as reinforcing agent. These are the most
widely used reinforcing fibres in advanced composites. Although the manufacturing
technology for carbon fibres are complex, but are more amenable to large scale production
than other fibres used for advanced composite. Advance composites are those having
superior performance compared to conventional structural materials such as steel and
aluminium alloys. Carbon fibres have very useful engineering properties [66].
Carbon fibres with extremely high modulus can be prepared by carbonisation of organic
precursor fibres followed by graphitisation at very high temperature (>2500 C). Generally
the special textile polymeric fibres (which can be carbonised without melting ) are used
as precursor. Most widely used precursor fibre is polyacrylonitrile. Polyimide, phenolic
resin, polyvinyl alcohol, cellulose, pitches can also be used [68].
The density of the carbon fibre varies in the range of 1.6-2.0 g.cm-3. Varieties of carbon
fibres are available. Two major types of carbon fibres are (i) The high tensile strength but
medium Youngs Modulus (HT) fibres (200-300 GPa) and (ii) high Youngs Modulus
fibre (400 GPa). Carbon fibre reinforced polymer matrix composites are mainly used in
the defence related aerospace industries and in the sporting goods industries. Epoxy
resin is the most commonly used matrix with carbon fibre. Polyimides, polysulphones,
polyether ether ketones, polyesters and certain thermoplastics are also used [68].
Fabrication methods like hand lay up technique, filament winding, vacuum moulding,
sheet moulding, pultrusion are used for making carbon fibre composites. The properties
of these composites like other composites depend on the interfacial bond strength. Carbon
fibre is an inert material. As a result strong adhesion between carbon fibre and polymer
matrix is difficult. To increase the adhesion properties of carbon fibres the surfaces are
made rough by different processes. Oxidation or etching process in an acid increase
surface area and improve the wettability of the surface which increase the interlaminar
shear strength.
324
Novel Multifunctional Epoxy Resins
Glass fibre reinforced composites (GFRC) using epoxy resins have higher mechanical
strength, chemical and mechanical resistance than other thermosetting resins. Open mould
techniques and closed mould techniques are used for their fabrication. The use of GFRC
have gained importance progressively in railway engineering, building and constructions,
automotive industries, marine applications, aerospace industries and sports applications
[66, 67]
Soni and others have prepared some glass fibre composites using tetrafunctional epoxy
resins. The physical and mechanical properties of these composites are shown in the
Table 7.9.
Table 7.9 Some physical and mechanical properties of glass fibre reinforced
composites [15]
Properties TGDDM-DDM TGDDM-DDS
Fibre/resin content % (v/v) 70:30 70.5:29.5
3
Density (g/cm ) 1.85 1.83
Void content % (v/v) 3.05 3.68
2
Tensile strength (kg/cm ) 3800 4309
2
Flexural Strength (kg/cm ) 4906 5800
Interlaminar shear strength (kg/cm2) 162 173
For high performance composite materials carbon fibres are used as the main reinforcing
fibres. High strength and stiffness, light weight, improved fatigue resistance, good friction
and wear qualities, low thermal expansion and thermal and electrical conductivity combine
to make carbon fibre composites an attractive substitute for various metals, special alloys
and other materials. These performance characteristics in concern with composite structure,
offer a degree of versatility in design and fabrication that is unavailable in other materials.
Carbon fibre reinforced composites are used in aircraft and aerospace industries , military
aircraft, commercial aircraft, helicopters, missiles, spacecrafts, general aviation craft.
The use of carbon fibre in non-military application is limited because of the high cost of
the fibre as well as the fabrication. They are used in automotive, sports, civil engineering
such as industrial roller system, compressed natural gas tanks [69].
Properties of composite depends upon a number of factors such as the nature of the
resins, curing agents, modifiers, the relative amounts of resin and fibre present in the
325
Handbook of Polymer Blends and Composites
composite and the orientation of reinforcing agents. Table 7.10 shows the
physicomechanical properties of carbon fibre reinforced epoxy composites with different
fibre orientations [15].
References
1. H. Lee and K. Neville, Handbook of Epoxy Resins, McGraw Hill, New York,
1967.
3. Waterborne & Solvent Based Epoxies and Their End User Applications, Volume
2, Eds., G. Bianchini, R. Dyer, A. Fream, P. Heffer, S.P. McOwan, P. Oldring, I.
Royston and N. Tuck, John Wiley and Sons, London, UK, 1997, 2, 8.
326
Novel Multifunctional Epoxy Resins
10. Encyclopedia of Polymer Science & Engineering, 2nd Edition, Volume 6, Eds., H.
Mark, N. Bikales, C. Overberger, G. Menges and J. I. Kroschwitz, Wiley
Interscience Publication, New York, 1986.
16. S. R. Patel, Synthesis and Application of N-glycidyl Epoxy Resins, Sardar Patel
University, Vallabh Vidyanagar, Gujarat, India, 1992, [Ph.D. Thesis].
17. S. R. Patel and R. G. Patel, Journal of Thermal Analysis, 1993, 39, 229.
18. S. R. Patel, V. S. Patel and R. G. Patel, Thermochimica Acta, 1991, 182, 319.
327
Handbook of Polymer Blends and Composites
20. S. Suzue, S. Honda and H. Kondo, inventors; Toray Industries Inc., assignee;
JP62146916A2, 1987.
21. S. Suzue, S. Honda and H. Morimoto, inventors; Toray Industries Inc., assignee;
JP62146917A2, 1987.
22. R.H. Patel and R.G. Patel, Thermochimica Acta, 1990, 160, 323.
23. R.H. Patel and R.G. Patel, Thermochimica Acta, 1990, 173, 95.
24. K. G. Amin, K. J. Patel and R. G. Patel, Iranian Polymer Journal, 1997, 6, 227.
33. K. Katsuura and N. Inagaki, Journal of Applied Polymer Science, 1980, 22, 679.
34. A. Granzow and C. Savidas, Journal of Applied Polymer Science, 1980, 25, 2195.
328
Novel Multifunctional Epoxy Resins
40. T. F. Bradley and H. A. Newey, inventors; Shell Development Co., assignee; US,
2,716,009, 1955.
48. Y. Nemoto and A. Ito, inventors; Mitsubishi Petrochemicals Co. Ltd., assignee;
JP2041316, 1990.
49. N. Mogi and S. Narusa, inventors; Sumitomo Bakelite Co. Ltd., assignee;
JP03195723A2, 1991.
329
Handbook of Polymer Blends and Composites
54. R.H. Patel, Studies on Modified Epoxy Resins, Sardar Patel University, Vallabh
Vidyanagar, Gujarat, India, 1990. [Ph.D. Thesis]
55. S.R. Patel and R.G. Patel, Thermochemica Acta, 1992, 202, 97.
56. J. Daly, A. Britten and A. Garton, Journal of Applied Polymer Science, 1984, 29,
1403.
57. P.D. Mclean, R.F. Scott and A. Garton, British Polymer Journal, 1983, 15, 66.
58. H.K. Soni, R. G. Patel and V.S. Patel, Polymer for Advanced Technologies, 1993,
4, 509.
59. R.H. Patel and R.G. Patel, Die Angewandte Makromolekulare Chemie, 1988,
161, 89.
61. C.S. Swaffer, inventor; Imperial Chemical Industries Ltd., assignee; GB884578,
1961.
62. R.D. Patel, R.G. Patel and V.S. Patel, British Polymer Journal, 1987, 19, 37.
63. H.K. Soni, V.S. Patel and R.G. Patel, Thermochemica Acta, 1992, 196, 327.
66. M.S. Bhatnagar, Polymer Plastics Technology Engineering, 1993, 32, 53.
67. D. Feldman, Polymeric Building Materials, Elsevier Applied Science, London &
New York, 1989.
330
Novel Multifunctional Epoxy Resins
70. T. Shiobara and H. Shimizu, inventors; Shin-Etsu Chemical Industry Co. Ltd.,
assignee; JP04114056A2, 1992.
331
8
Flame Retardant Polyester Resins
Polymers play a paramount role in our day-to-day life being increasingly utilised in a
number of areas such as textile fabrics, home furnishing, domestic and industrial buildings,
appliances, transportation vehicles, etc. Flammability of polymeric materials is a problem
of primary importance as polymer systems become more sophisticated and performance
requirements become more stringent. Application of chemical substances to reduce the
fire hazard of materials has been known for a long time [1]. Manufacturers and consumers
pay serious attention to the flammability properties of polymeric materials not because
of the rapid expansion in the utilisation of synthetic materials, but mainly because of the
involvement of governments in establishing flammability standards. A wide variety of
techniques has been used to reduce the flammability of polymeric materials. The most
important are:
Flame retardants are divided into two groups: inert and reactive. The inert flame retardants
(additive-type) are the leading retardants in terms of production volume of plastics. Some
important flame retardants in the production of plastics are aluminium hydroxide,
antimony trioxide, non-halogenated phosphates, halogenated phosphates, chlorinated
hydrocarbons, bromine-containing substances, boron-containing substances etc. Such
additive type flame retardants are rarely introduced into the macromolecular structure
of polymeric material but introduced at the stage of when polymeric materials are
processed into articles without changing the technology of material manufacturing.
333
Handbook of Polymer Blends and Composites
Most of inorganic flame retardants are non volatile and form low-toxic gases during
decomposition and are more available and cheaper compared to organic one. Of the
halogen-containing compounds that belong to the inert flame retardant, the most widely
used are the inexpensive chlorine-containing low-molecular-weight aliphatic
hydrocarbons, used in combination with the Sb2O3 and other antimony compounds.
One of the new trends in the reduction of polymer material flammability is the use of
metallic complexes with organic ligands, containing phosphorus, nitrogen, oxygen and
other atoms with unshared electrons as flame retardants.
Fillers used to reduce the flammability of polymer materials are generally of three types,
inorganic-type fillers, noncombustible thermostable organic fillers and modified organic
fillers. Fillers can be dispersed with granular particles (sand, chalk, kaolin, etc.), flaked
particles (graphite, mica, talc, etc.) or fibrous particles (glass fibre, asbestos, etc.) or
porous particles (glass microspheres, vermiculite, perlite, etc.). In the majority of cases
the inorganic fillers are used. Since fillers are introduced into the polymer in vast quantities
(>20%), their presence reduces the relative fuel component content of the material [2].
Fire proof coatings can be divided into two groups (i) coatings ignitable with difficulty and
incombustible coatings and (ii) heat-insulating in combustible coatings. The first group includes
various polymer coatings of the paint and varnish type, and sheets and films that are less
combustible than the objects protected while the second includes coatings that reduce the
rate and the depth of heating through the protected surface under the effect of a flame.
Incombustible polymers are of two types (i) polymers that give off little or no fuel gases
during decomposition, e.g., inorganic and heterocyclic polymers such as polyphosphazines,
polysiloxane, etc. [3, 4], and organic polymers that liberate primarily incombustible gases
during decomposition such as polytetrafluoroethylene, polyperfluoro-alkylenetriazines,
nitrosofluorocarbon elastomers, etc. [5] (ii) polymers that do not decompose completely
as they form larger quantity of carbonised residue such as polyphenylenes, polyimides,
polybenzimidazoles, polyquinoxalines, polybenzimidazo-pyrrolones, etc.
334
Flame Retardant Polyester Resins
8.1.1 Polyesters
In the last few decades, the polyester resins have gained increasing importance due to a
wide range of applications in practically all major industries such as paints, varnishes,
textile fibres [6], printing inks, adhesives [7], wire insulation, moulding and packaging
materials [8], building materials [9], laminates, coating and composites [10], production
of boats, bath, shower units, automobile parts, railway carriage panels [11]. The
importance of thermotropic aromatic polyesters as a new class of materials is ever
increasing because in their liquid-crystalline state they can be spun or injection molded,
leading to oriented fibres or plastics of thermally stable high strength [12, 13], useful in
electrical and electronic parts and heat resistance container [14], photographic film and
magnetic recording tape [15].
The number of literature and patent references reveals the growing interest in imparting
flame resistance to polyesters, particularly for use in reinforced plastics. Among the
methods used to reduce the rate of combustion in polyester compositions are the addition
of non reactive inorganic and/or organic flame retardant additives, flame retardant
components in either the acid (or anhydride) or diol moiety of the molecule, and flame
retardant vinyl monomer or crosslinking agents.
Polyester fibres, fabrics and films are surface treated with reactive flame retardants by
soaking them in the flame retardant and subsequent thermal treatment during which the
flame retardant undergo either polycondensation or polymerisation reactions and their
solubility in water or organic solvent falls sharply. Fabrics designed for use as decorative
and upholstery materials and also carpet fabrics made of polyester fibres are treated
with latex of chlorine-containing polymers (polychloroprene, polyvinylidinechloride, vinyl
chloride copolymer).
Flame retardants protect polymers by either denying oxygen access to the burning polymer
or by poisoning the flames. Aluminum hydroxide (Al(OH)3) sets free water which douses
the flame. The inorganic fillers used to impart flame retardancy are generally alum,
antimony trioxide, borax, chalk, magnesium oxide or silica. These fillers are usually
335
Handbook of Polymer Blends and Composites
cheap and in many cases help to improve the mechanical properties and surface
appearance of the fabricated article. The use of these fillers is restricted to applications
whose opacity is not a problem. Fillers such as chalk and silica when used in high
percentage give only slight reduction in burning rate compared to borax and alum
which seem to be more effective. Antimony trioxide and magnesium oxide have little
effect on flammability and a halogen must be added with it for effective flame
retardancy. Approximately 25% chlorine or 15% bromine is required to obtain a
30.0 limiting oxygen index (LOI). The behaviour of a polymer in a fire is often
characterised by the LOI of the polymer. This index is defined as that volume fraction
of oxygen in a mixture of O2 and N2 which just allows a polymer to burn. Material
with LOI <22.5% are usually called flame retardent; those with LOI >27%, self-
extinguishing. For example, LOI are 14% for poly(oxymethylene), 17% for
poly(ethylene), 20% for cellulose, 25% for wool, 32% for poly(vinyl chloride), and
95% for poly(tetrafluoroethylene) [16].
Waddel and others [17] have reported that the oxygen index values have been increased
to 33.4 and 38.7 by the addition of 1 and 3 wt% antimony trioxide, respectively,
along with 25 wt% chlorine. While the same degree of oxygen index is obtained by
changing the ratio of chlorine:antimony trioxide to 20:1 and 12.5:2.5 (wt%). Under
conditions of combustion, volatile antimony oxyhalides are formed which blanket
the flame.
336
Flame Retardant Polyester Resins
Studies have been carried out to evaluate the fire-retardant properties of zinc
hydroxystannate relative to those of antimony trioxide in brominated polyester resin
(10% & 28% bromine) containing inorganic fillers such as ATH, Sb2O3, talc and CaCO3
[11]. It was observed that the effectiveness of ZnSn(OH)6 and Sb2O3 as flame retardants
depend on several factors including their incorporation level, the bromine content of the
resin, the chemical nature of the filler, and the filler loading. In this study Sb2O3 gave
optimum performance in ATH-filled resins, whereas ZnSn(OH)6 exhibited a marked
flame retardant synergism with anhydrous Al2O3. Calcium carbonate is detrimental to
flame retardancy, talc and anhydrous alumina are reasonably effective. Limiting oxygen
index values are given in Table 8.3.
337
Handbook of Polymer Blends and Composites
Table 8.2 Oxygen index and smoke generation data of Atalac 711-05A resins
with flame retardants
Oxygen index Smoke
Flame retardant
% generation %
None 33.0 8.4
1% Sb2O3 34.5 14.0
3% Sb2O3 42.0 16.2
5% MoO3 36.0 5.4
7% MoO3 39.0 4.3
1% Sb2O3 plus 1% MoO3 39.5 3.9
1% Sb2O3 plus 2% MoO3 40.0 7.1
8% Alumina trihydrate plus 2% MoO3 40.0 1.2
8% Alumina trihydrate plus 2% ammonium
38.5 1.0
dimolybdate
Table 8.3 Effect of inorganic synergists on the flammability of unfilled and filled
polyester resins
LOI values
Filler
Synergist (3 phr)
None Sb2O3 ZnSn(OH)6
(100 phr)
10% Br 28% Br 10% Br 28% Br 10% Br 28% Br
None 24.4 32.6 32.0 41.0 46.0 53.5
ATH 38.2 44.5 41.8 66.4 65.9 67.1
Al2O3 32.3 33.5 33.0 43.8 53.5 59.6
Talc 31.4 38.8 34.5 54.6 57.1 54.6
CaCO3 - - - 29.3 37.6 32.6
ATH: aluminia trihydrate
phr: parts per hundred
Calcium sulphate dihydrate [24] is another hydrated salt that has been used to confer
flame retardancy. This works by blanketing the flame with an inert gas (water vapour),
but also by an endothermic decomposition of the calcium sulphate dihydrate.
338
Flame Retardant Polyester Resins
Recently Nyacol Nano Technology, Inc., [25] offers a non aqueous colloidal antimony
pentoxide, Sb2O5, dispersion as a synergist with halogenated flame retardents for use in
unsaturated polyester and vinyl ester resins, e.g., NYACOL, APE1540 and NYACOLAB40
flame retardant additives. These are dispersions of nano-sized (colloidal) antimony pentoxide
particles (40% by wt.) in a liquid halogen-containing polyester (50% by wt.) and vinyl ester
resins. The oxide particles are nearly 1/100 the size of powdered antimony trioxide. APE1540
is used in high temperature cure system using cobalt-MEKP type cure systems. AB40 can be
used in high temperature cure systems or with disubstituted peroxide cure systems. AP1540
or AB40 can be substituted for antimony trioxide in formulations involving any polyester or
vinyl ester resin system containing bromine or chlorine. Advantages of antimony pentoxide
over a conventional antimony trioxide system are:
Better penetration into the structure for improved dimensional consistency and no
spongy laminate feel.
The dispersions can be added to the resin with simple stirring. They will not settle
and concentrate at the bottom of the container. In filament winding operations,
APE1540 assures more reliable operation without settling in the resin trough.
Boeing [26] reported that a degree of flame retardancy can be achieved in polyesters by
incorporating sodium silicate with asbestos powder, lithopone, kaolin, diatomaceous
earth talc, silica gel, cement, zinc oxide, pumice, calcium sulphate or various pigments.
The addition of red phosphorus to polyethylene terephthalate (PET) has been reported to
decrease its flammability [27]. The LOI values were increased; for 0, 4, and 12% phosphorus
the LOI numbers were 20.4, 27.5 and 32.6, respectively. Tsuyoshi and others [28] have
339
Handbook of Polymer Blends and Composites
reported a red phosphorus flame retardant and nonflammable resinous composition where
a fine red phosphorus powder is subjected to surface modification treatment.
Other inorganic materials suggested as flame retardants for polyesters [29-31] are
ammonium sulphate, ammonium hypophosphate, ammonium dibasic phosphate,
ammonium borate, ammonium biphosphate, ammonium chloride, potassium carbonate,
potassium silicate, aluminum chloride, sodium borate, sodium silicate, sodium arsenate,
sodium acetate, sodium tungstate, calcium chloride, zinc phosphate, zinc borate, zinc
oxide, magnesium hydroxide.
A number of organic compounds containing halogen and phosphorus are used as additives
to impart flame retardancy to polyesters. Phosphorus-containing flame retardents oxidise
in fires to non-volatile phosphorus oxides which form a glass-like coating on the surface
of the substrate; oxygen can no longer access the polymer. Phosphorus oxides also react
with water to give phosphoric acids which catalyse the splitting-off of water. Bromine or
chlorine-containing organic compounds generate halogen radicals which combine with
radicals produced by the burning polymer; this action interrupts the kinetic chain and
poisons the flame [16].
Chlorinated paraffins have been used to impart flame resistance to plastic compositions
including polyesters. They are quite effective, primarily by release of hydrogen chloride
at combustion temperature and are still prevalent today owing to good overall economics.
As a class, they have limited thermal stability and are not recommended for temperatures
above 175 C. The Chlorowaxes are chiefly used in polyvinyl chloride (PVC), some
unsaturated polyesters and some polyolefins.
340
Flame Retardant Polyester Resins
Williams [32] reported flame retardant polyesters using various derivatives of halogenated
bisphenol-A (Figure 8.2).
Triphenylphosphine oxide has been evaluated [35] as a flame retardant for PET. The
compound was found to be effective in an oxygen atmosphere but not in a nitrous oxide
341
Handbook of Polymer Blends and Composites
atmosphere. These results were typical of flame retardants acting in the vapour phase.
Deshpande and others [36] have also reported that when triphenylphosphine oxide was
used as an additive, it may simply volatilise and function in the vapour phase. However,
when it was part of the polymer backbone, volatilisation may not have been completely
possible, and then, in some way, the retardant cloud enhanced the residue formation.
Some other halogen and phosphorus containing organic compounds that have been
considered as flame retardants for polyesters are:
Most halogenated flame retardants are used in conjuction with antimony compounds
(antimony oxide, antimony oxychloride or organic antimony compounds) in a true
synergistic relationship. This significantly reduces the total amount of flame retardant
needed to meet requirements of flame retardancy. Some of these combinations are as follows:
342
Flame Retardant Polyester Resins
The thermal properties of polymers and blends determine the effect of flammability
reduction in polyester materials to a large degree. This is why polyetheylene terephthalate
(PETP) fibres with the addition of 10 mol.% hexabromobenzene, 0.5 wt% Sb2O3, and
0.5 wt% TiO2 have a higher LOI (28.1%) than when the hexabromobenzene is substituted
by an equal amount of decabromodiphenyl (LOI = 22.6%) [59].
Inclusion of the flame proofing element into the monomer portion of the polyester
has come under intensive study. Among the elements used in the preparation of reactive
flame retardant components, phosphorus and halogens have been most widely used,
either singly or in combinations with other elements, such as antimony. A large number
of halogen-containing, reactive diols, polyols, anhydrides and other functional group
containing intermediates are known. Among these the tetrachloro- and tetrabromo-
phthalic anhydrides [60] and especially chlorendic acid and anhydride are the
important ones. The use of the latter in self-extinguishing polyester resin systems has
been described by numerous investigators [61, 62, 63]. Palm and others [64] and
Ohse and Cherdron [65] described flame resistant polyesters based on dichloromethyl
-lactones. Polymers containing the CCl3 group are most effective in imparting flame
retardant characteristic properties to polymeric systems. Unsaturated polyesters
containing the trichloromethyl group, polychloroester, has been prepared using 3,3,3-
trichloropropylene oxide (TCPO) [66, 67, 68]. Other halogen containing alkane diols,
343
Handbook of Polymer Blends and Composites
such as chloropropane diols, mono-, di-, and trichlorobutane diols [69], and
chloromethyl propanediol have received attention. In order to confer flame retardancy
to an unsaturated polyester formed by reaction of difunctional acids and
pentaerythritol dichloride and pentraerythritol trichloride, its fire-resistance laminates
having good mechanical properties [70].
Another diol based on aromatic diphenols has been used for the preparation of flame
resistance polyesters. As mention earlier, the tetrabromobisphenol-A is also an
important intermediate to form several speciality brominated flame retardants. The
bis(hydroxy ethyl ether) of tetrabromobisphenol-A formed by reaction with 2 moles
of ethylene oxide, is a true diol because the two primary hydroxyl groups are much
more reactive than the hydroxyls attached to the benzene ring. This diol can be reacted
with a dibasic acid to form a linear polyester:
Jacson and others [71] described photochemical chlorination where chlorine was
substituted on the gem-dimethyl groups of the polyterephthalate and polycarbonate
of 2,2,4,4-tetramethyl-1,3-cyclobutanediol, 2,2-dimethyl-1,3-propanediol and the
polycarbonate of 4,4-isopropylidenediphenol (bisphenol-A). They reported that
polyesters are self-extinguishing when it contained about 28-30% chlorine (three
chlorine atom per polyester unit).
344
Flame Retardant Polyester Resins
and Stepien [73] using hexachloro-endo methylene tetrahydrophthalic acid. The Diels-
Alder adduct of cis-4-cyclohexene-1,2-dicarboxylic anhydride with hexachlorocyclo-
pentadiene, 2,3-dicarboxy-5,8-endomethylene-5,6,7,8,9,9-hexachloro-1-
2,3,4,4a,5,8,8a-octahydronaphthalene anhydride (Figure 8.4), has been incorporated
as the fire-retardant component in polyester compositions [74].
Figure 8.4
345
Handbook of Polymer Blends and Composites
Diol derive from tetrachloro-m-xylene [90] Phosphorus containing aromatic diol [91]
Phosphorus containing aromatic diols [92- Phosphorus containing diacetate of diol [95]
95]
346
Flame Retardant Polyester Resins
An intensive study has been made on inclusion of flame proofing element into the monomer
portion of the polyesters. Substitution of flame retardant element for portion of the styrene
is also feasible but is economical. Naturally, the halogen-containing styrenes have been
investigated as monomers with unsaturated polyesters. For example, Prince [97] reported
comparative study of ortho-bromo styrene and para-bromostyrene flame retardancy of
polyester at low and high flame temperature. It is shown that the first isomer is more
effective at the lower temperature and vice versa at the higher temperature. Dibromostyrene
is claimed as a fire retardant for polyesters [98] and is marketed in the UK [99]. It is
described as a mixture of 2,4- and 2,5-dibromostyrenes which is extremely reactive. For
development of its full self-extinguishing characteristics, the manufacturer recommends
that 5% antimony oxide be utilised with 6% in the finished resin. A typical formulation is:
Great Lakes Chemical, Europe has recently introduced polybrominated styrene flame
retardants, PBS-64 and PBS-64HW. PBS-64HW is a higher molecular weight version of
PBS-64. The products have a higher flame retardant efficiency than an existing product.
PBS-64 is effective as a flame retardant in high-temperature polymers, including polyamides
and polyesters, where good thermal stability, high flow, and colour are important.
347
Handbook of Polymer Blends and Composites
electrical properties are at a premium. Mitsuto and others [102] reported the flame
retardant polyester having brominated aromatic ring in the side chain
(polytribromostyrene) as flame retardant, with improved elongation and impact strength.
Yosho [103] prepared fire and heat resistance polyester compositions with excellent
moldability using rubber modified polystyrene and poly(pentabromobenzyl acrylate),
and are useful for electric and machinery parts.
[110]
348
Flame Retardant Polyester Resins
[111]
[112]
The global market for flame retardants remains divided along differences in regional fire
safety, health, and environmental standards in the US, Europe, and Japan. Suppliers still
anticipate growing conversion to halogen-free flame retardant systems. Such systems are
increasingly specified by multinational equipment makers motivated by legal and voluntary
restrictions. Phosphorus-type additive meets a progressive non-halogen technology for
flame retardants. Phosphorus flame retardants activate in the condensed or solid phase
in the polymer and break down into phosphoric acid, causing plastic to char. The protective
char insulates the unburned plastic from the flame and heat source and suppresses volatile
gases. These additives are seen as less toxic than halogenated systems, but are viewed as
less efficient and lower performing versus brominates.
Machio and others [114] of Mitsubishi Chemical Corporation Japan, have used phosphate
(1-40 parts), melamine cyanurate (1-40 parts), and other compounds with polyester
resin to get fire resistance composition. Yashida and others [115] of Asahi Chemical
Industries have developed fire resistance halogen free polyester composition.
Flame retardant and halogen free-flame retardant finishing are available for polyester-
based synthetic fibre materials [116]. Nicca Chemical Company, Japan has reported the
reaction product of phosphorus oxychloride and phenyl phosphoric dichloride with
349
Handbook of Polymer Blends and Composites
R1-14 = H, alkyl
Toshio and others [118, 119] reported fire-, heat- and hydrolysis-resistance polymer
composition with good mechanical properties using liquid crystalline polyester,
phosphorus compound (Figure 8.8), melamine cyanurate and glass fibre.
Certain additives used in flame retardant polyester resin compositions are halogen-free.
Tin additives such as zinc hydroxy stannate and zinc stannate are used both in halogenated
and halogen-free polyester resin composition [120]. Basic aluminum oxalate (BAO) high-
temperature, non-halogenated flame retardant filler is based on alumina trihydroxyde.
This can be used in polyesters processed up to 330 C without degrading its flame
retardancy.
A number of firms use zinc borate which show self extinguishing formulations. In systems
of this type, partial replacement of antimony oxide at 1:1 ratio is used as the point of
departure. US Borax reports that Firebrake 2B (zinc borate) is non-toxic by ingestion or
350
Flame Retardant Polyester Resins
dermal application. It is not a primary skin irritant or corrosive material and it is not an
eye irritant. Magnesium hydroxide is also used [121] as a halogen free additive. It is a
smoke suppressant.
A non-halogenated flame retardant system, Exolit AP740TP for polyester gel coats protects
plastics against flames and also reduces smoke density and heat release. This halogen
free material is white, free-flowing powder that mix easily with gel coats. The gel coat
begins to foam when exposed to heat.The heat insulating capacity enables a carbon-
foam layer to develop in the gel coat to protect the polymer layer below.
8.1.8 Applications
Polyester resins have gained increasing importance due to wide range of applications
(Section 8.1.1) and serious attention has been taken to the flammability properties of
these materials. Polyester fibres with reduced combustibility is achieved using reactive
flame retardant during synthesis, by including additive-type admixtures in the polymer
melt before processing, or by special treatment of the product, e.g., flame retardant spun
bond nonwoven textiles comprises sandwiching of a polyester fibre containing
phosphorous with other thermoplastic polymer at a temperature 30 C lower than the
melting point of the thermoplastic polymer and then applying high pressure [122, 46].
The fire resistance polyester fibres are manufactured by treatment of polyester fibres
with solution containing [P-(NH2)2:N]m (m = 3-6) and heating at 100 C . The fibres
are useful for safety nets at construction sites, nets used in sports, etc., [123]. The wash
fast fire proofing agents containing brominated cycloalkanes (1,2,5,6,9,10-
hexabromocyclododecane) finely dispersed in water by using 1 self emulsifiable polyester
and are used to finish polyester fabrics [124].
Copolyesters with reduced flammability have been derived through the use of halogen-
containing diols or phosphorus-containing diols or by using dibasic acids and their esters.
The introduction of flame retardants into polyester melts makes it possible to prepare
materials that are more resistance to different treatments (laundering, dry cleaning) than
materials that have been coated by the same flame retardant [125].
A wide variety of flame retardant polyester resin compositions are reported. Some of
them are reviewed here.
351
Handbook of Polymer Blends and Composites
Jun [127] has reported fire retardant polyester potting composition useful for solenoids,
sensors, etc., composed of polyester (PET), fillers, other polymers and 5-25% brominated
fire-retarding agent (Pyrochek 68PB, Pyrochek LM).
Tetsuya [128] has reported laminated substrate material for printed wiring board showing
good adhesion, chemical resistance, bending resistance and fire resistance using
compositions of polyester, brominated phenol, novalak epoxy resin, polyisocyanurate,
tertiary amine and Sb2O3. Hitoshi [129], reported non-halogenated fire resistance resin
composition for electric cable. Koji and others [130, 131] have used non-halogenated
fire resistance resin compositions for electric and automotive parts using polyester,
phenoxy resins and red phosphorus.
The casting of large parts such as septic tanks, bath tubs and sinks and flat sections
simulating marble.
To produce cylindrically shaped items with high burst strengths. Railway tank cars
with a capacity of over 22,000 gallons. Filament wound reinforced plastic gasoline
storage tanks.
Filament wound pipes for the oil industry and rocket casing for military use.
Auto head lamp mountings, fender extensions, window frames and hood scoops.
Formulations of sheet moulding compounds (SMC) which are used for manufacturing
exterior automobile parts such as fenders, doors, tailgates and similar exterior car
body parts which have a surface appearance resembling a finished sheet metal surface.
Corrosion resistant industrial equipment, piping, ducts, blowers and chemical resistant
lining in chemical plants.
352
Flame Retardant Polyester Resins
Corrosion resistant floorings made of resins mixed with concrete are used in
chemical plants.
Household equipment such as mirror-frames, air coolers and washing machine bodies
Decorative small inner garden made of fibre reinforced polyester decorations such as
artificial rocks, lanterns and bamboo leaves as interior displays of hotels and
restaurants.
Chokes (Blasts) used in tube light fittings filled with unsaturated polyester resin.
There are a large number of fire tests, which may be divided into various categories. The
sub division of fire tests are:
Fire test
Some flammability test methods generally used for polyester are shown in Table 8.5.
353
Handbook of Polymer Blends and Composites
354
Flame Retardant Polyester Resins
References
5. H. J. Fabris and J. G. Sommer, Rubber Chemistry and Technology, 1977, 50, 3, 523.
355
Handbook of Polymer Blends and Composites
10. T. Weng, A. Hiltner and E. Baer, Journal of Macromolecular Science A, 1989, 26,
1, 273.
11. R. S. Bains, P. A. Cusack and A. W. Monk, European Polymer Journal, 1990, 26,
11, 1221.
18. R.C. Nametz, Industrial & Engineering Chemistry, 1967, 59, 5, 99.
21. ASTM D635, 1990, Standard test method for rate of burning and/or extent and
time of burning of self supporting plastics in a horizontal position.
356
Flame Retardant Polyester Resins
25. Nyacol APE1540 and Nyacol AB-40, Nyacol Nano Techonolgy Co.,Inc.,
Ashland, MA, USA, www.nyacol.com/ape1540-ab40-1-98.htm.
34. M. Ravey, Journal of Polymer Science: Polymer Chemistry, 1983, 21, 2, 375.
36. A. B. Deshpande, E.M. Pearce, H.S. Yoon and R. Liepins, Journal of Applied
Polymer Science, Applied Polymer Symposia, 1977, 31, 257.
42. R. Sckell, inventor; Chemische Fabrik Kalk GmbH, assignee; DE1191569, 1965.
357
Handbook of Polymer Blends and Composites
46. N. Hiroshi, inventor; Nikka Chemical Ind. Co., Ltd., assignee; JP7070924, 1995.
47. I. Hiroaki , inventor; Inoue Senko KK, Japan, assignee; JP8001077, 1996.
48. J. R. Leebrick and J. J. Cassidy, inventors; M and T Chemical Co., Inc., assignee;
BE627662, 1963.
51. B. T. Hayes, W. J. Read and L. H. Vaughan, Chemistry & Industry, 1957, 1162.
52. R.S. Cooper, inventor; Diamond Alkali Co., assignee; US2664411, 1953.
54. D. Miture and A. Noryuki, inventors; Teijin Ltd., assignee; JP0959495, 1997. check
57. C. J. Gebhart and G. E. Gleim, inventors; The Goodyear Tire & Rubber Co.,
assignee; DE2428533, 1975.
60. Y.C. Chae, W. M. Rinehart, C.S. Shull, R.A. Cass and R.J. Rohrbacker, Presented
at the 22nd Annual Technical Conference, SPI Reinforced Plastics Division,
Washington, D.C., 1967.
358
Flame Retardant Polyester Resins
61. P. Robitschek and C. T. Bean, Industrial & Engineering Chemistry, Design &
Process Development, 1954, 46, 8, 1628.
62. P. Robitschek and C.T. Bean, inventors; Hooker Electrochemical Co., assignee;
US 2,779,700, 1957.
63. P. Robitschek and C.T. Bean, inventors; Hooker Electrochemical Co., assignee;
US 2,779,701, 1957.
65. H. Ohse and H. Cherdron, Die Makromoleckulare Chemie, 1966, 95, 283.
76. P. L. Kinson and C. M. Orlando, Journal of Applied Polymer Science, 1979, 23,
1, 155.
77. M. Borr, K. E. MacPhee, and M. Kulda, inventors; Dominion Rubber Co. Ltd.,
assignee; CA681805, 1964.
359
Handbook of Polymer Blends and Composites
80. J.E. Dereich and A. Riihemaki, inventors; Diamond Alkali Co., assignee;
GB792368, 1962.
81. B.J. Bremmer, inventor; Dow Chemical Co., assignee; US2162615, 1964.
83. M. Borr and K.E. MacPhee, inventors; Dominion Rubber Co. Ltd., assignee;
CA663542, 1963.
84. M. Borr and K.E. MacPhee, inventors; United States Rubber Co. Ltd., assignee;
DE1149899 1965.
87. M. Wismer, E. E. Parker and R. E. Park, inventors; Pittsburgh Plate Glass Co.,
assignee; US3060146, 1962.
88. M. Wismer, E. E. Parker and R. E. Park, inventors; Pittsburgh Plate Glass Co.,
assignee; GB912944, 1962.
89. M. Wismer, E. E. Parker and R. E. Park, inventors; Pittsburgh Plate Glass Co.,
assignee; DE1143200, 1962.
91. H. Kondo, M. Sato and M. Yokoyama, European Polymer Journal, 1981, 17, 6,
583.
92. M. Sumie, M. Tetsuo and M. Nobuhiro, inventors; Nippon Ester Co. Ltd.,
assignee; JP61106657A2, 1986.
93. M. Tetsuo, U. Atsuko, I. Takayuki and T. Keizo, inventors; Japan Ester Co. Ltd.,
assignee; JP 02 11,623, 1990.
360
Flame Retardant Polyester Resins
94. S. Osami, K. Minoru, S. Junichi and F. Eiji, inventors; Unitika Ltd., assignee; JP
04 08,758, 1992.
95. M. Tetsuo, U. Atsuko, I. Takayuki and T. Keizo, inventors; Japan Ester Co. Ltd.,
assignee; JP0211624, 1990.
97. M. Prins, G. Marom and M. Levy, Journal of Applied Polymer Science, 1976, 20,
11, 2971.
98. L. Williams, inventor; Peter Spence and Sons Ltd., assignee; GB986634, 1965.
99. Dibromostyrene - Its Uses and A Flame Retardant Agent for Polyester Resin,
Technical Bulletin, Peter Spence and Sons Ltd., England.
104. P.E. Hoch, inventor; Hooker Chemical Corp., assignee; US3154591, 1964.
105. B.W. Nordlander and W.E. Cass, Journal of the American Chemical Society 1947,
69, 2679.
106. A. Toy and L.V. Brown, inventors; Victor Chemical Works, assignee; US2586884,
1952.
107. T. Okada, N. Tonami and Y. Miyake, inventors; Toyo Katsis Industries Inc.,
assignee; JP13697, 1960.
110. E. C. Hurdis, J. F. Petras and H. Romeyn, inventors; United State Rubber Co.,
assignee; US2891985, 1959.
361
Handbook of Polymer Blends and Composites
113. N.G. William, Modern Plastics International, McGraw-Hill Co., 1999, 84.
116. J. Sano, K. Ichihashi, H. Saito, and H. Banko, inventors; Nicca Chemical Co.
Ltd., and Kanebo Ltd., assignees; US 5,614,573, 1997.
122. N. Nobuo, M. Atsushi, and Y. Yasuhiro, inventors; Unitika Ltd., Japan, assignee;
JP0913254, 1997.
124. N. Hiroshi, Inventor; Nikka Chemical Ind. Co. Ltd., assignee; JP0770924, 1995.
125. E.L. Lawton and C. J. Setzer in Flame Retardant Polymeric Materials, Eds., M.
Lewin et.al., Plenum Press, New York, 1975, Chapter 4, 193.
126. D. Mitsure and A. Noryuki, inventors; Teijin Ltd., Japan, assignee; JP0959459,
1997.
362
Flame Retardant Polyester Resins
128. T. Tetsuya, inventor; Nitto Denko Corporation, Japan, assignee; JP08130368, 1996.
130. Y. Koji, M. Hideki and I. Hiromitsu, inventors; Toray Industries, Inc., Japan,
assignee; JP154307, 2000
131. Y. Koji, M. Hideki and K. Akinori, inventors; Toray Industries, Inc., Japan,
assignee; JP169681, 2000
132. ISO 871, 1996, Plastic - Determination of ignition temperature using a hot air
furnace.
133. ASTM D1929, 1977, Test method for the ignition of plastics.
134. ISO 1182, 1990, Fire tests, building materials, non-combustibility tests.
135. ISO 4589, 1996, Plastics Determination of burning behaviour by oxygen index,
Part-1, Guidance.
136. ISO 4589, 1996, Plastics Determination of burning behaviour by oxygen index,
Part-2, Ambient temperature test.
137. ASTM D2863, Measuring the minimum oxygen concentration to support candle-
like combustion of plastics (oxygen index)
138. ISO 4589, 1996, Plastics Determination of burning behaviour by oxygen index,
Part-3, Elevated temperature test.
139. ISO 12992, 1995, Plastics Vertical flame spread determination for film and
sheet.
141. BS 2782, Part, 1 Method 140A, 1992, Determination of the burning behaviour
of horizontal and vertical specimens in contact with a small flame ignition source.
142. UL94, Test for the flammability of plastics materials for parts in devices and
appliances, Underwriters Laboratories USA.
363
Handbook of Polymer Blends and Composites
145. BS 2782, Part-1, Method 140B, 1993, Determination of the burning behaviour of
flexible specimens in contact with a small flame ignition source.
146. ASTM D4804, Test method for determining the flammability of non-rigid
plastics.
148. IEC 707, 1981, Determination of the flammability of solid electrical insulating
materials when exposed to an igniting source.
149. ASTM D5025, Specification for a laboratory burner used for small scale burning
tests on plastics materials.
150. A.J. Hammerl, SPI Division of Reinforced Plastics, 17th Annual Technical
Conference, Proceedings Section 12h, Chicago, 1962.
151. ASTM D2863, 1970, Flammability of plastics using the oxygen index method.
153. ASTM E662, 1979, Test for specific optical density of smoke generated by solid
materials.
154. NES 711, 1981, Determination of smoke index of the products of combustion
from small specimens of materials.
364
9
Cure Kinetics of Vinyl Ester Resins
M. J. M. Abadie
Introduction
The need for composite materials destined for chemical engineering or for marine
applications is increasingly evident, in particular for corrosive environments. In recent
years, numerous attempts have been made to obtain a suitable matrix for these materials.
Initial work concerned the most classic and economic matrix such as unsaturated polyester
(UP), standard polyesters (ethylene glycol/terephthalic acid) and bisphenol-A fumarate
type polyesters. For all these matrices, the results showed a low stability of these materials
after cure towards ageing. Standard cured polyesters are well known for their weak chemical
resistance in particularly towards hydrolysis, and the epoxies by the difficulty of their
processing caused by the high viscosity, long cure cycle and the toxicity of some hardeners
[1]. On the other hand, cured polyesters of the bisphenol-A fumarate type have been accepted
for a long time for their good chemical resistance, but their mechanical performance is
poor, being similar to that of the standard polyesters, thus limiting their use [2].
Many chemical processes around the world today use chemicals that are very aggressive
to the traditional materials used for construction of chemical plants, such as steel, titanium
and other metals. Such processes include, amongst many others, metal refining, pulp and
paper production, chlorine production and flue gas desulphurisation.
Vinyl ester resins have been found to overcome many of this drawbacks and can be
fabricated with the same processes as used with conventional polyesters. Recent reviews
[3, 4] reveal that they are becoming very important in new industrial applications such
as coatings, printed circuit boards, metal foil laminates, building materials, ship building,
automotive parts and fibre reinforced composites [5]. Such developments are not only
due to the good chemical properties of the cured resins but also to the oportunities
offered to the fabricator to combine the mechanical properties of epoxies with the easy
processing in use of unsaturated polyesters. Furthermore, their low molecular weight in
comparison with that of unsaturated polyesters allow them to incorporate a high ratio
of fillers (60 to 70% by weight) and also good wetting of fibres [6].
For 34 years, vinyl ester resins have been used, in glass reinforced plastics (GRP) or more
generally in fibre reinforced plastics (FRP), to combat the corrosive nature of the chemicals
365
Handbook of Polymer Blends and Composites
found in these processes. The successful use of a variety of composite products has been
demonstrated in hundreds of different chemical environments. Such chemicals include
strong acids, bases, solvents, oxidising agents, hypochlorite as well as very hot gasses.
Often these environments are associated with high temperatures and high mechanical
forces, e.g., pressure, vacuum or wind loadings.
FRP has proved itself to be one of the most cost effective materials of construction for
process plant engineering. FRP combines corrosion resistance and high temperature
performance with acceptable investment costs compared to other solutions such as rubber
lined carbon steel and exotic metals such as stainless steels, high nickel alloys and titanium.
Epoxy vinyl ester resins have established a pre-eminent position as the resins of choice in
composites for the construction of chemical plant due to their excellent resistance to a
wide range of chemicals and to their outstanding combination of thermal and mechanical
properties. These very positive properties are the result of the molecular structure of the
resins and is also strongly dependent on the catalyst system used (% of each component,
environment, etc.) which should be dosed carefully.
9.1 History
Although various types of vinyl ester resins were prepared in laboratory-scale quantities
during the late 1950s, it was not until 1965 that they were commercially introduced first
by Shell Chemical Company under the trade name of EPOCRYL resins and then in 1966
Dow Chemical Company introduced DERAKANE resins. These resins were identified
as epoxy (meth)acrylates and were shown to provide chemical resistance superior to that
of the premium (chemical-resistant) unsaturated polyesters. In 1977, Interplastic
Corporation and Reichhold Chemical Company introduced vinyl ester resins under the
CoRezyn and Corrolite and later NORPOL DION product designations, respectively.
At the same period, DSM coined the ATLAC vinyl ester designation.
Before curing, vinyl ester resins are thermoplastic and consist of a polymer backbone
with an acrylate (R = H) or methacrylate (R=CH3) termination R-[-O-C(=O)-C(R)=CH2]2
with R derived from an polyester resin, epoxy resin or urethane resin. Although vinyl
ester resins have sometimes been classified as polyesters, they are typically diesters that
(depending on the polymer backbone) contain recurring ether linkages.
After curing, vinyl ester resins are thermosetting resins forming a three-dimensional network
structure but those based on epoxide resins are of particular commercial significance.
366
Cure Kinetics of Vinyl Ester Resins
The epoxy vinyl ester resins are mainly a backbone of epoxide, resulting from the of
polycondensation reaction of bisphenol-A and epichlorohydrin and terminated at both
ends by acrylate or methacrylate functions; they are named bisphenol-A epoxy-based
acrylate or methacrylate vinyl ester resin (see Figure 9.1), i.e., Derakane 411 (Dow
Chemical Company). Some of them may be produced from phenol resins which have
been modified by acrylic or methacrylic acid and which are called phenolic-novolac
epoxy-based vinyl ester resin (see Figure 9.2), i.e., Derakane 470 (Dow Chemical
Company). When the bisphenol-A structure contains bromide atoms they are called
terabromo bisphenol-A epoxy-based vinyl ester (see Figure 9.3), i.e., Derakane 510
(Dow Chemical Company).
Epoxy vinyl ester resins can be produced from almost any epoxy resin depending on
the properties needed. The bisphenol-A based grades have good all-round performance,
balancing excellent mechanical properties with good thermal resistance. Novolac-based
epoxy resins will give higher thermal resistance, tetrabromo bisphenol-A will improve
the fire resistance due to the bromine content, or even a rubber modified epoxy resin
can be used to give high impact resistance.
Figure 9.1 Bisphenol-A epoxy-based (meth)acrylate vinyl ester resin Derakane 411
from The Dow Chemical Company
367
Handbook of Polymer Blends and Composites
The non epoxy vinyl ether resins have a polymer backbone of either a polyester
chain formed by the condensation reaction of fumaric acid and bisphenol-A (see
Figure 9.4), i.e., ATLAC 382 (DSM) or by sequences of urethane groups coupled to
Figure 9.4 Bisphenol-A fumaric acid-based polyester ATLAC 382 from DSM
368
Cure Kinetics of Vinyl Ester Resins
Epoxy vinyl ester resins are based on epoxy resins, which are modified in such a way that
they can be cured by the same method as traditional UP resins, i.e., via free radical
mechanism with styrene as a co-curing monomer.
An epoxy vinyl ester resin chemistry will take as its basis the structure of a typical
bisphenol-A type.
One mole of epoxy resin (see Figure 9.6) usually obtained by he polycondensation of
bisphenol-A and epichlorohydrin in presence of sodium hydroxide - is reacted, in the
presence of catalysts, by the addition of ethylenically unsaturated carboxylic acids with
two moles of methacrylic acid to give an epoxy vinyl ester resin.
Methacrylic acid
369
Handbook of Polymer Blends and Composites
The acid groups react with the epoxy groups to form ester linkages, leaving two reactive
vinyl bonds at the terminal position of each molecule (see Figure 9.7). The reaction of
acid addition to the epoxide ring (esterification) is exothermic and produces a hydroxyl
group without the formation of by-products, e.g., as in polyesterifications where water
is produced. Hence the name epoxy vinyl ester. It should be noted here that the hydroxyl
group formed along the backbone of the chain is responsible for the good adhesion
properties of the epoxy vinyl ester resin due to the presence of hydrogen bonds. Therefore
aramid or carbon fibres may be used with epoxy vinyl ester resins.
Appropriate diluents and polymerisation inhibitors are added during or after esterification.
The epoxy vinyl ester resin is then dissolved in styrene (30 to 50 wt%). Styrene monomer
offers the advantages of low cost, desirable reactivity and many years of formulating
experience in the field. In fact the solvent is also considered as a coreactant in the
curing reaction due to its double bond which takes part in the crosslinking reaction.
While styrene is the most common monomer used to dilute vinyl ester resins, other
monomers which have been used include, divinyl benzene, vinyl toluene, chlorostyrene,
alpha-methylstyrene, t-butyl styrene or dicyclopentadiene (see Table 9.1).
370
Cure Kinetics of Vinyl Ester Resins
The effect of styrene content on the corrosion resistance of vinyl ester resins has been
reported [3]. In contrast to polyester resins, vinyl ester resins have good corrosion
resistance at low monomer levels.
Vinyl toluene diluted vinyl ester resins were developed for use in electrical applications.
These resins were shown to have less water absorption, higher flexural strength and
electrical strength retention versus heat ageing than a conventional polyester resin diluted
in styrene. Vinyl toluene diluted vinyl ester resins offer the advantages of easer
processability (especially in matched metal die moulding) and lower cost than epoxy
resins in electrical applications. Vinyl toluene is less volatile and has a higher flash point
than styrene. The former property is important in view of the push for lower monomer
emissions during fabrication and processing.
Dicyclopentadiene acrylate-diluted vinyl ester resins have the advantage of longer shelf life
than conventional styrene-diluted vinyl ester resins. Dicyclopentadiene acrylate is more
compatible with vinyl ester resins than styrene, which is advantageous since higher levels
of dicyclopentadiene acrylate are required to produce comparable viscosities. A disadvantage
of dicyclopentadiene acrylate monomer is that it yields resins which are more brittle.
The catalytic system consists of two parts: an initiator called catalyst and an activator called
promotor, each of them having a specific role in the curing reaction and being dependent.
371
Handbook of Polymer Blends and Composites
The catalyst is a peroxide and the choice of peroxide catalysts is determined by the
particular resin in question and the temperature at which it is to be cured. Generally
methyl ethyl ketone peroxide (MEKP) is used for room temperature curing and benzoyl
peroxide (BPO) or t-butyl perbenzoate is used for elevated temperature curing. MEKP
(60% MEKP in dimethyl phthalate) is generally used in a concentration varying between
1 to 5 wt% for room temperature curing.
The activator used is generally a cobalt salt (cobalt naphthenate or preferably cobalt
octoate due to its higher purity) used at a concentration of 0.2 - 0.3 wt%. It reacts
according the following redox scheme:
where only the alcoxy radicals RO formed during the oxidation of the cobalt salt
are active in radical polymerisation whereas the peroxide radicals ROO are engaged
in chain peroxidation reactions generating peroxide ROOH which are recycled
according to:
Note that BPO with N,N-dimethyl aniline (DMA) as an accelerator may also be used for
curing vinyl ester resins at room temperature.
Thus the room temperature cure can be effected without the application of external
heat. DMA or other aromatic tertiary amines may also be used to further accelerate the
MEKP curing system. MEKP with a high dimer content is more reactive than low dimer
MEKP in curing vinyl ester resins, while the opposite is true with polyester resins.
Criteria for determining the catalyst system and judging the degree of cure include exotherm
temperature, residual monomer, physical properties, working time and development of
hardness or tack-free state. The exotherm temperature should be high enough to cure the
resin but not so high as to cause cracking of the resins. This is particularly important in
cast parts of large mass and thickness where the heat dissipation will be slower. Residual
monomer will also deleteriously affect physical properties and corrosion resitance.
372
Cure Kinetics of Vinyl Ester Resins
Oxygen in the air will inhibit the complete cure of an exposed vinyl ester resin
surface. This inhibited surface may vary in depth, and may result in reduced
weatherability, poor chemical resitance and/or premature failure. This problem is
overcome by preventing or reducing contact of the curing surface with air (see
infusion technique).
9.4.4 Additives
The ideal inhibitor would give infinite shelf life while not interfering with, or
retarding, the peroxide-catalysed cure of the vinyl ester resin. Frequently the catalyst
system and the curing temperature for a particular application will determine the
choice of inhibitors as with polyester resins. Periodic aeration of vinyl ester resin
has also been found to be helpful in extending the shelf life.
The superior properties of epoxy vinyl ester resins compared to UP resins can be readily
appreciated by comparing the structures of the resins. The epoxy vinyl ester resins,
when cured, have only terminal crosslinking sites which means that the whole length
of he molecule is available, and able to, stretch and uncoil when stressed. Thus epoxy
vinyl ester resin based composite mouldings are tough and resistant to thermal and
mechanical shock. On the other hand, the UP resins crosslink along the length of the
entire chain, e.g., in Figure 9.8 there would be between four and seven crosslink points
on each backbone molecule.
373
Handbook of Polymer Blends and Composites
This high level of crosslinking leads to the more brittle nature of the cured resin. Also
due to the molecular shielding that occurs on the backbone, many of the double bonds
are unable to react at all during the crosslinking reaction. These unreacted double
bonds are potential sites for chemical attack and breakdown during the lifetime of the
FRP structure. The epoxy vinyl ester resins also have a highly aromatic structure which
gives excellent thermal resistance whereas the UP resins are mainly aliphatic in nature,
which leads to only moderate thermal performance.
The secondary hydroxyl groups on the backbone of the epoxy vinyl ester resins
(produced by the epoxy/acid reaction) also assist in producing composites with very
good mechanical performance. These hydroxyl groups have an affinity to the hydroxyl
groups at the surface of glass fibres. Thus, during composite production, the resin wets
out the glass fibre very well, producing a composite with very good mechanical
performance, both static and dynamic.
The corrosion resistance of resins is also controlled by the chemical structure of its
backbone. The UP resins, due to the relatively large number of ester groups, are
susceptible to attack by hydrolysis. Epoxy vinyl ester resins, with only two ester groups
per molecule (for resins that begin life as difunctional epoxies), are inherently more
resistant to chemical attack.
374
Cure Kinetics of Vinyl Ester Resins
kd1
ROOH + Co ++
RO + HO- + Co+++
kd2
where kd1 and kd2 are the rate coefficients for the formation of the alkoxy RO and
peroxy ROO radicals, respectively.
The alkoxy radical being much more reactive to ethylenic monomers than the peroxy
radical [7], one can therefore describe the rate of initiation by:
Assuming through the consumption and regeneration of Co++ species, that a steady state
is achieved, the equilibrium concentration of Co++ attained can be given by:
[Co++]eq=kd2[Co++]0/(kd1+kd2) (9.2)
The substitution of the term [Co++]eq in equation (9.2) by that given in equation (9.1) gives:
d[RO]/dt=kd1kd2[Co++]0[ROOH]/(kd1+kd2) (9.3)
Alternatively, the rate of polymerisation of the monomer is given by the relationship [8]:
where [M] and [Mn] are the monomer and radical concentrations, respectively, f the
initiation efficiency, Ri the rate of initiation which was shown in equation (9.1), and kp
and kt are the propagation and termination rate coefficients, respectively.
The combination of equation (9.3) and equation (9.4) indicates that the rate of
polymerisation should increase with increased cobalt and MEKP levels (equation
(9.5)).
375
Handbook of Polymer Blends and Composites
-d[M]/dt=kp[M](fkd1kd2[Co++]0[ROOH]/kt(kd1+kd2))1/2 (9.5)
Commercial resins are supplied containing a phenolic stabiliser which, while present
only in small amount, leads to an observable induction period before polymerisation.
This inhibitor, Z, is added to scavenge adventitious peroxy and alkoxy radicals (R*),
and that the rate of the inhibitor radical (RZ*) is not to promote polymerisation.
kz
R* + Z RZ* (9.6)
The relationship between induction period, inhibitor concentration and the rate of
initiation can be derived as follow [9]. The radical concentration during the induction
period is given by:
d[R*]/dt=Rikz[Z][R*] (9.7)
[R*]=Ri/kz[Z] (9.8)
from which an induction time can be derived in which Z0 is the initial inhibitor
concentration:
Z0
t= ++
(9.10)
k d [Co ]0 [ROOH ]
9.7 Kinetics
Kinetics of polymerisation and crosslinking reactions have been studied thanks to the use of
the Trombomat apparatus (from PRODEMAT SA, France) which is described in detail [10,
11, 12]. A mobile system, constituted by a bar terminated by a calibrated ball (diameter, =
16 mm), is immersed into the resin and vibrated by a pendular movement (see Figure 9.9). As
376
Cure Kinetics of Vinyl Ester Resins
the reaction system proceeds, the viscosity increases as a consequence of increase in molecular
weight and the amplitude of the pendulum decreases. Forces on the ball are proportional to
the viscosity of the medium. This technique shows the sharp increase of viscosity (see Figure
9.10) which may correspond to gelation when vitrification of the medium occurs after gelation.
On the other hand, one can follow simultaneously the state change of a reaction medium
from a liquid state to a gel state and thereafter to an infusible and insoluble solid. The second
phenomenon is the exotherm engendered by the crosslinking reaction. The progress of the
reaction can be evaluated by the quantification of parameters such as:
gel time (tg), determined by the curve of viscosity versus time (see Figure 9.10),
reactivity time (tr), determined by the curve of temperature versus time (see
Figure 9.11),
polymerisation time (tp), related to the above times by the following relationship
(see Figure 9.11):
tp = tg + tr
377
Handbook of Polymer Blends and Composites
378
Cure Kinetics of Vinyl Ester Resins
All these parameters are strongly influenced by sample volume and surface area. The
system has been optimised for 130 g and a resin density of 1.3, the optimal surface area
is 60 cm2 (r = 4.38 cm) [13]. All experiments are conducted in air atmosphere. Data from
this technique are shown in Figures 9.12, 9.13, 9.14, 9.15, 9.16 and 9.17.
Figure 9.12 Gel time (viscosity technique) versus wt% of catalyst (MEKP) measured at
25C (initial temperature) for different cobalt salt concentrations
Figure 9.13 Viscosity versus time for epoxy vinyl ester and unsaturated polyester resins
379
Handbook of Polymer Blends and Composites
Figure 9.14 Temperature versus time for epoxy vinyl ester and unsaturated
polyester resins
Figure 9.15 Gel time (viscosity technique) versus wt% of activator (cobalt salt)
measured at 25C (initial temperature) for 1.5 wt% of catalyst (MEKP)
380
Cure Kinetics of Vinyl Ester Resins
Figure 9.16 Heat flow (DSC data) versus temperature for different wt% of activator
(cobalt salt) using 1.5 wt% of catalyst (MEKP)
Figure 9.17 Gel time (viscosity technique) versus wt% of catalyst (MEKP) measured at
different initial temperatures and for 0.2 wt% of cobalt salt
381
Handbook of Polymer Blends and Composites
The Derakane 411-45 (The Dow Chemical Company) has been accelerated and catalysed
at different concentrations of cobalt salt (NL51P) and organic peroxide (Butanox LPT)
which varied for the activator from 0.15 to 0.30 wt% and for the catalyst from 1.5 to
3.0 wt%.
Figure 9.12 shows the variation of the gel time versus catalyst weight percentage for
different cobalt levels. It is observed that, for a given wt% of activator, the gel time
decreases when the concentration of the catalyst increases. This is in agreement with the
prediction of equation (9.5).
In fact, an increase in catalyst level results in a faster polymerisation rate and can give
a network rich in residual double bonds, the catalyst not being totally consumed. The
presence of this unreacted catalyst could have significant consequences for the final
properties of the material [14]. On the other hand, a slow reaction rate which
corresponds to a low level of catalyst would increase the role of inhibition of oxygen
and therefore affect the copolymerisation reaction and ultimately modify the final
structure of the network.
It is interesting to note that, in the same experimental conditions (wt% of same catalyst
and same activator), the gel time of an epoxy vinyl ester resin is higher that the one observed
with a standard unsaturated polyester due to the different amount of double bonds along
the backbone of the chemical structure (see Figure 9.13). The same reasons explain why
UP resins are more exothermic than epoxy vinyl ester resins (see Figure 9.14).
In order to investigate the effect of activator level, the Derakane 411-45 was cured at a
range of cobalt concentrations varying from 0.1 wt% to 0.9 wt% by weight of the resin
and at a constant concentration of catalyst of 1.5 % by weight.
Figure 9.15 shows that increasing the cobalt salt concentration in the range 0.2 - 0.5
wt% reduces the gel time. Above this range and up to 0.9 wt% the gel time continues to
increase reaching a plateau. This was confirmed also by the differential scanning
calorimetry (DSC) results (see Figure 9.16), where the maximum in exotherm shifts to
higher temperatures after the value of 0.5 is reached. However, it is clear from these
results that cobalt species play a double role. They accelerate the reaction in its initial
stages encompassing the gelation process which is in agreement with our results in the
range 0.1-0.5 wt% and which is also in accord with equation (5), but at higher conversion
rates, they act as a retarder.
382
Cure Kinetics of Vinyl Ester Resins
High concentrations of transition element catalysts are well known to inhibit free radical
reactions. This could be explained by reactions such as:
where initiating and propagating radical chain can be consumed by such a reaction with
Co++ [13].
The Derakane 411-45 (45 wt% of styrene) activated at 0.2 wt% of cobalt salt and
catalysed at different ratios of MEKP has been studied at different initial temperatures:
15 C, 20 C and 25 C.
Figure 9.17 shows that the gel time of the resin depends strongly on the initial
temperature at which it is performed. The increase in temperature has the effect of
decreasing the gel time.
On the other hand, for a given increase in temperature, the exotherm peak increases
(see Figure 9.18) whereas the polymerisation time (tp) decreases (see Table 9.2), i.e.,
the reaction is more exothermic. Therefore, for 3% of peroxide and 0.2% of cobalt,
the temperature achieved is 85 C for a reaction starting at 15 C whereas it goes up to
120 C when used at 25 C. These results are similar to those obtained for unsaturated
polyester [9, 10].
A regression analysis of the gel time data was performed and a comparison of actual and
predicted times for 25 C is given in Figure 9.19. This analysis showed that the
dependencies are approximately 3/2 order in accelerator and first order in catalyst
concentrations. The experimental error in gel time is about 15%. For each temperature
a proportionality constant (equivalent to a rate coefficient) can be determined. Figure
9.20 gives a fit of the reciprocal of gel time with the product [ROOH]*[Co]3/2 at 25 C.
Similar fits at 15 C and 20 C then allows an estimation of the activation energy from
the slope of the arrhenius plot shown in Figure 9.21. This analysis gave an activation
energy of 64 kJ K-1 mol-1 which is similar to that expected for catalysed decomposition of
the peroxide initiator [8]. A statistical analysis of all the data fitted to the three variables,
catalyst, cobalt and temperature gave a correlation of 0.96.
383
Handbook of Polymer Blends and Composites
Figure 9.18 Exotherm peak versus wt% of catalyst (MEKP) determined at different
initial temperatures for 0.2 wt% of cobalt salt (thermal device)
384
Cure Kinetics of Vinyl Ester Resins
Figure 9.19 Actual and predicted gel times at 25C for various catalyst concentrations
(MEKP) for different wt% of activator (cobalt salt) : a, b, c & d of 0.15 wt%, 0.2
wt%, 0.25 wt% & 0.3 wt%, respectively
Figure 9.20 Rate coefficient determination from the fit of gel time to the product of
catalyst (MEKP) and activator (cobalt salt) at 25C
385
Handbook of Polymer Blends and Composites
A DSC (Dupont 912) was used to measure the exotherm rate during polymerisation.
DSC was also used to obtain the total heat of reaction by heating to 200 C. Residual
heats of reaction were obtained by heating cured specimens to 200 C. All the reactions
were conducted in aluminium pans under a static air atmosphere. Sample weights were
10 to 20 mg with an empty pan as a reference. Samples (resin in bulk) were heated from
room temperature up to 200 C in a dynamic mode with a heating rate of 10 K/min to
detect the residual reactivity. A time, to, is defined as the time where just after the last
adjuvant (catalyst) is added, the resin is put in the pan, sealed and then heated. The
average total heat of the reaction measured at this time and at different concentrations
of catalyst was 356.9 20.5 J/g. This enthalpy was used as a reference and considered to
correspond to 100% of polymerisation (no residual reactivity). On the other hand, this
value was approximately the same as this given by Suziki and others [15] for the
copolymerisation of the blend styrene/methyl methacrylate, 67.3 kJ/mol (360 J/g). It
should also be noted that it was independent of the added catalyst level.
The amount of residual reactivity [16], mainly unreacted styrene, determined at time t
is given by the relationship:
where Ht is the heat of reaction measured at time, t and Ho the heat measured at time, to.
386
Cure Kinetics of Vinyl Ester Resins
An example of the effect of catalyst concentration on the residual reactivity after cure is
shown in Figure 9.22. It can be seen that the amount of residual reactivity decreased
with increasing amount of catalyst up to 2.5 - 3 % by weight of the resin. Within the
concentration range 2.5 - 3%, the residual reactivity level after two months at room
temperature lay in the range 8.5 to 6.4%. However, 2.5% of catalyst should permit a
reasonable cure.
Figure 9.22 Residual chemical functions as function of wt% of catalyst (MEKP) for
0.2 wt% of activator (cobalt salt). Reactions conducted at initial temperature of 20C
(DSC data)
In recent years throughout Europe and North America, there has been growing concern
over the emission of styrene monomer into the workplace which has led to the
introduction of legislation intended to reduce workers exposure to styrene. The
387
Handbook of Polymer Blends and Composites
workplace concentration limits have been set in European countries and the USA as
10-20 ppm (15 minutes).
The effects of styrene on human health have been extensively studied. At very high
concentrations styrene can produce, in certain individuals, effects similar to those seen
with alcohol, i.e., dizziness, nausea and fatigue. When exposure stops there is complete
recovery, with no evidence of cumulative effects. Regarding the potential long-term effects
of styrene exposure, the majority of data supported the position that based on current
knowledge, styrene does not warrant regulation as a human carcinogen.
For many years paraffin wax has been used as an additive to standard laminating resins
for the final resin rich top-coat layer on GRP fabrications. These mixtures are applied in
order to prevent cure inhibition due to atmospheric oxygen, thereby promoting a more
thorough cure in the laminate. This is achieved by the formation of a wax rich layer
which reduces the penetration of oxygen into the resin surface. At the same time, this
wax barrier also reduces the amount of styrene emitted into the atmosphere. However,
these simple resin/wax mixtures are not stable and the phases separate with time. Also,
the high concentration of wax formed at the surface reduces the secondary bonding
strength of any further laminates applied over the top of this wax top-coat. Low styrene
emission (LSE) resins are not a new idea. Most LSE resins available today use the principle
of paraffin wax addition to reduce styrene emissions, with varied success in respect of
the key requirements of storage stability, secondary bonding and performance. Some
resins appear cloudy in both liquid resin and solid cured forms due to the wax dispersions,
thus impairing visual inspection of finished GRP laminates.
Dows LSE resin technology is based on a similar idea for reduction of styrene monomer
emissions, but at the same time, has successfully overcome problems of transparency,
phase separation and secondary bond strength and without significant variations in
mechanical performance and chemical resistance.
Airborne concentrations of below 20 ppm were only attainable by use of LSE resins
combined with ventilation and air extraction from the work place. As LSE resin could
not go below a concentration of 10 ppm of styrene emission, professionals have developed
a new technology which permits zero styrene emission, i.e., resin-infusion processes
including Seeman Composite Resin Infusion Moulding Process (SCRIMP), Ultraviolet
Vacuum-Assisted Transfer Moulding (UV-VARTM) or Resin Injection Recirculation
Method (RIRM) which are all environmentally friendly techniques. In these technologies,
briefly stated, a strong vacuum is used to saturate, or infuse, a dry lay-up with resin in a
388
Cure Kinetics of Vinyl Ester Resins
single-sided (male or female) mould. These new processes, which are vacuum-assisted
fabrication methods may solve the problems encountered by builders of large complex
composite pieces such as tanks, scrubbers and vessels or FRP boats from the complexity
of air-quality regulations.
Developed as the result of year of research on the shop floor and in the laboratory, the
process has proven to be a superior method for producing high quality composite parts
made from a wide range of fibre and resin combinations. All commercial fibres, core
materials and any resin in the range of 0.05-1 Pa will SCRIMP with outstanding results.
To cover the wide range of available resin systems, the process can be run at room or
elevated temperature.
Similar in concept to resin transfer moulding (RTM), but requiring only one tool side
and a simple vacuum bag, SCRIMP routinely produces large (185 m2) parts, using both
single skin and cored construction, and highly comples 3-D trussed parts weighing up to
1,360 kg.
In addition to his high integration level, the process is highly attractive owing to the
following characteristics:
low investment: in addition to the standard marine-type tool (about US$90/m2), the
second most expensive equipment is a vacuum pump;
automation: requires about the same labour as lay-up manufacture until the vacuum
bagging stage but no labour is necessary after this stage (automated impregnation).
In all cases, the resulting composite material properties directly compare to properties
that had only been achievable in highly controlled, expensive autoclave processes.
These techniques have created the need for resins which should have a long gel time,
passing from 45 minutes for conventional manual rolling during hand lay-up up to at
least 90 minutes of gel time for infusion. Therefore the catalytic system contains beside
promotor (cobalt salt), accelerator (dimethylamine) and catalyst (peroxide), a gel time
retarder such as acetyl acetone (2,4-pentanedione) to achieve a broad range of open or
working times. Some recent works have demonstrated the interest to use free radical
389
Handbook of Polymer Blends and Composites
9.9 Conclusion
Above all the main use of the epoxy vinyl ester resins is to combine the high performance
attributes of their parent epoxy resins with the handling and cure advantages at room
temperature of unsaturated polyester resins.
The catalytic system and cure conditions used for fabrication of reinforced resin equipment
are very important factors to control. Measurement of gel time, exotherm and residual
reactivity is a useful technique for optimising process conditions and for quality control.
The advantage of a given system may be exploited for specific fabrication techniques
and circumstances for obtaining chemically resistant and handling applications of
reinforced plastics.
With commonly used MEKP and cobalt octoate systems, the level of peroxide has a
great influence on gel time, exotherm peak and residual reactivity, mainly styrene, more
than does the level of activator.
Although the gel time is an important factor which should be known by the industrialist,
its measurement is not easy. It depends essentially on four parameters: catalyst, activator
and temperature levels, and the hygrometry of the medium. Thus at 25 C, for a level of
activator between 0.2% and 0.3% and with 2% catalyst, the average gel time is between
30 and 50 minutes.
Finally and from the industrial point of view, it is preferable to obtain a long gel time by
reducing the level of activator rather than by reducing the level of the catalyst; the
temperature can be also reduced. In reality it is the temperature which governs these two
parameters. Generally, in winter resins are activated at 0.3% in order to allow use in the
temperature range (11-19)C whereas in summer they are activated at 0.2% for use at
(21-29)C [14].
It is very important to indicate that the exothermic nature of the crosslinking reaction
intervenes as a post cured process relaxing residual strains.
Finally, there are residual mechanical properties of the composite after thermooxidative
ageing and their relationship with the matrix and the fibre which should determine
satisfactory use in processing. The chemistry of the matrix is the key point of any strucutral
390
Cure Kinetics of Vinyl Ester Resins
composites and has a direct influence on the homogeneity of the composite and its
mechanical properties.
Airborne styrene concentrations of below 50 ppm are possible by use of LSE resins on
their own, 20 ppm with the addition of a large and expensive ventilation system. To
meet the exposure limit of zero ppm, the infusion technique has proven to be, in the area
of composite manufacturing, a superior, high-volume method for the production of quality
parts in an environmentally sound fashion; it saves labour and time with dry lay-up,
creates a healthier and cleaner work environment, eliminates costly exchange of heated
air, produce laminates with superior properties (high fibre to resin ratio up to 80%) and
is inherently repeatable.
References
391
Handbook of Polymer Blends and Composites
11. M. J. M. Abadie, Presented at Eurocoat 96, Genoa, 1996, Vol. II, p.43.
15. M. Suziki, H. Miyama and S. Fujimoto, Journal of Polymer Science, 1958, 31,
212.
392
10
Cure Monitoring
R. A. Pethrick
10.1 Introduction
Thermoset resins are extensively used in the fabrication of composites and adhesive
bonded structures [1-5]. The mechanical properties and durability of these structures
depend critically on the polymerisation chemistry having been allowed to progress to a
state of near completion [1]. An undercured material will often have inferior mechanical
properties compared with those of one, which has undergone an optimal cure procedure.
Studies of cure monitoring can be divided into two groups. Firstly, those studies which
are concerned with the development of a fundamental understanding of the cure
chemistry and, secondly, methods developed for use in the control of the processes
involved in the fabrication of real structures.
There are wide ranges of different amine and epoxy resins used commonly in composite
manufacture [1, 2]. The epoxy resin system is typical of most curing systems; a low
molar mass, multifunctional molecule undergoing progressive reaction to produce higher
molar mass species. The initial structure of the epoxide will influence its viscosity,
reactivity and ability to form a crosslinked network. The base epoxy resin, diglycidyl
ether bisphenol A (DEGBA), is a solid for n = 0 and as the value of n is increased, it
becomes a mobile and then a sticky liquid, Figure 10.2.
393
Handbook of Polymer Blends and Composites
394
Cure Monitoring
395
Handbook of Polymer Blends and Composites
Epoxy novolac, tetrafunctional epoxides and cycloaliphatic resins are usually viscous
liquids. Reaction of a tetrafunctional epoxide with a primary amine will lead to a highly
crosslinked matrix used in high performance applications, whereas reaction between an
epoxide and a monofunctional primary amine will produce an essentially thermoplastic
material. Primary amines and anhydrides are difunctional and secondary amines are
monofunctional. To optimise the properties, the epoxides and amines used in commercial
resins are blends of various monomers, selected to achieve maximum stability in storage,
an appropriate time to gelation [tgel] and a desired glass transition temperature (Tg). It is
usual to blend 4,4-methylene dianiline with polymethylene dianilines to reduce the rate
of cure and give a stable pot life. The pot life is the time between mixing and the observed
increase in the viscosity of the mixture at a particular temperature. Solid polyamides can
be mixed with an epoxy resin to form a system with several months stability at ambient
temperature, but when heated to about 170 C will cure in several minutes [6-10].
The characterisation of the cure in epoxy resins, polyesters resins, polyurethanes and
similar materials lies in defining the nature of the chemical reactions occurring. The
degree of crosslinking is a function of the stoichiometry of the monomers involved, and
the extent of the reaction achieved during cure. The reaction of the curing agent with the
epoxy resin must result in a three-dimensional network only when one of the components
has a functionality greater than two, and the other component has a functionality of not
less than two. Aliphatic systems usually cure at room temperature, whereas aromatic
containing amines require elevated temperatures often in excess of 150 C.
With an excess of the epoxide, a prepolymer is formed which will have a higher molar
mass than the starting epoxide and can undergo further reaction with amines to form a
three-dimensional network. As the stoichiometry approaches one equivalent of epoxide
per equivalent of reactive amine sites, the molar mass will approach infinity and a three-
dimensional network polymer is obtained. As the ratio of the curing agent to epoxide is
increased a linear polymer (thermoplastic) and finally, with an excess of curing agent an
amine-epoxide adduct is obtained. Similar behaviour can be observed in polyurethanes,
polyester resins and similar thermosetting systems, Figure 10.4.
At a molecular level, the cure of an epoxy resin, or similar crosslinking system, will
generate molecules of ever increasing size and this process will continue until an infinite
network is formed. As the chemical reactions proceed, the physical nature of the resin
changes from a Newtonian fluid to a gel and then possibly to a glassy solid. Gelation
396
Cure Monitoring
Figure 10.3 Various types of structure generated during the early stages of cure of an
epoxy resin. The numbers indicate the stoichiometry of the components required to
create the structure on the right hand side of the figue
397
Handbook of Polymer Blends and Composites
Gillham [4] has proposed a time dependent phase diagram to describe the cure process
and has called it a Time-Temperature Transformation (TTT) diagram, Figure 10.5.
398
Cure Monitoring
The diagram is a plot of the times required to reach gelation and vitrification during
isothermal cure, respectively, and has four distinct phases: liquid, gelled rubber, ungelled
rubber and gelled glass. There are also three critical temperatures, which are defined as:
Tg, the maximum glass transition temperature of the fully cured system.
gel Tg, the isothermal temperature at which gelation and vitrification occur simultaneously.
Tgo , the glass transition temperature of the freshly mixed reactants.
During isothermal cure at a temperature between gel Tg and Tg, the resin will first gel
and then vitrify. Gelation retards macroscopic flow and changes the distribution of
material throughout the matrix. Phase separated structure at a micron level can be
generated as a consequence of differences in solubility and is locked in at this stage.
Vitrification is the locking in of structure at the nanometer level and loss of the free
399
Handbook of Polymer Blends and Composites
volume required for local chain motion. Devitrification is usually associated with
depolymerisation and chain scission generates chain ends, which plasticise the matrix
and lower the glass transition temperature.
For composite fabrication, the ungelled glassy state allows the creation of prepreg. The
heating of the monomers to incorporate the fibre will advance the state of cure of the
matrix, increase its viscosity and average molar mass. However, cooling the prepreg to
below gel Tgel effectively inhibits reaction allowing the material to be kept for periods of
up to six months to one year and still be capable of being successfully cured. The prepreg
will still be able to undergo flow on heating and take up the desired shape before gelation
occurs. Formulations can be processed as solids, e.g., moulding compositions, when Tgo
> ambient temperature; used as liquids, e.g., casting fluids, when Tgo < ambient
temperature. The gel Tg is critical in determining the upper temperature for storing
reactive materials to avoid gelation. The pot life is not uniquely defined, but corresponds
to the point in time when an appreciable increase in the viscosity of the material can be
detected. The use of pre-reacted mixtures raises the question of how to control their
processing, when the state of cure is unknown and is central to cure monitoring.
In composite systems, shrinkage stresses due to volume contraction of the resin develop
when adhesion to the fibres can be identified after gelation above gel Tgel and before
vitrification below Tg. The tensile stresses generated in the resin affect the composite
properties and can lead to fibre buckling [11, 12]. The matrix material on cooling may
contain internal stresses as a consequence of the frozen-in distribution of states dictated
by the cure temperature. These internal stresses can be relaxed as a consequence of local
plasticisation by chain ends and may lead to crazing and crack formation. Cure of finite
specimens can lead to higher temperatures being created inside the structure than at the
surface as a consequence of the exothermic nature of the reaction. At higher temperatures
cure can lead to the outside vitrifying before the inside. In this latter case, internal
stresses develop as the inside contracts relative to the vitrified outside due to contraction.
Cooling cured material from above Tg will lead to the outside vitrifying before the
inside and internal stresses will be minimised by cooling very slowly, i.e., annealing.
Prolonged isothermal curing at Tcure, below Tg would lead to Tg = Tcure, if the reaction
were quenched by the process of vitrification. In practice, the observed value of Tg is
higher because of the combined effect of advancement of cure during the process of
measurement and the effects of the exotherm. Full cure is attained most readily by
reaction above Tg and more slowly by curing below Tg to the full cure line on the TTT
cure diagram. Full cure is not a unique state and depends upon the time temperature
reaction path. In many of the commonly used systems, there are a number of competing
chemical reactions possible with different activation energies and hence complex reaction
400
Cure Monitoring
patterns are possible. Furthermore, material properties depend on the reaction path for
physical reasons, such as density of the polymerised system depending on the
polymerisation pathway. The highest density matrix material is formed about 30 C
below the optimum cure temperature for dicyandiamide cured systems; these materials
having significantly lower electrical conductivity than those formed at or about Tg.
The limiting viscosity in the liquid state is controlled by gelation above gel Tg and by
vitrification below gel Tg. At gelation, the weight average molar mass and zero shear rate
viscosity become infinite, although the number average molar mass may still be very low.
Viscosity in the vicinity of vitrification below gel Tg is described by the Williams Landel
Ferry (WLF) equation [13]. It is possible from knowledge of the way in which the various
species grow during the polymerisation process to be able to predict the viscosity of the
reacting system and these theoretical models can help with the development of an
understanding of the relation between molecular structure and the cure process [14].
In the fabrication of composite structures, a number of factors have to be taken into account:
If the initial viscosity of the resin in the early stages of cure is too low, then fibre
settling will occur and resin will flow out of the edge of the moulding.
Too high a viscosity will not allow compaction of the structure when pressure is
applied to the prepreg. A non-uniform distribution of fibres through the thickness of
the structure and fibre bucking can result from too high a viscosity.
Thermal induced stresses arise when the cooling is too quick and it is often desirable
to undercure and subsequently anneal the structure to remove stresses.
Pressure is applied to the matrix before gelation to achieve the optimum densification
of the composite.
The time to gelation can be calculated from reaction kinetics and the conversion at gelation
from the Flory theory [15]. The time to vitrification can be computed from reaction
kinetics [14, 15, 16], which increase with Tcure. In the absence of diffusion control, the
general kinetic equation describing the cure reaction is
dX E
= A.exp A f ( x ) (10.1)
dt RT
where X is the extent of reaction, EA the activation energy, R is the gas constant and T is
the temperature of the cure process. The function f(x) describes the mechanistic detail of
401
Handbook of Polymer Blends and Composites
the cure process and will change with the system being investigated. The times to gelation
can be computed versus temperature assuming that the simple Flory theory is obeyed
[15]. The effects of diffusion control are evident as the gelation time is approached in
terms of deviations between experiment and theory. The principle characteristics of the
cure process are defined in terms of the gelation and vitrification times. If the initial
concentration of the two reactants, A and B, are CA0 and CB0, respectively, the degree of
cure can be described by
C A0 C A
C* = (10.2)
C A0
where C* is the degree of conversion and equals zero when there is no reaction and equals
one when all of A is reacted. However, it is not easy to monitor reacted and unreacted As
and Bs during the curing reaction in a thermoset system as a function of time, unless the
reaction is carried out under infrared or Raman spectroscopy. The reaction exotherm can
be observed directly using differential scanning calorimetry (DSC). When samples of an
unreacted thermoset are investigated in a DSC, the amount of energy released during the
cure process can be measured, QT. Every crosslink produced will release a small amount of
energy in the form of heat and the degree of cure can be defined as:
Q
C* = (10.3)
QT
dQ
Q= dt .dt
0
(10.4)
DSC data is commonly fitted to empirical models that accurately describe the curing
reaction. Hence, the rate of cure can be described by the exotherm, dQ/dt and the total
heat released during the curing reaction QT as:
.
dC* Q
= (10.5)
dt QT
With the use of equation (10.5), it is possible to take the DSC data and obtain a model to
fit the kinetic data. For example, the curing kinetics for many heat activated cure materials,
vinyl esters and unsaturated polyesters can be described by:
402
Cure Monitoring
E
dC*
( )
n
= k0 .exp RT .C* m 1 C* (10.6)
dt
where E is the activation energy, R is the gas constant and k0, m and n are constants that
are obtained by fitting the equation to the DSC data. The cure reaction is very slow at
first, then increases and slows down again towards the end of the reaction. On the other
hand the mixing of activated cure materials such as polyurethanes will instantly start
releasing heat after the mixing of the two components has occurred. The model in this
case has the form:
E
dC*
= k0 .exp RT ( 1 C* ) (10.7)
dt
The kinetics are best obtained using a combination of DSC data and spectroscopic
observation of the change in concentration of a particular band with time.
A well known problem in thick section components is that inefficient heat transfer can
lead to curing gradients and become more complicated and difficult to analyse since the
temperature and curing behaviour are highly dependent on both the mould temperature
and the geometry of the part being processed [17-19]. A thick part will result in higher
temperatures and a more complex cure distribution during processing and even thermal
degradation. A relatively easy way to check temperatures that arise during moulding
and curing or demoulding is desirable. A one-dimensional form of the energy equation
that includes the exothermic energy generated during cure can be solved as follows:
T 2T .
.C p = k 2 +Q (10.8)
t z
assuming the material is confined between two mould halves at equal temperatures, the
use of a symmetrical boundary condition at the centre of the part being fabricated is valid:
T
=0 at z=0 (10.9)
z
and T = Tm. At the mould wall, Barone and Caulk [20] have used the finite difference
technique and a six constant model that represents dC*/dt, for a curing sheet of a thermoset
403
Handbook of Polymer Blends and Composites
404
Cure Monitoring
polymer. Figures 10.6 and 10.7 show a typical temperature and degree of cure distribution,
respectively, during solidification of a 10 mm thick sheet of thermoset material.
In Figure 10.6, the temperature rise resulting from the exothermic reaction is obvious
and increases in thicker parts and with increasing mould temperature. The time to reach
80% cure versus thickness of the moulding at various mould temperatures, Figure 10.8,
demonstrates the time at which the internal temperature exceeds 200 C due to the
exothermic reaction.
It is assumed that for this system, the resin matrix will begin to undergo thermal degradation
above 200 C and produce high residual stresses in the final moulding. These calculations
indicate that improper processing can result in a non-uniform cure distribution, which can
ultimately lead to voids, cracks or imperfections inside the moulding.
Figure 10.8 Cure times versus plate thickness for various mould temperatures, shaded
region represents the condition at which thermal degradation may occur
405
Handbook of Polymer Blends and Composites
At any instant of time, the resin velocities in the prepreg and in the bleeder cloth, which
is used to cover the part being moulded and takes up excess resin, may be represented by
Darcys law
S dP
V= . (10.10)
dz
406
Cure Monitoring
where S is the apparent permeability, is the viscosity and dP/dz is the pressure gradient.
The law of conservation of mass together with the Darcys equation gives the following
expression for the rate of change of mass M in the composite:
dM P Pu
= r AzVz = r Az Sc hc
dt c
dz
0
(10.11)
where r is the resin density, Az is the cross sectional area perpendicular to the z axis, hc
is the thickness of the compacted plies. Pu is the pressure at the interface between the
composite and the bleeder. The subscript c refers to conditions in the composite at
position hc. Accordingly, Pc is the pressure at hc and is the same as the applied pressure.
At any instant of time the resin flow rate through the composite is equal to the resin flow
rate into the bleeder cloth:
rAzVr=rAzVb (10.12)
The temperature and viscosity of the resin inside the bleeder are assumed to be independent
of position, but not time. Combining these equations yields:
Sb P0 Pb
r AzVz = r Az . (10.13)
b hb
where hb is the instantaneous depth of resin in the bleeder. Rearranging the above
equations yields the rate of change of resin mass in the composite:
dMr r Az Sc P P
= h 0 b
dt c
1 + G ( )
t
dz
0
(10.14)
Sc bhb
G(t ) = .
Sb h c
0
dz (10.15)
The mass of resin that leaves the composite and enters the bleeder cloth in time t is:
407
Handbook of Polymer Blends and Composites
t
dMr
MT =
0
dt
.dt (10.16)
t
1 dMr
hb =
r b Az
0
dt
.dt (10.17)
where b is the porosity of the bleeder cloth. The thickness of the compacted region is:
hc = n shl (10.18)
where hl is the thickness of one compacted ply and ns is the number of compacted prepreg
plies. The value of ns varies with time depending on the amount of resin that has been
squeezed out of the composite. Similar analysis can be carried out for flow parallel to
the tool plate [23] and the situation where both flows are possible has been considered
by Kardos and others [24].
For an arbitrary selected set of process conditions: heat and pressure, the variables inside
the composite can be calculated using the previous model. The procedure is then repeated
until a set of processing conditions is found which satisfies the following conditions for
thermoset matrix composites:
The temperature inside the material must not exceed a defined value at any time
during the cure so as to avoid the possibility of degradation.
At the end of the cure the resin content must be uniform and must have the desired value.
The cured composite must have the lowest possible void content.
The cured composite must have the desired thermal and mechanical properties.
The calculations require extensive numerical calculations but have been used satisfactorily
to predict the behaviour of a number of interesting situations. From the brief outline of
408
Cure Monitoring
the background theory and practice of curing of a composite material, it will be clear
that there are several critical stages in the optimum cure of a structure:
Monitoring the progress of the reaction to the finally cured state (optimum Tg and
usually low residual unreacted monomer).
Identification of the way in which the viscosity changes during the cure process. In
the case of a prepreg, the viscosity of the pre-reacted monomers will be high at room
temperature but will drop as the temperature is increased. However, during this
heating process the viscosity will increase as a consequence of the further reaction in
the system. A balance of these two effects is usually desirable in order to control the
final distribution of fibre in the composite.
The rate of increase of the viscosity should be such as to allow the optimum time for
compaction to occur. Definition of this time scale can be assisted by the use of the
computer models.
The temperature selected for the cure process must also reflect the effect, which the
mass and shape of the structure may have on the possible exotherm created by the
reaction process. Excessive heating will produce uneven cure, degradation of the
cured resin and a poor final product.
Post cure may be advisable to increase the final state of the cure and achieve
improvements in the final mechanical properties of the structure. The post cure and
the cooling cycle will dictate the extent to which residual stresses may be retained in
the final composite and this can influence its durability.
The above factors need to be considered and if possible monitored during the cure process
to achieve the optimum material properties.
A number of terms are used to define the cure process and ideally should be measured to
allow definition of the TTT diagram.
409
Handbook of Polymer Blends and Composites
These terms are essentially synonymous [25] and define the time period before the resin
mixture has achieved sufficient reaction to make its successful use questionable. The
definition is subjective but corresponds to a viscosity in the range 10-2 to 102 Pa-s implying
that a knowledge of the isothermal viscosity-reaction time curve would be a more precise
method of determining their correct values.
410
Cure Monitoring
411
Handbook of Polymer Blends and Composites
The VNC, shown in Figure 10.9, is usually close to its resonance condition and the
simplest model for the system is based on the concept of a mass suspended on a spring
with one end fixed. If the spring has a stiffness constant of k0, the system will satisfy the
following equation of motion:
2y
m + k0 y = 0 (10.19)
t 2
or
2y
+ 02y = 0 (10.20)
t 2
where m is the mass involved in the vibration and 0 is the natural frequency of oscillation
in radians per second of the oscillator with 2 = k / m
0 0
y = y0 sin( 0t ) (10.21)
412
Cure Monitoring
For free vibration in air the system will have a natural frequency o of k0 / m . For
forced vibration the system is capable of oscillating at any frequency dictated by the
driving frequency. When the driving force equals the natural frequency, resonance will
be achieved. The equation of motion of the forced vibration for the system is:
2y y
m + (C + ) + k0 y = f0 cos(.t ) (10.22)
t 2
t
where is the viscosity of the fluid, C is a shape factor for the probe in the curing fluid,
is the damping term for the internal friction in the vibrator and fo cos t is the imposed
periodic force of angular frequency . In this situation, the frequency is dictated by the
external applied voltage and is a variable quantity for the system.
The solution to equation (10.22) for the amplitude at resonance in terms of the output
voltage is:
1
a 4 4 + a 33 + a 22 + a1 + a0 = (10.23)
V2
where V is the output voltage and a0 to a4 are constants for a particular instrument. This
model is an over-simplification of the real situation; the parameters obtained should not
be accorded any physical significance and should be treated as unique to a particular
vibrating needle system. However, these relationships do provide a remarkably good fit
with experimental data confirming the ability to use the output as a direct indication of
the viscosity changes, which occur during the cure process.
The change in the resonance frequency can be modelled using an extension of the simplistic
models. The equation of motion of the mass, m, is given by:
2y
m + ( k0 + k1 )y = 0
t 2 (10.24)
(k0 + k1 ) / m = 02
The addition of a second elastic element increases the natural vibration frequency, hence
the frequency of resonance for forced vibration. Since a shift to higher frequency has not
been associated with viscous changes it may be inferred that observation of such a shift is
413
Handbook of Polymer Blends and Composites
diagnostic for additional elasticity in the vibrating system. The cure time is usually obtained
from a study of the output voltage as a function of time. The voltage will usually increase
to a maximum and then decrease to an asymptotic value which is take as being indicative
of the fully cured state. This device is used extensively for the characterisation of the cure
process. A more detailed discussion of the method and illustrations of its application to
polyurethane and polysulphide have been presented elsewhere [25, 31].
The principal difference between this approach and that of the VNC is that the device has
incorporated a linear variable differential transformer (LVDT) which allows the magnitude
of the vibration of the needle or paddle to be observed directly rather than inferred from the
voltage applied to the driving coils. The change in the amplitude of the motion of the paddle
can be obtained using a phase sensitive detector locked to the driving signal for the linear
motor, Figure 10.10. The dynamic range of the curemeter is defined by the spring and the
frequency of operation; for the current system it is possible to explore the range 10 to 105 Pa-
s. A weaker or stronger spring theoretically allows a lower or higher range to be investigated.
The demodulated output of the LVDT is compared with the driving force of the linear motor
(p1) using a phase sensitive detector. Both the in-phase and quadrature components of the
probe displacement are recorded and stored in a computer. The temperature within the
reaction vessel placed in the stainless steel sheath close to the probe is also recorded. A
mathematical analysis of the system is based on the following assumptions:
The amplitude of motion for the spring (p1) is fixed, irrespective of the probe amplitude.
In practice the difference in compliance between the motor suspension and the coupling
spring means the motor is practically independent of the probe amplitude. The motor
suspension has an effective compliance that is a factor of 100 times greater than the
compliance of the spring.
The mass of the probe is negligible, i.e., resonant behaviour of the probe and spring
can be neglected. The operating frequency is more than a decade lower than the
resonant frequency of the system so this assumption is justified.
Probe movement produces negligible changes in contact area between probe and
material. Some finite change in the area is inevitable, but with correct design of the
probe, the effects are minimised and would in any case show up as second harmonic
components in the probe displacement. Such harmonics are rejected by the phase
sensitive detector.
The force of the material on the probe is purely due to the viscosity of the fluid.
Stirring of the liquid is considered negligible.
414
Cure Monitoring
Equating the forces at the lower end of the coupling spring and neglecting the mass
effects of the spring and probe gives:
( )
k p1 (t ) p2 (t ) ) + *Cp2 / dt = 0 (10.25)
where K is the spring constant, p1 and p2 are the instantaneous displacements from
equilibrium at points p1 and p2, respectively, * is the complex viscosity of the material
and C is a geometric factor related to the contact area between probe and material.
415
Handbook of Polymer Blends and Composites
Since p1(t) is a sinusoidal function, the differential equation can be written in complex
notation as:
k( p1 p2 ) + j..Cp2 = 0 (10.26)
where the values of p1, p2 and are also dependent upon the frequency . Equating real
and imaginary components, assuming that is real, gives:
where p2 and p2 are the real and imaginary components of p2. Solving for p2 and p2 gives
[ (
p'2 = p1 / 1 + ..C / k2 )]
(10.28)
p"2 = p1 ...C / k1 + ( ..C / k)
2
The viscosity can therefore be obtained directly from observation of the movement
occurring at point p2 relative to movement at point p1. In the current system, the linear
motor is sufficiently strong for the amplitude of the motion at point p1 to be assumed
constant during the whole of the experiment.
From the above data it is possible to calculate shear moduli given the relationship
where f is the oscillation frequency of the probe. The compliance can be directly calculated
from the shear moduli.
416
Cure Monitoring
life at 120 C. The pot life is an arbitrary parameter and it is possible to define it as a 3-5%
reduction in an amplitude or increase in the viscosity of the media. Beyond this point, the
resin begins to build a structure that will be incapable of further flow and produces a lower
quality moulding. The point at which the calculated viscosity reaches a value of 104 Pa-s is
normally accepted as the gelation point. A more detailed discussion of the technique and
the related analysis of data have been presented elsewhere [31].
417
Handbook of Polymer Blends and Composites
Figure 10.12 Torsional braid apparatus for the measurement of cure in thermoset resins
418
Cure Monitoring
The pendulum is intermittently set into oscillation to generate a series of damped waves as
the material properties of the specimen change with temperature and/or time. The shear
modulus, G, is given by:
2
G = KI(2..f ) 1 + (10.30)
2.
where f is the frequency (Hz) of the oscillation, I is the moment of inertia, is the
logarithmic decrement, = ln (A / A ), Ai is the amplitude of the ith oscillation, and
i i +1
K is a geometric factor. Equation (10.30) can be approximated by:
G KI(2..f )
2
(10.31)
KI
.(2..f ) I(2..f )
2 2
G = (10.32)
The loss tangent tan = GG, where is also the phase angle between the cyclic stress
and cyclic strain vectors; it is a characteristic measure of the ratio of the energy dissipated
per cyclic to the maximum potential energy stored during a cycle.
G
= tan = = (10.33)
G .f
where is the damping coefficient. During a dynamic mechanical thermal analysis experiment,
the moduli can be monitored by observing the changes in the frequency and logarithmic
decrement () of the oscillations. In a torsional pendulum experiment, where the geometric
constants are known, absolute moduli are obtained. The data which is obtained is similar to
that obtained from the vibrating needle experiments and can be generated in terms of viscosity-
time profiles from which the gelation and vitrification points can be derived.
2u
Ts k.l = I. (10.34)
t 2
419
Handbook of Polymer Blends and Composites
Figure 10.13 Schematic of a typical rheometer capable of being used in cure monitoring
where Ts is the torque applied by the transducer, l is the angular displacement of the lower
plate, u is the angular displacement of the upper plate, I is the moment of inertia of the
plate and transducer and k is the stiffness of the transducer. The sample displacement is
measured as the difference between the values of for the upper and lower transducers,
[l-u]. The displacement of the upper and lower transducers can be expressed in the form
l = l . j.t
u = l j (.t )
(10.35)
where is the phase lag between the harmonic motions of the two plates. Substituting
these equations into equation (10.34) one obtains
Ts = u e (
j .t )
(
I 02 2 )
(
r = l u = l e j.t 1 l e j.t ) (10.36)
where r is the difference between the upper and lower amplitudes and 0 is the resonant
frequency of the transducer. The complex modulus, G, can then be calculated from:
G = C. =
2
(2
)[
Ts C .u .I o (cos u ) + j sin ]
= G + i .G
r (
1 + 2 2.cos ) (10.37)
420
Cure Monitoring
Typical results for the behaviour of a thermoset are shown in Figure 10.14. Several
guidelines for testing systems are available and the best known of these is ASTM D4473
[36]. This attempts to give quantitative information by identifying the following features:
The cross over point between the storage modulus G and the loss modulus G which
is one of the definitions of the gel point.
A tangent is constructed on the G curve in the region of the steepest ascent. The
intersection of the tangent with the steady asymptotic G value is often used as the
cure point.
Figure 10.14 The time dependence of the viscosity data for a typical resin system
421
Handbook of Polymer Blends and Composites
The local maximum in the phase angle if it exists is identified and called the gel
point of the material.
The minimum in the dynamic viscosity is often termed the maximum fluidity and
can be used in process control.
The variation for a typical system with identification of the appropriate turning points is
shown in Figure 10.14. A range of techniques is available to measure the viscosity
profiles and include the use of capillary rheometers [37-40].
Figure 10.15 Typical heat flow plot for a thermoset resin system
422
Cure Monitoring
which is DEGBA/diamino diphenyl sulphate, the cure starts at about 120 C reaches a
maximum at about 230 C and is complete by about 340 C. From this both the rate and
extent of conversion can be derived at a given time and temperature. In contrast to
rheological measurements, DSC does not directly detect the onset of gelation, so that the
two techniques provide complementary information on the curing process.
In order to interpret such data it is necessary to use a kinetic model. The simplest model
assumes that the rate is proportional to some function of the rate of conversion and that
the proportionality constant has an Arrhenius type temperature dependence:
x
= k.f ( x ) and k = A.exp( E / RT ) (10.38)
t
x
t
(
= ( K0 + K1X ) 1 X 2 ) (10.39)
where X is the fractional conversion and K0 and K1 are apparent rate constants [45].
Fairly good linearity can be achieved over the first 40% of the reaction, but curvature
indicating departure of the reaction from the proposed mechanism is observed in the
later stages of the reaction. Various attempts have been made to use DSC to quantify
cure, however its main application appears to be in the evaluation of the extent of reaction
and also the change of the Tg with the extent of conversion.
Electrical measurements have been used for many years to establish the nature of the
cured state of thermoset materials. The electrical conductivity of the material is usually
consequence of ionic impurities, which are left behind in the monomers after synthesis.
423
Handbook of Polymer Blends and Composites
For instance in epoxy resins, unless removed during production, a significant level of
chloride ions can usually be detected. As the matrix changes from being a liquid to a
solid so the contribution from these conducting species will be reduced and there is a
corresponding reduction in the electrical conductivity of the material. Such
measurements have been used for many years to assess the state of cure from rubber
and related systems. More recently, dielectric analysis and frequency dependent
electromagnetic sensing have been extensively used for cure analysis [4]. Dielectric
properties such as permittivity, loss factor and conductivity have been used to investigate
the glass transition, secondary transitions and rheological phenomena. Dielectric
analysis can detect vitrification during cure, but there are conflicting opinions about
its ability to detect gelation [46, 47]. Dielectric measurements are especially attractive
because they have the potential to investigate fundamental properties of curing
thermosets, as well as to perform on-line cure monitoring and process control for these
same thermosetting systems.
All of the approaches used [46, 48, 49] involve examination of the frequency dependence
of the real [] and imaginary [] parts of the complex permittivity [*]. The principal
difference between the approaches lies in the design of the sensor configuration. Unless
the electrode structure is a parallel plate configuration then the form of the field
penetration into the resin has to be considered. It is usual to use microelectrodes,
which may be inserted into the structure or located, in an area where they present no
potential problem. The permittivity and loss can be obtained as follows:
C ( ) G ( )
( ) = 0
( ) = (10.40)
C .C 0
Both the real and imaginary parts of e* can have an ionic and a dipolar component:
= d + i = d + i (10.41)
The dipolar component d arise from rotational motion of molar dipoles associated
with the resin system. The ionic component I arises from the constrained translational
diffusion of charge that is either trapped at the electrode surfaces or within the
microstructure of the curing resin. In addition the loss component contains a
contribution from the ionic conductivity which can be expressed by:
( )
ohm1cm1 = 0 ...( )
(10.42)
0 = 8.854 x10 14.C 2 . J 1 .cm1
424
Cure Monitoring
The ionic component of the loss at higher frequencies can be subtracted to reveal the
dipolar contribution. The analysis of the system is difficult because in a reacting matrix
the nature of the dipole is changing with time and the simplistic assumptions used to
model the relaxation behaviour are often erroneous because of these changes in molecular
dipole content. However, since the media has a dominant effect on the rotational motion
and relaxation of the dipoles present the changes in the dielectric spectrum are easily
interpreted in a simplistic manner.
425
Handbook of Polymer Blends and Composites
and it is only when subsequent reactions have taken place that the material will be
transformed into its glassy vitreous state, characteristic of the complete cured
thermoset. During this process the dipoles will be slowed down and this is evidenced
by the peak moving from high frequencies to low frequencies as cure time proceeds.
426
Cure Monitoring
The glass transition temperature has long been recognised as being a dynamic process
and its movement with the extent of cure is clearly seen in Figure 10.16. We can
define a point at which vitrification occurs within the context of the time window
for the observation and it is typical to use a frequency of 1 Hz as reference. When
the material is completely cured the dielectric constant becomes independent of time
and frequency. However, there may be a residual dielectric loss, which reflects the
level of ionic impurity within the matrix material.
The usual cure monitor micro sensors have a limited frequency range and are also
very sensitive to the ionic conductivity contribution between the electrodes. The
latter sensitivity makes them useful for sensing the changes in the viscosity. A
typical sensor output for a micro sensor system is shown in Figure 10.17. The
point of reaction onset (A) is reflected in the sensor output by a drop in the rate of
change in the low frequency ionic mobility trace. The drop reflects the fact that
the onset of the reaction causes a decrease in the resin fluidity. This drop is consistent
with the large drop in the dipolar processes clearly seen in Figure 10.17. The point
of maximum flow and the magnitude of (b) of the fluidity (viscosity minimum) at
this time is determined by measurement of the time of occurrence and the magnitude
of a maximum in the low frequency trace. Timing of this event is critical for the
application of pressure to the composite in order to achieve optimum consolidation
of the resin matrix.
Figure 10.17 Typical output from a dielectric sensor used in monitoring cure
427
Handbook of Polymer Blends and Composites
Attempts have been made to correlate the output of the dielectric sensors with the
changes in the viscosity and an approximately linear variation has been observed, Figure
10.18. The deviation of the data from the ideal linear relationship may in part be a
consequence of the changes in mobility being sensed as the viscosity is increased. Using
the correlation plots it is possible to re-plot the dielectric data in terms of an effective
viscosity, Figure 10.19. This data has to be used with caution as the data is reflecting
a microscopic viscosity and it is the macroscopic viscosity, which is important in many
of the flow issues with composite manufacture.
Figure 10.18 Plot of the change in the permittivity with the log of the viscosity
428
Cure Monitoring
Figure 10.19 Calculated viscosity changes for a curing system obtained from
dielectric measurement
Thermally stimulated discharge (TSD) measurements can be used to investigate the vitrified
state of the curing matrix. In the experiment the sample is vitrified at a very low
temperature, typically by isothermal cure just above, gel Tg. The resultant matrix will
have a low Tg and during the experiment, which involves the cyclic heating of the sample
to above Tg application of polarisation cooling to low temperature and then observation
of the discharge currents, which arise on heating, the sample will have advanced its state
of cure. A typical trace for such a system, Figure 10.20 contains a low temperature peak
which is associated with large scale motion of the polymer backbone, (Tg) and a higher
temperature space charge peak which reflects the ionic mobility of the matrix. As cure
proceeds the location of the Tg peak increases in temperature and the amplitude of the
space charge peak decreases reflecting the increasing cross-link density of the matrix
reducing the ionic mobility. In principle, these effects can be investigated with conventional
dielectric measurements, however the TSD method is much simpler to apply to the
characterisation of the vitrified state.
429
Handbook of Polymer Blends and Composites
Figure 10.20 The thermally stimulated discharge traces for TETA/DEGBA. The traces
are displaced on the y axis to allow a cleared presentation of the shifts in the curves as
the cure progresses
(Reprinted with permission from John Wiley & Sons Inc., C.G. Delides, D. Hayward,
R.A. Pethrick and A.S. Vatalis, Journal of Applied Polymer Science, 1993, 47, 11,
2040. Copyright 1993, John Wiley)
430
Cure Monitoring
431
Handbook of Polymer Blends and Composites
Figure 10.21 Typical dielectric constant and loss for DEGBA/DDM cured at 80 C
(Reprinted with permission from John Wiley & Sons Inc., C.G. Delides, D. Hayward,
R.A. Pethrick and A.S. Vatalis, Journal of Applied Polymer Science, 1993, 47, 11,
2040. Copyright 1993, John Wiley)
432
Cure Monitoring
Figure 10.22 Typical dielectric constant and loss for DEGBA/TETA cured at 30 C
(Reprinted with permission from John Wiley & Sons Inc., C.G. Delides, D. Hayward,
R.A. Pethrick and A.S. Vatalis, Journal of Applied PolymerScience, 1993, 47, 11,
2040. Copyright 1993, John Wiley)
433
Handbook of Polymer Blends and Composites Cure Monitoring
Figure 10.23 Typical dielectric constant and loss for DEGBA/TETA/16.2 wt%
CTBN cured at 30 C
(Reprinted with permission from John Wiley & Sons Inc., C.G. Delides, D. Hayward,
R.A. Pethrick and A.S. Vatalis, Journal of Applied Polymer Science, 1993, 47, 11,
2040. Copyright 1993, John Wiley)
434
Cure Monitoring
Figure 10.24 Typical dielectric constant and loss for DEGBA/DDM/16.2 wt%
CTBN cured at 80 C
(Reprinted with permission from John Wiley & Sons Inc., C.G. Delides, D. Hayward,
R.A. Pethrick and A.S. Vatalis, Journal of Applied Polymer Science, 1993, 47, 11,
2040. Copyright 1993, John Wiley)
435
Handbook of Polymer Blends and Composites
after vitrification may be interpreted in terms of the phase separation of the epoxy
component dissolved within the CTBN at the point at which it phase separates from
the main matrix system. The cure of these occluded phases within the occluded
phase will change the mobility within the CTBN occluded phase and this will be
reflected in terms of a shift in the location of the MWS feature on a dielectric axis. It
is apparent that the MWS feature provides very useful information on the way in
which the morphology develops in these systems. Since the dielectric data is providing
information on the volumetric average, the data provides a very useful complement
to the assessment obtained through electron microscopy, which reflects a section
through the distribution of phases present. Since in thermoplastic modified systems
the phase behaviour is often complex it is not surprising to find that the distribution,
shape and detailed composition can vary with the conditions used in the cure process.
The CTBN example indicates that the extent to which post cure has been carried out
will influence the internal structure of the separated phases and can lead to changes
in the mechanical properties of the final cure composite and is therefore of interest in
terms of control processing.
436
Cure Monitoring
(a) (b)
(c) (d)
(e) (f)
Figure 10.26 Scanning electron micrographs (SEMs) of fracture surfaces for the DEGBA/
DDM/CTBN system. Amount of CTBN (wt %): (a) 0; (b) 5.2; (c) 10.6;
(d) 16.2; (e) 16.2; (f) 22.6 (Reprinted with permission from John Wiley & Sons Inc.,
Authors Journal of Applied Science, 1993, 47, 11, 2040. Copyright 1993, John Wiley)
437
Handbook of Polymer Blends and Composites
1. Change in the composition of the matrix causes the occluded phase to change from
being polyether sulphone dispersed in a spherical particle structure, progressively
to change its shape to elongated and sausage like domains, going through a co-
continuous phase to a phase inverted structure where the epoxy now will have a
spherical form.
2. The size of domain structure depends on the functionality and molecular weight of
the polyether sulphone added to the system.
3. The nature of the morphology depends on the temperature and speed of cure.
Analysis of the MWS feature obtained from dielectrics indicates that the nature of the
size, shape and total volume fraction of particles within the structure changes and this is
supported by electron microscopy measurement of cured sections. The distribution of
particles in the material reflects the nature of the phase separation process occurring in
the material and is therefore sensitive to the rate and conditions used in the curing process.
It is seen that this change in distribution is reflected in changes in mechanical properties
and is important in relation to obtaining the required characteristics in the final matrix
of the material. Control of the morphology is important in maintaining the quality of
the final composite material. The changes in morphology also have an effect on the
surface roughness and this can be critical in terms of the durability of the composite and
the ability of those materials to wet the fibres.
438
Cure Monitoring
Figure 10.27 Examples of real time dielectric traces measured at 180 C: (a) Dielectric
constant, (b) dielectric loss for MY0510/4,4-DDS/30% 5003P, (c) Dielectric constant,
(d) Dielectric loss for MY0510/4,4-DDS/30% RT-PS(A)
(a) & (c) reproduced with permission from A.J. MacKinnon, S.D. Jenkins,
P.T. McGrail and R.A. Pethrick, Macromolecules, 1992, 25, 13, 3492, Figure 3.
439
Handbook of Polymer Blends and Composites
Figure 10.28 Further examples of real time dielectric traces measured at 180 C: (a)
Dielectric contant, (b) dielectric loss for MY0510/4,4-DDS/30% 5003P, (c) Dielectric
constant, (d) Dielectric loss for MY0510/4,4-DDS/30% RT-PS(A)
440
Cure Monitoring
The molecular modelling of the cure process is not as well developed as the description
of the bulk heat and mass transfer processes described above. Recently, there have
been attempts to generate simple molecular based models, which allow the simulation
of the cure process from molecular based assumptions. Classical [64-68], percolation
[69-72], kinetic [73, 74] and Monte Carlo simulations [75-77] have all been used to
predict the point at which gelation occurs in a thermosetting system. Gelation has
been extensively investigated theoretically [78-82] and good agreement with experiment
is observed and allows the Monte Carlo approach to be used for the prediction of
chain growth in curing systems.
Initially the monomers will behave as simple monomer liquids and which will almost
invariably exhibit Newtonian behaviour with the addition of a high frequency
retardation process [83]. The retardation times for simple liquids are typically of the
order of nanoseconds and viscosities have values in the range 0.5 to 10 Pa-s. The
monomers used in the typical thermoset system will be expected to initially behave like
a simple liquid and it is only at relatively high degrees of conversion that truly polymeric
behaviour will be observed.
Classically, the viscoelastic properties of linear low molecular weight polymer melts
can be described by the Rouse theory in which the viscoelastic behaviour is a function
of the molar mass of the chain and its interaction with environment [84-86]. For
higher molecular weight polymers, coil entanglement can occur and the rheological
behaviour is best described by a reptation type motion [86, 87]. It is essential that
the equations are able to accurately predict the rheological behaviour for small molecule
species as well as high molecular weight materials. The analysis of rheological behaviour
usually concentrates on the readily accessible low frequency domain and hence on
high molecular weight species, low molecular weight species have usually been ignored.
In the description of a curing system, the initial behaviour will be dominated by these
low molecular species and hence it is crucial that the validity of the theories for such
materials be established. In reactive systems, it is very difficult to obtain experimental
data over a broad frequency range and hence the simulations have been checked against
stable polymers of known structure.
441
Handbook of Polymer Blends and Composites
SIMULATION OF THE
REACTION SYSTEM
Calculation of the number, topography and
size of various species present
RHEOLOGICAL BEHAVIOUR -
as a function of extent of reaction
A Monte Carlo modelling approach is is used to predict the molecular weight distribution
and topography of the species at various degrees of conversion [88]. This calculation
gives a complete description of the connectivity of each of the original monomers to
other monomers in the system and the structures generated may range from oligomers to
linear or branched chain polymers. A Monte Carlo approach avoids generating an
analytical model of the polymerisation and essentially mimics the real process by picking
pairs of monomers at random and reacting them together according to set of
predetermined reaction rates. In principle any number of different monomers and an
arbitrary number of reaction rates can be dealt with, the only proviso is that for an
accurate description of the system it is essential that a sufficiently large number of each
species be included in the initial reaction mixture. DryAdd is a commercial simulation
package [88], which has been modified in this case to give the connectivity of each
monomer. The software allows the inclusion of different reactivities for species with
defined functionality, different proportions of each species and allows for change in
reactivity as the reaction proceeds.
442
Cure Monitoring
Linear polymers will be studied to explore the high frequency limiting viscosity of low
molecular weight polymers and narrow molecular weight distribution polystyrenes will
be studied.
Branched or star polyisoprenes will simulate the type of structure formed in the
intermediate stages of reaction in gelling systems.
The high frequency modulus G*() of a number of simple liquids and low molecular
weight polymers [85, 89, 90] have been shown to be described as follows:
1
G* ( ) = (10.43)
J* ( )
1 JDC
J * ( ) = J + +
iDC (1 + i ) (10.44)
r
where DC is the low frequency limiting viscosity value, JDC is the magnitude of the
relaxing component of the compliance, J is the high frequency limiting value of the
compliance, r is the longest retardation time of the process and is a distribution function
varying from 0 to 1 which describes the width (skewness) of the retardation process.
The relaxation time is dictated by the nature of the process, which for low molecular
weight materials is whole molecule rotation but in oligomers takes the form of an internal
rotation. Equation (10.44) can be recast as:
1 JR
J* ( j ) = J 1 + +
i M (1 + i ) (10.45)
r
443
Handbook of Polymer Blends and Composites
Je = J + J r (10.46)
1 2 J
J* (i ) = J 1 + + (10.47)
i M (i M )
1/ 2
This equation has the merit that from measured values of J and it is possible to
describe the behaviour of a simple liquid system. In the simulations described here it is
assumed that this form is applicable up to the point at which the first normal mode
appears. The limiting effective low frequency value will then be assumed to correspond
to the highest normal mode of the polymer chain. Data for a wide range of simple
liquids including bis (m- (m- phenoxy phenoxy)phenyl) ether, di (isobutyl) phthalate, di
(n-butyl) phthalate, isomeric propyl benzenes, squalane and trichloroethylphosphate all
fit this theory [83]. Studies of polybutenes of molecular mass up to 2700, acrylates in
the range 4.5 x 103 to 7 x 104 and tetra(2-ethylhexyl) silicate all exhibit simple liquid
behaviour and can be fitted to the above equations.
RT
N
i j
G * ( )Rouse =
M 1 + i
j =1 j
(10.48)
where the values are the relaxation times of a series of N modes. The relative values
are given by the eigenvalues of the N x N matrix A where:
444
Cure Monitoring
1 0 0
1 L 0 0 0
1 2 1 0 L 0 0 0
0 1 2 1 L 0 0 0
A =
L L L L L L L L
0 0 0 0 L 1 2 1
0 0 0 0 L 0 1 1
N
= 0 a M2 (10.49)
m0RT
where o is the monomer friction coefficient, Na is Avogadros number, mo is the repeat unit
weight, R the gas constant, T the absolute temperature and M the molecular weight. Above
Mc it is necessary to include in the calculation the effects of chain entanglement or reptation.
The reptation motion of an entangled polymer molecule may be described by the relationship:
G0 8 Td
G* ( ) =
5 j
j =1
4 2
1 + iTd / j 2
(10.50)
where G0 is the terminal modulus corresponding to the value in the tail of the relaxation,
Td is the terminal relaxation time defined in terms of the steady state flow viscosity
and has the form:
5 2(0) 1
Td =
8G0 j
j odd
4 (10.51)
The form of the constitutive equations describing the behaviour of entangled polymer
systems still remains an area of some controversy [91, 92], however the above formulation
describes accurately the observed behaviour for linear narrow molecular weight
polydimethylsiloxanes [93, 94] and will be used here.
445
Handbook of Polymer Blends and Composites
It will be assumed that the modulus terms are additive [86], at least in the limit of relatively
low molecular weights for which the reptation contribution is small or absent. The total
modulus for the system is then be given by:
For a system in which there is a molecular mass distribution then the calculation will
need to be performed for all the species present and summed over all the molecules
present and over all frequencies.
The contributions to G*() which vary with molecular weight (equations 10.48 and
10.50) have to be summed. For each molecular mass a large number of Rouse modes
will be produced and this calculation has to be repeated a number of times to include the
effects of the molecular weight distribution. This process is very inefficient and the
approach adopted is to calculate the distribution function for the relaxation process
over a discrete set of times logarithmically spaced in time and then convoluting the
distribution with the relaxation function:
RT
( ) =
*
(10.53)
M 1 + i
446
Cure Monitoring
The step interval of the distribution function was adjusted to give as close an
approximation to the true function as required and in practice 1/50th of a decade
was used. Application of this approach to the 10,200 molar mass polystyrene,
Figure 10.28.
The predictions show a good agreement between experiment and theory and indicate the
importance of the Davidson Cole contribution. In the case of the polystyrene of molecular
weight (Mw) 580, the plot obtained is purely that of the Davidson Cole contribution
indicating that at this molar mass the polymer behaves as a simple liquid and would be
Newtonian up to a (normalised) frequency of the order of 107 Hz or greater. The higher
molecular masses have an increasing contribution from the Rouse mode and the total
viscoelastic behaviour can be described by a simple combination of the Davidson Cole and
Rouse contributions. This contribution is likely to be a dominant for low molar mass
species. The polystyrene was the only polymer studied to exhibit a 97,000 Mw Doi Edwards
contribution. It is therefore reasonable to assume that in most cure type situations, unless
the linear dimensions of the growing chains exceed that of Mc, the models used need not,
at low degrees of conversion, include the effects of chain entanglement. This comparison
validates the form of the equations as accurately describing the behaviour of linear low
molar mass polymers.
Figure 10.28 Predicted variation of viscosity vs. frequency during during the formation
of a thermoplastic.
Reproduced with permission from D. Hayward, R.A. Pethrick, B. Eling and E.
Colbourn, Polymer International, 1997, 44, 3, 248, Figure 3. Permission is granted by
John Wiley and Sons Ltd., on behalf of the SCI.
447
Handbook of Polymer Blends and Composites
Growth of a linear polymer system can be simulated using a 400 MW linear aliphatic di-
isocyanate and a 600 MW linear diol. The viscosity frequency variation for the predicted
molar mass distribution was calculated at 10% steps between 20% and 90% conversion,
Figure 10.29. Initially the high frequency component makes a dominant contribution to
the viscosity, but the viscosity remains Newtonian for all practical frequencies. However,
as the reaction proceeds, the Rouse contribution increases and the viscoelastic behaviour
becomes significant at lower and lower frequencies. In this system, the molar mass
distribution stays below Mc up to about 85% conversion and the Doi Edwards contribution
has been omitted from the calculation. Typical values of Mw for the onset of reptation
behaviour are given by Graessley [92] as in the range 20,000-80,000.
Figure 10.29 Frequency dependence of the viscosity for a simulated linear reacting system
(Reprinted with permission from D. Haywood, R.A. Pethrick and B. Eling and
E. Colburn, Polymer International, 1997, 44, 3, 248. Copyright Society of Chemical
Industry. Permission is granted by John Wiley & Sons Ltd. on behalf of the SCI)
The predicted real component of the viscosity at 10 rad.s-1 was then plotted as a function
of % conversion, Figure 10.28. This type of characteristic is observed when thermoplastic
is generated and is marked by a rapid increase in the viscosity as high degrees of conversion
are approached (see Figure 10.30).
448
Cure Monitoring
The above model can adequately describe the behaviour of a linear polymer, but in systems
in which gelation is the result of the formation of chemical crosslinks rather than physical
entanglement effects it must be assumed that the growing chains do not approximate to
a simple linear chain. In the initial stages the polymer chain will resemble a star branched
polymer and the Rouse behaviour of such molecules, and indeed any arbitrarily branched
system, has been predicted by Forsman [96] using a graph theory approach. A network
may be represented by a numbered series of nodes and a numbered series of chains
joining the nodes, where the numbering is arbitrary, i.e., for a network of six nodes and
five chains and a matrix, G can be constructed according to the rules:
this then be used to construct a matrix A defined as A = GT.G. As in the linear case the
relaxation times can be calculated from the eigenvalues of the A matrix. The G and A
matrices for the above branched system diagram are:
449
Handbook of Polymer Blends and Composites
0 0 0 0 0 0
0 0 0 0 0 0
0 0 1 0 0 1
0 0 1 0 0 0 0 3 0 1 0 1
0 0 2 0 0 1
G = 0 0 0 1 1 0 A =
0 1 0 1 0 1 0 1 0 2 1 1
0 0 0 1 1 0
0 1 0 0 0 0
0 1 0 0 0 0 0 1 1 1 0 2
and for comparison the A matrix for a 6 chain linear system is:
1 1 0 0 0 0
1 2 1 0 0 0
0 1 2 1 0 0
A =
0 0 1 2 1 0
0 0 0 1 2 1
0 0 0 0 1 1
By solving for the eigenvalues of the generalised matrix the relaxation times of the modes
of the arbitrary network can be calculated. These eigenvalues are then used with the
equivalent of the Rouse equation and require addition of two extra components; the Doi
Edwards equivalent for the branched chain (if the masses exceed Mc) and the high
frequency limiting behaviour.
As an initial test of the application of the programme for a reactive system, a calculation is
performed on a 36 element chain with various configurations, Figure 10.31. Data for a
linear chain are compared with those for a cross (4 x 9 units) and a star (3 x 12 units). The
linear chain exhibits the lowest frequency Rouse mode. The modes in the star molecules are
shifted to higher frequency as expected, the longest and hence lowest frequency mode requiring
a node at the centre of the polymer. Similarly the cross molecule has its lowest mode at even
higher frequency. All three molecules exhibit the same high frequency Davidson Cole type of
process, which makes a relatively small contribution to the total viscoelastic response.
450
Cure Monitoring
Figure 10.31 Calculated viscosity against frequency for linear, cross and star 36
element polymers
Reproduced with permission from D. Hayward, R.A. Pethrick, B. Eling and E.
Colbourn, Polymer International, 1997, 44, 3, 248, Figure 6. Permission is granted by
John Wiley and Sons Ltd., on behalf of the SCI.
The above calculations indicate that it is becoming possible to predict from a molecular
basis the viscoelastic behaviour of a curing system and hence in theory describe the
expected TTT curves for an unknown system. In practice, this is not yet possible but it
could be that in the near future such an approach could be viable.
10.20 Conclusions
Both from the point of view of the study of model systems and also monitoring cure in a
real situation, the use of dielectric and viscosity measurements allow definition of the
gelation and vitrification points. The extent of reaction is best defined either using direct
spectroscopic observation of the loss of a characteristic infrared or Raman vibration or
by measurement of the change of the reaction exotherm as a function of time. Studies of
the vitrified state and the loss of trapped gel phase are most conveniently carried out
using thermally stimulated discharge measurements. The data on cure of thermoset
resin systems is best represented using the TTT diagram. Use of a combination of Monte
Carlo simulation of the cure reaction combined with viscoelastic modelling can provide
useful insight in to the connection between the extent of reaction and the nature of the
viscosity - time profile for a particular system. Cure of thermosets is becoming well
451
Handbook of Polymer Blends and Composites
understood and the effects of phase separation add an important dimension to the
definition of a particular matrix system.
Acknowledgments
I wish to acknowledge all the people who have been involved in this work and include D.
Hayward, I. McEwan, E. Pollock, S.D. Jenkins, P.T. McGrail, and A.J. MacKinnon. The
research has been sponsored by EPSRC, ICI plc, MOD and AFSOR Contact.
References
8. Epoxy Resins: Chemistry and Technology, 2nd Edition, Ed., C. A. May, Marcel
Dekker, New York, 1988.
10. R. S. Bauer in Applied Polymer Science, 2nd Edition, Eds., R. W. Tess and G. W.
Poehlein, American Chemical Society Symposium Series No.285, American
Chemical Society, Washington, DC, 1985, 931.
452
Cure Monitoring
12. J. D. Ferry, Viscoelastic Properties of Polymers, 3rd Edition, John Wiley & Sons,
New York, 1980.
15. P. J. Flory, Principles of Polymer Chemistry, Cornell University Press, Ithaca, New
York, 1953.
16. M. T. Aronhime and J. K. Gillham, Journal of Coatings Technology, 1984, 56, 7/8,
35.
17. T. A. Osswald and G. Menges, Materials Science of Polymers for Engineers, Carl
Hanser, Munich, 1995, 226.
18. T. A. Bogetti and J. W. Gillespie, Presented at the 45th SPI Conference, 1990,
Washington, DC, Paper No.19A.
20. M. R. Barone and D. A. Caulk, International Journal of Heat and Mass Transfer,
1979, 22, 1021.
21. A. C. Loos and G. S. Springer, Journal of Composite Materials, 1983, 17, 2, 135.
453
Handbook of Polymer Blends and Composites
26. D. Durand in Polymer Yearbook 3, Ed., R. A. Pethrick, Harwood, London, 1984, 229.
29. ASTM C679-87 (1997), Standard Test Method for Tack-Free Time of
Elastomeric Sealants.
30. ASTM D2377-00 Standard Test Method for Tack-Free Time of Caulking
Compounds and Sealants.
32. Polymer Laboratories, Thermal Sciences Division, The Technology Centre, Epinal
Way, Loughborough, LE11 0QE, UK.
33. Rheometric Scientific Ltd., Surrey Business Park, Weston Road, Kiln Lane,
Epsom, KT17 1JF, UK.
35. R. W. Hall, Presented at Flow and Cure of Polymers - Measurement and Control,
RAPRA Technology, Shawbury, UK, 1990, Paper No.9.
36. ASTM D4473-95a Standard Practice for Measuring the Cure Behaviour of
Thermosetting Resins Using Dynamic Mechanical Procedures.
38. M. Nasir and P. Albihn, Presented at Flow and Cure of Polymers - Measurement
and Control, RAPRA Technology, Shawbury, UK, 1990, Paper No.2.
39. F. S. Baker and R. E. Carter, Presented at Flow and Cure of Polymers - Measurement
and Control, Rapra Technology, Shawbury, UK, 1990, Paper No.10.
454
Cure Monitoring
41. J. M. Barton and K. Dusek, Epoxy Resins and Composites, Advances in Polymer
Science, No. 72, 1985, 112.
57. R. W. Sillers, Proceedings of the Institution of Electrical Engineers, 1937, 80, 378.
455
Handbook of Polymer Blends and Composites
64. P. J. Flory, Journal of the American Chemical Society, 1941, 63, 3083.
69. D. Stauffer, Journal of the Chemical Society, Faraday Transactions, 1976, 272, 1354.
70. E. Donoghue and J. H. Gibbs, Journal of Chemical Physics, 1979, 70, 2346.
72. Y. Kwan and B. E. Eichinger, Journal of Chemical Physics, 1984, 80, 3877.
73. Y. Kwan and B. E. Eichinger, Journal of Chemical Physics, 1984, 80, 3885.
74. H. J. Hermann, D. Stauffer and D. P. Landau, Physical Review Letters, 1982, 49,
412.
456
Cure Monitoring
81. A. L. Efos and B. Shklovskii, Physica Status Solidi (a), 1976, 76, 47580.
82. M. Rubinstein, R. H. Colby and J. R. Gillmor, Polymer Preprints, 1989, 30, 1, 81.
83. A. J. Barlow, A. Erginsav and J. Lamb, Proceedings of the Royal Society, 1967,
A300, 356.
84. H. Yamakawa, Modern Theory of Polymer Solutions, Harper & Row, New York,
1971.
85. J. D. Ferry, Viscoelastic Properties of Polymers, John Wiley & Sons, New York,
1970.
86. M. Doi and S. F. Edwards, The Theory of Polymer Dynamics, Clarendon Press,
Oxford, 1986.
88. The DryAdd Polymerisation and Network Modelling Package, Oxford Materials
Ltd, Cheshire Innovation Park, PO Box 1, Chester, Cheshire, UK.
89. W. R. Gray, G. Harrison and J. Lamb, Proceedings of the Royal Society, London,
1971, A356, 77.
91. R. H. Colby, Journal of Polymer Science, Part B, Polymer Physics, 1996, 34, 16, 2665.
92. W. W. Graessley, Journal of Polymer Science, Part B, Polymer Physics, 1996, 34,
16, 2663.
457
Handbook of Polymer Blends and Composites
458
11
Curing and Bonding of Composites using
Electron Beam Processing
Daniel L. Goodman and Giuseppe R. Palmese
11.1 Introduction
Curing is done near room temperature, allowing the use of low cost, low temperature
tooling such as wood, plaster or foam.
EB-curable resins and adhesives have long shelf lives and can be stored at room
temperature. They are typically one component and solvent-free.
For large parts that are inconvenient or impossible to fit in an autoclave, EB processing
may be the best low-temperature curing alternative. Using a portable electron beam
system, the curing equipment can be brought to the part, rather than vice versa.
459
Handbook of Polymer Blends and Composites
Figure 11.1 Composite manufacturing processes suitable for EB curing include (a)
filament winding, (b) pultrusion and (c) prepreg layup with vacuum bag. Other
methods (not shown) include VARTM and automated tape placement. Parts are
transferred via cart or conveyor to be cured under an EB magnetic scanner. In line or
in situ curing processes are under development for pultrusion and automated tape
placement. This is shown in (b) and discussed in Section 11.6.
(Figure reprinted from [1] with permission of T. Walton, Aeroplas International Corp.,
Hollis, NH, USA)
Figure 11.2 High energy electrons are produced in the accelerator (sometimes known
as the EB gun), transported through the air, and then penetrate deeply into the
materials. The depth of penetration is proportional to the energy, measured in millions
of volts (MeV). The dose is a measure of deposited energy. Typical doses for curing
composites or adhesives are in the range 50-200 kGy.
460
Curing and Bonding of Composites using Electron Beam Processing
Adhesive bonding coupled with autoclave curing is widely used in the aerospace industry
to produce large composite parts with complex core structures. EB-curable adhesives
have the potential to replace thermally cured aerospace adhesives in bonding large
integrated structures in fewer steps, at reduced cost and without autoclave processing.
The automotive industry uses adhesives to bond composite parts and structural assemblies.
For ground vehicle fabrication, EB bonding offers high throughput and command-cure
bonding near room temperature. EB-curable adhesives can be used to bond composites
to metal substructures. The process avoiding de-bonding during the cool-down cycle
that can occur due to differences in thermal expansion of the metal and composite.
Although electron beam curing and bonding of composites is an active area of research,
the technology has not yet been widely adopted in industry. This is partially due to the
conservative nature of the aerospace and automotive industries. Pre-qualification has
begun as part of the industry-wide Composite Affordability Initiative (CAI) programmes,
but there will be many years between this work and the use of the materials in production.
This is because few new aircraft designs are planned, the design cycles are extremely
long and the amount of composite used on aircraft may actually be decreasing. In the
automotive industry, work done by research groups on new technologies, though
promising, are not necessarily adopted for production. The reason that novel automotive
concepts are not produced is often due to market forces, rather than to technical issues.
461
Handbook of Polymer Blends and Composites
EB-curable adhesives have made significant strides in the last few years. Their properties
meet the requirements of automotive applications. The strength of the best EB-curable
paste adhesive is close to that of industry-standard aerospace systems. But these adhesives
cannot be used in the hottest or coldest aerospace environments, and are just beginning to
be used on secondary structures. In some cases, aerospace companies find it more convenient
to have their adhesive supplied as a film rather than as a paste. However, EB-curable film
adhesives do not currently possess the strength needed for any aerospace application.
There are several reasons why the properties EB-curable composites and adhesives do
not match those of state-of-the-art aerospace systems. One reason is that the resin
chemistry (either free radical or methacrylated and cationic epoxy-based) produces
different polymer structures than heat-curable systems. Another oft-cited reason for
reduced composite properties is reduced fibre-matrix interface strength due to
incompatibility between EB-curable resins and fibre finish. These issues and current
research to improve material are discussed in Section 11.4.
11.1.3 Background
Electron beam composite curing and bonding is a subset of the field of radiation processing
of polymers. Electron beam irradiation is a widely used commercial process. In the
electronics industry, EB processing is used to crosslink wire and cable insulation, to
produce heat-shrinkable tubing and to make polymer devices such as resettable fuses for
automotive and portable electronics [3, 4]. Information on radiation processing of
polymers, including the curing of inks and coatings by UV light or EB, is available from
the Radtech organisation [5]. Electron beam irradiation is also used to sterilise medical
products, increase the strength of supermarket plastic bags and crosslink rubber in tyres.
The French firm Aerospatiale was the first company to use electron beam composite
curing in a production environment. Aerospatiale fabricated large filament-wound rocket
motor cases and cured the resin with EB [6]. They reduced the time to cure a rocket
motor from about a week to less than eight hours, while producing lower residual stresses
in their part. This application is described further in Section 11.2.1.
Encouraged by this success, the US Defense Advanced Research Project Agency (DARPA)
decided to fund work on the use of electron beam curing to lower the cost of manufacturing
aircraft composites. The two largest DARPA-sponsored programmes were the Affordable
Polymer Composite Systems (APCS) programme led by Northrop Grumman during 1994-
1998 [7, 8, 9] and the Integrated Airframe Technology for Affordability (IATA) programme
led by Lockheed Martin from 1995-1997 [10]. Both programmes produced demonstration
aircraft parts using EB curing and bonding. The APCS programmes produced significant
cost savings with the use of EB in a variety of aircraft parts. The APCS programmes and
462
Curing and Bonding of Composites using Electron Beam Processing
its results are described in Section 11.2.2. The IATA programmes designers were
encouraged to be forward-looking and futuristic. They asked how novel integrated designs
using EB curing and bonding as an enabling technology could radically reduce fabrication
costs. Their results are summarised in Section 11.2.3.
The most recent large aerospace programme to examine EB composite curing and bonding
is the CAI, a cooperative programme between the US Air Force, US Navy, Lockheed
Martin Corporation, Boeing and Northrop Grumman. One of the CAI demonstration
articles is an EB-cured and bonded keel-duct interface. The part is based on an IATA
design, similar to that shown in Section 11.2.3. CAI has also tested many of the
commercially available EB-curable resins and adhesives and is creating benchmarks,
comparing the systems to thermal systems on aircraft or spacecraft, that have flown or
are planned to fly. The CAI data was not available for this review, but will soon be
available from the companies participating in the testing [11]. A material selector CD-
ROM with processing data is also planned for release in the future.
The high production rate achievable with EB processing is an important advantage for
automotive production. As early as 1992, researchers at Chrysler were considering the
possibility of a completely EB-cured and bonded automotive body [12]. This concept is
described in Section 11.3.1. They decided that such extensive use of an unproven technology
was too large a technological leap. The Composite Concept Vehicle (CCV) was instead
produced from resin injection moulded thermoplastic [13, 14, 15]. One possibility for
bonding this car is with EB-curable adhesive, as described in Section 11.3.2. Other
applications of EB curing to ground vehicle structures are described in Section 11.3.3.
EB-curable resins and adhesives for aerospace and automotive applications cure by either
cationic or free-radical polymerisation. Aerospace programmes have primarily adopted
cationic resins, because of their higher use temperature and low shrinkage. Oak Ridge
National Laboratory has coordinated the largest development effort in cationic resins
and has licensed their materials to several commercial suppliers [16]. Free radical resins
and adhesives have an advantage in that they cure more quickly than cationics and are
not inhibited by nucleophilic (alkaline) materials on fibres or surfaces [12]. Cationic,
free-radical and interpenetrating network EB-curable resins for composite matrices are
described in Section 11.4.1. Section 11.4.2 describes the current status and the
requirements for EB-curable composites, primarily in the aerospace industries. Additional
research planned to improve EB-curable resins are summarised in Section 11.4.3. The
current status of EB-curable adhesive development is described in Section 11.5.
Several new facilities devoted exclusively to EB composite curing and bonding have
recently begun operation. The capabilities and cost of these facilities are described in
Section 11.6. Section 11.7 contains conclusions and expected future directions in EB
composite curing and bonding.
463
Handbook of Polymer Blends and Composites
The French company Aerospatiale has manufactured solid propellant rocket motors
for more than 20 years. The cases were traditionally made from filament wound carbon
fibres with heat-cured epoxy resin. Beginning in 1979, they looked for ways to increase
production speed and to reduce thermal stresses during cure and cool-down. Researchers
found that a combination of EB and X-ray curing reduced their curing time from four
days to less than eight hours while keeping the structures near room temperature [6].
Because appropriate EB-curable resins did not exist, they developed an acrylated
formulation with the low viscosity needed for filament winding [17].
Aerospatiale was the first to realise the need for new fibre sizings (sizings and finishes
are coatings applied to the fibres to improve adhesion and processing) to improve the
adhesion of EB-curable resins to carbon fibres. They developed specialised coupling
agents that improved the fibre-matrix interface by creating covalent bonds between
the fibre and the resin [18, 19]. The hydroxyl groups on the carbon fibres react with an
isocyanate group on the sizing, and the hydrogen bonds on the sizings acrylate group
are free to crosslink with the resin during EB curing. A standard rocket motor test
(burst strength test) is to fill the cylindrical cavity with fluid and increase the pressure
until the casing fails. Aerospatial found that motors incorporating EB-curable resins
and their improved sizings had burst strengths higher than their conventional heat-
cured structures.
Following Aerospatiales lead, several groups have recently used electron beam curing
to fabricate filament wound aerospace components. A US Army-sponsored programme
investigated the feasibility of developing faster, more economical manufacturing
techniques for filament wound composite motor cases for small diameter tactical rocket
motors [24]. They produced thin-wall pressure vessels from a variety of EB-curable
resins and demonstrated the use of low cost foam tooling. They achieved burst pressures
that compare favourably with their baseline thermally cured pressure vessels. Figure
11.4(a) shows the pressure vessels during EB curing. Figure 11.4(b) shows one of their
pressure vessels after burst testing.
464
Curing and Bonding of Composites using Electron Beam Processing
(a) (b)
Figure 11.4 (a) Filament wound pressure vessels during EB curing. (b) After burst testing.
(Reprinted from [24] with permission of R. Foedinger, DE Technologies, King of
Prussia, PA, USA)
A group in Italy patented an in situ curing process that combines layup and EB curing in
one system and has developed a system containing a low energy electron beam accelerator
coupled to a two-axis filament winder for in situ curing of pressure vessels [20, 21].
Their layer-by-layer curing process is applicable to a variety of composite production
processes and is capable of producing parts of arbitrary size. Using their layer-by-layer
process, the group was able to achieve interlaminar shear strengths equal to that obtained
when curing their structures all at once using a high energy (10 MeV) electron accelerator.
Using a similar process, a US group has coupled a low energy electron beam gun to an
465
Handbook of Polymer Blends and Composites
automated tape placement machine for in situ curing [22, 23]. Both systems are shown
and described in more detail in Section 11.6.
A 1998 programme at the NASA Marshall Space Flight Center (MSFC) examined the
use of EB curing to correct manufacturing problems with their 267,000 N-thrust
FastTrack rocket motor. They were experiencing difficulties due to a thermal expansion
mismatch between the filament wound graphite and the silica-phenolic tapewrap at the
throat of this one-piece composite nozzle. They showed that EB curing was a practical
alternative, as long as the dose rate (and therefore heating due to EB dose and resin
exotherm) was kept low. The part produced is shown in Figure 11.5.
Figure 11.5 Filament wound nozzle cured by EB near room temperature to avoid cracking
due to thermal expansion mismatch between the silica (centre) and graphite overwrap.
(Photograph courtesy of William McMahon, NASA Marshall Space Flight Center,
Huntsville, AL, USA)
The DARPA-sponsored APCS programme examined many of the key issues facing the
aerospace industry as it began to adopt EB curing and bonding. The overall goal of the
APCS programme was to see if EB curing could significantly reduce the recurring and
non-recurring cost of aerospace composites. Northrop Grumman personnel evaluated
EB-curable materials, developed new processes, accessed EB cost benefits, and produced
demonstration assemblies including a full-scale article. Although new material
development was outside the programme scope, resin suppliers produced novel systems
to meet Northrop requirements.
466
Curing and Bonding of Composites using Electron Beam Processing
and a modified epoxy. These were compared with a heat-curable Cytek-Fibreite 3501-6
baseline. The cationic epoxy CAT-B resin (Applied Polymerics, Benecic, CA) had the best
overall properties [9]. CAT-B demonstrated a dry glass transition temperature (Tg) of
180 C, close to the 190 C baseline. Composite lamina and laminate properties were
measured on woven and unidirectional material forms. The CAT-B properties approached
those of Cytek-fiberite 3501-6 except in compression and shear strength. Northrop
attributed this to poor resin-fibre adhesive, as indicated by dry fibres in the failed region
of the specimens. It is believed that a fibre sizing or a coupling agent compatible with the
EB-curable resin can improve the interface.
The culmination of the APCS programme was the fabrication of a full-scale demonstration
article. Northrop Grumman selected the aft centre fuselage assembly of the F/A-18E/F
fighter plane containing engine inlet ducts, frames, fuel cells, keels, dust-to-side skin
webs and side skins. A wide range of part forms, resins and fabrication processes were
used to demonstrate all the EB curing and bonding processes examined during the
programme. Materials and processes used included:
Woven AS4 (Hexcel)/Cat-B prepreg and IM7 (Cytec-Fiberite)/RB-47 tape with hand
layup.
Resin transfer moulding of woven AS4 with VEB-2 (applied Polymerics) and CAT-
B resins.
467
Handbook of Polymer Blends and Composites
Northrop Grummans materials and assembly methods for the EB-cured fuselage assembly
are shown in Figure 11.6. Figure 11.7 shows the assembly undergoing EB irradiation of
adhesive bond lines during final curing steps. After assembly, the fit of the frames to the
duct showed a maximum gap of less than 1 mm.
Figure 11.6 The aft centre fuselage assembly of the F/A-18E/F fighter was selected by
Northrop Grumman to demonstrate a variety of EB curing processes and materials.
EB-curable materials used included CAT-B cationic epoxy prepreg and Loctite 334
adhesive. Syncore is a foam core with glass beads (high modulus).
(Figure courtesy of R. Vastava, Northrop-Grumman Co. El Segundo, CA)
Figure 11.7 Duct assembly undergoing EB curing as part of the APCS programme. The
EB gun (at right) is directed along the adhesive bondline to cure the duct splices.
468
Curing and Bonding of Composites using Electron Beam Processing
Cost studies constituted a significant portion of the APCS programme. Northrop conducted
cost trade studies for composite monolithic details, sandwich and integrally stiffened structures,
and various assemblies. They found that the cost savings potential of EB processing varied
with the structure, complexity and manufacturing process, and ranged from approximately
10% for simple details with direct process substitution up to 60% for a complex assembly.
The cost estimates included the cost of EB accelerator systems, buildings with the required
features, part handling systems, operating, maintenance and repair. Typical EB facilities
costs are described in Section 11.6. Capacity analysis for each facility and return-on-
investment analysis were also performed. The pay-back period naturally depends on the
volume of composite parts processed. Northrop found a two-four year payback period
for the assumptions of a single programme using EB, a seven year equipment life, an
equipment acquisition value of $5M and a total of 54,000 to 90,000 lbs of composite
processed. Typical return-on-investment curves are shown in Figure 11.8.
Figure 11.8 Return on investment as a function of savings for an electron beam aerospace
composite processing facility. Based on data collected by the APCS programme [8].
469
Handbook of Polymer Blends and Composites
structures. IATA selected for design review the wing/fuselage section of the Joint Agency
Strike Fighter/Assisted Take Off and Landing (JAST/ASTOVL) aircraft. because it is the
most costly portion of the airframe and the most structurally challenging for composites.
The IATA programme strategy reflects where production dollars are spent during
manufacture of high performance aircraft. For prototypes or short production runs, cost
is dominated by the tooling (up to 70% of final costs), which is related to the number
and complexity of the parts required to produce the final structure. For larger production
runs, cost is dominated by the recurring costs for fabrication and assembly (up to 65%
of total costs). The primary fabrication cost elements are material placement and
processing, which can be 40-70% and 25-35% of fabrication costs, respectively, depending
on the number of parts and their complexity. Unlike metallic parts that can be fit and
assembled economically using mechanical fasteners, the mechanical assembly of composite
structures is expensive, and is the largest contributor to scrapped parts.
To reduce costs, the IATA programme examined integration of parts for reduced part count
and combinations of manufacturing processes including advanced fibre placement, EB-cured
prepreg and resin transfer technologies (RTM, VARTM). They also examined co-processing
technologies and unique methods for assembly that eliminates the extensive use of fasteners.
Among the results of the programme were many detailed designs, several demonstration
manufacturing parts and preliminary material specifications for EB-curable resins and
adhesives. One of the IATA fuselage section designs is shown in Figure 11.9. A
EB-curable
Vartim tri-resin adhesive bond line
Keelson E-beam (typical OF14)
cured: 2 required EB-curable adhesive bond
line (typical OF7)
Figure 11.9 An IATA design utilizing EB curing and bonding to reduce costs. (a) The
VARTM tri-resin keelson (central structural element of an aircraft) is EB-cured and
rotated into place. (b) An integrated EB-cured tape-placed upper skin is lowered into
place. EB-curable adhesive bond lines are also shown.
(IATA figures courtesy of D. Sidwell, Lockheed Martin Corp, Palmdale, CA)
470
Curing and Bonding of Composites using Electron Beam Processing
demonstration part, shown in Figure 11.10, was fabricated to verify the concept design
and provide a check on cost estimates.
A windshield frame and arch for the T-38 Talon (Northrop-Grumman) is one of the few
EB-cured aircraft components that has flown. The T-38 Talon is the US Air Forces primary
supersonic jet training aircraft. Two windshield assemblies were fabricated, substituting
EB-cured materials for their thermally cured analogues [26].
The windshield frame is made from four components that are assembled into one
windshield frame and one windshield fairing. The four components are the fairing,
innerskin, bulkhead doubler (a thermal shield for the bulkhead) and outerskin/arch. The
first three components are sufficiently thin (< 5 mm) to be easily penetrated by electrons
with energy of less than 10 MeV. The outerskin/arch is quite thick, and required the
deeper penetrating ability of X-rays to cure the adhesive through the stainless steel. The
prepreg frame after lay-up and during cure is shown in Figure 11.11. The frame survived
471
Handbook of Polymer Blends and Composites
Figure 11.11 An EB-cured and bonded windshield frame and arch for the US Air
Force T-38 supersonic trainer. (a) Prepreg layup, (b) During EB cure.
(From [21], used with permission)
bird impact testing with damage comparable to frames fabricated using conventional
processing [27].
The T-38 is one of several projects that required deeper penetration than is possible
using direct electron irradiation. Researchers at Aerospatiale, Oak Ridge National
Laboratory and at NASA MSFC have also used X-rays to cure thick filament-wound
structures. The effects of X-rays and of electrons on radiation-curable materials are very
similar, since both cause polymerisation via secondary electrons. However, X-ray curing
is much slower than direct EB curing.
Placing a thin, high-density water-cooled metal target in front of the electron beam
generates the X-rays. Up to five percent of the EB energy can be converted to X-rays
in this manner. The low conversion efficiency together with X-ray absorption
throughout a large volume is the reason why X-ray curing is a slow (and expensive)
process.
Although the majority of research funding for EB curing and bonding has been for
aerospace applications, the technology offers significant advantages for ground vehicle
fabrication. A primary advantage of EB curing for automotive applications is cure speed.
An entire composite automotive frame can be cured in less than two minutes using
472
Curing and Bonding of Composites using Electron Beam Processing
currently available high-power electron accelerators [12]. However, curing is only a small
portion of the technical challenge of automotive assembly with EB. Moulding, tooling
and assembly issues for EB-curable automotive parts are major challenges, but have
received little attention to date.
The first substantial work in this area was in 1991 by Billiu [28, 29]. Billiu devised a
method to form a hollow section fibre-reinforced composite by injecting a two-part
resin that reacts to form a thermoplastic. Using supersonic tack welding, the sections
can be fusion welded together and then the B-staged composite can be cured to a
hard thermoset with electron beams. Billiu brought his ideas to the attention of the
Chrysler Corporations advanced technology group, who were investigating ways to
make an all-composite car body. In theory, an entire car can be EB cured, even the
adhesive and the paint. Chrysler studied the idea and designed an automotive production
facility that included injection moulding and EB curing.
The Chrysler advanced technology group determined that there are too many technical
risks to this approach, and instead decided to produce an all-composite thermoplastic
car frame using injection moulding [13, 14, 15]. The result is the CCV shown in Figure
11.12. The CCV body is composed of six large injection-moulded pieces made of glass-
reinforced polyethylene terephthalate (PET). Four of these sections are shown in Figure
11.13. The CCV inner- and outer-frames are bonded with adhesive to themselves and
to a steel chassis. EB curable adhesive can be used to bond these parts together. This is
described in Section 11.3.2.
473
Handbook of Polymer Blends and Composites
Figure 11.13 Inner and outer frames are bonded together to form the rigid CCV
body. One method to bond the CCV frame to itself and to the steel frame is with EB-
curable adhesive. Other thermo-plastic components (not shown) include the roof
and undercarriage.
The baseline materials for bonding together automotive body panels are heat-curable
adhesives. Conventional heat-curing adhesives require many minutes to bond. Total
bonding time includes the time to heat the material at the bond-line plus the time to cure
the adhesive. Plastic automotive bodies such as the CCV are effective insulators, preventing
heat from reaching the bond-line. Oven or forced hot-air temperatures must be limited
to avoid deforming or melting the thermoplastic. These effects combine to give hot-air
bonding times of many minutes, whereas automotive manufacturers prefer cycle times
under two minutes.
Some of the bonds in the CCV join the plastic to a steel frame chassis. After joining with
heat-curable adhesives, some sections of these bonds and near-by composite can crack
when cooled to room temperature. The cracking is due to the thermal expansion mismatch,
caused by the much larger contraction of the plastic than the steel during cooling.
To avoid these problems with heat-curable adhesives, some manufacturers have formulated
adhesives that cure at room temperature. Once mixed, these systems have a very short
working life before they harden. Such systems are well suited to bonding of small areas,
but are very difficult to use on large areas such as automotive bodies. Designers would
prefer a room temperature command cure, which provides extremely rapid curing once
the adhesive, parts and tooling are aligned in place. Curing of adhesive with electron
beams provides this processing flexibility.
474
Curing and Bonding of Composites using Electron Beam Processing
Automotive adhesives for structural bonding are much more flexible (although not nearly as
strong) as those used to bond aerospace structures. It is difficult to maintain tight dimensional
tolerances in moulded parts, so the bond-line thickness varies considerably in the CCV (and
in automotive applications generally.) The adhesive for the CCV thus has the additional roles
of shock absorber and of caulk or sealant. Researchers at Chrysler and Science Research
Laboratory, Somerville, MA (SRL) found that for the CCV, EB-curable adhesive requirements
included a lap shear strength of 3.5-5 MPa, an elongation to failure of more than 100%, and
an immediate cure strength of more than 50% of final strength. They developed adhesives
that met these specifications and demonstrated bonding of CCV panels [30, 31]. Although
the CCV is not being produced (due to market issues and difficulties in part moulding),
Daimler-Chrysler is considering EB curing for other automotive parts [32].
The US Army has sponsored several projects to determine whether EB curing has advantages
for military vehicle fabrication. In one of these programmes, side skirt armoured panels for
the Composite Armoured Vehicle (CAV) Advanced Technology Demonstrator (ATD) were
fabricated using EB curing [31]. The CAV is the first armoured combat vehicle to use a
fully integrated armour and structural composite hull [33]. Figure 11.14 shows a picture
of the CAV with the location of the EB-cured armoured panel indicated. The demonstration
Figure 11.14 Side skirts for the CAV were fabricated using EB composite curing. The
seven-layer panels consist of composite, rubber, adhesive, ceramic tile, electromagnetic
interference (EMI) shielding and signature protection.
475
Handbook of Polymer Blends and Composites
showed that multiple layers, including base composite, ceramic tile armour and signature
protection, could be cured together at low temperature using EB.
The EB-cured demonstrator panels performed as well in all respects as side skirts fabricated
using conventional resins and heat curing at United Defense L.P., the prime contractor
on the CAV-ATD. Figure 11.15 shows a panel after ballistic testing.
Figure 11.15 EB-cured CAV panels after ballistic testing. The hexagonal pattern is
shine-through of electrons at ceramic tile edges [31].
High energy electrons generate ionic species, free radicals, and molecules in excited
states capable of initiating and sustaining polymerisation. Figure 11.16 summarises
the pathways for induction of polymerisation by EB irradiation. Depending on the
chemistry of the resin system being irradiated, polymerisation can occur by free radical
as well as ionic mechanisms. The polymerisation of acrylic/methacrylic systems, maleic
and fumaric polyester resins, maleimides, and thiol-ene systems proceeds via free-radical
mechanisms without initiators. EB-induced polymerisation of nitroethylene proceeds
via anionic mechanisms, and epoxies are polymerised cationically with the appropriate
photoinitiator under EB irradiation. Of these systems, free-radical-cured systems based
on acrylate and methacrylate functionality and cationically cured epoxies catalysed
using diaryliodonium or triarylsulphonium salts have shown the most promise for
composite applications.
476
Curing and Bonding of Composites using Electron Beam Processing
Figure 11.16 High energy electrons in EB-curable resins and adhesives generate ionic
species, free radicals, and molecules in excited states that initiate and sustain
polymerisation.
The rate of EB initiation can be expressed as Ri=IeGF, where Ie is the electron beam
intensity or current, and G and f are efficiency factors for the production of initiating
species and their initiating efficiency. The value of G depends on the chemical structure
of the irradiated material. Unsaturated materials have G values hundreds of times larger
than saturated moieties.
The formation of radical initiating species occurs readily for vinyl-containing monomers
such as acrylates and methacrylates. Polymerisation proceeds via propagation steps that
occur by the successive addition of vinyl-containing monomers to the active radical centre.
Termination occurs as with thermal systems by several processes including combination,
disproportionation, chain transfer, and radical trapping. The rate of initiation and
polymerisation in EB-cured systems is determined by the dose rate. The polymerisation
induced by high dose rate EB curing can differ significantly from polymerisation initiated
via thermal decomposition of initiators, because the high concentrations of initiating
radicals in EB curing changes the balance between propagation and termination.
477
Handbook of Polymer Blends and Composites
The mechanisms of EB-induced cationic chain polymerisation are not well understood.
However, investigations conducted on the thermal curing of epoxy resins by cationic
mechanisms form a basis for understanding cationic polymerisation of epoxies by
EB irradiation.
Several authors have investigated the thermal curing of epoxy resins by cationic
polymerisation [40, 41, 42, 43]. Such polymerisations are comprised of initiation,
propagation and termination steps. Chabanne and others have investigated the mechanisms
associated with cationic homopolymerisation of epoxy initiated by BF3-amine catalysts
[44, 45 ,46]. The salt RNH3+BF4- is thought to activate the epoxy monomer by forming an
oxonium ion active centre. Propagation has been reported to proceed via two competing
mechanisms known as the activated chain end (ACE) or activated monomer (AM)
mechanism. In the ACE mechanism, growing chain cyclic tertiary oxonium ions are formed,
and polymerisation proceeds by chain addition of monomer. On the other hand, the AM
mechanism proceeds via the addition of molecules containing hydroxyl groups to the
activated monomer. This is accompanied by a charge transfer to regenerate an activated
monomer that continues to react. During the initial stages of reaction and in systems with
low initial concentration of hydroxyl groups, the ACE mechanism is predominant. However,
the presence of hydroxyl groups, i.e., water, polymer chain ends, and co-monomer, can
favour the AM mechanism. Other reactions include the chain transfer reactions in which
478
Curing and Bonding of Composites using Electron Beam Processing
an activated chain end reacts with a hydroxyl moiety, water, or alcohol. This transfer
reaction via an hydroxyl-containing compound results in an inactive oligomer.
Figure 11.17 The formation of an interpenetrating network: The first network forms
at room temperature, and EB irradiation forms a second network, increasing Tg.
479
Handbook of Polymer Blends and Composites
There are several advantages to this type of resin. First, the cure shrinkage shortcoming
associated with traditional acrylate/methacrylate systems is reduced via the physical
constraints imposed by the interpenetrating network and a lower concentration of double
bonds. These systems have a total shrinkage of 57%, with most of this shrinkage
occurring during the C-staging process. Second, the systems maintain good processability
via viscosity control by manipulation of reactive diluent concentration and type. Third,
C-staging allows for EB irradiation after removal of the part from its mould or tooling.
Finally, the IPN structure results in materials with inherently higher fracture toughness.
Some of the drawbacks associated with these systems include relatively low Tg compared
to aerospace-grade epoxies and the need to control the temperature during the staged
curing process.
Most EB-cured composite parts have been fabricated from prepreg or by VARTM. Resins
for these two processes are described in this section.
The mechanical properties of three industry standard prepreg resin systems are given in
Table 11.1. The amine-cured epoxy resin 3501-6 was developed to operate at temperatures
up to 177 C and is recommended for general-purpose structural applications. Cytec-
Fiberites 977-3 amine-cured epoxy has a 177 C dry service capability and a 132 C wet
capability. It is a toughened system designed for use in primary and secondary aircraft
structures or any application which requires impact resistance and good hot/wet
performance. Hexcels 8552 is an amine-cured toughened epoxy resin system
recommended for structural applications requiring high strength and damage tolerance.
It was developed to operate in environments up to 121 C.
Also included in Table 11.1 are the properties of some representative EB-cured systems.
CAT-B is an untoughened cationically-cured epoxy resin representative of cationic systems
developed by the Oak Ridge National Laboratory-led consortium [16] and evaluated
under the APCS programme [7, 8, 9]. The resin designated RB-47 (Applied Poleramics,
Benecia, CA) is a developmental free-radical-cured methacrylate prepreg system supplied
by Aerospatiale that was also evaluated during the APCS programme.
Data in Table 11.1 shows that the Tg of the methacrylate system is lower than that of the
other systems and that the tensile strength of RB-47 is significantly lower. CAT-B has a
dry Tg roughly equivalent to values for the thermally cured systems. CAT-B also possesses
480
Curing and Bonding of Composites using Electron Beam Processing
equivalent tensile strength and tensile modulus to the untoughened 3501-6. The major
shortcoming associated with this cationically cured system, however, is low fracture
toughness.
481
482
also exhibits reduction of compressive and interlaminar shear strength relative to the
3501-6 baseline.
The properties of unidirectional IM7 composites based on the cationic EB-cured resin
systems EB-4 and E-B6 show similar reduction of properties compared to 977-3 and 8552
systems. Compressive and interlaminar shear strength are significantly deficient in these
systems. Thus there is a well-supported trend that carbon composites based on cationically
cured epoxies possess poor compressive and interlaminar shear strength. This is perhaps a
result of the brittle nature of EB-cured cationic epoxy systems and also the existence of a
weak interphase between the carbon fibre and the EB-cured resin system.
VARTM is a liquid moulding process that is driven by vacuum pressure. The reinforcing
preform is typically formed around one-sided tooling, and a vacuum bagging technique
is used to seal the system. Impregnation is carried out using low-permeability media to
rapidly direct flow over large surface areas. This technique can be used for manufacturing
large structures such as roadway bridge decks and large on-deck ship structures like
mast enclosures. This fabrication technique is also under evaluation for cost reduction in
the manufacture of aircraft structures, as part of the Air Force-led CAI programme. A
major requirement of such systems is that they possess a viscosity below 0.5 Pa-s with
adequate working time for processing conducted at ambient temperature.
The properties of several thermally commercially available VARTM resins are given in
Table 11.3. The VARTM resin systems being considered for this comparison are materials
having a service temperature around 100 C. The three thermally-cured reins are all
epoxy-based systems. The Resinfusion 8611 system manufactured by Ciba Geigy and
the Shell 863/Lindride 6k (Londau, Edison, NJ) systems are amine- and anhydride-cured
systems, respectively. Both need to be mildly heated for optimal viscosity and therefore
require processing temperature 2030 C above ambient conditions. SC-15 is a two-
phase toughened amine-cured epoxy supplied by Applied Poleramics Inc., that possesses
a sufficiently low viscosity to allow for room temperature VARTM processing.
The EB-curable systems were developed under the US Army Research Laboratory
programme, Non-Polluting Composites Repair and Remanufacturing for Military
Applications and are based on the epoxy-methacrylate IPN systems described earlier.
VCCM4 is a two-phase toughened system based on VCCM1.
The higher fracture toughness of the IPN systems compared to the Shell 862/Lindide
system.is apparent from the data in Table 11.3. Toughening using second-phase tougheners
483
484
results in a 2.5 fold GIC increase for VCCM4 compared to VCCM1. Toughening is
accompanied by an acceptable viscosity rise, practically unchanged Tg, and slightly
improved flexural properties.
The thermally cured toughened SC-15 has the highest values of GIC and KIC but lower Tg
and modulus values. In general, epoxy-based VARTM resins do not perform as well as
the systems formulated for prepreg applications. This is a direct result of the requirement
for low viscosity at ambient temperature. Thus, the IPN-based EB-curable systems
outperform many of the commercially available thermally cured VARTM systems and
possess a very good balance of properties. These include control over viscosity, fracture
toughness, Tg, stiffness, and strength. IPN based EB-cured toughened VARTM resin
systems with higher Tg (180 C) are currently under development.
485
Handbook of Polymer Blends and Composites
prepreg materials standard to the aerospace industry. The property data for the thermally
cured Shell 862/Lindride 6k system were obtained under the DARPA-sponsored Bridge
Infrastructure Renewal Programme for the design and construction of a carbon-fibre
Composite Army Bridge.
Generally, the composite properties of the EB-cured materials are significantly better
than those of the Shell 862/Lindride 6k system. The EB-cured systems have much better
compressive properties 675.6-758.4 MPa for VCCM resins compared with 550 MPa
for the thermal system. This is perhaps a result of the greater fracture toughness of the
IPN EB-cured resins. In addition, the interlaminar shear strength for the EB systems is
slightly greater than that of the thermally cured system. Carbon-fibre composites fabricated
by VARTM using VCCM1 and VCCM4 also compare favourably with autoclave-
processed 3501-6 based systems.
Much of the early materials development work to date has taken what could be termed
a trial and error approach since initially key factors that affect the performance of EB-
cured resin systems and to some extent what properties they influence were not well
understood. The limitations of todays EB-cured resin systems and composites are clear.
Cationically-cured epoxy systems generally do not possess sufficient fracture toughness.
Composite properties that are sensitive to resin performance such as compressive strength,
suffer accordingly. In addition, interlaminar shear strength of EB-cured composites is
much lower than is acceptable for standard thermally cured epoxies. This is due not only
to the low fracture toughness of the EB system but also to poor interfacial strength
between carbon fibre surfaces that have been optimised for thermal cure of epoxy-amine
systems and not EB-cured epoxies and acrylate/methacrylate composites.
The EB curing community has recognised these limitations, and several research and
development programmes are underway to address these issues. Two examples are the
Army Research Laboratory-led (ARL) Strategic Environmental Research Development
Program (SERDP) programme and a second phase of the Oak Ridge National
Laboratory-led consortium. The ongoing ARL programme is working to improve
fracture toughness, a key barrier to the widespread insertion of environmentally-friendly
EB resin systems in current and future military composite applications. The Oak Ridge
National Laboratory-led consortium has been established to address the issues relating
to poor interfacial performance.
The ARL programme has taken advantage of the C-staging capacity of interpenetrating
network systems to control phase separation of a toughening additive prior to irradiation.
486
Curing and Bonding of Composites using Electron Beam Processing
The goal is to develop a microstructure that provides toughness. The short polymerisation
time for traditional cationic and acrylate/methacrylate systems makes it difficult to control
phase separation and microstructure during EB processing. Further improvements of
toughness for these materials will require a more complete understanding of EB cure
mechanisms and the associated kinetics and thermodynamics of polymerisation.
Historically, much work was carried out to improve the interfacial strength between carbon
fibres and thermally cured epoxy systems. Eventually the proper surface treatments and
sizing materials were developed to optimise this critical aspect of composite performance.
The solutions relied initially on trial and error and more recently have been based on a
fundamental understanding of the chemical and physical processes that occur in the
interphase region. It is well known that fibres have the potential to perturb polymer network
structure in the vicinity of the surface. New mechanisms influencing interfacial behaviour
may be associated with EB cure. These range from chemical incompatibility to physical
phenomena linked only to irradiation. Investigations by the Oak Ridge National Laboratory-
led consortium are addressing these issues by developing sizings and coupling agents to
improve interfacial performance with cationically cured systems. The success of much of
this and similar work will depend on attaining a more complete understanding of the
processing-structure-property relationships of EB-cured epoxies.
Adhesives that cure using electron beams are a subset of radiation-curable adhesives,
which also include systems that cure by ultraviolet light (UV). Until about 1995, all
commercially available EB curing adhesives were developed for UV. During the past 5
years, there has been considerable effort to develop EB-curable adhesives for aerospace
and automotive applications. The aerospace work is in response to the results of the
APCS and IATA programmes that showed that new EB-curable adhesives are needed to
reduce fabrication and assembly costs and to increase design freedoms. There has also
been demand from the CAI programme, which has tested and ranked many commercially
available EB-curable adhesives.
The first experiments on EB-curable adhesives began 20 years ago. Campbell and others
[57] recognised the advantages to include fast curing, reduced air pollution due to 100%
reactive formulations, reduced energy consumption and reduced residual thermal stress.
His group evaluated both acrylate and bismaleimide formulations for bonding 7075-T6
aluminium to itself. Lap shear strengths up to 14 MPa were obtained for the acrylate-
based adhesives (Loctite Company). They found up to 90% strength retention after
immersion in water, jet fuel and acetone at room temperature for 500 hours and after
500 hours in air at 177 C.
487
Handbook of Polymer Blends and Composites
Campbell and Brenner also studied crosslink density effects in epoxy acrylate adhesive
formulations and found that a difunctional reactive diluent resulted in a 20% higher lap
shear strength than the same epoxy acrylate diluted with a monofunctional reactive diluent
[58]. The EB-cured formulations have up to about 5% greater strength than the same
formulations cured with benxoyl peroxide at 80 C.
Other groups have studied electron beam curing of adhesives more recently, with results
indicating that commercially available UV-curable acrylate adhesives bonded to both thermally
and EB cured carbon fibre composites are insufficient for aerospace applications. Lopata
and others [59] studied some of the same Loctite adhesives as Campbell and others, as well
as more advanced Loctite adhesives. They achieved lap shear strengths up to 21 MPa on
sanded or chromic acid-treated composites. Scanning electron microscopy results suggest
that a surface primer used on carbon fibre composites can improve the adhesive bond strength.
Farmer and others [60] studied EB cured epoxy paste adhesives for bonding the Air Force T-
38 supersonic jet trainers composite windshield frame, composed of 36 adhesive-bonded
sheets of 301 hard, 0.41 mm thick stainless steel. The materials were all cationic epoxies
developed originally for composite matrices by the Oak Ridge National Laboratory-led
Cooperative Research and Development Agreement (CRADA) [61]. Their results, summarised
in Figure 11.18, indicated that the EB curable epoxies came within 13% of the thermally
cured adhesive and met the minimum requirements for this application.
Figure 11.18 (a) Lap shear strength of cationic adhesives on stainless steel substrate
for bonding of the T-38 windshield. (Application described in Section 11.2.4). (b) Lap
shear strength with various scrim cloths and Adhesive 11L. The T-38 windshield has a
minimal acceptable bonding strength of 21 MPa. The adhesive labelled EA is the
Hysol EA9628 thermally-cured baseline. (From reference [60], with permission)
488
Table 11.5 Property comparison of aerospace and EB-curable adhesives. The Cytec-Fiberite adhesives are
widely used higher-temperature curing systems. Hysol EA 9396 is the heat curable baseline for this work.
Loctite 334 was the strongest commercially available EB-curable adhesive as of 1996 [9]. The most recent EB-
curable acrylate adhesives meet the 9394/9396 baseline [61]. EB-curable film adhesives require considerable
However, there are several difficulties with using cationic epoxies as adhesives. A lap
shear result of 21 MPa is frequently not sufficient for aerospace applications. Even if this
value is acceptable at ambient temperature, it leaves no margin for reduction in strength
that typically occurs as the temperature increases. Cationic epoxies are inhibited by
nucleophilic compounds often added to adhesives to improve properties as well as found
on the surfaces of cured composite resins and primers. This makes it difficult to improve
their adhesive properties and restricts their range of application. Two free-radical and
eight cationic adhesives were recently tested by the CAI programme. The average cationic
epoxy lost 70% of its strength when bonding a (nucleophilic) 977-3 composite substrate
(Cytek-Fiberite), whereas the free-radical systems maintained full strength.
As of 1996, the Loctite 334 adhesive was considered the best commercially available EB-
curable paste adhesive based on work by Lopata and others [59] and unpublished
Northrop-Grumman internal R&D (IR&D) data. Loctite 334 was used to bond a variety
of EB-cured composites in the APCS programme, including co-cured sub-assemblies and
a full-scale composite fuselage assembly [7, 8]. The properties of Loctite 334 are listed in
Table 11.5. Table 11.1 also shows the properties of commercially available adhesives
including Hysol 9394/6, a heat-curable system that has been qualified by Lockheed Martin
for bonding large cryogenic tanks [25, 62 ,63].
A recent US Army-sponsored programme has led to the first EB-curable adhesives with
properties needed for aerospace applications. The adhesives developed by Science Research
Figure 11.19 Lap-shear strength of one-part (acrylate) and two-part (epoxy acrylate)
EB-curable adhesives on aluminum substrate, compared with a heat-curable baseline.
Testing parameters are Hot: 82 C, Cold: -55 C, Hot-Wet: Water immersion at 90 C
for 24 hours, thermal cycling from 194 C to 140 C [61].
490
Curing and Bonding of Composites using Electron Beam Processing
Laboratory Inc., the ARL and the University of Delaware, are of three types: one-
component paste adhesives, two-component paste adhesives and film adhesives. The
one-part adhesives and film adhesives are based on free radical chemistry. The two-part
dual curing adhesives have an interpenetrating network due to both thermal reaction
and EB-crosslinking. Such a hybrid material is similar in concept to the B-stageable EB-
curable IPN resins described in Section 11.4.1.3 [50] and the dual-curing adhesive used
for bonding of the Chrysler CCV described in Section 11.3.2 [30]. The properties of
these systems are shown in Figure 11.19 and are comparable to the baseline thermally
cured system (Hysol 9394).
Electron beam equipment has been used for the past 50 years for radiation processing of
materials. The many types of EB accelerators and their economics for polymer crosslinking
are described in [64, 65]. With the exception of Aerospatiales system for curing rocket
motors, EB facilities used for composite curing have been designed for other uses, i.e.,
irradiation of wire, tubing and film or sterilisation of medical products.
The scan horn in Figure 11.20 sweeps a curtain of electrons back and forth, while the
transport system moves the part under the horn. This facility uses a cart-on-track system,
while others use conveyor belts to move the product through the concrete radiation-
shielding maze. The concrete wall thickness scales with beam energy, and is typically
1.5-3 meters thick for beam energies in the range of 3-10 MeV. Safety interlocks prevent
Figure 11.20 Sketch of a filament winding device coupled with a low energy electron accelerator.
(Reprinted from Reference [20], by permission from the Society for the Advancement
of Material and Process Engineering (SAMPE)
491
Handbook of Polymer Blends and Composites
personnel from entering the shielded vault while the machine is in operation. Ozone gas
is generated when electrons pass through air and then exhausted from the curing area by
high-speed blowers.
Researchers have proposed and demonstrated other facilities for EB curing and bonding
beside the traditional closed maze. Because narrow concrete walls and a fixed scan horn
place limitations on the size and shape of composites to be cured, Aerospatiale and the
Science Research Laboratory have built large shielded rooms with movable accelerators.
These facilities are accessible to large composite parts through sliding doors or removable
concrete blocks.
Cart or conveyor handling systems do not provide precise positional control, and make
it difficult to attach sensors to parts during cure. A new facility dedicated to composite
curing research uses an accurately positioned table that moves back-and-forth under the
beam to overcome these problems [66]. This will allow the use of in situ cure monitoring
and other in-line measurements.
Constructing new EB curing facilities is very expensive. Cost analysis performed under the
APCS programme estimated $5 million for a facility capable of meeting their production
needs, with the majority of that cost due to items other than the electron beam system [9].
One way to lower this cost is to install low energy (0.2-0.3 MeV) EB systems into existing
facilities and use in situ or layer-by-layer curing. X-ray shielding requirements at these
energies are 2-6 cm of lead or 10-30 cm of concrete blocks. This approach is being used for
filament winding [20] and for automated tape placement [22, 23].
Figure 11.20 shows a sketch and a photo of a filament winding system coupled with
a low energy (0.25 MeV) electron accelerator. The system is enclosed in lead, allowing
personnel to stand next to the machine during operation. Figure 11.21 shows an
Automated Tape Placement (ATP) system coupled with a portable, low energy (0.25
MeV) electron gun. The EB gun is attached to the ATP head, and the entire assembly
is attached to a four-axis gantry system. The ATP system is shielded with a wall of
portable concrete blocks that can be easily installed or removed. The wall can be
seen in the background.
Although the in situ EB curing systems shown in Figure 11.21 are easier to install and
cost less than high-energy systems, obtaining good composite properties are more difficult.
To obtain good consolidation, researchers often use a warm debulk and vacuum bag
pressure prior to high energy EB curing. The ATP head applies heat and pressure for only
a very short time, proving little room for error. Good tape quality with full fibre wetting
is critical to producing in situ cured composites without voids. This is because unlike
conventional autoclave curing, there is no time for resin flow to fill tape voids, and thus
flaws must be avoided during tape fabrication. Voids can also occur if the top layer
492
Curing and Bonding of Composites using Electron Beam Processing
Figure 11.21 In situ or layer-by-layer EB curing allows use of lower energy electron
beam systems. It is thus possible to add the EB gun and localized X-ray shielding to
existing composite facilities. An in situ curing system at NASA MSFC in Huntsville,
AL, USA, laying and curing tape is shown. The EB gun is on the left and the ATP head
is in the foreground. A concrete block x-ray shielding wall is in the background.
springs back due to insufficient memory before it is cured. The degree of spring-back as
a function of EB cure dose, as well as the tack, drape and memory of the uncured tape
are determined by the viscoelastic properties of the EB-curable resin.
Despite these challenges, initial results from in situ curing systems are encouraging.
Researchers using the filament winding system shown in Figure 11.20 report interlaminar
shear strength equal to high-energy EB-cured systems [20]. Researchers using the ATP
systems have found good interlaminar strengths and lower dose requirements than
expected. A heat post-cure was found necessary to achieve full glass transition temperature,
especially when very low (20 kGy) doses were used [22, 23, 67].
Electron beam curing and bonding is a promising new technology that offers technical
and cost advantages for fabricating aerospace and automotive structures. Technical
advantages for aerospace structures include curing at room temperature using low-cost
tooling, and the ability to fabricate large integrated structure including structures too
large to fit inside autoclaves. Studies by aerospace companies have shown potential cost
savings of 10-60% by using EB curing and bonding of integrated structures. For
automotive fabrication, advantages of EB processing include high production rates and
near-room temperature curing and bonding.
493
Handbook of Polymer Blends and Composites
Researchers have spent a significant amount of effort during the past decade to develop
this technology. The majority of this work has been to develop processes for manufacturing
aerospace structures. Some of this work combines EB curing with prepreg and VARTM
composite fabrication techniques. These methods have been used to demonstrate large
integrated structures such as fuselages, and wing skins. Other demonstrations have
combined filament winding with EB curing to produce rocket motors and large filament-
wound pressure vessels.
Considerable work has also been spent to formulate EB-curable materials. EB-curable
resins and adhesives have been developed with properties close to those needed for
aerospace applications. EB-curable adhesives with the properties needed for automotive
bonding are now available.
Several companies have also developed specialised electron beam equipment for curing
and bonding of composites. For example, in situ EB curing systems using low-energy
electron guns with a minimal amount of X-ray shielding are being added to existing
composite facilities.
Despite these successes, EB curing and bonding has not yet been widely adopted in industry,
and the technology remains primarily an area of research. For example, research areas
include toughening of cationic resin systems, improvement of interfacial properties, and
extending use temperature ranges.
In 1997, the Air Force technical monitors for the APCS and IATA programmes evaluated
the potential of EB for aerospace manufacturing [2]. Their conclusions provide a
blueprint for the continued development and commercialisation of EB processing.
Although there has been some additional work in the last several years, especially in
the areas of material testing, pre-qualification and novel EB facility development, their
conclusions remain valid:
494
Curing and Bonding of Composites using Electron Beam Processing
environments are required. Structural properties must be sufficiently defined such that
usable weight estimates are possible. Properties must be evaluated for material
performance at acceptable extremes of material and process parameters. And finally,
for the materials and process to reach a stage where designers can confidently
incorporate it, statistically significant data must be generated. Such a development
programme represents a significant investment.
References
2. F. Abrams and T.B. Tolle, International SAMPE Symposium, 1997, 42, 548.
6. D. Beziers and B. Capdepuy, International SAMPE Symposium, 1990, 35, April, 51.
495
Handbook of Polymer Blends and Composites
15. http://www.media.Chrysler.com.
496
Curing and Bonding of Composites using Electron Beam Processing
32. A. Rich, Daimler-Chrysler Liberty and Technical Affairs, Madison Heights, MI,
USA, personal communication, April 2000.
34. C. E. Hoyle and J. F. Kinstle, Radiation Curing of Polymeric Materials, Eds., ACS
Symposium Series, American Chemical Society, Washington DC, 1990.
35. S. P. Pappas in Radiation Curing: Science and Technology, Ed., S.P. Pappas,
Plenum Press, New York, 1992.
37. J. V. Crivello, M. Fan, and D. Bi, Journal of Applied Polymer Science, 1992, 44,
1, 9.
44. P. Chabanne, L. Tighzert and J-P. Pascault, Journal of Applied Polymer Science,
1994, 53, 6, 769.
497
Handbook of Polymer Blends and Composites
51. U. P. Dalal, Formation of IPN Via Thermal and Radiation Curing, Masters
Thesis, University of Delaware, Newark, DE, USA, 1999.
55. Hercules Graphite Fiber Handbook, 1985, Bacchus Works, Magna UT.
56. Final Technical Report, Bridge Structure Renewal Program, Defense Advanced
Research Project Agency, MDA972-94-3-0030, Arlington, VA.
498
Curing and Bonding of Composites using Electron Beam Processing
499
12
Composites: At the Turn of Century
12.1 Introduction
Military technology has always lead to civilian spin-offs. The Internet is a living example
of Advanced Research Projects Agency Network (ARPANET), which started as a military
communication network in the United States. Likewise the composite technology also
gained momentum due to World War II (see Chapter 1).
In this chapter, applications that are mostly catering to the needs of the global society
and represent the high-tech end of the spectrum are discussed. We have attempted to cite
emerging areas which give increased emphasis and provide future directions for growth.
In doing so replication of information has been avoided as far as possible.
With this constraint, the focus has been narrowed to a few selected examples which are
by no means comprehensive but represent the state-of-the-art.
While aircraft and military applications of composites comprise only 0.8% of the total,
civil applications account for the most (19.7%) in the USA. In the European market on
the other hand, the use of glass reinforced plastics (GRP) is as high as 56% and that of
reinforced thermoplastics (RTP) is 19%.
Macromolecules offer many possibilities for creating smart molecules. Smart molecules
are designed (Figure 12.1) to receive a stimulus, transmit or process it, and then to respond
by producing a useful effect. New capabilities exist for the predictive design of polymer
molecules from the rapid advances in computer modelling of the structure and properties
of large molecules. Powerful chemical methods are available for the creation of polymeric
modules which can be assembled into smart molecules.
501
Handbook of Polymer Blends and Composites
The combination of molecular modelling, tunnelling and atomic force microscopy, and
state-of-the-art synthesis gives us a new dimension for fabricating intelligent composites.
Complicated molecules can now be modelled easily and quickly. Individual parts of these
molecules can be synthesised and then assembled together using the techniques of
contemporary polymer synthesis. Direct observation of molecular scale features with
new molecular resolving microscopes provides other useful information about the design,
assembly and performance of these smart molecules [2]. This project brings all these
things together. The early results on all three fronts of this project are most promising.
1. Attachment to the fibre - a weak link that breaks as soon as a pre-determined amount
of separation occurs.
2. A piezochromic group that changes its colour when the separation occurs.
3. A loose loop of polymer which maintains a useful level of attachment even after the
fibre and the matrix have separated by an anticipated amount.
12.4 Nanocomposites
Periodically, a new field of science and technology emerges that is not merely evolutionary,
but promises revolution in many aspects of our lives. Nanotechnology is emerging as a
502
Composites: At the Turn of Century
new field that has the potential to transform health care and medicine, biotechnology,
manufacturing, energy, and information processing. The essence of nanotechnology is
the creation and utilisation of materials and devices at the level of atoms and molecules
and the exploitation of unique properties and phenomena at the nanoscale.
Scientists know that the catalytic, mechanical, electronic, optical, and other properties of a
material are significantly and favourably altered when that material is fashioned from nanoscale
building blocks. For example, nanocrystalline copper is up to five times harder than
conventional micron-sized copper. And ceramics, which normally are brittle, can be made
more easily deformable if their grain size is reduced to the low nanometer range. Such improved
properties can also be produced in nanocomposites, in which the building blocks, for example,
nanoscale metal particles or nanometer thick filaments or sheets of a ceramic, are dispersed
in a matrix of another material, such as a polymer [3, 4, 5]. Because the building blocks of a
nanocomposite are nanoscale, they have an enormous surface area, and therefore there are a
lot of interfaces between the two intermixed phases. The special properties of the
nanocomposite arise from the interactions of its phases at the interfaces.
In 1989, NASA initiated a programme with industry to develop technology for a high-
speed civil transport (HSCT) Mach 2.4, 1600 mph. This was the second government-
503
Handbook of Polymer Blends and Composites
funded programme in the United States for high-speed commercial airplane technology;
the government funding was about $1.6 billion over 10 years. Several anticipated benefits
are low cost, low density, flame retardance and outstanding mechanical properties.
The programme was managed at the NASA Langley Research Center, it was a joint
activity between industry and NASA; the major portion of funding came from the
government. The programme was divided into different phases; the environmental activity
began in 1989 and individual discipline technology development began about 1993.
In about 1988, Boeing began to evaluate commercially available adhesive and composite
materials such as bismaleimides, cyanates, and thermoplastic polyimides using internal
research and development funds for a Mach 2.4 HSCT. Each material failed to meet the
requirements of the proposed HSCT because of problems with one or more of these
performance factors; processability, long-term thermo-oxidative stability (durability),
solvent resistance under stress, and microcracking after thermal cycling. In 1991, work
was started with funds from the NASA Advanced Composites Technology Programme
to accelerate work on developing structural resins. Using information from years of
internal work in cyclobutene chemistry, Dow Chemical Co., developed a maleimide -
benzocyclobutene monomer that seemed to have many of the features required for a
high-temperature adhesive and composite matrix. However, maleimide - benzocyclobutene
carbon fibre composites lacked durability; they lost an average of ~15% of their
mechanical properties after 5,000 h of aging at 177 C in air.
NASA recognised that existing materials could not meet the requirements of a Mach 2.4
HSCT and began funding more intensive structural resin development in the latter part
of 1993. In mid-1994, a contract was awarded to Boeing with Lockheed Martin Marietta
(now Lockheed Martin), McDonnell Douglas, and Northrop Grumman as major
subcontractors. The airframe materials and structures part of the HSR programme had
a proposed budget of $350 million over 8 years.
Most of the resin selection effort focussed on amide oligomers terminated with reactive
groups. Low-molecular-weight molecules have lower melt viscosities and accordingly better
processability than high molecular weight molecules of the same composition. Processability
is one of the keys to success for a structural resins. Whatever the performance or cost of a
material, if large parts cannot be fabricated reproducibly, the material is little value.
From several possible reactive groups, the phenylethyl group was selected because it
appeared to offer the best overall combination of properties: a wide processing window,
long room temperature (RT) shelf life and an excellent performance for the cured polymer.
During the search for an acceptable resin at NASA Langley, a product of the reaction of
3,3, 4,4-oxydiphthalic anhydride (ODPA), 3,4-oxydianiline, and 4-(3-amonophenoxy)-
504
Composites: At the Turn of Century
During the course of this work, progress was made with other materials. Phenylethyl-
terminated imide materials with very low melt viscosities and good melt stability were
developed for fabricating composite support structures, e.g., frames, via resin transfer
moulding (RTM) and resin infusion.
12.6 Micromoulding
With the naked eye, the mouldings from a microsystem are in a scientific sense scarcely
visible, but their importance for example in medicine, computers, telecommunications,
the automotive and watchmaking industries is becoming ever greater. Out of market
demands for ever small plastic mouldings, the progressive miniaturisation of products, a
completely new branch of technology has developed, micro-engineering. Battenfeld has
demonstrated new, innovative paths using the microsystem.
12.6.1 Benefits
505
Handbook of Polymer Blends and Composites
12.7 Completely Recyclable Fabric Insert Moulding [9] (for Auto Interiors)
There has been a dramatic increase in the use of fabric insert moulding in injection
moulding developments, by the automotive industry. Several factors involving both
consumer taste and manufacturing issues are spurring new innovations in this technology.
Moulders who want to serve the automotive market are adding fabric insert moulding to
their portfolios, as the major automotive builders seem to do a maximum amount of
business with a minimum number of suppliers.
About a dozen Toshiba machines with clamp forces ranging from 310 to 3750 tons are
already being used in North America for fabric moulding, and many more in Asia.
Figure 12.2 Multi-stage injection sequence keeps melt pressures low in the cavity to
prevent damage to the inserted fabric
506
Composites: At the Turn of Century
However, the crux of the problem in fabric insert moulding is moulding the fabric without
damaging it. This generally requires a low-pressure approach for both clamping the
mould and injecting the resin. Each moulding system manufacturer differs somewhat in
its approach to controlling and delivering that pressure.
Rapid prototyping (RP) is a new area of technology involving the automated fabrication
of geometrically complex dimensionally accurate solid objects directly from computer-
aided design (CAD) files. Objects are constructed in a layer-by-layer, additive fashion by
computer-controlled machinery. This technology, also referred to as solid freeform
fabrication, is generic in nature. A single machine can be used to build an infinite variety
of parts without part-specific tooling or machine reconfiguration; a 3D CAD file is all
that is needed. There are several different types of commercially available RP machines
that differ primarily in the type of feed material that can be accommodated. They are
used to produce full-scale design prototypes for products such as aircraft and automotive
components, medical devices, consumer goods, and jewellery. These prototypes cannot
usually be used as fully functional parts because they are constructed from low-grade
materials such as paper, low-strength plastic powder, or wax.
507
Handbook of Polymer Blends and Composites
The University of Delaware, Center for Composite Materials (UD-CCM), USA, have
developed the VARTM process. VARTM offers numerous cost advantages over traditional
RTM via lower tooling costs, RT processing, and scalability to large structures. Recent
advanced technology demonstrators such as the Advanced Enclosed Mast Sensor (AEMS)
System and the Composite Armored Vehicle (CAV) have shown the potential of VARTM
technology for the low-cost fabrication of large-scale structures requiring thick-section
sandwich construction and hybrid multifunctional integral armor. A barrier to the insertion
of this technology is the lack of a science base, sensors, and control, resulting in costly
trial-and-error process development and a high degree of part-to-part variability in
production. Intelligent VARTM fabrication is an enabling technology for the large-scale
adoption of composites for various composite applications. Since it is reliable and
repeatable, low-cost manufacturing is required to reduce processing times and increase
the overall quality of the system.
VARTM is a composites manufacturing process that involves the lay up of dry reinforcing
fibres in fabric tape, or bulk form as a preform in a mould and impregnation of the
preform with liquid resin using negative pressure, i.e., vacuum, followed by cure and
demoulding. The advantages of the VARTM process include the requirement for only a
single-sided, non-heated mould, which greatly reduces tooling costs and initial investments.
In addition, large parts can be moulded rapidly, and they can be infused rapidly, and the
use of resins that cure at room temperature such as vinyl esters can greatly reduce costs
by eliminating the need for large ovens. VARTM is a completely closed system that traps
volatile organic compounds (VOC), reduces the need for solvents, and results in less
scrap than other processes.
508
Composites: At the Turn of Century
Tooling costs are substantially lower than for similar processes, such as closed-mould
RTM, because only a one-sided mould is required. Multiple fibre layers are placed in
the mould and covered with a layer of bagging film. An airtight seal is then formed
between the mould and the bagging film with tacky tape. A vacuum is applied, drawing
the bagging film, and fibre matrix, tightly to the mould. Drawn by the vacuum, resin is
then allowed to flow into and fill, the fibre matrix from a separate inlet.
12.10 Monitoring End of Cure Using Fibre Optics and Ultrasonics [12]
Advanced composites can be fabricated by laminating multiple prepreg plies into the desired
shape and then curing in an autoclave with simultaneous application of proper heat and
pressure. The knowledge of cure process is very important in order to obtain fully cured and
high quality composites at reduced production costs. In situ sensors capable of monitoring
the cure process are therefore desirable.
For a comparison, a new cure monitoring system being developed by using conventional
broad band ultrasonic transducers with Extrinsic Fabry-Perot Interferomeric (EFPI) sensors
(optical devices which can sense minute dimensional changes in objects caused by deformation)
are concurrently used to detect the local strain development in the composite caused by
temperature variation, chemical shrinkage and process-induced stresses. The results obtained
from the two sensors were compared in order to evaluate their suitability for in situ monitoring
of autoclave processing [12]. Discussions were detect the end of cure, relative to the traditional
degree-of-cure method using the differential scanning calorimetry (DSC) technique.
12.11 GMT
New applications for these materials are tending towards larger parts integrating more complex
geometric features as for example in the sub-floor of the Mercedes A-class. Reducing part
thickness is also a major priority in order to decrease part weight and volume. To produce
larger and/or thinner parts, higher moulding pressures are required; thus the development of
easy-flowing materials is a major driving force for the GMT industry, as a means of limiting
509
Handbook of Polymer Blends and Composites
equipments costs. The flowability of GMT should be improved while maintaining mechanical
and part quality performance. The development of composites for compression flow moulding
such as thermoplastic GMT or thermoset sheet moulding compound (SMC), requires a
quantitative understanding of the relationship between the properties of the composite and
those of its constituent materials. The properties and the processability of the composite can
be tailored to application requirements by careful matching of the fibre structure, the matrix
properties and the interface between the two phases.
510
Composites: At the Turn of Century
But commercial compounders have yet to embrace wood fillers, despite its enticing
properties, lower weight, faster cycle and lower cost, because it just isnt very easy to do.
It takes a different screw mixing configuration in a co-rotating twin-screw extruder to
distribute wood fibre gently - it is extremely heat sensitive. Overheat it, and it just
chars, said Mike Ford, Vice President of Marketing of Natural Fibre Composites in
Baraboo, Wisconsin. Ford became interested when he did market research for the USDA
lab, and he is the first fully commercial compounder. They have been operating for eight
months and have commercialised several applications already, including a PP compound
for injection moulded coat hangers, which look like wood and are the colour of wood.
Natural fibres uses a Davis-Standard twin-screw extruder with capacity for 310,810
tons/year and dries the wood and also needs to dry the pellets, unless they are used
within a few weeks. Global Resource Technologies in Madison, Wisconsin started up a
year ago, using a Reifenhauser twin-screw extruder modified with special venting to
remove moisture from the melt before its pelletised. Phoenix Color & Compounding in
Sandusky, Ohio, has worked with wood fibre the longest (7 years) and commercialised a
pelletised concentrate that is 85% wood fibre in 15% proprietary binder. They plan to
work with one or more of the wood-fibre compounders to commercialise new products.
Materials are light weight and have 30% faster cycle times than unfilled polymer, but
they are different to work with.
Compounds of homo and copolymer PP with 20% loadings of wood fibre from scrap
pine and demoliation wood are comparable to compounds with 40% talc, fibreglass and
calcium carbonate, except for weaker unnotched Isod properties.
Significant progress has been made over the last several years in the development of
advanced composites, especially for structural applications in aircraft. The majority of
the work has been focussed on thermosets such as epoxies or addition type polyimides as
matrix materials. However, because of requirements for better mechanical properties
and solvent resistance, various thermoplastics are being actively evaluated.
One of them, PEEK (see Figure 12.4), a highly aromatic semi-crystalline thermoplastic,
is currently receiving considerable attention. Although it has a high melting point, PEEK
can be processed on conventional extrusion and injection moulding machines. Because
of its balanced properties, PEEK may find increasing use in applications such as electrical
and electronic parts, military equipment, wires and cables in nuclear plants, underground
railways, and oil wells, as well as advanced structural composites for aircraft.
511
Handbook of Polymer Blends and Composites
Figure 12.4 Structure of PEEK (a) formula (b) unit cell and (c) spherulite
512
Composites: At the Turn of Century
longitudinal sections of a high strength fibre, the EMU are linked together in wrinkled
sheets oriented roughly to the fibre axis. Longitudinal bending forms elongated voids
between the carbon sheets. In transverse sections, carbon layers exhibit an intricate
structure resulting from their strong radius of curvature. It is thought that the transverse
radius of curvature is due to the heteroatoms (such as nitrogen atoms), if the model
drawn outcropping the external surface of the fibre is considered. From detailed
observations of transverse cross-sections, polyacrylonitrile (PAN)-based carbon fibres
exhibit three different structures at their surface.
Kevlar is the first successful LCP fibre which has found widespread application as a
reinforcement for composites. Other well-known LCP polymers are Vectra fibres from
Hoechst Celanese and Xydar from Amoco. Structure of a typical LCP is shown in Figure
12.5. Properties of LCP fibres are similar to Kevlar but have an upper end-use ceiling of
approximately 350 C. LCPs are strong candidates for future applications in composites.
513
Handbook of Polymer Blends and Composites
There has been a dramatic increase in the use of cured thermoset materials in the last 20
years. Applications in the automotive, electrotechnical and building industries have led
to the worldwide breakthrough. Important aspects have been the outstanding relationship
between performance and price, high strength and stiffness and the possibility of
integrating different functions in one part. The typical service life of such technical part
is between 10 and 20 years.
A total worldwide market of 3 million tonnes had emerged by 1992, 40% of it in Europe
and the US and 20% in Japan. Most of these materials are made with a thermoset matrix.
Thermoplastic materials have shown a significant increase in the last few years and already
have 29% share of the European market. The thermoset matrixes are mainly of epoxy,
polyurethane or unsaturated polyester resins. The reinforcing fibres are primarily glass
fibres, with a much smaller volume of carbon and textile fibres also being used. Typically,
SMC are used for laminated sheets and bulk moulding compounds (BMC), mainly for
injection moulding. During processing, these materials will flow readily in a mould and
their mechanical properties can be tailored to the desired application.
The supposed disadvantage of thermosets, namely that they do not melt, proves to be an
advantage in particle recycling since parts from different raw-materials suppliers and part
manufacturers are compatible with each other and can thus be re-used together. In addition,
painted or pigmented parts, dusty and contaminated post-consumer parts are not very
critical as they can be re-used without washing. This makes the logistics of re-use much
simpler and it creates the volume of material necessary for achieving economic viability.
Just as much of the recycling success with plastics uses the recovered waste to produce
new products different from the original use, so it is with glass and other waste (crushed)
glass bottles and asphalt. In effect, the glass is substituted for conventional rock material
to provide a matrix-filler. Many US States now require at least 10% of highway materials
mixtures to be shredded tyres. Some success has also been achieved in mixing up to 20%
flash waste in aluminum which produces a relatively high-strength composites.
Carvalho and others [19] describe an effort to develop a plastic and glass (glastic) waste
composite prototype with similar commercial applications. This glastic composite prototype
consists of a mixture of eight coarse, polymeric powders, and three different weight proportions
of coarse-ground waste glass, 15%, 30%, and 45%. These prototypes have been prepared in
similar form to commercial clay bricks which are used as a standard of reference.
514
Composites: At the Turn of Century
The strategy of this research was to eliminate the cost of sorting individual polymers
from the waste stream, by moulding unsorted thermoplastic waste (with added glass).
This approach mandates that applications of the product should not require high tensile
strength. Applications such as brick veneer for residential construction, curtain walls,
or patio and landscaping paving stones would meet the requirement. The Glastic
composite prototype materials all outperformed standard clay bricks in compression
strength testing, although it is not envisioned that the prototype would be a substitute
for commercial brick in construction applications. These composite prototypes
demonstrate the potential for utilising plastic and glass waste mixtures in other novel
systems such as highway materials compositions which could include asphaltic based
compositions including shredded rubber. The potential applications of such material
systems should receive more attention as attention as alternatives to the extraordinary
landfill problems worldwide.
Thermoplastic composites are competing with innovative systems to reverse a trend back
to steel in bumper beams. The most novel offer is the industrys first I-section bumper
beam, developed for the 1999 Mitsubishi Galant.
Composite Products, Inc., has developed a manufacturing system [20] that incorporates
delivery of long glass fibre reinforced PP from the extrusion stage to the compression
moulding of the part. The process disperses the long glass fibre into the resin without
breakage for optimum strength, appearance, and lower cost for door frame rails for the
Chrysler Jeep TJ.
Burning of polymer matrix composites in post crash aircraft fibres generates a complex
mixture of combustion products comprised of gases, organic vapours, and particulate
matter including airbourne carbon fibres. There is concern among the fire fighting,
investigative, and mishap response communities that an unusual health hazard is
posed by this combination of combustion products. The nature and potential hazards
of acute exposure to airborne carbon fibres from fire and explosion involving advanced
composites materials have been studied. Data from fire tests and crash-site
investigations suggest that a small fraction of the fibres released in fires are respirable
and can be inhaled deep into the lungs. Most of the carbon fibres produced in fires
are 2-10 times larger than the critical fibre size associated with asbestos toxicity, and
their concentration is well below Occupational Safety and Hazards Act (OSHA)
515
Handbook of Polymer Blends and Composites
recommended levels for chronic exposure. At issue however are the toxicological
effects of adsorbed combustion products. Chemical extraction shows that a larger
number of toxic organic compounds are adsorbed on these fibres, several of which
are known carcinogens and mutagens in animals. At the present time there is no
conclusive evidence linking airbourne fibres from burning composites to any unusual
health hazard. However, no toxicological studies have been carried out to assess the
long-term health effects from exposure to a single high dose of fibrous particulates
and any synergistic interactions with the organic chemicals.
It is very important to document the on-site exposure among the personnel responding
to crash site to determine the specific protective gear requirements. Guidelines and
procedures have been developed by the US Airforce (USAF) for adequate protection
of the various response personnel and handling and disposal of composite wreckage.
These guidelines should be made available to civilian and airport fire fighting and
rescue personnel that are often the first to respond to postcrash fires.
The advantages of composite materials for marine structures, notably their high specific
properties, resistance to degradation in water, and flexible fabrication to produce special
shapes, become even more valuable when underwater applications are considered. For
composites used underwater at the present time, there are three types of application:
516
Composites: At the Turn of Century
a) Oceanographic
b) Submarine
c) Offshore
The first are often simply space frames designed to provide a stable support for
instrumentation. Recently a composite lander has been developed at Ifremer Plouzan,
France. This is a glass/epoxy structure, 1.5 m high, which is designed to support a range
of probes and instrumentation on the ocean floor. A major reason for the resplacement
was the corrosion products from the aluminium, which interfere with some of the
measuring devices. This type of support frame can be made by hand lay-up. It must be
able to withstand accidental impact, but the loading during service is not very severe.
517
Handbook of Polymer Blends and Composites
The DCN has used glass/resin composites since the 1970s to build some of its ships.
There are three large categories of organic matrices which are currently used: epoxy,
vinylester and most of all polyester. Most of the time, these materials are reinforced
with a high density glass woven roving, i.e., 800 g/m2. One of the DCNs current
concerns is to improve the fire behaviour of ship structures by building them from
fire resistant materials with minimal changes in the usual building technology, i.e.
wet process. For many external applications, the use of halogenated resins systems,
in place of a traditional resin can be considered. This solution presents the advantage
that the processing is not basically modified. The impact on staff and on equipment
of the thick toxic and corrosive smokes evolved during a fire will be limited because
of the localisation of these materials: external sections of superstructures and external
parts of hull.
518
Composites: At the Turn of Century
The properties of this polymer can be attributed to its chemical structure. The aromatic
rings are dominant components in the DGEBPA composition, and are mostly
responsible for the strength, for the resistance to high temperatures and radiation of
this material, resulting from its absorption and dissipation of energy, without bond
disruption.
The level of radioactivity is one of the key elements in the design. Since the container
must maintain its structural integrity for at least 500 years, means of assessing the
effects of the total dose rate corresponding to this lifetime is to put in place. Based on
existing works on epoxy and epoxy/carbon composites, these systems could take the
total dose generated by low level waste (LLW). However, the use of radiation absorption
substances as filling material in the container would decrease the dose reaching the
container wall to acceptable values.
Dental composites belong to the class of aesthetic restorative materials to fill cavities,
build up broken tooth, bond orthodontic brackets, etc. These must have adequate
strength and wear resistance to withstand the masticating and biting forces. One of the
main life limiting factors is the marginal leakage of food and bacteria through the gap
between the composite and enamel/dentine due to inadequate bonding.
Dental composites consist of two major components with a resin matrix based on
bisphenol A-glycidyl methacrylate (bis-GMA) and quartz or glass inorganic filler with
additives for activating the polymerisation. Colouring pigments are also added to
aesthetically match the shade with the natural tooth. The fillers impart strength, wear
resistance and also reduce the shrinkage of the composite during in situ polymerisation
in the oral environment. A number of bonding agents have been developed to ensure
integration of the tooth.
519
Handbook of Polymer Blends and Composites
The economics of the medical use of materials is quite different from that of most
other applications. The major costs in medicine are the facilities: hospital, equipment,
etc., and the labour: surgeons, anaesthetists, and other specialists, together with nurses
and support staff. The amount of material in a medical application is unusually quite
small, less than 1 kg, so the cost of even the most expensive materials is usually
insignificant compared with the other costs. In addition, there is the time and suffering
of the individual requiring the device, and use of expensive composites can often assist
speedy recovery.
A new material cannot be accepted immediately for medical use. A device designed for
internal prosthetic use has to be checked for unfavourable reactions, both short-term
and long-term. The same is true for external (orthotic) devices, though the requirements
are generally much less stringent.
Fibre glass is rapidly supplanting wood, leather, and steel for braces (or callipers) for
arms and legs. The braces are stronger, lighter, more comfortable, and less noticeable.
No straps are needed, and improved cuffs, having air holes, and padded with washable
foam, can easily be produced.
Fibre glass is also used for artificial legs and arms. It may soon be supplanted by
carbon-polymer because this has better specific properties. An important advantage
for reinforced polymers is their easy mouldability, in view of the wide range of shapes
and sizes required. The prostheses are made from plaster moulds of the parts, using
woven reinforcement with open, easily shaped weaves. These prostheses can very easily
be made to mimic the appearance of human limbs.
Internally, it is hoped to replace Vitallium (a cobalt alloy with 20% chromium and a
few per cent of molybdenum) which is used for bone splits, hip and knee joints, etc.,
with carbon-polymers. A great advantage of the composite is the possibility of matching
the modulus of the bone material. This prevents the de-stressing, and consequent
dissolution of the bone adjacent to the joint, that follows the use of the high modulus
metal. Further work is needed here to solve problems of inflammation and necrosis of
surrounding tissue due to the presence of the polymer.
520
Composites: At the Turn of Century
Recent observations suggest that plaited continuous carbon fibre can be used to make
new tendons. The body eventually replaces the carbon with collagen to produce a
natural tendon. Carbon-polymethyl methacrylate is being used for a bone cement. It
has superior fracture resistance and greater durability under load than the unreinforced
polymer. Animal tests have shown its usefulness in fixing internal splints and other
prostheses, and its use is being extended to human subjects.
The combination of excellent mechanical properties and the very low x-ray absorption
characteristics of carbon fibres is providing a unique opportunity for the development
of better x-ray analysis and treatment devices. The angiographic technique (which
involves the injection of a radio-opaque fluid into the bloodstream, for the location
of growths and foreign bodies) needs x-ray pictures to be taken in quick succession
(up to six per second). The film has to be placed and moved on very quickly, but held
precisely. The key to doing this has been the development of carbon-epoxy compression
plates. These can be very precisely moulded from the prepreg tape, and the plate
combines excellent fatigue properties with low x-ray absorption and good elastic
properties.
The computed tomograph x-ray body scanning technique is used to provide a three-
dimensional view of the body. This requires a support which provides great rigidity,
so that a person lying on it does not deflect it significantly, and at the same time it
must have low and consistent x-ray absorption for radiation passing in any direction
in a plane. A carbon-epoxy foam sandwich structure has been developed for this
which is highly successful, allowing x-ray resolutions that were not previously
attainable. A less exacting requirement is for thereapeutic x-ray table tops. These
also are made with carbon-epoxy foam sandwiches.
In these X-ray applications, no other material has been found which comes anywhere
near the excellence of carbon polymers.
521
Handbook of Polymer Blends and Composites
The most evident advantage of polymer concrete has been its wide range of applicability.
Another advantage is the possibility of producing a material with controllable properties.
In the case of polymer concretes it is possible to develop tailor-made materials for
particular applications. Understanding the nature of polymer concretes is the best way
to rationally prepare for and make the most effective composites. The main problems
of material engineering of polymer concrete materials in comparison with other concrete-
polymer composites have been dealt with.
The same even in the range of the same type of concrete-polymer composite (Figures
12.7 and 12.8).
All types of concrete-polymer composites can be reinforced. Steel bars are used as
reinforcement only with PIC and very rarely with polymer cement concrete. Various
fibres, mainly steel, but also glass, polymeric, and carbon fibres, are employed for such
purposes in the case of polymer cement concrete and PIC. Several special types of
polymer concretes can be produced, such as lightweight concrete and nonshrink or
expanding polymer concrete.
522
Composites: At the Turn of Century
523
Handbook of Polymer Blends and Composites
524
Composites: At the Turn of Century
brought down the cost of FRP materials into an affordable range, so that their application
for civil engineering infrastructure has become both competitive and a reality. It is generally
agreed that the new generation of high-performance composites in construction will play a
key role in these materials have demonstrated outstanding strength-weight and stiffness-
weight ratios, enhanced durability, increased resistance to corrosion, abrasion, chemicals,
and fatigue, reduced construction time, life-cycle to efficiencies, increased ductility, and
ability for self-diagnosis and self-healing (smart materials).
Composites are also used in civil engineering structures, albeit to a very limited extent.
And their use has been limited to specific types of work. Presently they are being used
mainly for rehabilitation and retrofit of existing bridges, and also for rehabilitation of
old buildings. The retrofit work involves mainly two components: to increase the flexural
capacity of beams and girders, and thereby enhance the load-carrying capacity of bridge
superstructures, and seismic retrofit of bridge columns. This work is done typically using
carbon fibre reinforced polymer (CFRP) or glass fibre reinforced polymer (GFRP) strips
of wraps. Such applications can be found worldwide.
The use of composites versus conventional building materials is discussed. For example
a bridge to be repaired, located in Lucerne, Switzerland was completed in 1969. It is
designed as a continuous, multispan box beam with a total length of 228 m. The damaged
span of the bridge has a length of 39 m. The box section is 16 m wide, with a central,
longitudinal web. Core borings were performed a couple years ago to mount new traffic
signals. In the process, a prostrating tendon in the outer web was accidentally damaged,
with several of its wires completely severed by an oxygen lance. As a result, the granting
of authorisations for special, heavy convoys across the bridge was suspended until after
completion of the repair work. Since the damaged span crosses Highway N2, the traffic
lanes in the direction of Lucerne on this highway had to be closed during the repair
work, which could therefore only be carried out at night.
Adhesively bonded steel plates have been used for the post-reinforcement of concrete
structures for over twenty years. Several problems are however associated with the
technique, namely the difficult handling of the heavy steel plates, a risk of corrosion at
the steel/adhesive interface, and the difficulty in forming satisfying butt connections,
which are necessary due to the limited available lengths of steel plates. These problems
lead to the research project Post reinforcement of concrete structures with carbon fibre-
reinforced epoxies at the Swiss Federal Laboratories for Materials Testing and Research,
EMPA, in Dubendorf. The goal of the project was to determine the feasibility of replacing
the steel reinforcing plates with featherweight carbon fibre reinforced epoxy laminates.
525
Handbook of Polymer Blends and Composites
Laminates of this material are characterised by a low density, very high strength and
stiffness, an excellent fatigue behaviour, and an outstanding corrosion resistance. The
reinforcement is usually performed with 1-2 mm thick, unidirectional (all fibres oriented
in a given direction) CFRP laminates which are adhesively bonded to the structure. The
feasibility of effectively reinforcing concrete with carbon fibre composite laminates was
originally demonstrated at EMPA.
CFRP are fifty times more expensive, per kilogram, than the steel used to this date for
the reinforcement of existing structures. Do the unquestionably superior properties of
CFRPs justify their high price? When one considers that, for the repair of the Ibach
bridge, 175 kg of steel could be replaced by a mere 6.2 kg of CFRP, the high prices
suddenly no longer seem so outrageous. Furthermore, all the work could be carried out
from a mobile platform, thus eliminating the need for expensive scaffolding.
The trend in selection of packaging materials in the last several years has been towards
the use of lighter weight materials and the replacement of rigid containers with flexible
or semi-rigid structures. Composite materials of various types have played a primary
role in these changes.
Composite packaging materials fall into two categories. Structures combining paper or
aluminum foil with plastic are designed to obtain from this combination of materials,
properties either impossible or uneconomical to obtain with one material alone. While
not really new, these materials have proliferated in recent years. Brick-pack structures
used in aseptic packaging, for instance, have replaced tin-plated steel cans with laminated
structures containing paper, aluminum foil, and low density polyethylene. Metallised
films in bag-in-box structures have replaced cans for bulk food distribution, and have
replaced glass for wine.
Composite materials containing plastic and paper alone do not pose any special concerns
for incineration beyond those posed by single resin plastics. Aluminum-containing
composites may be a pollution concern.
526
Composites: At the Turn of Century
References
6. I. S. Chun and D. H. Reneker, Bulletin of the American Physical Society, Series II,
1994, 39, 705.
110. T. A. Bogetti and J. W. Gillespie, Journal of Composite Materials, 1992, 26, 5, 626.
12. Advanced Technology for Design and Fabrication of Composite Materials and
Structures: Applications to Automotive, Marine, Aerospace and Construction
Industry, Eds., G. C. Sih, A. Carpinteri and G. Surace, Kluwer Academic
Publishers, Dordrecht, 1999, 1-14.
14. T. C. Stening, C. P. Smith and P. J. Kimber, Modern Plastics, 1981, 86, November.
527
Handbook of Polymer Blends and Composites
21. S. Gandhi, R. Lyon and L. Speitel, Journal of Fire Sciences, 1999, 17, 1, 20.
23. F. Le Lay and J. Gutierrez, Polymer Degradation and Stability, 1999, 64, 3, 397.
25. N. Taly, Presented at the 43rd International SAMPE Symposium, Anaheim, CA,
USA, 1998, 1229.
528
Contributors
M. Arnautu
Gh. Asachi Technical University of Iasi
Department of Macromolecules
Bd. D. Mangeron, no. 71
6600 IASI
Romania.
Vasile Blascu
Faculty of Textiles and Leather Technology
Gh.Asachi Technical University of Iasi
53 Dimitrie Mangeron Boulevard
RO 6600 Iasi
Romania
E. Ya. Davydov
Emanuel Institute of Biochemical Physics
Russian Academy of Science
Kosygin St., 4
Moscow 117977
Russia
Daniel L. Goodman
Science Research Laboratory
Electron Solutions Inc.,
Somerville,
MA 02143
USA
539
Handbook of Polymer Blends and Composites
Aurelia Grigoriu
Faculty of Textiles and Leather Technology
Gh.Asachi Technical University of Iasi
53 Dimitrie Mangeron Boulevard
RO 6600 Iasi
Romania
Yukun Hua
College of Wood Science & Technology
Nanjing Forestry University
Nanjing 210037
PR China
Byung-Sun Kim
Composite Materials Laboratory
Korea Institute of Machinery and Materials
66 Sangnam-dong
Changwon
Korea 641-010
Shiushichi Kimura
Materials and Structures Laboratory
Tokyo Institute of Technology
4259 Nagatsuta, Midori
Yokohama 226-8503
Japan
Anand K. Kulshreshtha
Indian Petrochemical Corporation Ltd
Vadodara 391346
India
Giuseppe R. Palmese
University of Delaware Center for Composite Materials
Newark DE 19716
USA
Manishkumar P. Patel
Department of Chemistry
Sardar Patel University
Vallabh Vidyanagar - 388120
Gujarat
India
540
Contributors
Ranjanben G. Patel
Department of Chemistry
Sardar Patel University
Vallabh Vidyanagar - 388120
Gujarat
India
Rasmika H.Patel
G.H. Patel Institute of Materials Science
Sardar Patel University
Vallabh Vidyanagar - 388120
Gujarat
India
V.S. Patel
Sardar Patel University
Vallabh Vidyanagar -388120
Gujarat
India
R.A. Pethrick
Department of Pure and Applied Chemistry
University of Strathclyde
Thomas Graham Building
295 Cathedral Street
Glasgow G1 1XL
Marcel Popa
Gh. Asachi Technical University of Iasi
Department of Macromolecules
Bd. D. Mangeron, no. 71
6600 IASI
Romania.
Cornelia Vasile
Romanian Academy
P.Poni Institute of Macromolecular Chemistry
41A Gr. Ghica Voda Alley
RO 6600 Iasi
Romania
541
Handbook of Polymer Blends and Composites
Eiichi Yasuda
Materials and Structures Laboratory
Tokyo Institute of Technology
4259 Nagatsuta
Midori
Yokohama 226-8503
Japan
542
1
An Overview of Composite Fabrication,
Design and Cost
Anand K. Kulshreshtha
1.1 Introduction
Various conventional methods of composite fabrication are as old as 40 years and have
their origins in the World War II. During this period, technology in all spheres of life has
jumped by leaps and bounds which has had its impact on the way these methods are
used today. It is now easy and clear to demarcate these methods between conventional
and new. What is new today, e.g., robotic moulding, has its roots in the old (hand-lay
up, lamination, autoclave, filament winding, pultrusion, etc.) and the boundary between
them is fuzzy and not sharp.
The following advances in composite technology have taken place during the
intervening era:
The five resins currently used for the production of composites, other than for research
and development purposes are epoxies, polyesters, phenolics, silicones and polyimides.
5
Handbook of Polymer Blends and Composites
Table 1.1 gives information on filler-resin combinations and various composite fabrication
processes, while Table 1.2 give properties, applications and advantages of several
thermoset resins.
6
An Overview of Composite Fabrication, Design and Cost
1.3 Prepregs
7
Handbook of Polymer Blends and Composites
mixers. Major melting target is by heat conduction rather than by shear. The processing
ease and ultimate properties are determined by how close the fibres and resin are in
contact during moulding.
Prepreg tapes comprise unidirectional fibre rovings spread into a web of even thickness
and impregnated with a resin system formulated to give the material handleability and
processability (Figure 1.1). Multi-layers of prepreg can be laminated into composite
materials with a high level of fibre reinforcement (CF 65% by volume) and multidirectional
properties are achieved when the individual layers are aligned at different angles to the
longitudinal axis. For example, a composite with a lay up construction of 0/+60/-60
has pseudo-isotropic performance with equivalent mechanical properties in several
directions. The use of prepreg therefore enables component manufacturers to tailor
composite materials specifically to meet operating stresses, and reinforcement can be
introduced in precisely those directions in which it is required (Figure 1.2). Material
usage is thereby minimised, contributing further to the weight advantages over
conventional components. Effect of process parameters and dependence of properties on
the flexibility of prepregs are shown in Figure 1.1 and given in Table 1.3.
8
An Overview of Composite Fabrication, Design and Cost
Figure 1.2 gives an overview of design possibilities for making a composite product via
the prepreg route
9
Handbook of Polymer Blends and Composites
The bond strength of a laminate is determined by the extent of contact between skin
panels and core material. A method for splicing and shaping sandwich panels is shown
in Figure 1.3. The types of adhesive found in most regular use in industry range from
ductile polyurethanes to the stiffer and more brittle epoxies. The range of stress/strain
properties for these materials are typically, 10 MPa shear strength and 150% shear strain
for ductile acrylic, to 60 MPa shear strength and 2.5% shear strain for a brittle epoxy.
The shear modulus of adhesives is small compared to most engineering materials, but
strains to failure are much larger. The cohesive strength of adhesives is low and therefore
the ability to support tensile loads is also low. Adhesive joints must therefore be designed
to avoid or minimise tensile loads. As a consequence, joints that have a tendency to
cleavage or peel are to be avoided. In contrast, adhesives have high shear strength. It
would be unrealistic, however, to design a joint that is loaded in pure shear only, but this
should be the ultimate goal. Therefore to ensure maximum performance, careful attention
must be paid to the joint geometry (Figure 1.4) and the choice of adherent/adhesive
material combination [1]. The efficiency of composite products formed is a function of
laminate design (Table 1.4).
10
An Overview of Composite Fabrication, Design and Cost
Filament wound
Continuous 0.77 214 124 58.0
(unudirectional)
Cross laminated
Continuous 0.48 136 50.0 36.8
fibres
Cloth laminated
Continuous 0.48 136 29.6 21.8
fibres
Mat laminated
Continuous 0.48 136 39.4 29.0
fibres
Chopped fibre
Non-
systems 0.13 41.8 10.3 24.7
continuous
(random)
11
Handbook of Polymer Blends and Composites
Despite evidence showing that thick monolithic laminates are extremely damage tolerant,
there will always be the potential requirement for embodying a major repair. Unusual
examples of historical events resulting in major damage include:
12
An Overview of Composite Fabrication, Design and Cost
It must be remembered that unlike the majority of the current composite parts, removal
or replacement of a wingbox is not an option. Hence it must be recognised that a
structure of this type will require major repair at some time and the challenge exists to
find techniques which restore adequate and durable strength to the structure within a
timescale that causes minimum downtime for the operator.
When reinforcing fibres are dispersed in a rigid polymeric matrix, any stress applied
to the polymeric matrix is transmitted to the fibres as well. This distributes the load
along its length and adds greatly to the modulus, strength, impact resistance, and
dimensional stability of the total plastic product.
13
Handbook of Polymer Blends and Composites
Simple shapes, such as flat and corrugated sheet and boat hulls, are made by impregnating
liquid polyester resin into woven glass fabric, laying the assembly in a low-cost mould and
applying mild pressure during room-temperature relation and preferably final oven cure.
More complex shapes are made by using a nonwoven mat of random glass fibres.
High-speed production, for mass markets like transportation, can be achieved by using
more costly matched moulds resembling light-weight compression-moulding equipment.
Major applications are in boats, transportation equipment, construction, and tanks.
The production of such reinforced thermoset plastics is obviously more laborious and
costly than most plastics processing, and this has certainly limited its growth. The use of
short glass fibres (0.1 cm or less) in thermoplastic injection-moulding resins like
polystyrene and nylon permits normal, high-speed, economical compressing, and produces
high modulus, strength, and dimensional stability parts. This creates a new class of plastic
materials that are competitive with die-cast metals in many applications.
Aside from glass fibres, a great deal of kraft paper is impregnated with partially
polymerised phenolic resin, stacked, and compression-moulded into laminated boards
that find wide use in electrical construction and, especially with a decorative melamine
topcoat, in countertops, cabinets, table tops, and furniture.
Many of the techniques for processing polymer composites are the same as for the unfilled
polymers, but require some modification, possibly because of the higher viscosities which
need higher moulding pressures to give the same output. Some processing techniques,
however, are specific to the fabrication of composites or to one particular class of
composite, because of some special requirement. Some indication of both classes is given
in Table 1.6.
Composite fabrication techniques are classified according to the complexity of part shape.
For simple shapes rapid manufacturing techniques, e.g., stamping, injection moulding,
are used while for fabricating large, complex parts even hand lay-ups are used.
Compression moulding, which is as old as the history of composite materials being used for
the fabrication of phenolic resins, is therefore the oldest forming technique. In this, the
composite in powder form is poured into the cavity of a mould. It softens on heating, and
pressure is then applied to fill the mould. Viscosity first decreases and then increases as
curing proceeds. In a typical moulding, a pressure in the range 20-60 MN/m2 is applied for a
period of 1-2 minutes. The mould is then cooled to a temperature so that the rigid shape of
14
Table 1.6 Fabrication processes for fibre composites
Process Reinforcement Vol.% Method Common Resin Typical Application
1. Hand lay-up CSM or woven mat 13-30 Resin impregnation of dry fibre with Polyester GRP boat hull
brush and roller
2. Spray-up Chopped rovings 15-30 Direct application of chopped Polyester
rovings and resin
3. Compression Chopped or 10-60 SMC, BMC, wet resin plus CSM Polyester, epoxy and Medium-sized complex
Moulding continuous fibres and phenolic powder. Pressed to shape in a phenolic preform, mouldings (e.g. lorry
fillers closed mould between matched dies preperg, moulding body panels)
4. Injection Short fibres 5-20 Conventional injection moulding of Polyester resin. Several Numerous small-to-
Moulding precompounded granules, BMC, thermo-plastics and related medium domestic
the moulded object is retained on ejection from the mould. The technique has the disadvantage
that some material is lost as flash, since the mould must be overfilled to allow pressure to
build-up; it is also slow compared with, say, injection moulding. On the other hand it is
clearly useful for short-runs which would not really justify the use of injection moulding.
SMC plays an important role in the reinforced plastics industry. For automotive
applications, it is used to produce exterior skins with a surface comparable to that of
steel but with significant weight reduction. It also provides greater versatility than steel
in the intricacy of parts produced. SMC remains more expensive than steel but is also
noncorrosive. The trend by automobile producers toward fewer units produced for each
model and the subsequent increase in tooling costs is working in favour of SMC. Significant
growth in SMC use is predicted over the next several years.
16
An Overview of Composite Fabrication, Design and Cost
SMC, as the name suggests, is material produced in sheet form that is then moulded into
composite articles (Figure 1.6). These articles include automotive exteriors and structural
members, bathtubs and shower stalls, electrical applications such as computer housings,
housings for tractors and small machinery, doors trays and others. The material
characteristically consists of a thermoset polyester-styrene matrix with inorganic fillers
and chopped-strand glass fibre reinforcement. The mixture undergoes thickening, and
the sheet is in solid form at the time of moulding.
SMC evolved from a need for an easy-handling moulding compound that can produce
parts with physical properties comparable to those of preform and similar mat-based
composites. The major expense is incurred by tooling and presses, with compounding
machines relatively inexpensive.
1.5.3 Tooling
Thermoset injection moulding has brought new levels of productivity and quality to
industry over the past ten years. By replacing compression moulding techniques, injection
17
Handbook of Polymer Blends and Composites
processing has reduced the time and labour traditionally associated with thermosets.
Still, when compared with thermoplastic processes, these heat cure operations can
demonstrate certain wasteful and inefficient characteristics.
Recent tooling innovations have addressed several of potential drawbacks involved with
thermosets. Moulders were formerly guaranteed a scrap percentage on every shot. Cold
runner manifolds now eliminate cured runner material. Cold manifolds also bring sprue
volumes down to an economical minimum. The growing popularity of gate cutting systems
further demonstrates the efficiency possible with properly designed thermoset tools.
Secondary operations are reduced significantly or eliminated by having runner separation
occur in the tool. Gate, rubber or sprue marks are also less prevalent on parts moulded
with gate cutters.
Dielectric monitoring can be performed on a shot to shot basis. With the proper equipment
package, todays moulding machines can be directed to cycle based on the cure rate of a
material. This monitoring system optimises cycles by reducing cure time to its minimum
potential. Scrap incurred by undercuring parts is eliminated while needless overcure does
not consume press time.
These innovations teamed with proven mould design techniques give thermoset injection
moulders the process advantages associated with thermoplastic moulding. No longer is
material the only cost savings possible when opting for thermosets.
Mould Release
Easy release of the part from the mould after completion of the curing cycle is normally
accomplished by lubricating the mould with waxes or silicones. At other times, such
substances as zinc stearate or paraffin wax are added to the resin mix in order to facilitate
the removal of parts. The latter are called internal lubricants.
The development of RRIM systems with better properties has followed the use of reinforcing
fillers in urethane, especially glass in bead and fibre form. Both glass beads, hammer milled
glass are commonly used (0.8-2 mm random length) and chopped strand glass (1.5-3 mm
precision lengths) which are processed as slurries in the liquid urethane components. As
good wetting and preferably bonding between urethane and glass matrices is considered
desirable, the glass is often surface treated with chemical coupling agents. However, with
longer fibres and higher loadings, e.g., the 3 mm fibres and greater than 6 ph of the 1.5 mm
variety, the fibres are distributed in both polyol and diisocyanate components due to viscosity
limitations. Settlement and high viscosity can be overcome by raising the polyol and
18
An Overview of Composite Fabrication, Design and Cost
Over a period of a vehicles life, considerable savings in fuel are possible by the use of PU
lightweight plastic parts replacing other structural materials. Table 1.7 compares the
properties of typical unreinforced with reinforced RIM urethanes. It can be observed
that in the same hardness range the flexural modulus of the RRIM system is nearly twice
that of the RIM system.
The principal advantage of RIM is the ability to process low-viscosity liquid reactants,
typically 0.1-1.0 Pa-s, using low temperature and pressures, particularly during the
mould-filling stage. This means that small-scale hydraulic equipment and lighter weight
and lower cost moulds can be used, facilitating short production runs and prototype
applications. Low-viscosity reactant systems also facilitate composite materials
production, so called reinforced RIM (RRIM) composites (Figure 1.7), in which
continuous fibre reinforcement mats are preplaced in mould cavities prior to injection.
Capital investment and operational costs in RIM are therefore much less than those
for conventional injection moulding.
19
Handbook of Polymer Blends and Composites
RRIM is similar to RIM in its intensive resin mixing procedures and its reliance on fast
resin reaction rates. It is also similar to resin transfer moulding (RTM) in using preforms
that are placed in the cavity of a compression mould to obtain optimum composite
mechanical properties.
20
An Overview of Composite Fabrication, Design and Cost
RTM is a low pressure moulding process, where a mixed resin and catalyst are injected
into a closed mould containing a fibre pack or preform. When the resin has cured the
mould can be opened and the finished component removed (Figure 1.8).
The process is suitable for simple GRP mouldings through to complex structural
components for the automotive and aerospace industries. Typical product size can range
from a few cm2 up to 20 m2.
A wide range of resin systems can be used including polyester, vinylester, epoxy, phenolic
and methyl methacrylates, etc., combined with pigments and fillers including aluminium
trihydrates and calcium carbonates if required.
The fibre pack can be either glass, carbon, aramid or a combination of these. There are
a large variety of weights and styles commonly available. The production volume of
RTM relative to conventional methods is shown in Figure 1.9.
21
Handbook of Polymer Blends and Composites
22
An Overview of Composite Fabrication, Design and Cost
23
Handbook of Polymer Blends and Composites
Filament content in this process can be up to 80% by weight so that high stiffness and strength,
in the reinforcement direction, are possible. The object may be finished with a gel coat.
1.5.8 Pultrusion
Another generic automated process developed for the composite industry is pultrusion,
which also made its appearance in the 1950s and has evolved into a number of variations,
most of which today utilise programmable controls.
24
An Overview of Composite Fabrication, Design and Cost
The process consists, essentially, of those elements depicted in Figure 1.11. Resin-wet
reinforcements are drawn into the system through squeezeout bushings (a heated mould
(die) through which resin-dripped continuous filament tow is pulled at a constant rate),
which remove excess resin, and are optionally preheated by dielectric radio frequencies
(5-100 MHz) or microwave frequencies (915-2450 MHz) or, in cases where conductive
reinforcements such as graphite are included, by induction preheating. Thermoset
pultrusion is done with a single long die with polymerisation taking place entirely within
the residence time in the die. The roving package then enters a heated forming/curing die
and the cured stock is pulled out of the die by suitable pulling devices.
In heavy sections, throughput rates were dramatically increased when high frequency
preheat was introduced to pultrusion equipment. Experimentation ultimately led to the
augmented cure invention; this optimum synergistic combination of cure methods first
achieves instantaneous preheat throughout the fibre/resin mass, and then pulls the mass
immediately into a heated steel die. Since the material preheat phase, heating the steel die
is really more a means of preventing cure heat loss than a means of introducing additional
heat. The first production application for the pultrusion process and equipment was
power pole lanyards for the electrical industry. Pultruded products soon covered a wide
range in the tremendously broad market for electrical bar stock that is characterised by
unidirectional reinforcement and constant cross section.
25
Handbook of Polymer Blends and Composites
The inflatable mandrel has several benefits, which have led to increasing use of this
method. When necessary, the mandrel can often be deflated and removed. In most
applications, however, the inflatable mandrel can be left in as a liner for the
composite material. Applications such as fuel tanks require this type of liner. Another
benefit of the inflatable mandrel is the ability to add pressure as the winding
progresses. This prevents the collapse of the inner fibres, which sometimes occurs
as the outer wraps are wound onto the vessel. Increasing the internal pressure causes
the inner wraps to stay in tension and not tend to buckle. The inflatable mandrel
also allows the vessel to be cured with an internal pressure that is equal to the
expected operating pressure of the part, thus eliminating possible stress cracking
during initial pressurisation.
26
An Overview of Composite Fabrication, Design and Cost
The technique involves a combination of prepreg roving and cloth, together with dry
roving or tow, which is fed through two short heated dies to compact the laminate.
The tackiness of the prepreg bonds the dry roving to it and produces a composite
which can be handled. The resin, however, is not actually cured at this stage.
The resulting pultruded pre-form can be put into cold storage for final moulding at a
later date. When it is needed it can then be brought to room temperature and placed in
position in the mould. If it is to be moulded to another part, this latter is also placed in
the mould. A normal autoclave moulding procedure follows in which the re-form and
part are cured.
Thermoplastic pultrusion uses high technology matrices which have high softening
temperatures and high physical properties with relatively low melt viscosities. Compared
with thermosetting polymers, thermoplastics matrices will greatly improve the toughness
of the composite. Much work has been undertaken using standard prepregs such as
the solvent applied polysulphone, and poly(ether sulphone) polymers in combination
with glass, Kevlar or carbon roving tow, tape or cloth reinforcements. These component
parts are fed through short, heat and chill, dies to be pultruded into profiles.
Hazards
Neural networks can give a prediction of autoclave cure in less than a second [2], while
such a prediction would take hours from numerical models.
27
Handbook of Polymer Blends and Composites
They can provide rapid on-line prediction of cure parameters, leading to the possibility
of continual, well-defined, real-time control. A neural network based controller can
compensate in real-time for unexpected occurrences and prepreg property variation.
Composites of the same or higher quality than from conventional cycles can be
consistently produced with the neural network controller, even when unexpected
occurrences and prepreg property variation are not included.
A perfectly prepared meal can easily be spoiled by cooking it badly. For best results,
the cook needs to ensure that the oven is set to provide the right thermal conditions.
The same applies to an aeroplane, or aerospace components, cured in an autoclave
except that the control problem is more complex.
The autoclave cure process is used in the production of advanced polymer composites,
primarily for the automotive, marine, and aerospace industries. Historically, autoclave
cure has been used to produce composite parts of high quality. One major hindrance
to the production of composites through autoclave cure is that the process is largely
trial-and-error, leading to much waste and parts with less-than-optimum properties.
An autoclave is, in essence, an insulated pressure vessel that allows heat, pressure,
and vacuum to be allied to a laminate. The laminate is laid by hand onto a tool plate
prior to cure and surrounded by a lay-up assembly composed of numerous parts
including bleeder material which absorbs excess resin, and vacuum bag which renders
the system air-tight. Each layer of the laminate is a section of prepreg tape, composed
of a thermosetting resin matrix (commonly epoxy) and fibrous inclusions (including
glass, carbon, and aramid fibres).
Prepreg manufacturers provide estimates of the proper cure cycle, estimates developed
through experience and trial-and-error. Such manufacturer-defined cure cycles do
not take part thickness and shape, unexpected occurrences during the process, or
variations in prepreg properties (such as resin content, molecular properties, and
slight inclusions) into account. While manufacturer-defined cure cycles are widely
used during autoclave cure, an improved method of determining cure cycles is
necessary to ensure high product quality.
Control algorithms are used to accomplish the following goals: to provide an estimate
of the proper heater temperature adjustment at discrete time steps, and to control
autoclave pressure based on the monitoring of resin viscosity.
28
An Overview of Composite Fabrication, Design and Cost
The raw materials used are graphite, aramid fibre or glass fibre in the form of a
unidirectional or woven tape. The tape is impregnated with an epoxy based resin.
Goldsworthy Engineering, USA, has developed a two phase system. In the first phase,
the tape is prepared and in the second phase, the tape is placed. The material from the
supply spool is measured, cut, angled on both the ends and transferred to a new backing
paper. The cut tape on its new backing paper is rewound on a new supply spool and the
composite cassette is placed on the tape placement machine.
1.5.16 Pulforming
This process was developed for the NASA shuttle stiffener ring. Reinforcing fibres are
drawn through a tank for resin impregnation and then through a die. Pulforming may be
curved or straight. A composite hammer handle is made on a straight pulforming
equipment. A BMC, in the form of an extended rope on a supply, is fed through a
measuring device that measures the length to provide the desired volume. The length is
cut and compressed into a fold around a single travelling roving strand at uniform intervals.
A press travelling at line speed forms the BMC pieces into the shape of changing volume.
All primary reinforcing fibres encapsulate the entire fibre with the formed BMC. These
are curved by heated split female dies.
Similarly automatic machines have been fabricated to make automotive leaf springs.
Different special injection moulding equipment and/or procedures are used to provide
different methods of controlling reinforcement lengths and orientations with no or
minimum fibre damage.
More popular now [3] is injection-compression moulding (ICM) that is also called coining,
injection stamping, etc. With this technique, reinforcement in different forms can be
strategically located in a matched mould cavity to meet reinforced plastic orientation
with proper impregnation and other capabilities (Figure 1.12). The mould partly resembles
a compression mould where a male plug fits into the female cavity. The reinforcements
can be located in the cavity by the different insert methods.
29
Handbook of Polymer Blends and Composites
Other methods include use of continuous reinforced web or tape that is positioned between
mould after each moulding cycle. ICM is a simplified variant of injection and compression
moulding. An essential difference lies in the manner in which contraction occurs during
cooling in mould; shrinkage, melt flow stresses, etc., can be better controlled to the
extent of being eliminated.
Another special injection moulding system example is the patented British Technology
Group SCORIM process. It can control orientation of fibres. A special head at the end of
the injection moulding barrel splits the melt flow into two streams. During the holding
stage in the cavity, two hydraulic, cylinders alternately actuate pistons, compressing the
melt in the cavity in one direction and then the other.
30
An Overview of Composite Fabrication, Design and Cost
The quality of FRP products depends on the compound mixture quality and dimensional
reproducability of moulded parts. An example of this is the cavity transfer mixer for the
efficient distribution of compound additives in extrusion an the injection moulding of
compact discs. Similarly a refinement of co-rotating twin-screw extruders is taking place
for dispersive mixing of polymer compounds and specially for heavily filled polymers.
The combination of the compounding extruder and auxiliary devices such as gear pumps
and mixing devices are now providing for compounding/profile extrusion and also for
direct blending injection moulding. The advantage of these combinations include:
Single heat history - important with heat sensitive materials and reinforcement.
Reduced fibre breakage and molecular degradation
Elimination of extra handling of raw and intermediate materials
Reduced labour cost.
This technique is applicable to polymer matrix composites, thereby providing for longer
continuous operation of parts at higher temperatures.
The versatility of this technique has been enhanced by use of multi-live feed moulding by
the single feed point. In multi-live feed moulding the single feed from the injection moulder
31
Handbook of Polymer Blends and Composites
is split into a number of feeds [4] in which each one is capable of supplying pressure to
the cavity independently of the others.
The moulder consists of three parts: the injection moulder barrel, the two live-feed head,
the mould. The two packaging pistons are driven by an auxiliary hydraulic power pack
containing two separate hydraulic pump and valve systems. The hydraulic feeds to the
two pistons are controlled by a microprocessor.
There are three options for packing with the device: the pistons can be pumped back and
forth at the same frequency but with a phase difference of 180 C, or in phase or both
held down under static pressure. The effect of the use of the first option is to shear the
melt within the mould cavity while it is solidifying. The effect of the second option is to
compress and decompress the melt in the cavity at the set frequency. The third option
applies a static packing pressure to the cavity.
Isolation of the multi live-feed device and mould cavity from the injection moulding
machine screw and barrel allows much higher pressures to be applied to the solidifying
melt than those used conventionally.
Artificial neural networks (ANN) are an information processing technique that emulates
the neural reasoning behaviour of biological neural systems.
Neural network approach was applied in building a process model for quality control in
injection moulding. The inputs to the networks are the process parameters and the outputs
are the quality characteristics. After the supervised learning, these process models can
32
An Overview of Composite Fabrication, Design and Cost
project the moulding parameters onto the quality characteristics for on-line quality forecast
or quality control. Figure 1.13 shows the structure of a neural network for the quality
prediction of moulded parts.
Figure 1.13 Structure of a neural network for quality prediction of moulded parts
1.6.6 Co-Injection
In the co-injection process, two different process, two different materials fed from
separate barrels are joined in a manifold so that one material can be encapsulated
within the other.
Materials economics are a major benefit of co-injection. It allows use of regrind, either
your own, or post-consumer plastic, as an inner, or core material as a way to cut
materials costs.
33
Handbook of Polymer Blends and Composites
There are other benefits to co-injection as well. One example is a refrigerator door handle
being moulded by co-injection. The inner material is foamed polypropylene and the surface
is a glass-reinforced polypropylene. As the blowing agent in the inner layer expands in the
mould cavity it presses the outer layer into a contour of the mould giving the part its shape.
In effect, that approach gives you the benefit of gas-assist, without using gas-assist moulding.
Because less material is crystallising in the core (foam), savings in cycle times can be achieved.
Co-injection machine uses a smart pin, which moves to different positions, controlling
the flow of core and surface material in a sequential manner. The barrels come together
in a Y configuration, minimizing potential for materials hangup.
The new technology uses a hybrid multi-component injection moulding machine developed
by Demag Ergotech GmbH. The machine consists of a standard thermoplastic moulding
machine retrofitted with a rotary platen and a liquid silicone injection unit. The process
consists of four operations:
The Ergotech machine uses hot-runner delivery of the primary material, with the silicone
injection unit mounted at a 90 angle to the primary unit. An isolation plate prevents
transfer of too much mould heat into the silicone cold-runner system. The machines
NC4 control provides integrated control of the silicone and hot-runner heaters and other
process parameters.
The process could benefit customers by eliminating assembly costs currently associated
with silicone overmoulding of thermoplastics. Additional benefits are expected to include
elimination of process variability associated with assembly operations and reduced
34
An Overview of Composite Fabrication, Design and Cost
material costs from increased design flexibility. Technology had been commercialised by
the end of 1998 with customers in the automotive, telecommunications, or consumer
products industries.
Advance USA licenses the Strata technology to thermoforming processors and uses it
for its own manufacturing operations. In late 1998 the company began production
of glass-reinforced thermoformed parts at its facility in New Stanton, USA. An $8
million capital investment included a four-station rotary machine with a 10 m2 forming
area and a urethane foam injection system.
The new Stanton installation is producing 100,000 parts/year with the Strata
processes, 20,000 of which are reinforced thermoformed camper tops. The campervan
roofs are made from a coextruded sheet. Thermoformed skins, along with a glass
preform, are inserted into a matched mould, then injected with urethane foam to
create the sandwich structure.
The conventional injection moulding process is the basis for low pressure processes. The
technique of in-mould surface decoration involves injection moulding and hot stamping
in one single operation. Here, the objective is to produce flat components of a multilayer
structure cost-effectively. It can be carried out in a conventional injection press with
some additional controls. Before the cycle starts, the cut textile blanks or any other in
the mould decorating sheet is positioned in the moving half. The cavity is partially filled.
The formation of the moulding is then accomplished by the closure of the motion of the
moving half of the mould. In this process a thin thermoset coating is injected onto an
injection or compression mould.
35
Handbook of Polymer Blends and Composites
One of the difficult problems to overcome in ISD is to avoid weld lines in multigated
parts where melt fronts from different gates meet. The surface of the moulding shows
a crease at the welded marks. In order to reduce this effect or even to get rid of it
altogether, a cascade control system for hot runner nozzles has been developed. The
objective is to open and close the gates sequentially. For instance, initially, the gate at
one extremity will be opened until the melt reaches the point immediately below the
adjacent (second) gate. The second gate is now opened till the melt reaches the third
gate and so on. By this, the creation of otherwise unavoidable weld lines between two
can be prevented. A typical application of this system of control is in the production of
trim for car roof pillers.
The use of fusible core technology to mould parts of complex geometries is being
revived in the automotive industry for moulding air intake manifolds. The traditional
manufacturing route for these items had been the aluminium die cast process. The
fusible core technique involves lifting a temporary core (cast out of tin bismuth alloy
of a low melting point) in an injection mould, using a robot. During melt injection,
the surface of the core forms the internal contour of the moulded part. The component
is ejected along with the core, which is subsequently melted out, recast and reused.
Apart from weight reductions, Nylon air intake manifolds moulded by this technique
offer additional advantages like reduced air resistance and elimination of secondary
machining.
36
1.7.5 Alpha 1 Moulding Machine
This machine is equipped with two injection units, a long stroke vertical press and shuttle
table, allows combinations of compression moulding, gas-assisted injection moulding
and stamping. GE Plastics is using Alpha 1.
In low pressure moulding the mould cavity is filled at low speed through large gates with
a controlled pressure profile in the shape of a broad inverted U. The melt temperature
profile is controlled by adjusting screw speed and flow resistance during plastication.
LPM works better with low viscosity semicrystalline materials and is not suitable for
thin wall parts. Slow injection, lower melt and higher mould temperatures reduce residual
stress to allow demoulding at a higher temperature to maintain or reduce cycle times.
For larger parts, low clamp force can be achieved using multiple gates with programmed
opening. Lower clamp force allows use of self clamping moulds and multistation injectors.
Laminate moulding involves moulding plastic at low pressure directly behind textile,
film, or metal.
In liquid gas injection moulding, a volatile liquid is injected at low pressure into the melt
and the vapourises to form hollow channels in the part. The liquid condenses and is
absorbed in the part.
Commercial success requires fast reaction and efficient preform formation. Cycle time
for thermally active systems can be decreased by using higher mould temperatures and
heating the preform.
Thermoformable mat heated by infrared heat to melt the binder and pressed into shape by
one or two moving platens while supported by a holdslip edge clamp to reduce wrinkling.
37
Handbook of Polymer Blends and Composites
Low pressure and temperature processing by RLCM allow the use of inexpensive light-
weight tools, especially for prototyping. RLCM allows customising reinforcement to
give desired local properties and part consolidation in complex 3D geometries.
Part geometry can be considered in terms of function where the part is located on a scale
ranging from enclosure (containers or panels, often with cosmetic surfaces) to load bearing,
and complexity which allows for geometric complexity ranging from symmetric (planar
or axisymmetric) to fully three-dimensional designs.
Robots are ideally suited to applications involving repetitive, cyclic operations. The robot
will ensure consistent cycle times and will always accurately follow the programmed
paths. Inspection of parts can be carried out by vision, weight and check stations, which
signal to the robot production of good or bad parts. The robot will continue its normal
path for good parts and can be programmed to follow a different path for rejects. Robots
are equipped with counters to count numbers of rejects.
Cartesian robots - also known as beam or gantry, always with linear Axes, horizontal,
vertical and demould. Usually specified with 1 or more rotational axis. Mounting of
the robot is normally platen or floor fixing. Access to the mould is normally top or
side entry.
Mouldings are generally deposited at the rear/front or end of the moulding machine. NB
Cartesian Robots can be equipped with a second vertical arm for sprue removal from 3
plate tools.
Scara robots - Robots whose arms have two parallel rotary joints to provide
Compliance in a plane (can have from three to five axis). SCARA=Systems
Incorporating Selective Compliance Articulated Robotic Arm.
38
An Overview of Composite Fabrication, Design and Cost
A novel type of advanced manufacturing process control system has been developed
which not only controls the machinery but addition uses advanced sensors,
mathematical models, and heuristic reasoning to determine automatically what the
best processing parameters should be.
The prototype system utilises sensors placed in situ in the component and an advanced
kinetic model to monitor physical and chemical changes in the raw materials. In
real-time this system controls and adjusts all relevant parameters.
The overall material cure control system has three main components: a laser fibre-
optic spectroscopic sensor and material analysis module; materials and processing
software module (IBM-PC) and processing equipment self-directed control module
(workstation). The system performs a remote analysis of the chemical and physical
constitution of the material being processed in an autoclave. It interprets the data
using a combination of heuristic reasoning and mathematical models and uses this
information to provide self-directed control of the composite cure operation.
Based on this sensor input and predictions of the model, the materials and process
module makes a decision on what changes should be made in the prevailing conditions
and sends control signals to the process equipment in the plant to adjust control
parameters on equipment controllers.
Increased quality at reduced fabrication cost (20-40% reduction in cure cyclic time).
39
Handbook of Polymer Blends and Composites
Table 1.10 lists the faults produced by different fabrication methods in fibre-reinforced
and unreinforced thermosets.
40
An Overview of Composite Fabrication, Design and Cost
Taking carbon fibre as an example, the price band becomes considerably lower as
the tow size is increased from 3000 filaments to, say, 20,000 filaments. Multiplicity
of reinforcement forms and content affects costs, using glass fibre as an example
(Figure 1.14).
Figure 1.14 Relative costs of glass reinforcements for different forms of fibre
Cost of resins increases from general resins (epoxies) to speciality resins (polyimides,
PEEK, LCP, etc.) due to reduction in volumes of resin manufactured (Figure 1.15).
41
Handbook of Polymer Blends and Composites
Figure 1.15 Prices versus heat deflection temperature for commercial resin matrices
Taking pultrusion as a product example, the cost clearly goes up in proportion to size
and complexity of the part dominated by the design (Figure 1.16). Design and fabrication
benefits of composites over metals are well known.
Tooling cost can be very high in comparison to the cost of material inputs. This is illustrated
in Figure 1.17 which gives US Army data for the fabrication of rocket motor cases made
from various materials, including composites. This also applies to epoxy glass and hybrid
composites used in the production of nozzles, exhaust cones, blast tubes, radomes, etc.
Use of epoxy tooling in slow RTM and that of steel tooling for fast RTM results in the
higher part cost, see Figure 1.18 where component costs are shown separately as material/
tooling/labour/capital/energy costs.
However, tooling costs are drastically reduced (Figure 1.19) as a result of switch over
from aluminium to carbon-epoxy resin in a fairing made for wing of an aircraft (resulting
in 31% cost saving!).
42
An Overview of Composite Fabrication, Design and Cost
Figure 1.16 Design influence on the price per pound of pultruded shape
43
Handbook of Polymer Blends and Composites
44
An Overview of Composite Fabrication, Design and Cost
Figure 1.19 Relative costs of a fairing for a wing pivot fitting using aluminium and a
carbon-epoxy composite
The overall cost of fabrication of any combustion product, e.g., an aircraft, is a complex
function of a large number of variables (Figure 1.20) such as design and manufacturing,
etc., and drops considerably as it reaches from the concept-to-planning-to-production stage.
Figure 1.21 and Table 1.11 show that fabrication cost is dependent on the cost of inputs
(raw materials energy, tooling, labour, machinery, waste and the production volume). In
addition, this figure gives a very useful comparison of all these cost inputs for composites
and various polymers and elastomers.
45
Handbook of Polymer Blends and Composites
46
An Overview of Composite Fabrication, Design and Cost
47
Handbook of Polymer Blends and Composites
48
An Overview of Composite Fabrication, Design and Cost
Taking a pultruded product as an example, Figure 1.24 shows that cost per foot depends
on the purchase volume and scaling up of fabrication.
The cost of fibre raw materials determines the fabrication cost of composites for various
reinforcements, especially different types of glass, aramid and carbon fibres.
Cost savings from composites technology (Figure 1.25) shows the specific cost per unit
strength for metals, alloys, and various composites. The carbon fibre composites are
the most expensive, due to the very high cost of carbon fibre. But then carbon fibre
composites are used in fighter-bomber aircrafts, e.g., stealth, missiles, rockets, satellites,
space shuttle and space stations. The cost/benefit level is 5,000 dollars per kg due to
weight saving in spacecraft and 500 dollars per kg in aircraft (Figure 1.26), where 15-
30% savings in load result. Apart from weight savings no other material comes close
to carbon fibre composites in very high strength and stiffness, outstanding fatigue
resistance and extremely low thermal expansion.
49
Handbook of Polymer Blends and Composites
Composites will replace metals and alloys in applications where no competition will
exist. Hand lay-up, and robotic moulding, will exist side by side of artificial intelligence
in machines and continued innovations and variants of existing fabrication methods and
elimination of production rejects and scrap will provide relief to polymer processors
especially those involved in making composites.
Even sky is not the limit because composite materials have been fabricated and deployed
in space.
References
50
An Overview of Composite Fabrication, Design and Cost
51
2
Liquid Moulding Processes
2.1.1 Introduction
US Navy contractors introduced it in the mid 1940s with the name of the Marco
method to manufacture long personnel boats [1]. Typically, in this method, a polyester
resin is drawn by vacuum application to infiltrate fibre glass reinforcement, placed
between two mould halves. This technology was also used to produce aircraft and
missile nose cones using the pressure to drive the resin rather than the vacuum.
Processes recognisable as RTM (see Figure 2.1) were developed and patented with a
little commercial success during the 1950s and the early 1960s finding some interest
during the 1970s for marine, commodity goods and recreational parts. It excited
attention and significant development in the 1980s for the structural parts of aircraft,
military, automotive applications and high performance sporting goods. In fact, RTM
enables processors to satisfy the requirements of the aeronautical and the automotive
industries, assuring cost savings and performance improvements over the traditional
hand lay-up methods.
53
Handbook of Polymer Blends and Composites
2.1.2 Advantages
RTM is one of the most efficient and attractive processes of high performance composite
materials with low cost manufacturing. The major benefit over other polymer composites
manufacturing techniques is the separation of the moulding process and the composite
consolidation from the design of the fibre architecture. This allows the designer to combine
different kinds of fibre preform and to produce parts with tailored properties to fit specific
mechanical requirements. In fact, liquid moulding enables attainment of high levels of
microstructure control and part complexity compared with processes like injection
moulding and compression moulding (see Figure 2.2), where the initial fibre distribution
changes as consequence of the fibre orientation during the mould filling.
54
Liquid Moulding Processes
Further RTM is performed at low temperature and pressure. The resin injection at low
pressure allows avoiding fibre deformations due to the resin movement that can induce
fibre wash out. This inconvenience is present in the similar manufacturing technique
Structural Reaction Injection Moulding (SRIM), where the resin is injected under high
pressures and inertial forces have a dominant role during the filling [2].
The investment costs depend on the preform manufacturing and tooling design, and the
geometry of the desired part. Generally the tooling costs are lower than the costs relative
to other liquid moulding processing as reported in the Table 2.1 that compares some
aspects of injection, compression and RTM.
55
Handbook of Polymer Blends and Composites
The mechanical properties of the finished part depend on the type of reinforcement. In
the case of the injection moulding only chopped fibres can be used, leading to lower
strength of the part.
Preform manufacturing;
Resin injection;
Cure;
Part removal.
The first phase, preform design and manufacturing, involves the assembly of the fibre
reinforcement into a preform that is shaped, oriented and located into the mould cavity.
After the mould closing, a low viscosity resin is injected. The resin, usually a thermoset
polymer, wets the fibres and, ideally after the complete mould filling, polymerises to
form the composite part. To activate the cure reaction, the mould and the preform generally
are heated. The air is displaced and escapes from vent ports located at the end of the
mould. Once the part consolidation is complete, it is extracted.
56
Liquid Moulding Processes
In RTM manufacturing many types of fibre reinforcement can be used: carbon, glass (E,
S, S-2, or A), Kevlar fibres, available commercially in roving or twisted into yarns. The
fibre selection depends on the mechanical specifications and environmental requirements
of the desired part and on economical criteria. To form the preform, the dry fibres can be
assembled in several ways: random mats, woven, 2-D and 3-D braiding.
The fibre orientation and the preform geometry have influence both on the structural
properties of the part and on processing issues. To ensure high quality of the final part,
it is necessary to wet completely the preform with the resin. The ease of the resin flow
through the preform and, as a consequence, the cycle time, is strongly affected by the
fibre compaction and orientation. Further, some handling, compressibility and drapability
of the fibre preform are required to cut and stack the fibre layers, to properly distribute
and to adapt the reinforcement to the mould cavity.
Table 2.2 shows the characteristics of the simplest reinforcing fibre forms:
57
Handbook of Polymer Blends and Composites
Other kinds of fibre reinforcement include knit, 2-D and 3-D braided commonly used
for particular applications, for example to manufacture one-piece preforms. Sometimes,
to improve the properties and the processing capabilities, different kinds of reinforcement
are combined, i.e., random mat layers placed between woven layers.
Thermoforming;
In the traditional cut and place preforming, the fibres (chopped or continuous strand
mats, unidirectional fabric, 2-D weaves, 3-D knit) are cut in layers of the desired shape.
Then the layers are assembled manually and packed into the mould cavity.
In the thermoforming process, a fibre continuous strand mat is deformed under pressure
and stretched to drape the mould surface. To link the fibres, a thermoplastic binder is
usually used. At the start, the mat is heated above the melt temperature of the binder
causing the fibres to slide. After forming, the system is cooled at environmental temperature
allowing the binder to become rigid and give strength to the mat.
58
Liquid Moulding Processes
The 2-D Circular Braiding is a textile process where the fibre yarns are twisted to form
the fabric that has a complex shape in most cases. In the past, it was used typically, to
manufacture parts with high aspect ratio (sporting goods such as baseball/softball bats,
golf club shafts, snow skies). In the last year, in the aircraft and aerospace industry the
braided/RTM structures are gaining a great success satisfying the specific requirements
of ballistic tolerance, low weight, high stiffness and economics [3]. The experience achieved
in commercial applications allows applying this technology in aerospace field with low
manufacturing costs over the traditional hand lay-up.
Several resin systems are available for RTM: polyesters, vinyl esters, urethanes, epoxy,
phenols, and bismaleimides. Generally the automotive and infrastructure industries use
low cost and performance systems, while for aeronautical and military applications higher
performance resins are preferred. To ensure good processing conditions and high final
properties, the resin should have to satisfy some requirements:
fibre adhesion;
low volatility;
quick gelation;
To avoid displacement of the fibre reinforcement during the flow, the resin viscosity
cannot be too high, further to assure the full impregnation of the fibre preform before
the complete mould filling, the resin viscosity must be not too low. Before the injection
into the mould cavity, the resin is mixed with the catalyst by a static or dynamic mixing
59
Handbook of Polymer Blends and Composites
system. The injection system consists of positive displacement pumps, one for the resin
and one for the catalyst, able to transfer the reacting system with the correct mass ratio
to the injection gate. Figure 2.5 shows a schematic of a static injection system.
Pump selection is strongly affected by the resin system chosen to inject. In fact, resins,
such as unsaturated polyesters, vinyl esters, acrylic, bisphenol, exhibit a high resin/catalyst
ratio ranging between 100 and 25, while epoxy systems are characterised by low resin/
catalyst ratio in the range 13.3. This induces similar displacement pumps with adjustable
stroke for epoxy systems, while in the case of polyester resins the catalyst pump size is
smaller. Further, since the cure reaction is activated by heat, the mixing of the vinyl esters
and polyester resins with the initiator can occur in the holding tank. However, highly
reactive systems such as epoxies and urethanes have to be kept separate.
accurate control of the resin/catalyst ratio especially for highly reactive resin;
The tooling system design has to be based on the characteristics of the desired part, the
annual production and economic criteria, costs and durability. The shape and the
dimensions of the part affect the mould size, which is also fixed by specifics such as:
60
Liquid Moulding Processes
injection gates;
air vents, to expel the air and cure reaction gas products;
The quality of the final part depends on the correct definition of the mould and its
elements, which are designed and optimised to account for the chemo-rheological
behaviour of the resin and the permeability preform. Another crucial parameter is the
tooling material that influences the costs, the mould durability, the surface quality and
the heat transport phenomena during the manufacturing cycle. In fact, the mould thermal
conductivity and inertia determine the resin-heating rate and, as a consequence, the cure
reaction activation; while the mould thermal expansion controls the shrinkage and the
final characteristics of the part. The most common mould materials include: steel,
aluminium, glass and epoxy.
heating platen press, the mould is positioned between the two heated platens of a
press, in this case, since the heat transfer occurs with conductive properties, the heating
can take place slowly;
integral heating, the mould is equipped with pipes where an heating fluid (hot water
or oil) flows;
Due to the increasing diffusion and use, modelling and numerical simulation of composites
manufacturing processes such as RTM and its derivative processes are becoming more
important allowing control and management of the product quality by designing and
optimising the process. In fact, as in the other composite manufacturing technologies, in
RTM, the processing cycle and the performance of the finished part are strongly coupled.
In fact, the microstructure integrity and the quality of the desired part depends on both
processing parameters and designing factors such as material selection, mould design,
injection temperature and pressure, resin cure conditions, that affect the manufacturing
cycle too. For example, the fibre preform design determines not only the thermomechanical
61
Handbook of Polymer Blends and Composites
requirements of the part but also the filling time, influencing the ease of the resin flow.
Similarly, the optimisation of the mould temperature distribution, the injection temperature
and pressure levels must be based on the purpose of short cycle time, the complete wetting
of the fibre bundles, the absence of the preform deformations and the removal of air
bubbles. Obviously, in addition to these criteria, economical considerations and cost
analysis have to be included playing the major role (see Figure 2.6).
Due to the interaction of different factors and the complexity of the process, process
optimisation, tooling and preform design are often performed by trial and error methods
leading to time consuming and expensive adaptations of the mould and the preform
during the starting period.
62
Liquid Moulding Processes
voids. Before the mould is built, the designer can optimise the filling and the curing stage
by controlling the processing parameters (the injection pressure, the placement of the
gate and the mould temperature) and thus reducing the tooling costs.
However, the production cycle is complex involving different physical and chemical
phenomena. The resin flow is coupled to heat transport, chemical reaction and rheological
changes. The mould filling, in fact, doesnt take place under isothermal conditions. Since,
generally, the resin is injected at a lower temperature than the mould, heat transport
from the mould walls and the fibre perform, to the resin occurs. Further, the curing
reaction is highly exothermic. For fixed mould geometry and fibre preform, the flow
pattern is affected mainly by the resin viscosity that depends on the temperature and on
the degree of the cure. At the start, owing to the heating from the mould, the resin
viscosity may drop by orders of magnitude; when the curing reaction starts, the viscosity
increases as the resin solidifies.
Finally, to successfully predict the filling behaviour and the consolidation stage, the process
modelling should incorporate correctly the coupling between heat transfer, resin cure
and flow, accounting for voids and permeability sub-models.
The resin flow through the fibre preform packed into the mould is a typical example of
flow trough a porous media. It can be examined at different levels depending on the
different scales of flow resistance. In fact, due to the structure of the fibre preform and
the interaction between the liquid and solid phase, the flowing resin encounters various
kinds of flow resistance. A fibre preform is manufactured by weaving tows (called throats)
of fibres that contain thousands of filaments (called junctions) that are intertwined and
twisted with the fibre tows. This structure induces a heterogeneous flow field that can
cause different undesired effects such as [4]:
race tracking (phenomenon induced by the imperfect adherence between the mould
walls and the fibre mat) at the length scale of the fibre mat, promoted by the non
uniform contact between the mat and the mould walls;
spherical macrovoids formation in the interstices of the tow structure at the length
scale of the fibre tows;
cylindrical air bubbles or microvoids formation at the length scale of the fibre
filaments, occurring in the case of fast filling when the resin flows faster around the
fibre tows than in the interstices of filaments.
63
Handbook of Polymer Blends and Composites
Therefore, different models, microscopic and macroscopic, have been used to describe
the resin flow. The main focus of this chapter is to present the macroscopic approach
being able to predict the flow front location, the mould filling time and the pressure
distribution during the mould filling.
The macroscopic flow behaviour is well described by one-phase model based on Darcys
law that is well established to model the flow through porous media. Darcys law relates
the fluid velocity linearly to the pressure gradient by the resin viscosity and the permeability
of porous medium. Darcy deducted this law empirically in 1856 measuring the pressure
drop and the flow rate for water flowing through a column of sand [5]. He found the
following relationship:
V=K (2.1)
L
V = K (2.2)
The K coefficient depends both on the fluid properties and the porous matrix. It is possible
to separate the two effects [5], writing it as:
kg
K= (2.3)
where g/ are the properties of the fluid phase and k is the permeability, function of the
solid phase structure (size and arrangement of the pores, tortuosity, porosity).
For horizontal flow and constant density, the equation (2.2) is:
k
V= p (2.4)
64
Liquid Moulding Processes
Equation (2.4) is the most common form of Darcys law. It is valid for laminar flow, i.e.,
Reynolds number (based on the averaged diameter of the grains of the porous matrix)
ranging between 1 and 10.
In the case of flow of non-Newtonian fluids through porous media, more complicated
expressions of Darcys law are available, accounting for the dependence of the viscosity
on the wall stress [6, 7, 8].
The pressure distribution and the velocity field can be obtained by combining the
continuity equation and Darcys law, that represents the momentum balance equation in
the case of flow of Newtonian fluids in anisotropic, homogenous porous media, when
both inertia and long-range viscous effects are neglected [9]. Assuming that the resin is
uncompressible and the solid phase is stationary, i.e., the resin flow doesnt deform the
fibre preform, the continuity equation has the common form:
v = 0 (2.5)
Substituting Darcys law, Equation (2.4), into the continuity equation gives the governing
equation for the pressure distribution:
k
p = 0 (2.6)
Since typical RTM parts have a thin-shell geometry, the thickness is much smaller than
the other dimensions (see Figure 2.7), the flow in the thickness direction is usually
neglected. In the isothermal case and for a Newtonian fluid, the two-dimensional flow
model is commonly used [10, 11, 12, 13, 14]:
2 2
p p
K x 2 + Ky 2 = 0 (2.7)
x y
The tensor permeability is symmetrical everywhere for homogeneous media, otherwise locally,
so it is always possible to find a coordinate system coinciding with its principal axes.
65
Handbook of Polymer Blends and Composites
Once the pressure distribution is known by solving Equation (2.7), the relationship
between the filling time and the wet length can be obtained by integrating Darcys law.
In the case of one-dimensional flow, Equation (2.7) has an analytical solution, in particular
the pressure field and the filling time are:
x Q0
p( x ) = p0 1
L p( x ) = A K (L x)
x
t = A L
2
t = L
fill
2p0 K x fill Q0
respectively, where L is the cavity length and A the cross sectional area of the gate.
During the mould filling race tracking phenomena and edge flow can occur causing
undesired dry spot formation. These effects are induced by the non-uniform and perfect
contact between the packed fibre preform and the mould walls. In fact, its difficult to
cut the fibre preform to the precise shape of the mould and, consequently, a gap can exist
between the fibre preform and the mould edge (see Figure 2.8).
66
Liquid Moulding Processes
This gap is a preferential flow channel for the resin which moves faster through the porous
medium due to the smaller flow resistance. Conclusively, different velocities Vm and Vg, in
the porous medium and in the gap region, respectively, generate the racing of the resin near
the mould walls and such defects in the part, as dry spot and macro voids formation [15].
These effects can be modelled with two different approaches. In the first approach, due to
the absence of the porous medium in the gap region, the flow behaviour is modelled on the
basis of the Navier-Stokes equation, using, in the porous region, the Brinkmann extension
of Darcys law to ensure the velocity and stress continuity at the interface of the two
regions. This model involves many numerical complications; so another route is preferred
in the recent literature. The gap region is often treated as a porous medium with an equivalent
permeability, modelling the flow pattern with Darcys law [16]. Different expressions of
the equivalent permeability as function of the gap dimension and the fibre preform structure
are available [17, 18]. This approach has been verified experimentally and the results are
in good agreement with those obtained from numerical resin flow profiles.
To properly model the physics of the RTM process, the resin flow model has to be coupled
to the analysis of the heat transport phenomena accounting for the changes of the resin
viscosity, both for the temperature and the exothermic cure reaction advancement. In fact,
the mould filling is strongly affected by the heat transfer with the mould wall and by the
polymerisation chemical reaction that influences the resin viscosity in the opposite way.
A correct simulation of heat transfer would be useful to address some critical issues such as
premature and not uniform resin cure. To obtain parts without defects and without
degradation of the polymer, the polymerisation reaction would have to occur after complete
fibre impregnation and, further, to develop uniformly without strong exothermic peaks.
67
Handbook of Polymer Blends and Composites
Two different thermal models have been investigated to evaluate the temperature evolution
in the mould cavity during the filling. In the most general approach the heat convection
between the flowing resin and the fibres is accounted for by treating separately the two
phases (two-phase model) [19, 20, 21]. Since the convective heat transfer coefficient
between the two phases has to be evaluated experimentally and is related to the specific
fluid-fibre preform system, this model is complicated to implement. However, in RTM
the resin flow is slow which assumes a local equilibrium between the flowing resin and
the fibres. At each point of the filled cavity, the heat transfer coefficient between the resin
and the fibres is very high, allowing us to assume the same temperature for the two
phases that, in terms of energy balance, are depicted as one phase, the composite.
Assuming that the local equilibrium model and two-dimensional resin flow are valid, the
energy balance for the filled cavity has the following form including the conductive, the
convective and the generation terms [22, 23, 24, 25, 26]:
c c pc
Tc
t ( ) ( )
+ rc pr (v T ) == kc T + r H r 1 Vf (2.8)
where the suffixes r and f relate to the properties of the resin and the fibres, respectively,
& is the rate of the reaction, Hr is the heat generated by the curing reaction, Vf is the
fibre volume fraction, , cp and k are the density, the heat capacity and the thermal
conductivity, respectively.
( )
c c pc = rc pr 1 Vf + f c pf Vf (2.9)
and the effective thermal conductivity can range between the two expressions:
(
kc = kr 1 Vf + kf Vf) (2.10)
1
=
(
1 Vf
+
Vf )
(2.11)
kc kr kf
for the parallel and perpendicular heat flow modes between the resin and the fibres, respectively.
Note that, since the heat conduction along the thickness direction is significant as the in-
plane convection, the heat transfer needs to be analysed in three dimensions, conversely
to the resin flow which is usually examined only in the plane due to the thin geometry.
68
Liquid Moulding Processes
In some cases, to account for the local fluctuations in the fluid velocity and temperature
away from their average values, an additional conduction term, named the dispersion
term, is included in the energy balance [19, 24]. It is possible to show by a scaling analysis
that the dispersion term has a relevant effect on the heat transfer when the dimensionless
number Graetz (representing the relative importance between the convection, the flow
direction, and the conduction in the thickness direction) and the ratio between the Graetz
and the microscale Peclet number (ratio between the microscale convection and the
conduction) are large.
To solve Equation (2.8) the boundary conditions at the gate, at the mould walls and at
the flow front have to be set. Generally, Dirichlet conditions are considered assuming:
It is worth noting that the last two boundary conditions are not realistic. In fact, the
mould wall temperature could be affected by the heat transfer phenomena inside the
cavity especially in the case of non-metal moulds. For this reason, some researchers are
developing more reasonable conditions at the mould walls [23]. For example, modelling
the mould with a heating element located at a distance behind the mould wall [22].
Regarding the flow front, setting the flow front temperature equal to the dry fibre
temperature can be a valid approximation near the end of the filling process but not at
the beginning when the temperature gradients between the resin and the fibre mat are
high. This can lead to an incorrect evaluation of the filling time owing to greater heating
of the incoming resin that induces lower values of the resin viscosity and higher velocity.
To solve the energy balance Equation (2.8) and determine the temperature evolution, the
rheological and kinetic models need to be specified:
= ( ,T ) (2.12)
= ( , T , t ) (2.13)
describing the dependence of the viscosity on the temperature and degree of cure and the
cure reaction rate, respectively.
Many empirical models exist depending on the specific resin curing system and its reaction
mechanism.
69
Handbook of Polymer Blends and Composites
To determine the resin conversion, the mass balance equation of the reaction product
(cured resin) needs to be specified:
1
+ v = Df + Dd +
t 1 Vf (2.14)
where Df is the molecular diffusion that is typically negligible compared to the other
terms. To account for the dispersion effects induced by the porous medium tortuosity, an
addition diffusion coefficient Dd is included that is associated with the deviation of the
concentration and velocity from the average values [9].
2.3.7 Permeability
One of the most critical parameters in RTM flow simulations is the fibre mat permeability.
In fact, the right permeability knowledge is useful to describe correctly the mould filling
and the fibre impregnation, allowing to control the complete fibre wetting and the ensure
an high quality production.
Currently several researchers have conducted experiments to characterise the fibre preform
[27, 28, 29, 30, 31] and to find permeability expressions as a function of the fibre volume
fraction, fibre diameter and arrangement [32, 33]. The most common models are: the
capillary model and the flow across cylinder array model. Both models have also been
extended to shear thinning fluids [34, 35].
Capillary model
The capillary model considers the porous medium as a system of parallel capillaries with
diameters estimated in terms of the hydraulic radius of the system. Using the Hagen-
Poiseuille equation to describe the flow in the capillaries and correcting the velocity with
the porosity and the tortuosity of the solid matrix medium, the Blake-Carman-Kozeny
equation for the permeability can be obtained:
1 Rf2 3
K=
L
2
4 (1 ) 2
K0 e (2.15)
L
70
Liquid Moulding Processes
where Rf is the fibre radius, the porosity, Le/L the tortuosity and K0 is the shape factor.
The term K0(Le/L)2 is called the Kozeny constant. For random media with porosity in the
range 0.4 to 0.7, it has a value of 4-5. Equation (2.15) is valid for slow Newtonian flow
through porous media over a limited range of porosity. It cannot be used in the case of
low porosity and high porosity. For low porosity, Equation 15 doesnt predict that the
transverse permeability to the fibre arrangement is zero for the maximum packing fraction.
This means that the transverse Kozeny constant would be infinity. For high porosity the
capillary representation is not geometrically correct.
Further, for transverse flow to unidirectional carbon fibre bed it is found [27] that Equation
(2.15) fits the experimental data well in the porosity range 0.25-0.5 (the ratio between
the transverse and axial permeability is 1/19).
In the case of the flow perpendicular to aligned fibres the permeability can be evaluated
schematising the porous medium as an array of aligned cylinders and calculating the
drag resistance across them. This approach has the advantage to be extended to every
kind of fibres arrangement. To find the closed form solution, extreme arrangements of
the fibre spacing have been considered: aligned cylinders with very low and very high
porosity, solving in the first case, the motion equation within a representative cell of the
closed cylinder packing and, in the second case, thought to be independent cells
approximating the porous medium [2, 37].
Two kinds of experimental flow tests are widely used: one-dimensional and two-
dimensional that can be performed at constant injection pressure or at constant flow
rate. Table 2.3 summarises the features of the two methods.
71
Handbook of Polymer Blends and Composites
The permeability evaluation is carried out applying Darcys law to the experimental data
[38, 39, 40], in particular:
Q0 x fr
Keff = in the case of constant flow rate (2.16)
A p
2
x fr
Keff = in the case of constant injection pressure (2.17)
2p t
where xfr is the flow front position, p the pressure drop between the inlet and flow
front, t the time. It should be noted that Equation (2.16) can be used for saturated and
unsaturated flow experiments, being in the first case the cavity fully filled, while in the
second case the flow front moves.
This kind of test can provide the effective permeability in one direction, so it is helpful to
determine the through-thickness component permeability of transversely orthotropic
preforms or to characterise the random fibre reinforcements. In the first case, in fact, all
the fibres are in parallel planes and one of the principal components of tensor permeability
is perpendicular to the fibre plane, while, in the case of random mats, the flow has not a
preferential path being isotropic the permeability.
In the simple case of two-dimensional tensor permeability, the effective permeability and
the principal component K1 and K2 are related by the relationship:
72
Liquid Moulding Processes
2
K2
1
K K1
Keff = K1 cos 2 ( + ) 1 + 2 tan 2 ( + )
K1 K2
K1 (2.18)
( 2
)
tan ( + ) + 1
where is the principal orientation angle of the fibre perform and is the angle between
the mat orientation and the flow direction. Performing three one-dimensional flow test
experiments at different orientations, the permeability tensor can be completely
characterised [2].
The race-tracking effects can compromise the accuracy of the permeability measurement
with this method. In fact, due to the larger flow along the mould edges, the permeability
estimation could be greater than the real permeability of the preform.
The radial flow experiment is the most popular method, allowing the complete estimation
of the in-plane permeability tensor in a single quick test [41, 42, 43, 44, 45]. In fact, the
visual observation of the flow front shape allows the determination of the principal
orientation. Generally the liquid resin spreads developing a circular or an elliptical flow
front, in the case of isotropic or anisotropic fibre preform, respectively. For the elliptical
front case, the eccentricity of the ellipse gives directly, the principal axes of the tensor
permeability.
In these kinds of measurements, the need is to minimise the flow resistance in the transverse
direction to attain the flow only in the plane, so a hole is usually cut at the centre of the
fibre preform. Further, since the shape of the injection gate affects the flow front profile
evolution, large radial dimensions of the mould are used to permit to the flow front to
reach its steady state orientation.
The permeability estimation is obtained by Darcys law written for radial flow:
K rr dp
vr = (2.19)
dr
In the following section the procedure is reported for isotropic and orthotropic fibre
reinforcement.
73
Handbook of Polymer Blends and Composites
Isotropic permeability
Substituting Equation (2.4) in the continuity equation written in polar coordinates, the
La Place equation is obtained in the form:
d 2p 1 dp
+ =0 (2.20)
dr 2 r dr
Integrating the equation and applying the boundary conditions at the inlet port, radius
R0 and at the flow front Rfr, the flow front position as function of the time can be obtained.
Table 2.4 reports the solutions for the both cases of constant injection pressure and
constant flow rate.
4 K rr p0 t
F= 2
R0
Q0 Q t
2
Rfr Rfr pi = ln 1 + 0 2
with F = 1 + 1 4 hK rr hR0
2 ln
R0 R0
In the case of constant injection pressure, the plot of the F coefficient as a function of the
time gives a straight line, from which slope the permeability Krr is then computed. Similarly,
in the case of the constant flow rate the slope at long times of the inlet pressure pi as
function of lnt allows the estimation of Krr.
Orthotropic permeability
The case of orthotropic permeability tensor may be simplified transforming the pressure
solution Equation (2.7), into an equivalent isotropic system, where it takes the form of
74
Liquid Moulding Processes
the La Place Equation (2.5). With this procedure, the co-ordinates of generic point (x, y)
are related to the transformed coordinates (xe, ye) by:
1/ 4 1/ 4
K K
xe = 2 x and ye = 1 x (2.21)
K1 K2
while the equivalent permeability Keq is related to the two components of the tensor
permeability by:
Keq = K1 K 2 (2.22)
The equations reported in the Table 2.4 with Keq instead of Krr allow its estimation.
Finally, considering that in the equivalent isotropic system the flow front becomes circular
and the relationship between x and y in the original system is:
K2
y= x (2.23)
K1
the plot of the flow front position along the minor axis of the ellipse versus the flow
front position along the major axis gives a straight line with slope K 2 / K1 . This last
estimation together with the equivalent permeability computation enables to characterise
completely the in-plane tensor permeability.
The current experimental flow front tests provide the permeability tensor estimation
separating the in-fibre plane and the through thickness data. This can be inadequate for
the new kinds of fibre reinforcements that are becoming more complex. For intricate
arrangement of the fibres the through thickness direction could not correspond to the
third principal component of the tensor permeability. For this reason some authors are
developing experimental one-dimensional flow test frameworks to identify completely
the permeability [39, 46].
An alternative and promising technique is measurement with fibre optic sensors. The
fibre optic sensor embedded into the fibre preform can be used to monitor the flow front
advancement. After removing the short segment (2 mm) of cladding layer, the fibre optic
is embedded inside the preform. The laser light is transmitted through the optical sensor
75
Handbook of Polymer Blends and Composites
and the light intensity at the end of the optical fibre is recorded. When the flowing resin
reaches the fibre core, the light intensity drops significantly. From the variations of the
light intensity the flow front position can be deduced.
Ahn and others [47] used a three-dimensional sensor grid embedded into glass fibre
preforms to monitor the flow front advancement. The test enables the principal directions
of the permeability to be found by the constructing of the ellipsoidal resin front as function
of the time. The fibre optic measurements were compared with the flow visualisation
data for different fibre volume fractions and were found to be in good agreement.
This technique would be useful to control in situ the fibre wetting by the resin movement
monitoring.
Resin Film Infusion (RFI) has been identified as an alternative cost-effective manufacturing
technology to the conventional prepreg technique and RTM. In fact, the production
cycle is usually performed in the autoclave where, by imposing a temperature and pressure
cycle, a hot-melt thermoset resin impregnates a dry textile preform enclosed in a vacuum
bag and cures to form the composite in a single step process. Unlike the conventional
autoclave process, where the need is to manufacture the prepreg lay-up, in the case of
RFI this requirement is eliminated reducing the costs and the undesired contact of the
operator with the resin. Further, RFI requires only one tool side the other being a polymeric
flexible bag [48], as a consequence, RFI is more economical than RTM, which becomes
expensive for the manufacturing of large parts with high fibre volume fraction. Figure 2.9
shows a schematic of a typical RFI element.
The resin film is placed on the surface of the tool plate, its thickness depends on the resin
mass required to infiltrate completely the fibre lay up that is placed on the top of the
resin film. An aluminium-perforated plate coated with a fibreglass weave is placed over
the assembly to absorb the excess of the resin. The whole system is entrapped with the
air pad and covered with a Nylon separating layer that prevents the resin escape and
allows the cure reaction gas products to be expelled and avoids the contact between the
assembly and the vacuum bag. The full vacuum is applied at the beginning of the cure
cycle and maintained for the whole process ensuring that all trapped air is removed. A
pressure and temperature cycle is imposed into the autoclave allowing the preform
compaction to a specified fibre volume fraction, the resin flow through the fibre and the
curing reaction activation.
76
Liquid Moulding Processes
In the recent years, RFI have gained popularity for industrial applications, arising
out of the need to optimise and control the processing cycle not by trial and error.
However, some authors have developed numerical simulations able to model the
resin infusion process [49, 50]. To provide accurate information about the processing
evolution and the final properties of the composite part, the modelling has to include
both resin flow, heat transfer and chemo-rheological model as in the case of RTM
and compaction analysis too. In fact, during the infiltration, due to the thermal
expansion of the tool and the external applied pressure, the fibre preform structure
can change, modifying its permeability; so to predict the final part thickness and
the fibre volume fraction, the compaction behaviour of the preform have to be
determined [51].
The most popular commercial form of RFI is the Seeman Composites Resin Infusion
Manufacture Process (SCRIMP) that was invented and patented in the late 1980s
by Seeman [52] and used for building large composite structure from boat hulls to
bridges to rail cars, etc. The SCRIMP technology (being low the solvent emissions)
enables the manufacturing of void free and high fibre content laminates in a cleaner
environment than the traditional methods.
77
Handbook of Polymer Blends and Composites
In the last few years, vacuum assisted techniques have gained considerable popularity
being a cost-effective manufacturing technology of large and complex parts and
offering relevant advantages in terms of part size and design. In VARTM a deformable
bag is used to cover a single cavity mould that contains the fibre lay up forming an
envelope that seals the tool-fibre system. The tooling (aluminium or composite) used
to support the structure is usually one-sided. In a typical VARTM sequence, at first
the vacuum is applied to expel the air from the fibre preform and to conform the
flexible sheet against the fibres and the mould surface, then the vacuum is used to
drive the resin to infiltrate the fibre preform. To promote the resin transfer into the
fibre reinforcement, a series of line sources are used being typically tubing or coiled
wire stretched into a spiral. Several authors investigated technological solutions to
accelerate the fibre impregnation such as multiple-port injection [53] and sensor-
based control system of the flow front [54].
The CIRTM [55] technology has been invented and developed to meet the
requirements of the manufacturing of composite integral armour for military ground
applications. CIRTM represents a promising technique expanding the VARTM
capabilities. In fact, even if VARTM is identified as a very cost effective technique
for the manufacturing of large composite parts with high fibre volume fraction and
excellent materials properties, it is used primarily with single resin system. Therefore,
in the case of multicomponent structures in which each layer has a specific purpose,
separate processing step and separate tools, multiple vacuum bagging system and
multiple post cure operations are required. Then, the different components are bonded
together, each inducing cost increases. CIRTM enables the injection of multiple resin
systems into a single step co-cure process eliminating the need for secondary bonding
operations, reducing the costs and improving the part durability and quality by
decreasing the risks of layers separation and mismatches. Figure 2.10 shows the
processing steps of the dual layered structure manufacturing for both VARTM and
CIRTM.
78
Liquid Moulding Processes
The SRIM process is an extension of the Reaction Injection Moulding (RIM) and the
Reinforced RIM (RRIM) technologies.
In the RIM manufacturing process two liquid components, typically an isocyanate and a
polyol, are mixed in a mixing chamber at high pressure by high-velocity impingement
and injected at low pressure into one or more cavities, where reactions occur exothermally.
The RRIM is a reaction injection moulding process where fibre reinforcement (chopped
fibres) is added to the liquid reactants.
Summary
In the last years, the liquid moulding processes have attracted the interest both of academic
researchers and industrial engineers for the production of structural polymer-based
composite materials which are employed in several fields (aerospace, automotive,
infrastructure and military industries). In particular, to control and optimise the RTM
composite manufacturing, great effort has been devoted to:
79
Handbook of Polymer Blends and Composites
Finally, it should expect that the RFI and VARTM technologies will gain significant
development and improvement due to the cost-effective features even to produce large
and complex parts.
References
5. J. Bear, Dynamics of Fluids in Porous Media, Dover Publishers, New York, USA,
1972.
10. A. W. Chan and S. T. Hwang, Polymer Engineering and Science, 1991, 31, 15, 1149.
11. Y-E. Yoo and W. Lee, Polymer Composites, 1996, 17, 2, 368.
12. M. K. Um and W. Lee, Polymer Engineering and Science, 1991, 31, 11, 765.
13. R. Gauvin and F. Trochu, Plastics & Rubber & Composites Processing &
Applications, 1993, 19, 13, 51.
80
Liquid Moulding Processes
14. D. R. Calhoun, S. Yalvac and D. G. Wetters, Polymer Composites, 1996, 17, 2, 251.
17. G. S. Beavers and D. D. Joseph, Journal of Fluid Mechanics, 1967, 30, 197.
19. R. Lin, L. J. Lee and M. J. Liou, Presented at ANTEC 91, Montreal, Canada,
1991, 815.
20. R. Lin, L. J. Lee and M. J. Liou, International Polymer Processing, 1991, 6, 4, 356.
21. R. Lin, L. J. Lee and M. J. Liou, Polymer Composites, 1993, 14, 1, 71.
23. T. J. Wang, R. J. Lin and L. J. Lee, International Polymer Processing, 1995, 10,
4, 364.
26. O. Mal, A. Couniot and F. Dupret, Composites Part A: Applied Science and
Manufacturing, 1998, 29A, 1/2, 189.
27. R. C. Lam and J. L. Kardos, Polymer Engineering and Science, 1991, 31, 14,
1064.
28. A. W. Chan and S. T. Hwang, Polymer Engineering and Science, 1991, 31, 16,
1233.
81
Handbook of Polymer Blends and Composites
38. Y. H. Lai, B. Khomami and J. L. Kardos, Polymer Composites, 1997, 18, 3, 368.
82
Liquid Moulding Processes
51. A. C. Loos and J. D. MacRae, Composite Science and Technology, 1996, 56, 3,
273.
54. S. M. Walsh and R. V. Mohan, Plastics Engineering, 1999, 55, 10, 29.
83
3
Use of Advanced Composite Materials in the
Construction of Suspension Push-Rods for a
Formula One Racing Car
M. D. Gilchrist
3.1 Introduction
Approximately 80% of a modern Formula One racing car is made from polymer matrix
composite materials. The majority of load-bearing components within a car rely on
carbon fibre-reinforced epoxy composites, which have high specific stiffness and strength
properties. Glass fibre and aramid fibre-reinforced composites are also used where a
structural component needs to have good damage tolerance and impact energy
absorption characteristics. A diverse variety of material architectures, including
unidirectional and woven fabrics and continuous and chopped fibres, and various
manufacturing processes such as autoclaving, thermoforming, resin transfer moulding,
vacuum forming and filament winding are available when designing a high performance
racing car. This diversity of materials and manufacturing processes permits design
engineers to customise the performance of a composite structure to a degree that is not
afforded solely by the selection of metals.
Many of the innovations that are first seen in Formula One racing cars are eventually
transferred to niche or broad market commercial vehicles and the use of composite materials
is no exception. Alternative fuel vehicles, including those powered by fuel cells, which are
now manufactured by every major car manufacturer in the world, rely heavily on the use
of advanced composite materials [1] despite the challenges of combining composites
processing operations with the predominant steel stamping and metal forming
manufacturing processes. General Motors 1992 gull-winged all-carbon fibre UltraLite
and Peugeots Tulip (five segment glass polyester over foam core body module) are two
such examples. The use of advanced composite materials in the construction of conventional
internal combustion engine powered cars is continually increasing. The latest S-Type Jaguar
uses a low density (550 kg/m3) 10% fibre reinforced reaction injection moulded (RRIM)
polyurethane system in the production of the interior door panels to satisfy stringent side-
impact protection standards [2]. The production of the moulded door panel in a single
process makes it possible to produce a set of four doors in a mere five minutes.
Other examples of composite materials that are now commonly used in commercial
production vehicles are readily found by considering different classes of composite
85
Handbook of Polymer Blends and Composites
(i) their low coefficients of thermal expansion that facilitate easy control of dimensional
clearances, between various moving parts,
(iii) their advantageous friction properties minimise the requirements for lubrication
between moving parts.
When MMC are used as replacement materials for steel and cast iron in production
engine applications, they are generally used because of their improved wear resistance,
their increased specific stiffness and their superior resistance to high cycle fatigue. It is
almost twenty years since aluminium-based MMC was used to manufacture pistons for
commercially produced Toyota diesel engines [5]. This particular composite piston used
a chopped fibre preform in the ring groove of the aluminium alloy in order to provide
local resistance against the wear and thermal fatigue of the piston. Other composite
pistons used SiC particulate reinforced aluminium, which has a lower coefficient of thermal
expansion than conventional Al-Si alloys. Consequently, this permits the use of smaller
clearances between a piston and the walls of its cylinder, and so leads directly to greater
engine efficiencies, power output and fuel consumption. The use of matrix reinforcements
also serves to increase the specific stiffness of a piston over one that is manufactured
without any reinforcement. More recently, the cylinder liners of the 2.3 litre engine in
the Honda Prelude were selectively reinforced: hybrid preforms of carbon and aluminium
fibres were infiltrated by molten aluminium [6]. Within the last decade other drivetrain
components, such as the connecting rod, have been candidate components for
manufacturing from MMC [7]. The use of lighter components when manufacturing the
piston/connecting rod subassembly can directly reduce the secondary dynamic inertial
forces that can occur, thus prolonging the useful service life of such critical engine
86
Use of Advanced Composite Materials in the Construction of Suspension Push-Rods
components. This is particularly true for smaller capacity engines. Reduced reciprocating
forces should also serve to reduce the loads that are applied to the crankshaft and to
reduce friction losses: consequently, this should enable increased fuel economy or engine
performance to be achieved [8]. However, despite these performance benefits, cost remains
an obstacle to the more widespread use of MMC in production vehicles.
Metal matrix composites, and in particular those based on aluminium matrices, are
suitable candidate materials for manufacturing gears, clutch parts, brake callipers,
turbocharger and supercharger compressors, pump housings, valves, brackets, pulleys,
and suspension pushrods and rockers. The physical characteristics for cast iron brake
systems are such that the cast iron can be replaced by aluminium-based MMC. In
particular, the high thermal conductivity and the good wear resistant properties of
aluminium MMC mean that brake drums and disk brake rotors can be made from these
lighter materials instead of heavier traditional cast iron. The static weight saving is
approximately 50% and this provides added dynamic benefits by reducing inertial forces,
increasing acceleration forces and reducing braking distances. Allison and Cole [8]
reported results of brake dynamometer tests, which indicated that MMC rotors are quieter
than cast iron rotors, they wear less, and they provide more uniform braking friction
throughout the braking cycle. Specific examples of commercially produced automotive
vehicles that use aluminium MMC disk brake rotors and brake drums would include
General Motors EV-I, the Plymouth Prowler and the Lotus Elise. Within automotive
racing applications, where improved performance takes priority over lower priced
materials, Hurley [9] has reported the use of aluminium MMC: brake callipers for Formula
One racing cars that are produced from an aluminium 2124/SiC/25p MMC are stiffer
than conventional cast iron and thus provide greater leverage and shorter stopping
distances. The use of aluminium MMC in the driveshaft of a vehicle capitalises on the
greater specific stiffness of these materials as compared to more commonly used aluminium
and steel. All driveshafts, irrespective of the material from which they are manufactured,
are constrained by the rotational speed at which they become dynamically unstable. The
critical speed depends directly on the inner and outer diameters and specific stiffness of
the driveshaft and inversely on the length of the driveshaft. In space restricted areas,
where it is inappropriate to increase the diameter of a driveshaft, the greater specific
stiffness of MMC offers attractive design alternatives to conventional metals: they permit
longer lengths of driveshaft at given diameters or reduced diameters at given driveshaft
lengths. For this reason, aluminium 6061/Al2O3 MMC have been used to manufacture
composite driveshafts for automotive vehicles.
Despite the steady increase over recent years and the projected future growth in using
composites in automotive applications, which are largely due to the specific physical
characteristics of the materials, a major breakthrough in manufacturing primary structural
parts in high volume automotive applications has not occurred. Most successful uses of
composites have been limited to non-structural and semi structural applications, and
87
Handbook of Polymer Blends and Composites
(i) Non-structural materials. These materials would broadly include short glass fibre-
reinforced polymers with reinforcements accounting for 10-50% by weight.
(ii) Semi-structural materials. These materials would typically be either glass mat
thermoplastics (GMT) and sheet moulding compound (SMC).
In commercial automotive vehicles, the use of polypropylene (PP) matrix composites has
become a ubiquitous success due to the excellent processing and mechanical characteristics
of PP, its good chemical and weather resistance, and its relative cost. High performance
PP composites with good physical properties have been developed in the last two decades
by chemically combining elastomeric (rubber) or inorganic (talc) fillers with PP to provide
a useful material with an extremely wide range of physical properties. The use of rubber
88
Use of Advanced Composite Materials in the Construction of Suspension Push-Rods
fillers provides a material that is both impact resistant and damage tolerant and has lead
to one of todays major applications of PP composites, namely, the automobile bumper.
The incorporation of talc fillers served to increase the rigidity of this material and
subsequently facilitated the development of large, wraparound PP bumpers.
Prior to the mid-1970s, advanced composite materials had only been used to manufacture
secondary and non-structural components in Formula One racing cars. However, the
failure of a rear wing at the 1975 Spanish Grand Prix, and the consequent death of four
spectators, dampened the budding enthusiasm of designers for composites. The current
extensive use of advanced composite materials in Formula One World Championship
motor racing can be traced back to the latter half of the 1970s when Colin Chapman of
Lotus designed the wing-car to have large downward thrust forces. This was achieved
by using large wing shaped under bodies on either side of the car. While this successfully
allowed greater speeds, the chassis width had to be reduced in order to incorporate these
large under wings. However, the reduced chassis width served to reduce the torsional
rigidity of the car and this was exacerbated by the increased cornering speeds afforded
by the under wings. The wing-car design proved so successful that within five years
after it was introduced, lateral acceleration forces of up to 3 g were being recorded; these
severe forces and the inadequacies of the suspension system were physically demanding
on drivers. The late 1970s saw the emergence of turbochargers that were capable of
providing thrust forces in excess of 1050 kW. This forced design engineers to consider
using the specific stiffness advantages and design flexibility of advanced composite
materials as primary structural load-bearing components instead of the high-performance
aluminium alloys that had previously been used.
The first major breakthrough in the use of composites in Formula One came with the
1981 McLaren MP4 car, the monocoque (shell structure of a car) of which was moulded
from a carbon fibre-reinforced epoxy polymer. The monocoque, which was manufactured
by Hercules in America, was moulded over a machined aluminium tool that was
subsequently removed in pieces through the cockpit opening. Unidirectional carbon/
epoxy was used as skins that surrounded an aluminium honeycomb core. This one-piece
construction proved so successful that it continued to be used by McLaren virtually
unchanged for six seasons. Gustav Brunner, in 1983, developed his ATS Formula One
car by moulding the monocoque in top and bottom halves, both parts being built in a
female mould. Not having to remove the moulding tool meant that the monocoque
could have a more aerodynamic and rounded shape and the cockpit opening could be
smaller than the McLaren car. However, this two-piece construction, which had a
circumferential joint horizontally around the monocoque, did not have the same structural
integrity as did McLarens one-piece.
Since the early 1980s the issue of joining and joints has become less problematic and
designers have been able to focus their attention on using composites to manufacture
89
Handbook of Polymer Blends and Composites
components other than primary structural parts but which are equally important to the
optimum performance of a car. These have included the gearbox, where torsional rigidity
is critical, and suspension components, where push-rods require high stiffness when setting
the suspension. Traditional metal suspension components have begun to be replaced by
composite materials that have greater specific stiffness and thereby provide designers with
greater control over the localised and overall stiffness of a complete suspension system.
The push-rod is the single component that most determines the stiffness of the suspension
system. However, it is to be noted that the change from metal to composites has not been
without problems for many Formula One teams. Williams, for example, replaced their
metal push-rods with composites but had to revert to metal after a series of rear suspension
failures during testing. Strength and energy absorption have also become important design
issues and these are made more complex with the use of increasingly sophisticated geometric
shapes and constructions.
The technical regulations governing the design of a Formula One racing car are defined at
the beginning of each racing season by the Fdration Internationale de lAutomobile [14],
the FIA, which is the ultimate governing authority for motor sport throughout the world.
These comprehensively address the bodywork, the dimensions, the weight, the engine, the
fuel, oil and coolant systems, the electrical systems, the transmission system, the suspension
and steering systems, the brake system, wheels and tyres, the cockpit, safety equipment and
safety structures, impact testing, roll structure testing, and static load testing. The key
parameters affecting structural design are those of material, geometry and weight. Constraints
are placed on the dimensions of the bodywork of a car (overall width and height must not
exceed 180 cm and 95 cm, respectively), the weight of a car (must not be less than 600 kg),
the maximum engine capacity (four stroke engines of not more than 3000 cm3, twelve cylinders,
and five valves per cylinder) and the size of the driver envelope within the cockpit. It is
estimated that 10 kg of excess weight can lead to a time difference of at least 0.1 second
around the length of a typical racing circuit; such a time penalty is critical in Formula One.
Static and impact design loads are specified for the safe operation of critical components of a
car, the roll bars and the drivers survival cell. All tests are required to be carried out in
accordance with FIA Test Procedures in the presence of an FIA technical delegate and by
using suitably calibrated measuring equipment.
A modern Formula One racing car has a central load-bearing structure that connects the
front and rear suspension systems. This structure consists of the monocoque, the engine, and
the gearbox casing. The driver, fuel tank and front suspension dampers are contained inside
the monocoque, while the engine is jointed to the back of the monocoque on four separate
studs. The gearbox casing is attached to the rear face of the engine. It is this three-piece box-
beam structure of monocoque, engine and gearbox that carries the inertial loads to the four
corners of the car. Various wing structures, under bodies, cooler ducting and bodywork are
attached to and arranged around this box-beam. The dynamic behaviour of this box-beam
structure exerts a major influence on the performance of a car during motion. While setting-
90
Use of Advanced Composite Materials in the Construction of Suspension Push-Rods
up a racing car at a circuit, subtle changes are made to the suspension elements in order to
modify its handling characteristics. Ideally, any small changes in the stiffness of a component
should be noticed in the balance of the car. However, this will not be the case if the structural
members transmitting the loads are insufficiently stiff. Therefore, the chassis members must
have good stiffness characteristics if the handling and speed around a circuit are to be optimal.
This Chapter describes the design, manufacture and structural response to compressive forces
of a set of carbon fibre-reinforced epoxy suspension push-rods. The structural behaviour
and specific stiffness of the push-rods is compared to that of equivalent weight steel and
aluminium push-rod designs and it is shown that advanced composite materials can provide
superior performance advantages when compared to conventional engineering metallic
materials. The ultimate catastrophic failure of the composite push-rods, which, in normal
design practice would occur at loads far in excess of operating loads, is examined and the
particular mechanisms of fracture and damage are detailed.
3.2.1 Design
In order to minimise the effects of wind-drag around the push-rods, it was decided to
utilise an aerofoil cross-section instead of a circular cross-section. Uniform and tapered
layups were used, the purpose of the taper being to increase the equivalent modulus
along the critical section of the push-rod and consequently, to increase the load at which
buckling would occur. Since the end sections of both the tapered and uniform layups
were identical, it was anticipated that the load at which compression failure should
occur would be identical for both types of push-rod. The push-rods were 650 mm long,
whilst the nominal wall thickness was 1.825 mm for the uniform push-rod, and varied
between 1.825 and 2.450 mm for the tapered push-rods. The major and minor external
dimensions of the airfoil axes were nominally specified at 38 x 18 mm.
One objective of this investigation was to establish the influence of the layup on the possible
buckling response of the push-rod. Since the principal in-service mechanical load on the
push-rod was uniaxial compression, it was necessary to maximise the number of 0 plies
within the stacking sequence in order to provide maximum uniaxial stiffness. A number of
cross-plies were necessary, however, to prevent longitudinal splitting of the push-rod. The
first stacking sequence that was considered was a uniform layup of (0/90; 09; 0/90),
i.e., two external 0/90 cross-plies of woven prepreg surrounding nine unidirectional 0
plies. This layup differs from that which is typically used in current Formula One design
only in that there are no tapered plies within the stacking sequence. As such, it was anticipated
that the ultimate mechanical response of this push-rod design would be a lower bound
limit and failure would be due to buckling.
91
Handbook of Polymer Blends and Composites
The remaining two push-rods were identically tapered along their midlengths and the
particular stacking sequence that was used was 0/90; 07; 90; 06; 0/90), i.e., two
outer 0/90 cross-plies of woven prepreg surrounding seven 0 plies, one 90 ply and six
0 plies. The taper was obtained by only placing some of the 0 and 90 plies along part
of the 650 mm length of the push-rods.
Commercially available laminate analysis software was used [15] to estimate equivalent
laminate properties and strengths from the precise ply properties of Table 3.1. The values
of these equivalent laminate properties are detailed in Table 3.2.
Table 3.2 Equivalent mechanical properties of uniform and tapered layups used
to manufacture the different push-rods
Equivalent Laminate Property Uniform Layup Tapered Layup
Longitudinal stiffness, E11 211 GPa 221 GPa
Transverse stiffness, E22 23.7 GPa 34.7 GPa
Poissons ratio, 12 0.115 0.068
Poissons ratio, 21 0.013 0.011
Shear modulus, G12 11.6 MPa 11.7 MPa
92
Use of Advanced Composite Materials in the Construction of Suspension Push-Rods
Using simple Euler buckling theory it was possible to predict the loads and strains
associated with buckling by approximating the airfoil cross-section with that of an ellipse:
2E
cr =
(KL / r)2
where E is Youngs Modulus
K is the buckling coefficient (0.7 for one end pinned and one end fixed; 0.1 for
both ends pinned)
L is the length of the push-rod
r, the smallest radius of gyration, is defined by
r = I/A
For push-rod 1 (PR1), classical laminate theory predicts a longitudinal modulus of 211
GPa (compare with Table 3.2). The moment of inertia is calculated from the half-major
and half-minor dimensions of the cross-section of the push-rod. These external dimensions
are 18.87 mm and 8.94 mm, whilst the corresponding internal dimensions are 17.05
mm and 7.12 mm, respectively. Consequently, the moment of inertia of PR1 is given as:
IPR1 =
4 (a 3
ext bext )3
aint bint = 5756mm4
Similarly, the cross-sectional area of PR1 is calculated as 148.6 mm2. In order to calculate
both the critical buckling load and the surface strains within the push-rod at this
corresponding level of load, it is necessary to determine the correct buckling coefficient, K,
for the particular boundary conditions that are applied to the push-rod ends. The nominal
boundary conditions, selected to represent what would typically occur in service, define
one end as pinned and allowed to rotate about this point and the other end as fixed and
93
Handbook of Polymer Blends and Composites
prevented from any displacements or rotations about this point. For such a case, the bucking
coefficient K = 0.7. However, the physical conditions within the constrained end of the
push-rod may not actually be quite as severe in reality and the design could reasonably be
considered insufficient to fully prevent rotation because of the fact that the push-rod is
hollow with a relatively thin wall section and the interior walls of the push-rod are not
prevented from rotating in towards the middle of the push-rod. The particular loading
arrangement used for the experimental tests (described in section 3.3) did not permit the
fixed boundary condition to transfer bending moments into the push-rod. Consequently, it
is more realistic to consider that the actual boundary conditions were closer to both ends
being pinned (K = 1) than to one end being fixed and the other being pinned (K = 0.7). On
this basis, the critical buckling load, Pcr = 57.9 kN if K = 0.7 and Pcr = 28.4 kN if K = 1. The
corresponding levels of surface strain would be 0.19% if K = 0.7 and 0.09% if K = 1. It
will be shown in section 3.4.2 that these values compare quite well with the average failure
strain of 0.10% and the failure load of 31.0 kN for push-rod 1.
The half-major and half-minor external and internal dimensions of tapered push-rods PR2
PR3 were 18.87 mm, 8.94 mm, 16.42 mm and 6.49 mm, respectively. The moment of
inertia and cross-sectional area were therefore calculated as 7064 mm4 and 195.2 mm2,
respectively. The critical buckling load Pcr = 74.4 kN if K = 0.7 and Pcr = 36.5 kN if K = 1.
The corresponding levels of surface strain are 0.17% and 0.08%. These values will be
shown in section 3.4.2 to compare well with the actual failure loads of the push-rods.
3.2.3 Manufacture
The autoclave curing cycle for the woven and unidirectional carbon/epoxy prepreg
involved a 1.5 hour cure at 125 C and 700 kPa with a heat-up and cool-down rate of 3
C per minute. When the temperature reached 125 C the vacuum was vented to
atmosphere. Pressure was then introduced and ramped at 100 kPa per minute to 700
kPa. When the pressure cycle was completed the pressure was ramped down at 50 kPa
per minute to 0 kPa, at which stage the vacuum was reintroduced.
Both the mould and the silicone mandrel were reused when manufacturing all three
push-rods and these were cleaned and degreased before being coated with release agent
94
Use of Advanced Composite Materials in the Construction of Suspension Push-Rods
(Freekote) prior to the plies of carbon/epoxy prepreg being wrapped around the mandrel
and placed within the mould. Figure 3.1 illustrates how the complete assembly was
vacuum bagged to evacuate air, solvents and entrapped volatiles from the laminate and
to allow the positive autoclave pressure to consolidate the laminate against the mould
surface. A breather cloth bagging assembly (absorbs any excess resin in the manufacturing
process) was used to absorb any excess resin flow and also to smooth out the sharp
corners of the mould, which could cause the vacuum bag to rupture under the high
autoclave pressures. A solid release film was placed against the mould walls to prevent
the breather cloth from sticking to the mould surfaces. Upon completion of the curing
cycle, the vacuum bag assembly was removed from the autoclave. The bag and breather
were discarded and the end plates were removed prior to the mould being opened. The
composite push-rod was then taken from the mould and the silicone mandrel removed
from the centre of the push-rod.
95
Handbook of Polymer Blends and Composites
hours to complete the curing process. The hollow silicone mandrel was then removed
from the mould and the copper pipe was extracted from the mandrel. No significant air
bubbles or voids, which would have made the mandrel unsuitable for manufacturing the
push-rods, were detected visually.
The GFRP slips were fabricated using the two halves of the mould. After spraying
release agent on both halves of the mould, six plies of GFRP were stacked in each
half of the mould. The two halves of the mould were covered in a release ply, covered
with a bleeder cloth (similar to breather cloth) and placed in a vacuum bag, which
was then sealed. The assembly was placed in the autoclave and cured using an
appropriate curing cycle.
Three push-rods have been tested statically to failure under compression using a
displacement mode of control on a 100 kN uniaxial servohydraulic fatigue machine
(Series 8501 Instron). The loading was introduced at both ends of a push-rod using
female end-fixtures which had been designed to provide boundary conditions that were
pin-jointed at the bottom and cantilevered at the top in order to simulate in-service
support conditions. This is shown schematically in Figure 3.3.
Surface strains, from gauges at three different positions on the push-rods, were
recorded using a data acquisition system which operated on a keypress sequence.
Strain gauges were aligned longitudinally and transversely along the length of the
96
Use of Advanced Composite Materials in the Construction of Suspension Push-Rods
Figure 3.3 Compression test setup used to establish the ultimate performance of the
carbon/epoxy push-rods
push-rods to measure the performance under compressive load. Two gauges (SG1
and SG2 of Figures 3.4(a)-(c)) were aligned axially at the midlength and on opposite
faces of the push-rods: these provided information on the presence of buckling, the
deviation from linearity in the mechanical response of the push-rod, and the fracture
strains. A third strain gauge (not included in Figures 3.4(a)-(c)) was aligned normal
to the first two gauges, close to the mid-length position of the push-rod, and was
used to calculate values of Poissons ratio.
97
Handbook of Polymer Blends and Composites
(a) (b)
(c)
Figure 3.4 Variation of surface strains with actuator load during testing of (a) PR 1,
(b) PR 2, (c) PR 3. Buckling is identified by the difference between the values of the
two surface strain readings, and starts with the onset of actuator load. Incipient
fracture is identified by the deviation of strain differences from linearity, which occurs
at approximately 90% of the ultimate failure load. The solid and dashed lines
represent average strains and strain differences, respectively.
98
Use of Advanced Composite Materials in the Construction of Suspension Push-Rods
All push-rods were loaded statically to failure by means of a displacement mode of control.
Load, displacement and strain values were collated at discrete increments of actuator
load. As the applied load was increased from zero, the response of the push-rods was
initially linear elastic. Figures 3.4(a-c) detail the variation of compressive surface strains
at the mid-length position on opposite sides of the three push-rods with increasing actuator
load. The strain responses deviated from linearity at approximately 90% of the final
failure load, although minor fracture events occurred before this deviation from linearity
in the first push-rod test (a minor fracture event occurred at 10.8 kN in Figure 3.4a).
This deviation of the strain difference, i.e., magnitude of strain difference between front
and back faces of the push-rods = SG2-SG1, from linearity, which occurred at
approximately 90% of the final failure load identified the onset of catastrophic fracture.
Ultimate failure of the first uniformly laid-up push-rod occurred some 40 mm from the
centre of the specimen whilst failure of the remaining three push-rods was concentrated
around the bottom ends of the specimens, close to the pin-jointed end-fixture of the
testing machine. Table 3.3 details the loads and strains at which ultimate failure and the
deviation from linearity occurred during the three push-rod tests.
The measured actuator loads and surface strains are presented in Figures 3.4(a-c) for the
three push-rods. The average load-strain relationship for all the push-rods is essentially
linear almost until fracture. However, the individual strain-gauge readings deviate from
linearity immediately with the application of load and this deviation continues to increase
directly with applied load up until failure.
99
Handbook of Polymer Blends and Composites
Table 3.3 identifies the maximum direct compressive strains which were measured during
each test and may be compared against the failure strains of the carbon-fibres of 1.5% [16].
While the maximum direct strain reading at failure of push-rod 1, i.e., 0.18%, is greater than
those recorded during the other three tests (this is to be expected since the strain gauge
position of this push-rod was closer to the failure site than in all other tests), this is considerably
less than the fibre failure strain. Consequently, failure of these components is considered to
be due to geometric and manufacturing factors rather than to material limitations.
A simple first mode of buckling was apparent along the length of the four test specimens,
with maximum lateral deformation, i.e., crest of the buckle, occurring close to the mid-
length of the push-rods. Buckling initiated with the application of load in all push-rod
tests, as can be seen from the deviation of the two sets of surface strain gauge readings
(Figures 3.4(a-c)) from the average compressive strain. The amplitude of the buckle
increased linearly in magnitude with actuator load until failure. No dial gauges were
used during the tests to quantify the amplitude of the buckle although this could be
estimated from the degree of bending that has been measured by the surface strain gauges.
The average failure strain and failure load of push rod 1, PR1, can be seen from Figure
3.4(a) to be 0.10% and 31 kN, respectively. These compare quite well with the previous
theoretical predictions (see also Section 3.2.1) for buckling with K = 1, i.e., Pcr = 28.4 kN
and strain = 0.09%. Similarly for push rod 2 (PR2) and push rod 3 (PR3), the theoretically
predicted load and strain with K = 1 (see also Section 3.2.1) are 36.5 kN and 0.08%,
respectively. These compare reasonably well with the experimental data shown in Figures
3.4(b) and (c) of average failure strain = 0.07% to 0.10% and failure load of 30 kN.
The compressive failure mechanisms that occurred in all four push-rods were similar although
failure of push-rod 1 occurred at a position close to the mid-length of the component whereas
failure was close to the pin-jointed end for the other three push-rods. The reason for this
different failure site is due to the fact that push-rod 1 was manufactured without any tapered
region in its mid-section, unlike the other two push-rods. The general appearance of the
fracture associated with the two tapered push-rods, PR2 and PR3, is shown in Figure 3.5.
The appearance of the fracture surface is different both around the perimeter of the push-rod
and through the thickness of the push-rod. The fracture is not uniformly compressive around
the perimeter and this is due to the different degrees of compressive strains that existed on
opposite sides of the push-rod. The lack of similarity of the through-thickness fracture features
is partly due to the variation of compressive strains and partly due to the different ply
orientations through the thickness of the push-rod. Figures 3.6 to 3.8 detail the compressive
100
Use of Advanced Composite Materials in the Construction of Suspension Push-Rods
failure sites that led to ultimate fracture of the push-rods as identified using scanning electron
microscopy. Many buckled and broken fibres and crushed fibre ends are evident in Figure 3.6
and the manner in which these fibres fractured is characteristic of compressive failure, i.e.,
fibre microbuckling and localised fibre fracture [17-20].
101
Handbook of Polymer Blends and Composites
Figure 3.8 Local view micrograph of fracture surfaces of fibre ends. Note the degree of
poor fibre-matrix adhesion where the fibres are debonded from the epoxy resin.
102
Use of Advanced Composite Materials in the Construction of Suspension Push-Rods
3.5 Discussion
A conventionally designed tubular steel cylindrical push-rod, which could be of 22 mm
external diameter and 2 mm wall thickness, would have a critical buckling load of Pcr =
31.1 kN and average surface strain of 0.12% for K = 1. The corresponding weight of
such a push-rod would be 0.613 kg (density = 7500 kg/m3). This compares against the
composite push-rods of the present chapter which would have similar buckling strength
and stiffness but which would be almost 70% lighter and would weigh approximately
0.203 kg (density 1600 kg/m3).
The average failure strain and failure load of push rod 1, PR1, can be seen from Figure 3.4(a)
to be 0.10% and 31 kN, respectively. These compare (see also Section 3.2.1) quite well
with the previous theoretical predictions for buckling with K = 1, i.e., Pcr = 28.4 kN and
strain = 0.09%. Similarly for push rod 2 and push rod 3, PR2 and PR3, the theoretically
predicted load and strain with K = 1 (see also Section 3.2.1) are 36.5 kN and 0.08%,
respectively. These compare reasonably well with the experimental data shown in Figures
3.4(b) and (c) of average failure strain = 0.07% to 0.10% and failure load of 30 kN.
While it is clear that composite suspension push-rods can be designed and manufactured
to provide equal stiffness and strength characteristics as conventional steel or aluminium
push-rods, and at the same time offer some degree of weight savings, it is necessary to
realise that other issues are important in deciding whether current metallic materials
could be replaced safely by high performance composites. The amount of energy
absorption by the suspension system in the event of an accident to the corner of a car, for
example, may well establish that plastic deformation occurring in wishbone tubes
(structural member, common part of suspension system) outperform a more brittle mode
of fracture in a composite system. In other cases, the reasons for using a composite
system might not be structural but could be due to aerodynamic factors where the
flexibility of manufacturing from composites permits greater geometric freedom.
3.6 Conclusions
Unidirectional and woven cross-ply carbon/epoxy composites were used to manufacture
suspension push-rods that could typically be used in a Formula 1 racing car. These were
subsequently loaded to failure under compression using end boundary conditions which
approximated those that would be used typically in service. Two push-rods had a tapered
mid-section consisting of 0 and 90 plies whilst an initial trial specimen was of constant
thickness along its length. The results of this investigation can be summarised as follows:
1. Fracture of the trial specimen (PR1) occurred close to the mid-length whilst fracture
in both other cases was close to the pin-jointed support in the loading frame. Buckling
occurred in all cases and this increased directly with the application of load.
103
Handbook of Polymer Blends and Composites
3. While these composite push-rods offer valuable weight savings over conventional
steel push-rods for Formula One motor racing without compromising strength and
stiffness, it is important that issues such as manufacturing complexity and
crashworthiness are evaluated thoroughly before conventional engineering materials
are replaced by high-performance composite materials.
Symbols
P actuator load
R smallest radius of gyration
Poissons ratio
stress
cr critical
11 longitudinal
22 transverse
Acknowledgements
The author would like to acknowledge the support provided by Messrs L. Curley and J.
Mallon of the National University of Ireland, Dublin, J. McQuillian of Jordan Grand
Prix and B. ORourke of Williams Grand Prix Engineering. Financial support provided
by Enterprise Ireland, Materials Ireland, and the National University of Ireland, Dublin
(Presidents Research Award) is gratefully acknowledged.
References
104
Use of Advanced Composite Materials in the Construction of Suspension Push-Rods
11. N. Svensson, R. Shishoo and M. D. Gilchrist, Polymer Composites, 1998, 19, 4, 360.
15. Laminate Analysis Program, LAP, 1991, Centre for Composite Materials,
Imperial College, London, SW7 2BY, UK.
16. D. R. Lovell, Carbon and High Performance Fibres Directory, 5th Edition,
Chapman & Hall, London, 1991.
17. M. D. Gilchrist and N. Svensson, Composites Science & Technology, 1995, 55,
2, 195.
105
4
Corrosion Resistance of Polymers, Polymer
Blends and Composites in Liquid
Environments
H. Hojo and K. Tsuda
Plastics, elastomers and composites are being widely used as construction materials
and also as coating or lining materials in corrosive environments. Their contribution
to the industries is quite significant, however, the materials in contact with the liquid
environment cause two kinds of degradation, one is a physical degradation and the
other is a chemical one.
When plastics are in contact with liquid environments, liquid molecules always
penetrate the materials. As illustrated in Figure 4.1, liquid molecules penetrate and
diffuse between the molecular chains of polymer, then the material cause swelling
which increase its mass and also volume but goes down the physical and mechanical
properties of the material.
If the polarities of plastics and liquid are different (polar and non-polar), the liquid
absorption reaches a equilibrium state after some exposure (Figure 4.1(a)). If the polarities
are the same such as polar to polar or non-polar to non-polar, after swelling the solvation
phenomenon occurs and polymer molecules are covered with liquid molecules, then the
molecular chains of polymers are dissolved physically because of loosening inter-molecular
bonding of the molecular chains (Figure 4.1(b)). Examples of the polarity of polymers
and liquids are shown in Table 4.1. Thus non-polar plastics show poor resistance to
hydrocarbon solvents that have low polarity. Crystalline thermoplastic polymers, however,
often show better performance for such applications. For example, polyethylene will
only dissolve in hydrocarbon solvents [1].
There are some exceptional intermediate plastics. For example, polyvinyl chloride (PVC)
shows polar behaviour in some degree to water, and unsaturated polyester (UP) has
polar group in the structure, but its framework is non-polar.
107
Handbook of Polymer Blends and Composites
The degradation of linear type thermoplastics is controlled mainly by this physical one,
however, in thermosetting plastics with three-dimensional structure (so called network
polymers) the amount of swelling is relatively low and depends on the degree of
crosslinking.
108
Corrosion Resistance of Polymer, Polymer Blends and Composites in Liquid Environments
As is well known, the swelling and the dissolution of thermoplastics occurs in organic
solvents, and the solubility parameter is used as a measure of solubility and swelling [2].
The solubility parameter is given by:
1/ 2
E d
= ( E / V )
1/ 2
= (4.1)
M
Figure 4.2 Relationship between swelling ratio and solubility parameters of various
liquids (orthophthalic type unsaturated polyester resin, 50 C)
109
Handbook of Polymer Blends and Composites
Polymers which have some special functional groups and/or bonds, tend to degrade by
chemical reaction in some environmental conditions, especially when in contact with
liquids [5]. This chemical degradation is considered as the corrosion phenomenon of
polymeric materials, as in metals. In a broad sense, corrosion is defined as the
110
Corrosion Resistance of Polymer, Polymer Blends and Composites in Liquid Environments
The probable corrosion reactions between reactive functional groups and bonds in
polymers and chemical environments are shown in Table 4.2.
In case of the resin without the ester bond in its structure, no corrosion occurs in acidic or
alkaline solution. Figure 4.3 shows an example of epoxy resin cured with menthane diamine
(MDA-ER, see Figure 4.11) immersed in sodium hydroxide solution at 80 C [6].
Figure 4.3 Relationship between retention of flexural strength of epoxy resin cured
with amine and immersion time in NaOH solution at 80 C
(: 10wt%, : 20wt%, : 30wt%, : 40wt%)
Of the corrosion reactions shown in Table 4.2, the most typical ones are hydrolysis and
oxidation.
a) Hydrolytic corrosion
Resins including ester bonds (symbol E in Figure 4.4) in main polymer chain and/or
crosslinking chain such as unsaturated polyester and epoxy resin cured with acid
anhydride, suffer chemical attack by acid or alkali, and scission of polymer chains occur
as the result of hydrolysis. The hydrolytic reaction by acid or alkali is expressed as,
111
Handbook of Polymer Blends and Composites
Figure 4.4 Schematic chemical structures of cured resins, (a) ortho or isophthalic type
unsaturated polyester resin, (b) novolak type vinyl ester resin, (c) bisphenol type epoxy
resin cured with methyl-tetrahydrophthalic anhydride (MTHPA)
Reaction by acid:
(4.2)
112
Corrosion Resistance of Polymer, Polymer Blends and Composites in Liquid Environments
Reaction by alkali:
(4.3)
When hydrolytic corrosion occurs, the number of ester bonds in the resin decreases and
carboxylate is produced as a corrosion product. The mechanism of hydrolytic corrosion
is proved by measuring IR spectrum before and after immersion of resin specimens [7].
Figure 4.5 shows IR absorption spectrum of three network resins before and after
immersion in 10wt% sodium hydroxide solution at 80 C. It is recognised that the ester
peak near to 1720 cm-1 becomes smaller and the carboxylate anion peaks at near 1570
cm-1 and 1440 cm-1 appear by immersion for all resins tested.
Figure 4.5 IR spectrum of resin surface before and after immersion in NaOH solution
at 80 C. (a) isophthalic type unsaturated polyester resin, immersed 504 h in 10wt%
solution, (b) novolak type vinyl ester resin, immersed 504 h in 10wt% solution, (c)
bisphenol type epoxy resin cured with methyl-tetrahydrophthalic anhydride
(MTHPA), immersed 360 h in 40wt% solution.
The hydrolytic reaction occurs also in ether, amide and urethane bonds as shown in
Table 4.3 [6, 8, 9].
113
Handbook of Polymer Blends and Composites
Ether
Polyether
Amide
Polyamide
Urethane
Polyurethane
b) Oxidative corrosion
Polymers having double bonds or ether bonds in their structure and also methyl, methylene
groups in fatty acids or phenyl groups, are open to attack by oxidising agents like O2, O3
and H2O2, and also strong oxidising acids.
In atmospheric environment, plastics are degraded gradually by the action of O2 and sunlight
(ultraviolet rays). This type of corrosive reaction called auto-oxidation occurs as:
(4.4)
114
Corrosion Resistance of Polymer, Polymer Blends and Composites in Liquid Environments
Oxidising agents such as NaClO, KMnO4, ClO2 attack the unsaturated position in resin
structures, and causes severe oxidative corrosion. Corrosion reaction of polymers by
NaClO is considered as:
(4.5a)
(4.5b)
(4.5c)
Solutions of oxidising acids like concentrated H2SO4, HNO3, H2CrO4 or H3PO4 also
corrode resins by oxidation other than hydrolysis. Figure 4.6 shows this behaviour of
vinylester resin immersed in three acidic solutions at pH 0.1 and 50 C [10].
Tensile strength decreases by two or three stages, and each stage corresponds to a different
mechanism of corrosion. Especially at the second stage, strength decreases with time,
and its degree depends on the type of acid and its oxidising strength. At the third stage,
the slope of the straight line for three acids are the same, then this stage of corrosion is
controlled by hydrolysis.
115
Handbook of Polymer Blends and Composites
Figure 4.6 Variation of tensile strength of novolak type vinyl ester resins by immersion
in pH 0.1 oxidising acidic solutions at 50 C
( : H3PO4, : HNO3, : H2CrO4)
(4.6)
where ~R is the main chain of resin and R1 stands for an alkyl group in an alcohol such
as CH3, C2H5, 1-C3H7, 1-C4H9, for example.
(4.7)
Figure 4.7 shows the weight change of orthophthalic type unsaturated polyester resin
versus root of immersion time in methanol and water [3].
In methanol, after rapid absorption, the weight tends to decrease and after drying the
immersed specimen, the weight loss can be clearly observed. This type of corrosion is an
example of penetration type corrosion shown in Figure 4.14 in Section 3.
116
Corrosion Resistance of Polymer, Polymer Blends and Composites in Liquid Environments
Figure 4.7 Weight change of orthophthalic type unsaturated polyester resin in the wet
or after drying, immersed in methanol (: wet, : dry) and water (: wet, : dry )
The corrosion rate of plastics and its composites are greatly dependent on the absorption
of water or environmental liquid and also reactivity.
The most remarkable difference in corrosion behaviour between metals and plastics is
the degree of liquid absorption into the materials.
An environmental liquid can easily penetrate into the plastics because the average distance
of the polymer chain is larger than that of an environmental liquid.
Table 4.4 shows the water absorption of various plastics measured by ASTM D-570
method [11].
Water absorption behaviour is different from plastics, and this is due to, at first, the
polarity. When the polarity of plastics and environmental liquid coincides, both molecules
attract each other, then the amount of absorbed liquid increase as described in Section 1.
117
Handbook of Polymer Blends and Composites
Besides the polarity of plastics, the following factors also affect water absorption:
degree of crosslinking
degree of crystallinity
The penetration behaviour of the liquid is evaluated by the diffusion coefficient and the
saturated liquid contents in the resin. The diffusion coefficient of liquid decreases with
increase of crosslinking density as shown in Figure 4.8 [12].
Moreover, the amount of liquid penetration into the plastics decreases, in general, with
an increase of degree of crystallinity [13].
The diffusion coefficients obtained experimentally from the weight change are shown in
Table 4.5.
These values are dependent on the kind of resin, temperature and filler, etc.
118
Corrosion Resistance of Polymer, Polymer Blends and Composites in Liquid Environments
119
Handbook of Polymer Blends and Composites
Generally, chain symmetry and strong intermolecular secondary valence bonds will assure
high glass transition temperature (Tg) and aid good packing of polymer chains, thus they
contribute to the chemical resistance. In contrast, the presence of randomly spaced pendant
groups or flexibilising groups decrease chemical resistance [8].
Although there are many exceptions, one can usually predict the resistance of plastics to
corrosives from knowledge of the polymer structure.
Table 4.6 shows the various physical properties of metals, ceramics and plastics
[14, 15, 16].
One can see easily from this table, that plastics are light weight, and have relatively low
thermal conductivity and low strength in comparison with other materials. And one of
the disadvantages of plastics in service is the low heat resistance, that is the service
temperature limit is below 100 C in common thermoplastics. However, the plastics that
show excellent heat resistance have been developed, they are known as engineering plastics
and the limit is about 150~200 C. In engineering plastics, the heat distortion temperature
exceeds 200 C and some plastics are durable at 300 C or over, they are called super
engineering plastics.
Roughly speaking, plastics show high resistance to acidic environments and as such are
able to be substituted for metallic materials, and they have low resistance for organic
solvents, for which metals show enough durability.
It can be seen from Table 4.6 that fluorocarbon plastics (PTFE) show good chemical
resistance to many environments. However, the fluoroplastics have problems in strength
and mouldability, they are not suitable for making large-scale equipment, and they are
usually used as a lining. For large scale structures, the corrosion resistant glass fibre
reinforced plastic (GFRP) is suitable.
120
Corrosion Resistance of Polymer, Polymer Blends and Composites in Liquid Environments
Table 4.6 Corrosion resistance and physical properties of various materials
Materials Corrosion resistance (RT) Physical properties
H2SO4 HCl HNO3 NaOH B
75% 35% 30% 48% 3
10 kg/m 3 -6
10 K -1
MPa W/m.K
PE + + - + 0.91-0.95 100-120 5.6-28 0.33
PVC + + + + 1.4 60 44.8 0.12-0.17
PTFE + + + + 2.16 100 24.1 0.24
Natural rubber - + - + 1.0-1.5 3-15 0.12-0.29
Carbon steel - - - + 7.85 11.7 400-500 49.9
Stainless steel - - + + 7.98 16.2 >520 16.8
Hastelloy C + + + + 8.89 11.3 >690 11.6
Titanium - - + + 4.5 8.8 >270 15.1
Alumina 3.0-4.0 6-9 17-34
(Corrosion resistance: + good - poor) ( density, coefficient of thermal expansion, B tensile strength, thermal
conductivity) RT: room temperature; Stainless steel: American Iron and Steel Institute (AISI) type 316
121
Handbook of Polymer Blends and Composites
122
Corrosion Resistance of Polymer, Polymer Blends and Composites in Liquid Environments
Unsaturated polyester resin (UP) is most frequently used as a matrix for FRP because it
is transparent and has excellent mouldability and good mechanical properties. By
controlling the amounts of styrene monomer, accelerator and catalyst, the resin can be
hardened in wide range of temperatures and time.
Corrosion resistance of this resin depends mainly on the number of ester bonds in the
resin. This is due to the fact that the ester group is degradable easily by hydrolysis, and
then the lower the concentration of ester group the higher the corrosion resistance.
The chemical structures of typical orthophthalic type UP are shown in Figure 4.9, and
other types of UP with different structures as the shaded part of ortho-UP in Table 4.7,
are also shown in the table.
This type of resin is most popular for the GFRP matrix, however, it is not considered as
a corrosion resistant resin because this it is degradated easily by hydrolysis in alkaline
solutions at normal temperatures, and water and acidic solutions at high temperatures.
The hydrolysis resistance of isophthalic type UP is higher than that of orthophthalic type
UP. Thus the resin is widely used for general corrosion resistant equipments and tanks.
Bisphenol type UP has good heat and corrosion resistances to many chemical liquid
environments. This is because of low numbers of ester bonds and steric hindrance to
liquid penetration around the ester bonding.
123
Handbook of Polymer Blends and Composites
This resin has excellent corrosion resistance to oxidising acids such as nitric acid,
nitric acid/hydrofluoric acid mixture, etc. The resin is used for example, as GFRP
lining for acid pickling tank of chrome plating [17].
Vinyl ester resins are made by reacting a liquid epoxy resin with unsaturated organic
acid, e.g., methacrylic acid, to terminate their molecules with reactive ester groups.
In the previously mentioned UP resins, carbon double bonds (C=C) located at random
on a molecular chain, and ester groups also exist repeatedly in a main chain. In vinyl
ester resins, however, both ends of the molecule have a double carbon bonding, and the
ester group exists adjacent to them.
There are a few unreacted double bonds, and the ester bonding is not included in a main
chain but they contain only one bond at each end of the resin molecule.
Therefore, the number of ester bonds is low, and this resin shows excellent corrosion resistance.
This resin is manufactured from standard bisphenol A-based liquid epoxy resin and
is typified by a tensile elongation at break of 5-6%, combined with a heat deflection
temperature (HDT) of 100-120 C. From the corrosion point of view, it resists well,
many aggressive chemicals [18].
This resin uses a novolak epoxy resin as its starting agent. The resulting vinyl ester
resin has higher crosslinking density than the bisphenol-A type resins. This provides
high resistance to penetration of chemicals into the matrix, and it has a superior
resistance to liquid environments. The heat resistance is also higher, and is HDT of
140150 C, and it is a little more brittle than bisphenol type resin [19].
Epoxy resin (ER) contains more than two epoxy groups (oxilane groups) in each molecule
chain. There are many types according to the chemical structure and molecular weight.
The epoxy group is reacted with hardener or curing agent, and then three-dimensional
crosslinked network polymers are formed.
124
Corrosion Resistance of Polymer, Polymer Blends and Composites in Liquid Environments
The most widely used epoxy resins are made by reaction with either bisphenol-A or a
novolak phenolic resin with epichlorohydrin. Aliphatic amines, aromatic amines and
acid anhydride can be used as hardeners.
(1) The resin shows a wide range of properties from liquid to solid states.
(2) The resin performance changes greatly with combinations of the resin and hardener.
For example, the resin cured with amines shows poor resistance to the acid, and that
cured with acid anhydride is weak to the alkali.
(5) The resin shows low shrinkage during and after the curing process, compared to
UP resins.
This resin has a furan ring produced from furfuryl alcohol (2-hydroxymethyl-furan) and
furfural (2-furaldehyde).
The resin is relatively brittle compared with UP resin, however it has excellent strength
at high temperatures.
For the properties of corrosion resistance, it shows high durability to acids and alkalis,
and also organic solvents such as esters, ketones, and halogenide hydrocarbons where
UP resins show poor resistance.
Therefore, this resin is one of the best plastics for using in a liquid mixture of a severe
corrosive acid and an organic solvent. However, it is necessary to note that the resin is
extremely weak for oxidising chemicals.
The resin is one oldest thermosets known as Bakelite, and is synthesised by the reaction
of the phenol and formaldehyde.
125
Handbook of Polymer Blends and Composites
This type of resin has a problem in the moulding process, because they are powder
(novolak) or high viscosity liquid (resol), and the polymerisation process is condensation,
and so the resin needs high temperatures and pressures.
At the beginning of 1970, the resol type phenol resin with high volatile and low viscosity
was developed and then used as FRP has increased rapidly. This is called as the second-
generation phenolic resin.
One of the most superior characteristics of phenolic resin is inflammability and also it
does not generate poison gases if it burn. Therefore, its usage in building and for equipment
for public transportation is being expanded.
In general, the resin shows relatively good resistance to acids but poor performance in
alkalis. Table 4.8 shows physical and mechanical properties of corrosion resistant resins
[19, 20].
126
Corrosion Resistance of Polymer, Polymer Blends and Composites in Liquid Environments
The corrosion behaviour and mechanism of polymers are discussed mainly with reference
to typical matrix resins for GFRP.
Isophthalic type UP (iso-UP) and bisphenol-A type ER shown in Table 4.7 were used.
Iso-UP was crosslinked by styrene, and epoxy was cured with phthalic anhydride (methyl-
tetrahydrophthalic anhydride : MTHPA) or menthane diamine (p-menthane-1,8, diamine
: MDA). The structures of the hardeners are shown in Figure 4.10, and the structures of
crosslinked resins are shown in Figure 4.11.
As mentioned in Section 2.3 the polyester resin (iso-UP)(a) has many ester bonds
(-COO-) within the main chain that are the attack sites by acid and alkaline solutions.
Bisphenol-A type epoxy resin has essentially no ester bond within the main chain but is
crosslinked by ester bonds in MTHPA-ER and by C-N bonds in MDA-ER.
Figure 4.12 shows the typical results of weight changes as a function of immersion time
[6, 21]. Basically, at first, weight increases by penetration of liquid into the resin, and
then decreases by dissolution of the resin into the liquid and finally increases again by
penetration of liquid into corroded resin. By penetration of liquid into the resin and/or a
hydrolysis reaction between liquid and resin, the strength of resin decreases as shown in
Figure 4.13.
127
Handbook of Polymer Blends and Composites
Figure 4.11 Chemical structures of cured UP and ER. (a) iso-UP, (b) MTHPA-ER,
(c) MDA-ER
128
Corrosion Resistance of Polymer, Polymer Blends and Composites in Liquid Environments
129
Handbook of Polymer Blends and Composites
By infrared spectroscopic analysis (IR), it was confirmed that corrosion occurred with
the hydrolysis reaction of ester bonding shown in Tables 4.2 and 4.3.
This form is observed when the corrosion product can be dissolved into the liquid at
the liquid contacted surface and the corrosion develops uniformly over entire surface.
This type of corrosion occurs when low molecular weight components are produced
by a corrosion reaction at both the main chain and the cross-linking chain. Corrosion
products dissolve into the solution directly from the surface.
1
x=
2
(h0 h) (4.8)
where h0 is the thickness of the original specimen, and h is that of the immersed one.
130
Corrosion Resistance of Polymer, Polymer Blends and Composites in Liquid Environments
Figure 4.15 shows the relationship between corrosion thickness and immersion time t. A
linear relationship is obtained:
x = k1t (4.9)
where k1 is a constant.
Part of the resin dissolves out and the residual part is left as a corrosion layer where
the initial chemical bonds are maintained.
Figure 4.16 shows examples of cross sections of the specimens corroded by this type.
The specimens were immersed in sodium and potassium hydroxide solutions [21].
131
Handbook of Polymer Blends and Composites
Figure 4.16 Cross section of iso-UP specimens immersed in 50 wt% NaOH solution,
500 h(a), and 60 wt% KOH solution, 71 h(b), at 80 C
(Reproduced with permission from H. Hojo, K. Tsuda and K. Ogasawara, Advanced
Composite Materials, 1991, 1, 55, Figure 6. Japan Society for Composite
Materials, Japan, 1991)
The boundary between the outside discoloured portion and the central unchanged
portion is clearly observed. It was found by infrared analysis that the corrosion was
observed only in the discoloured portion.
132
Corrosion Resistance of Polymer, Polymer Blends and Composites in Liquid Environments
This crosslinking molecular chain consists of highly stable C-C bonds, so those
molecular chains are preserved even after hydrolysing the ester bonds, and residual
corrosion products form a corroded layer.
Equation (4.10) means that the corroded layer makes a resistant layer to the
penetration of solution by diffusion and the reaction occurs at the interface between
the corroded layer and the non-corroded resin. This behaviour is the same as Wagners
parabolic law for high temperature oxidation of metals [23].
In contrast, if the same resins are immersed in potassium hydroxide solution, highly
swelled porous corroded layer are formed shown as Figure 4.16(b).
For potassium hydroxide the swelled and corroded layer does not behave as a barrier
for the diffusion of solution. Then the relationship between the corrosion thickness
and immersion time follows the rectilinear law the same as in Equations 10.410.9.
This agrees with the behaviour of metals which produces a rough oxide film in high
temperature oxidation.
This type of corrosion is observed in epoxy resin cured with amine such as MDA-ER
immersed in sulphuric acid [22]. The corrosion behaviour is characterised by a two-step
process in diffusion and reaction. The solution penetrates into the resin and reaches an
equilibrium state, and then the strength decreases owing to the initiation of a corrosive
133
Handbook of Polymer Blends and Composites
reaction as shown in Figure 4.18. In this type, the amount of saturated liquid absorption
is important as shown in Figure 4.19. And the penetration depth can be measured by
tracing sulphur atoms in sulphuric acid with an X-ray microanalyser. The depth y was
proportional to root of (t-to), where to is incubation time.
Figure 4.19 Relationship between retention of flexural strength and weight change for
MDA-ER in H2SO4 solution
( 50 C 10 wt%, 65 C 10 wt%, 80 C 10 wt%, 80 C 20 wt%)
134
Corrosion Resistance of Polymer, Polymer Blends and Composites in Liquid Environments
The same type of corrosion has been observed for orthophthalic type UP immersed
in boiling water. This type of corrosion is characteristic in plastics.
The forms of corrosion as shown in Figure 4.14 are determined by relative values of
reaction rate (R) and diffusion rate (D) as shown in Table 4.9.
When D<<R, reactant liquid which reaches virgin resin by diffusion is dissipated
immediately by reaction, then the surface reaction type or corrosion layer forming
type occur. The difference between both forms is caused by solubility of corrosion product
to liquid. If the product has high solubility, the surface reaction type is observed. On
the other hand, in case of low solubility, the corrosion layer forming type is observed.
If D>>R, reactant liquid penetrates into the resin with a large diffusion rate and the
corrosion reaction proceeds slowly. Therefore the reacted region is located in a part of
the liquid penetrated region. This behaviour is that of penetration type.
The number of repeated units (n) is also a factor that determines which of the three
forms is observed. In epoxy resins cured with phthalic anhydride, if n = 0 the surface
135
Handbook of Polymer Blends and Composites
reaction type was observed, while if n>1, the corrosion layer forming type was revealed
[25]. Because of the severe entanglement of the long polymer chain, the corrosion products
were difficult to dissolve.
The incompatible polymer blend which consists of a continuous phase and a dispersed
one can be expected to show unique corrosion behaviour. Because each phase shows
different durability to corrosion by environmental liquid [26, 27, 28].
The materials discussed are PA, PC, PAR, modified-polyphenylene ether (PPE), PBT and
polymer blends with these components; i.e. PA/PAR, PA/PPE, PBT/PC and PA/PC. The
durability of these components were confirmed experimentally and are listed in Table 4.10.
These systems all consist of continuous phase/dispersed phase.
The corrosion behaviour of polymer blends are roughly grouped into accelerated types
and improved ones [29].
The blends with easily degradable dispersed phase and high corrosion resistant
continuous phase are classified into this type of corrosion; these are PA/PAR (50/50),
PBT/PC and PA/PC (50/50) in alkaline solution. In these blends, the dispersed phase
is degraded first and then the continuous phase adjacent to the dispersed phase is
also degraded, though the continuous phase shows high resistance to corrosion when
136
Corrosion Resistance of Polymer, Polymer Blends and Composites in Liquid Environments
(a) (b)
they exist as a neat resin. Then many holes with larger diameter than that of the
original dispersed phase are generated by corrosion as shown in Figure 4.20 [26].
The other type of corrosion observed with incompatible polymer blends was the case
in which corrosion was reduced rather than accelerated. In this type, the interphase
showed a small effect on the corrosion rate.
PA/PAR and PA/PPE blends in acidic solution show this type of corrosion. The
degradable continuous phase includes corrosion resistant dispersed phase, and the
137
Handbook of Polymer Blends and Composites
Figure 4.21 Change of corrosion thickness with immersion time in 20% NaOH
solution at 50 C, ( PC, PC/PA(75/25) blend polymer).
The blends of corrosion resistant continuous phase and degradable dispersed phase
with low content were also grouped into this improved-type. In this case, the role of
interphase layer in blends on corrosion was negligible, then the corrosion rate was
dominated by the continuous phase. For a blend of PA/PAR (70/30) in alkaline solution,
138
Corrosion Resistance of Polymer, Polymer Blends and Composites in Liquid Environments
few weight changes were observed even after long-term immersion. This tendency was
observed when the dispersed phase, including the interphase layer, exists in isolation in
the continuous phase. So that, when dispersed phase content increases, the isolation
effect became unavailable, then the corrosion rate increased dramatically. This
phenomenon was observed with PA/PAR (50/50) in alkaline solution [29].
In general, the surface of the glass fibre is treated with coupling agent to increase the strength
of interface. Common couplers are organosilanes, and there are many different types are
available. The silane portion of the molecule is oriented towards the glass and the organic
portion is oriented towards the resin of the composite [30]. The interface is degradable by
hydrolysis [31], and liquid penetrates easily into the material through the interface by capillary
action. Figure 4.22 shows the effect of fibre orientation on corrosion rate [32].
139
Handbook of Polymer Blends and Composites
The unidirectional GFRP with matrix of bis-A type ER cured with MTHPA specimens
were immersed in sodium hydroxide solution. This system was the corrosion layer forming
type, and the thickness of the corrosion layer was measured in the direction of fibre axis
and the perpendicular direction. Corrosion thickness grows more quickly in the direction
of fibre axis than the perpendicular direction. The reaction between coupler and liquid is
also a factor to cause debonding of matrix from fibre and blister formation [33, 34].
b. Effect of acid
Acids leach out nonsiliceous ions preferentially, leaving highly porous silica glass.
Si-O M + H+ Si-OH + M+
c. Effect of alkali
Alkalis attack the silica network and concentrated alkali solutions dissolve the
glass itself.
Si-O-Si + OH Si-O + Si-OH
Adding Zr2O improves the durability to alkali by forming the insoluble film of
hydrated Zr2O.
Several types of glass fibre for composites are produced as shown in Table 4.12, and
most common is E glass [36]. This glass contains few alkali metal elements, and has
excellent water resistance. Thus it is used for electrical fields. C glass contains alkali
metal elements, and is widely used for acid resistant GFRP but has poor water resistance.
S glass is widely used in the aerospace industry because of its excellent strength. It also
has excellent corrosion resistance but is very expensive compared to E glass, thus it is
not used as an anti-corrosion matrix [37]. Alkali resistant (AR) glass shows excellent
resistance to alkali, it is used for GRC (glass fibre reinforced cement).
140
Corrosion Resistance of Polymer, Polymer Blends and Composites in Liquid Environments
Figure 4.23 shows the degradation behaviour of E and C glass fibre mat in 2 mol/l
nitric acid at 80 C [38]. C glass shows good resistance, but in E glass the weight
decreases severely because the component of E glass is dissolved into strong acid
[39]. The corrosion process is the removal of not only boron oxide but also calcium,
aluminium, and magnesium ions from the glass. By corrosion of fibre their strength
also decreases in strong acid with a spiral crack [40].
Figure 4.23 Weight change of E- and C-glass mat immersed in 2 mol/l H2SO4, 80 C.
( E-glass mat, C- glass mat).
141
Handbook of Polymer Blends and Composites
Examples of corrosion behaviour of GFRP are shown in Figure 4.24. There are
three stages of degradation, the first stage is due to the degradation of matrix, the
second one is due to that of the interface between matrix and glass fibre, and the
third one is due to fibre itself; the second stage degredation is especially predominant
[41]. The second and third stages begin earlier with increase of fibre content [42].
In general, corrosion behaviour of GFRP depends on the type of matrix resin and
glass fibre [43].
The durability of plastics to liquid corrosive environments have been discussed extensively
in the literature [44, 45, 46, 47, 48, 49, 51, 52, 53] and the technical sheets of manufacturers.
Tables 4.13 and 4.14 show corrosion resistance data for common plastics [47] and
engineering plastics [53]. Table 4.15 shows them for corrosion resistant resins [54,
55, 56].
Table 4.15 shows only a rough tendency of corrosion resistance, because corrosion
behaviour is dependent of temperature, concentration of chemicals and contact time
Figures 4.25 and 4.26 show the example of change of flexural strength with immersion
time [45, 52, 57].
142
Corrosion Resistance of Polymer, Polymer Blends and Composites in Liquid Environments
Table 4.16 gives more detailed data which include the effect of temperature and
concentration. Miasa and others showed these data with a plot of temperature versus
concentration as shown in Figure 4.27 [58].
143
144
Figure 4.25 Change of flexural strength of novolak type vinyl ester resin in 75 wt%
H2SO4 () and 35 wt% HCl solution ( ) at 40 C
145
Handbook of Polymer Blends and Composites
Figure 4.26 Change of flexural strength of ER and UP resins in 25 wt% acetic acid
solution at 99 C
146
Corrosion Resistance of Polymer, Polymer Blends and Composites in Liquid Environments
Figure 4.27 Temperature vs. concentration diagrams for three types of resin in HCl.
(Region A: satisfactory, B: questionable, C: unsatisfactory)
147
Handbook of Polymer Blends and Composites
The corrosion resistance of GFRP depends essentially on the corrosion resistance property
of resin used as a matrix. GFRP in contact with a corrosive chemical environment is
protected by the performance and also thickness of resin as a corrosion barrier.
The corrosion resistant GFRP structure of equipment such as tanks, towers, vessels and
pipes, is usually composed of a surface gel coat or resin-rich inner layer and an intermediate
resin-rich layer to prevent corrosion and also a high strength outer layer to preserve the
structure. A surface resin-rich inner layer thicker than 0.25 mm is made with resin more
than 85 wt% and acid resistant C-glass or synthetic surfacing mat (veil). The intermediate
layer is made of at least two layers of 450 g/m2 E-glass chopped strand mat impregnated
with 68~78wt% resin. The outer structural layer is composed of alternating laminates of
E-glass mat and woven roving layers impregnated with less than 70 wt% resin [59]. The
purpose of a surfacing veil or mat at inner layer is to prevent resin cracking during
moulding, impacting and temperature variation.
Generally the corrosion rate defined by an increasing rate of corrosion thickness, can
be expressed by:
dx
= k x C (4.11)
dt
148
Corrosion Resistance of Polymer, Polymer Blends and Composites in Liquid Environments
Figure 4.29 Example of resin lining and glass flake lining for steel equipment
149
Handbook of Polymer Blends and Composites
k = k0 exp( E / RT ) (4.12)
In the case of surface reaction type, the corrosion rate is independent of x, that is = 0
in Equation (4.11) , thus the Equation (4.9) can be obtained. In the case of corrosion
layer forming type, if the corrosion layer shows the resistance to penetration of liquid
and it increases with the thickness of the layer, the value of = -1 and the Equation
(4.10) can be derived. If the resistance of the layer to penetration of liquid can be neglected,
the same type of equation as Equation (4.9) holds. This behaviour was recognised for
iso-UP immersed in potassium hydroxide solution shown in Figure 4.16(b).
From the experiments under various concentrations and temperatures, the second,
following equations are obtained:
x = A4C t (4.15)
From these equations, the thickness of corrosion layer or dissolved part after immersion
at arbitrary concentration and temperature can be predicted. The equations of corrosion
rate are summarised in Table 4.17 [21, 60].
If the Equations (4.134.15) are obtained, the life based on flexural strength can be
estimated as follows [6, 61].
150
Corrosion Resistance of Polymer, Polymer Blends and Composites in Liquid Environments
3Pl
sB = 2 (4.16)
2bh0
where h0, b and l are thickness, width and span of the specimen before immersion, and P
is breaking load in flexural test after immersion respectively.
The true flexural strength sB is equal to that of the original resin sB0 thus:
3Pl
sB = sB0 =
2(b 2x)(h0 2x) (4.17)
2
In case of corrosion layer forming type, the strength of the layer was neglected. Therefore,
Equations (4.16) and (4.17) give the retention of flexural strength sB/sB0:
sB (b 2x)(h0 2x)
2
S= = 2 (4.18)
sB0 bh0
151
Handbook of Polymer Blends and Composites
Figure 4.30 Master curve of strength change of MTHPA-ER immersed in NaOH solution.
( 50 C, 10 wt%, 65 C, 10 wt%, 80 C, 10 wt%, 80 C, 20 wt%,
80 C, 30 wt%, 80 C, 40 wt%)
As the retention of strength sB/sB0: varies only with x, a master curve can be obtained by
plotting S with the right term of Equations (4.134.15). The example for MTHPA-ER
immersed in sodium hydroxide solution are shown in Figure 4.30.
By using this master curve, the retention of strength after long term immersion at any
concentration and temperature can be predicted.
For of resin in which the strength decreases abruptly after liquid absorption as shown in Figure
4.19, the saturated weight gain gives the useful information for the life prediction of resin.
The corrosion resistance of GFRP depends essentially on the performance of resin matrix,
however, there are many factors which will affect the corrosion behaviour and form of
the material. These factors are:
a) Reactive functional group and bond as shown in Table 4.2 in Section 1 and their
concentration [7].
152
Corrosion Resistance of Polymer, Polymer Blends and Composites in Liquid Environments
c) Number of repeated units, n, in Figure 4.9 in Section 3, molecular chain length [25].
d) Carboxylic acid
There exists carboxylic acid in anhydride cures epoxy and unsaturated polyester
resins. Corrosion rate normalised with ester concentration (dx/dt)(1/CE2) is
proportional to a concentration of carboxylic acid CCA2 as shown in Figure 4.31
[63]. CCA gives a degree of acidity of the resin.
Figure 4.31 Relationship between corrosion rate normalised with ester concentration
and concentration of carboxylic acid in bis-A ER with short main chain (), and long
main chain ( ). (80 C, KOH solution)
153
Handbook of Polymer Blends and Composites
e) Steric hindrance
Aromatic ring and bulky side group will prevent the corrosion of resin by covering
the molecular chain.
f) Degree of cure [63, 64] and crosslinking density [65] (Figure 4.8 in Section 4.2)
g) Residual substance
monomer, curing agent, crosslinking agent, reaction initiator of hardener [66].
When large temperature differences exists between inner liquid and outer atmosphere,
vapour diffusion inside the material are accelerated, this is water vapour diffusion [69].
154
Corrosion Resistance of Polymer, Polymer Blends and Composites in Liquid Environments
b) Concentration of solution
Figure 4.33 Effect of concentration on corrosion rate of bis-A ER cured with acid
anhydride. (Accelerator: none, amine, stannous octoate)
c) pH [7, 70]
d) Flow velocity
155
Handbook of Polymer Blends and Composites
Figure 4.34 Effect of flow velocity on amount of dissolved carbon from vinyl ester
resin immersed in 65 C, 25wt% NaOH solution.
( immersion (Re=0), Re=9.8 x 103)
e) Stress
The results of both the immersion test and the creep-rupture test are compared in
Figure 4.36. The retention of tensile strength versus immersion time in immersion
test and the ratio of applied stress to initial tensile strength versus rupture time in
creep-rupture test performed in sodium hydroxide solution, are presented [41,
75]. The behaviour is essentially the same independent of test methods, but the
transition time from the first stage (I) to the second one (II) is extremely affected
by stress condition under immersion testing. Therefore, stress accelerates the
corrosion of GFRP significantly. This stress corrosion phenomenon means that
conventional immersion test without loading are not always sufficient to evaluate
the corrosion resistance of the composite materials.
156
Corrosion Resistance of Polymer, Polymer Blends and Composites in Liquid Environments
Figure 4.35 Effect of creep applied stress () on retention of tensile strength for vinyl
ester composites (glass content:40.2 vol%) immersed in 65 C, 10 wt% NaOH solution
( =49 MPa, =98 MPa, =147 MPa)
Figure 4.36 Correlation between immersion test and creep-rupture test for vinyl ester
composites (glass content:40.2 vol%) in 50 C, 10 wt% NaOH solution.
( immersion test, creep-rupture test)
157
Handbook of Polymer Blends and Composites
Acknowledgement
The authors would like to acknowledge the contribution of Dr. Tetsuya Sakai for his
assistance in preparing the manuscript.
References
8. R. B. Seymour, Plastics vs. Corrosives, John Wiley & Sons, New York, 1982, 36.
9. M. Modesti and F. Simioni, Polymer Engineering and Science, 1996, 36, 17, 2173.
10. Koatsu Gas Hoan-Kyokai, Handbook of Corrosion Resistant FRP, Tokyo, Japan,
1983, 35.
158
Corrosion Resistance of Polymer, Polymer Blends and Composites in Liquid Environments
14. Report for Evaluation and Design of FRP Aimed to Apply to Equipment for
Preservation of Environment, Engineering Promotion Association, Tokyo, Japan,
Report, ENAA 1993-1, 1994, 9.
16. R. B. Seymour, Plastics versus Corrosives, Wiley Interscience, New York, USA,
1982, 131.
19. Practical Encyclopedia of Plastics, Ed., F. Ide, Sangyo Chosakai, Tokyo, 1993, 229.
20. Introduction to FRP, Ed., M. Miyari, Japan Reinforced Plastics Society, Tokyo,
Japan, 1992, 52.
23. The Corrosion Handbook, Ed., H. H. Uhlig, John Wiley & Sons Inc., New York,
1969, 16.
25. H. Hojo, K. Tsuda and K. Ogasawara, Reinforced Plastics (Japan), 1988, 34, 49.
159
Handbook of Polymer Blends and Composites
31. M. Basche in Interfaces in Composites, ASTM Stp. 452, American Society for
Testing and Materials, 1968, 145.
32. T. Tomiyama, Corrosion Behaviour of FRP Bolts using Notched Bar, Graduation
Thesis, Department of Chemical Engineering, Tokyo Institute of Technology, 1997.
34. Composite System in Materials, Ed., H. Hojo, Materials Science Society of Japan,
Shokabo, Tokyo, Japan, 1993, 195.
36. Introduction to FRP, Ed., M. Miyari, Japan Reinforced Plastics Society, Tokyo,
Japan, 1987, 56.
40. B. Das, B. D. Tucker and J. C. Watson, Journal of Materials Science, 1991, 26,
24, 6606.
41. H. Hojo and K. Tsuda, Presented at the 34th Annual Technical Conference,
Reinforced Plastics/Composites Institute and SPI, New Orleans, LA, USA, 1979,
Section 13-B.
160
Corrosion Resistance of Polymer, Polymer Blends and Composites in Liquid Environments
43. D. Santrach and R. Matzeg, Presented at the 46th SPI Composite Institutes
Annual Conference, 1991, Washington, DC, USA, 1991, 1-A/1.
45. Koatus Gas Hoan-Kyokai, Handbook for Corrosion Resistant FRP, Tokyo,
Japan, 1983.
47. R. B. Seymour, Plastics vs. Corrosives, John Wiley Interscience, New York,
USA, 1982.
50. The Possibility of Application of FRP for Main Process in Chemical Plants,
Engineering Promotion Association, Tokyo, Japan, Report No. ENAA 1991-1, 1992.
52. Guideline on FRP Tank for Tank Truck and Tank Container to Transport
Poisonous of Toxic Liquid, Japan Reinforced Plastics Society, Tokyo, Japan, 1993.
53. Technical Data Sheet, Chemical Resistance, Ensinger GmbH & Co., Nufringen,
Germany.
54. Koatsu Gas Hoan Kyokai, Handbook for Corrosion Resistant FRP, Tokyo,
Japan, 1983, 56.
55. The Possibility of Application of FRP for Main Process in Chemical Plants,
Engineering Promotion Association, Tokyo, Japan, Report No. ENAA 1990-7,
1991, 7.
161
Handbook of Polymer Blends and Composites
56. Vinyl Ester Resin, Ed., Research group of vinyl ester resin, Kagakukogyo-Nippo-
Sha, Tokyo, Japan, 1993, 317.
57. Vinyl Ester Resin, Ed., Research group of vinyl ester resin, Kagakukogyo-Nippo-
Sha, Tokyo, Japan, 1993, 258.
58. The Possibility of Application of FRP for Main Process in Chemical Plants,
Engineering Promotion Association, Tokyo, Japan, Report No. ENAA 1991-1,
1992, 59.
60. H. Hojo, K. Tsuda, M. Kubouchi and D.S. Kim, Metals and Materials, 1998, 4,
1191.
61. H. Hojo, K Tsuda and H. Ishizuka, Journal of the Japanese Society for
Composite Materials, 1991, 17, 6, 243.
63. H. Hojo, K. Ogasawara and H. Ishizuka, Reinforced Plastics (Japan), 1988, 34,
2, 49.
65. T. Akaoka and K. Watanabe, Reinforced Plastics (Japan), 1970, 16, 2, 70.
66. T. Sakai, S. Ono, M. Kubouchi, K. Tsuda and H. Hojo, Journal of the Materials
Science Society of Japan, 2000, 37, 5, 246.
69. S. Okuda and T. Iguchi, Advances in Organic Coating Science and Technology,
1979, 4, 257.
70. L. Prian and A. Barkatt, Journal of Materials Science, 1999, 34, 16, 3977.
162
Corrosion Resistance of Polymer, Polymer Blends and Composites in Liquid Environments
72. H. Hojo, K. Tsuda and D. S. Kim, Corrosion Engineering, 1989, 38, 12, 689.
73. D. S. Kim, K. Tsuda and H. Hojo, Journal of the Materials Science Society of
Japan, 1998, 35, 1, 32.
163
5
New Approaches to Reduce Plastic
Combustibility
G.E. Zaikov, S.M. Lomakin, S.V. Usachev, E.V. Koverzanova, N.G.
Shilkina and L.V. Ruban
5.1 Introduction
Our environment has mostly a polymeric nature and all polymers burn whether natural
or synthetic. The use of polymer flame retardants has an important role in saving lives.
There are four main families of flame-retardant chemicals:
4) Nitrogen-based organic flame retardants are used for a limited number of polymers.
Depending on their nature, flame retardants can act chemically and/or physically in the
solid, liquid or gas phase. They interfere with combustion during a particular stage of
this process, e.g., during heating, decomposition, ignition or flame spread.
Substitution of one type of flame retardant with another consequently means a change in
the mechanisms of flame retardancy. Halogen containing flame retardants act primarily
by a chemical interfering with the radical chain mechanism taking place in the gas phase
during combustion. High-energy OH and H radicals formed during combustion are
removed by bromine released from the flame retardant. Although brominated flame
165
Handbook of Polymer Blends and Composites
Phosphorus compounds mainly influence the reactions taking place in the solid phase.
By thermal decomposition the flame retardants are converted to phosphoric acid which
in the condensed phase extract water from the pyrolysing substrate, causing it to char.
However, some phosphorus compounds may, like halogens, act in the gas phase as well,
by a radical trap mechanism.
The interest in flame retarding polymers goes back to the nineteenth century with the
discovery of highly flammable cellulose nitrate and celluloid. In more recent times the
conventional large volume of plastics such as phenolics, rigid polyvinyl chloride (PVC)
and melamine resins possess adequate flame retardancy. By the 1970s the major flame
retardant polymers were the thermosets, namely, unsaturated polyesters and epoxy resins
which utilised reactive halogen compounds and alumina hydrate as an additive. There
was also a large market for phosphate esters in plasticised PVC, cellulose acetate film,
unsatureted polyesters and modified polyphenylene oxide. Alumina trihydrate (ATH)
was the largest volume flame retardant into unsaturated plastics.
166
New Approaches to Reduce Plastic Combustibility
In 1985-1986 a German study detected brominated dioxins and furans from pyrolysis of
a brominated diphenyl oxide in the laboratory at 510-630 C [1]. The relevance of these
pyrolysis studies to the real hazards presented by these flame retardants under actual use
conditions has been questioned. Germany and Holland have considered a ban or curtailed
use of brominated diphenyl oxide flame retardants because of the potential formation of
highly toxic and potentially carcinogenic brorninated furans and dioxins during
combustion. The issue spread to other parts of Europe where regulations were proposed
to restrict their use [2, 3].
167
Handbook of Polymer Blends and Composites
tricyclic aromatic compounds with similar physical and chemical properties, and both
classes are similar structurally. Certain of the PCBs (the so-called coplanar or mono-
ortho coplanar congeners) are also structurally and conformationally similar. The most
widely studied of these compounds is 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD).
This compound, often called simply dioxin, represents the reference compound for
this class of compounds. The structure of TCDD and several related compounds is
shown in Figure 5.1.
2,3,7,8-Tetrachlorodibenzo-p-dioxin 2,3,4,7,8-Pentachlorodibenzofuran
1,2,3,7,8-Pentachlorodibenzo-p-dioxin 3,3,4,4,5,5-Hexachlorobiphenyl
2,3,7,8-Tetrachlorodibenzofuran 3,3,4,4,5,-Pentachlorobiphenyl
These compounds are assigned individual toxicity equivalence factor (TEF) values as
defined by international convention (US Environmental Protection Agency (EPA), 1989).
Results of in vitro and in vivo laboratory studies contribute to the assignment of a relative
toxicity value. TEF are estimates of the toxicity of dioxin-like compounds relative to the
toxicity of TCDD, which is assigned a TEF of 1.0. All chlorinated dibenzodioxins (CDD)
and chlorinated dibenzofurans (CDF) with chlorines substituted in the 2,3,7, and 8
168
New Approaches to Reduce Plastic Combustibility
positions are assigned TEF values [1]. Additionally, the analogous brominated dioxins
and furans (brominated dibenzodioxins (BDD) and brominated dibenzofurans (BDF))
and certain polychlorinated biphenyls have recently been identified as having dioxin-like
toxicity and thus are also included in the definition of dioxin-like compounds. Generally
accepted TEF values for chlorinated dibenzodioxins and dibenzofurans are shown in
Table 5.1 [4].
169
Handbook of Polymer Blends and Composites
There are 75 individual compounds comprising the CDD, depending on the positioning of
the chlorine(s), and 135 different CDF. These are called individual congeners. Likewise,
there are 75 different positional congeners of BDD and 135 different congeners of BDF.
Only 7 of the 75 congeners of CDD or of BDD are thought to have dioxin-like toxicity;
these are ones with chlorine/bromine substitutions in, at least, the 2, 3, 7, and 8 positions.
Only 10 of the 135 possible congeners of CDF or of BDF are thought to have dioxin-like
toxicity; these also are ones with substitutions in the 2, 3, 7, and 8 positions. While this
suggests 34 individual CDD, CDF, BDD, or BDF with dioxin-like toxicity, inclusion of the
mixed chloro/bromo congeners substantially increases the number of possible congeners
with dioxin-like activity. There are 209 PCB congeners. Only 13 of the 209 congeners are
thought to have dioxin-like toxicity; these are PCB with 4 or more chlorines with just 1 or
no substitution in the ortho position. These compounds are sometimes referred to as
coplanar, meaning that they can assume a flat configuration with rings in the same plane.
In general, these compounds have very low water solubility, high octanol-water partition
coefficients, and low vapour pressure and tend to bioaccumulate.
Although these compounds are released from a variety of sources, the congener profiles
of CDD and CDF found in sediments have been linked to combustion sources [1].
The Hazards Substance Ordinance in Germany specifies the maximum level of chlorinated
dibenzodioxins and furans that can be present in materials marketed in Germany. This
has been extended to the brominated compounds. The two largest volume flame retardants
decabromodiphenyl oxide and tetrabromo bis-phenol A are said to meet these
requirements [2, 3].
170
New Approaches to Reduce Plastic Combustibility
The IPCS of the WHO has made several recommendations. Polybrominated diphenyls
production (France) and use should be limited because of the concern over high persistency,
bioaccumulation and potential adverse effects at low levels. There is limited toxicity data
on deca- and octabromodiphenyls. Commmercial use should cease unless safety is
demonstrated. For the polybrominated diphenyl oxides, a Task Group felt that
polybrominated dibenzo furans, and to a lesser extent the dioxins, may be formed. For
decabromodiphenyl oxide, appropriate industrial hygiene measures need to be taken and
environmental exposure minimised by effluent and emission control. Controlled incineration
procedures should be instituted. For octabromodiphenyl oxide, the hexa- and lower isomers
should be minimised. There is considerable concern over persistence in the environment
and the accumulation in organisms, especially, for pentabromodiphenyl oxide.
There are no regulations proposed or in effect anywhere around the world banning the
use of brominated flame retardants. The proposed EU Directive on the brominated
diphenyl oxides is withdrawn. Deca- and tetrabromo bis-phenol A as well as other
brominated flame retardants meet the requirements of the German Ordinance regulating
dioxin and furan content of products sold in Germany [6].
Polymer producers have been seeking non-halogen flame retardants and the search has
been successful in several polymer systems. Non-halogen flame retardant polycarbonate/
ABS blends are now commercial. They contain triphenyl phosphate or resorcinol
diphosphate (RD)P as the flame retardant. Modified polyphenylene oxide (General
Electric, Noryl) has used phosphate esters as the flame retardant for the past 15-20 years
and the industry recently switched from the alkylated triphenyl phosphate to RDP. Red
phosphorus is used with glass-reinforced Nylon 66 in Europe and melamine cyanurate is
used in unfilled Nylon. Magnesium hydroxide is being used commercially in polyethylene
wire and cable. The non-halogen solutions present other problems such as poor properties
(plasticisers lower heat distortion temperature), difficult processing (high loadings of
171
Handbook of Polymer Blends and Composites
ATH and magnesium hydroxide), corrosion (red phosphorus) and handling problems
(red phosphorus).
In this chapter the basic trends in the flame retardant hierarchy are discussed.
The main flame retardant systems for polymers currently in use are based on halogenated,
phosphorous, nitrogen, and inorganic compounds (Figure 5.2). All of these flame retardant
172
New Approaches to Reduce Plastic Combustibility
system basically inhibit or even suppress the combustion process by chemical or physical
action in the gas or condensed phase. To be effective, the flame retardants must decompose
near the decomposition temperature of the polymer in order for the appropriate chemistry
to occur as the polymer decomposes, yet be stable at processing temperatures.
Conventional flame retardants, such as halogenated, phosphorus or metallic additives
have a number of negative attributes. The ecological aspects of their application demands
the search of new polymer flame retardant systems. Among the new trends of flame
retardancy can be seen the use of intumescent systems, polymer nanocomposites,
preceramic additives, low-melting glasses, different types of char-formers and polymer
morphology modification [1]. However, it should be assumed tha their are close
interactions between the different flame retardant types in order to achieve a synergistic
behaviour. The block scheme of polymer flame retardant systems is given in Figure 5.2.
The brief explanations of the three major types of flame retardant systems (intumescent
systems, polymer nanocomposites and polymer organic char former), as well as intercalated
flame retardant system based on triphenylphosphine are presented in section 5.?.
The intumescent behaviour resulting from a combination of charring and foaming of the
surface of the burning polymers is being widely developed for fire retardancy because it
is characterised by low environmental impact.
173
Handbook of Polymer Blends and Composites
A condensed phase fire retardant mechanism is proposed for APP in Nylon 6 [18]. In
fact, an intumescent layer is formed on the surface of burning Nylon 6/APP
formulations which tends to increase the content of APP.
Thermal analysis has shown that APP destabilises Nylon 6, since the thermal
decomposition is observed at a temperature 70 C lower than that the pure Nylon 6
[18]. However, the intumescent layer effectively protects the underlying polymer from
the heat flux and therefore in the configuration of the linear pyrolysis experiments
the formulation, Nylon 6/APP (40%) decomposes more slowly than the pure polymer
[18]. These experiments prove the fire retardant action of the intumescent char. The
mechanistic studies of the thermal decomposition in the system Nylon 6 - APP show
that APP catalyses the degradation of the polymer and interacts with it forming
essentially 5-amidopenthyl polyphosphate:
174
New Approaches to Reduce Plastic Combustibility
175
Handbook of Polymer Blends and Composites
Table 5.2 Oxygen indices for high molecular weight Nylon 6 with added
melamine or its salts (LOI for pure Nylon 6 is 24) [20]
Compound Additive concentration, % wt.
3 5 10 15 20 30
Melamine - 29 31 33 38 39
Dimelamine phosphate - 23 24 25 26 30
Melamine - 24 25 25 30 32
pyrophosphate
Melamine oxalate - 28 29 - 33 -
Melamine cyanurate - 35 37 39 40 40
Melamine phthalate 34 48 53 - - -
176
New Approaches to Reduce Plastic Combustibility
To understand better the chemical reactions that are responsible for the intumescent
behaviour of APP-pentaerythritol (PER) mixtures, as model examples, a study of the thermal
degradation of pentaerythritol diphosphate (PEDP) was undertaken [24]. PEDP is a model
compound for structures identified in ammonium polyphosphate-pentaerythritol mixtures
heated below 250 C. Using thermogravimetric analysis (TGA) has identified five major
degradation steps between room temperature and 950 C and volatile products are evolved
in each step. The formation of the foam reaches a maximum at 325 C, corresponding to
the second step of degradation; foam formation decreases at higher temperatures. There
are no differences in the TGA or differential scanting calorimetry (DSC) curves in nitrogen
or air up to 500 C. Above this temperature, thermal oxidation leads to almost complete
volatilisation in a single step, which is essentially completed at 750 C. The elucidation of
the chemical reactions which occur upon degradation is easier if each step is studied
separately. The separation of the steps is accomplished by heating to a temperature at
which one step goes to completion, and the following reaction occurs at a negligible rate
177
Handbook of Polymer Blends and Composites
[24]. The chemical reactions which occur in the first two steps lead to the initial formation
of a char-like structure which will undergo subsequent graphitisation.
The first reaction is the elimination of water with the condensation of OH groups. This
overlaps with the elimination of organics when as little as 28% of the possible water has
been evolved. This involves essentially complete scission of the phosphate ester bonds
and results in a mixture of polyphosphates and a carbonaceous char. Three mechanisms
have been proposed in the literature for this reaction [25, 26]: a free-radical mechanism,
a carbonium ion mechanism, and a cyclic cis-elimination mechanism. The free-radical
mechanism was eliminated due to the lack of an effect of free-radical inhibitors on the
rate of pyrolysis [26]. The carbonium ion mechanism is supported by acid catalysis and
kinetic behavior and may compete with the elimination mechanism [25, 26]. The
carbonium ion mechanism should occur exclusively if there is no hydrogen atom on the
-carbon atom, as in PEDP, which is necessary for the cyclic transition state of the
elimination mechanism. The olefin is generated from the thermodynamically most stable
carbonium ion. Hydride migration or skeletal rearrangement may take place to give a
more stable ion of a carbonium ion of high reactivity is produced. After ring opening in
the ionic ester pyrolysis mechanism, a second ester pyrolysis reaction occurs, which could
also take place by the cis-elimination mechanism, as shown in Scheme 5.3. The formation
178
New Approaches to Reduce Plastic Combustibility
179
Handbook of Polymer Blends and Composites
180
New Approaches to Reduce Plastic Combustibility
The intumescent behaviour resulting from a combination of charring and foaming of the
surface of the burning polymers is being widely developed for fire retardance because it
is characterised by a low environmental impact. However, the fire retardant effectiveness
of intumescent systems is difficult to predict because the relationship between the
occurrence of the intumescence process and the fire protecting properties of the resulting
foamed char is not yet understood.
There is a strong correlation between char yield and fire resistance. This follows because
char is formed at the expense of combustible gases and because the presence of a char
inhibits further flame spread by acting as a thermal barrier around the unburned
material. Use of polymeric additives has been studied (polyvinyl alcohol (PVA) systems)
which promote the formation of char in polyvinyl alcohol-Nylon 66 system [27]. These
polymeric additives usually produce a highly conjugated system, aromatic structures
which char during thermal degradation and/or transform into crosslinking agents at
high temperatures:
181
Handbook of Polymer Blends and Composites
Scission of several carbon-carbon bonds leads to the formation of carbonyl ends. For
example, aldehyde ends arise from the reaction.
The identification of a low concentration of benzene among the volatile products of PVA
has been taken to indicate the onset of a crosslinking reaction proceeding by a Diels-
Alder addition mechanism [28]. Clearly benzenoid structures are ultimately formed in
the solid residue, and the infrared (IR) spectrum of the residue also indicated the
development of aromatic structures.
The idea of introducing PVA into Nylon 66 composition was based on the possibility of
high-temperature acid-catalysed dehydration [30]. This reaction can be provided by the
acid products of Nylon 66 degradation hydrolysis which would promote the formation
of intermolecular crosslinking and char. Such a system has been called synergetic
carbonisation because the char yield and flame suppression parameters of the polymer
blend of poly(vinyl alcohol) and Nylon 66 are significantly better than pure poly(vinyl
alcohol) and Nylon 66 polymers [31].
182
New Approaches to Reduce Plastic Combustibility
One reason for this phenomenon may be explained by the ability of PVA oxidised by
KMnO4-(polyvinyl ketone structures) to act as a neutral and/or monobasic bidentate
ligand (ligand with two donor atoms) [33]. The experimental results of others (IR
and electronic spectra) provide strong evidence of coordination of the ligand (some
metal ions Cu2+, Ni2+, Co2+, Cd2+, Hg2+) through the monobasic bidentate mode [34].
Based on the above the following structure can be proposed for the polymeric
complexes:
183
Handbook of Polymer Blends and Composites
Cone calorimeter combustion tests for PVA and PVA-ox (Table 5.3) clearly indicated
the substantial improvement of fire resistance characteristics for PVA oxidised by
KMnO4 in comparison with PVA. PVA oxidised by KMnO4 gives about half the peak
of heat release rate (RHR, kW/m 2), when compared with pure PVA. Even at 50 kW/
m2, the yield of char residue for PVA oxidised by KMnO4 was 9.1% [31].
The sample with PVA-ox displayed even better flame retardant properties due to the
catalytical effect of manganese-chelated fragments on the formation of char (Table
5.3). However, the less satisfactory correlation is given in the determination of total
heat release data (Table 5.3) [31]. Although, the cone measurements indicated a
decrease of total heat release for Nylon 66-PVA and Nylon 66-PVA-ox in comparison
with pure PVA, the sample of Nylon 66 with PVA-ox showed a higher value of total
heat release than Nylon 66 with PVA (Table 5.3). It has been qualitatively explained
this fact by the influence of a catalytical amount of chelated manganese-structure
incorporated in polymer on the smouldering of the polymer samples.
The flame out time for Nylon 66 with PVA-ox is larger than the flame out times of
Nylon 66-PVA and Nylon 66 (Table 5.4). The values of average heat of combustion
indicate the exothermal process of smoldering provided by chelated manganese-
structures (Table 5.4). The approximately equal amount of char yield for Nylon 66-
PVA and Nylon 66-PVA-ox has been found [31].
184
New Approaches to Reduce Plastic Combustibility
Table 5.4 Cone calorimeter data of the heat of combustion and the flame out
time for Nylon 66 compositions at a heat flux of 50 kW/m2
Composition Flame out time, s Average Heat of
Combustion, MJ/kg
Nylon 66 512 31.5
Nylon 66-PVA (80:20 wt.%) 429 25.1
Nylon 66/PVA-ox (80:20 wt.%) 747 29.5
Polymer-organic char former (PVA system) incorporated in Nylon 66 reduced the peak
rate of heat release from 1124.6 kW/m2 (for Nylon 66) and 777.9 kW/m2 (for PVA) to
476.7 kW/m2 and increased the char yield from 1.4% (for Nylon 66) to 8.7% due to a
synergistic carbonisation effect. The cone calorimeter was operated at 50 kW/m2
incident flux.
Polymer layered silicate (clay) nanocomposites are materials with unique properties when
compared with conventional filled polymers. Polymer nanocomposites, especially polymer-
layered silicates, represent a radical alternative to conventionally filled polymers.
185
Handbook of Polymer Blends and Composites
In the polymer industry there exists a need for new, more effective, and environmentally
friendly flame resistance polymers. Recent data on the combustion of polymer
nanocomposites indicate that they could be used for this purpose [36].
There are several proposed mechanisms as to how the layered silicate affects the flame
retardant properties of polymers [36]. The first is increased char layer that forms when
nanocomposites are exposed to flame. This layer is thought to inhibit oxygen transport
to the flame front, as well as, gaseous-fuel transport from the polymer and therefore
reduce the heat release rate of the burning surface. At higher temperatures, the inorganic
additive has the ability to act as a radical scavenger due to adsorbtion to lewis acid sites.
This may interrupt the burning cycle as radical species are needed to break polymer
chains into a fuel-fragments. The disordered nanocomposites also inhibit oxygen an
combustible fuel species transfer by increasing the path length of these species to the
flame front. The path length is dramatically increased due the surface area of the silicates
(approximately 700 m2/g for Na+ montmorillonite). There is also a high possibility of
alumina-silicate solid phase catalysis of polymer decomposition, which can dramatically
change the overall scheme of thermal degradation process kinetics.
The plots (Figures 5.3-5.5) compare the results obtained for Nylon 66 as such and an
intercalated Nylon 66 hybrid produced by using a Carver press to mix Nylon 66 with
5% by weight of Cloisite 15A (montmorillonite modified by ion exchange with dimethyl-
dithallow ammonium from Southern Clay Products Inc.) in inert nitrogen atmosphere at
260 C for 30 minutes. It can be seen that the RHR, displays a lower maximum peak in
the case of the nanocomposite (Figure 5.3), whereas the quantity of heat released (the
area under the RHR curve) is about the same for both products, suggesting that their
thermal degradation mechanism is the same [38]. The release of heat by the nanocomposite
over a longer period, however, points to its slower degradation. Figures 5.4 and 5.5 on
mass loss and specific extinction area illustrate the advantages of nanocomposite over
initial Nylon 66 fire behaviour.
186
New Approaches to Reduce Plastic Combustibility
Figure 5.3 Rate of heat release versus time for Nylon 66 and Nylon 66 nanocomposite
at heat flux of 35 kW/m2
Figure 5.4 Mass loss rate versus time for Nylon 66 and Nylon 66 nanocomposite at
heat flux of 35 kW/m2
187
Handbook of Polymer Blends and Composites
Figure 5.5 Specific extinction area (smoke) versus time for Nylon 66 and Nylon 66
nanocomposite at heat flux of 35 kW/m2
During the combustion test of the nanocomposite specimen, a carbon layer formed on
its surface from the start, grew over time and resisted the heat. Formation of a carbonised
layer on the surface of the polymer is a feature of all nanocomposites studied so far: the
pattern illustrated in Figure 5.6 has been reported for other nanocomposites based on
PS, PE and PP [38]. XRD and TEM examination of this residue has revealed an intercalated
nanocomposite structure [38].
188
New Approaches to Reduce Plastic Combustibility
Formally, the term intercalation describes accommodation of molecules and ions between
the expandable layers of two-dimensional inorganic compounds, and resultant products
are called intercalation compounds [35, 36, 40]. Hence, both inorganic and organic
guest species can be incorporated in inorganic matrices without aggregation. Polymer
nanocomposites can be seen as a result of this process. In some cases flame retardants
intercalated into inorganic layered structures can gain an advantage over original ones.
We used interlayer surfaces or interlayer spaces as organising media, and build up highly
functional system for triphenylphosphine (TPP-i) which can provide specific conditions
of triphenylphosphine pyrolysis.
Triphenylphosphine (TPP) has been used as effective flame retardant for some polymer
systems (especially for polystyrene) [41]. It is generally thought that flame-retardant
effect of TPP is due mainly to formation of P* radicals which act as free radical traps
around the flame. The intercalated/modified TPP (TPP-i) into a mineral structure of
kaolin indicated even a better flame resistance in a combination with PS. The presence of
char formation during polymer combustion gave to us an informal evidence of solid-
phase crosslinking and aromatisation reactions which can occur in the system TPP-i-PS
during high temperature pyrolysis. A difference has been found in the mechanism of
thermal degradation of TPP and TPP-i due to a physico-chemical behaviour of TPP-i
molecules which are placed between the crystal layers of inorganic structure of kaolin.
Namely, our idea was to create a flame retardant system (TPP-i) which eliminated P*
radicals into a gaseous phase and, on the other hand, to keep benzene radicals in the
solid phase during polymer degradation. Therefore, the mechanisms of TPP and TPP-i
thermal degradation are of great interest at the present study. Thermal degradation of
TPP and TPP-i was studied by means of DSC analysis and gas chromatography - mass-
spectrometry (GC-MS).
5.3.4.1 Experimental
Commercially available TPP was obtained from CHEMAPOL (Czeck Republic). The
Crystal kaolinite form of clay was obtained from Dneprodzerdjinsk mine (Republic of
Ukraine). Polystyrene was supplied by Samsung co., Nylon 6 was supplied by BASF Co.
DSC analysis spectra were obtained using Mettler TA 3000 at a heating rate 10 C/min
under air conditions. Intercalated TPP has been prepared from the melt of TPP (80 C) in
the presence of kaolin (1:1) for 72 hours [42]. Extensive pyrolysis of samples over one
minute were carried out in a laboratory pyrolyser in an air atmosphere at 400 C. The
189
Handbook of Polymer Blends and Composites
DSC curves of pure TPP and TPP-i are shown in Figure 5.6. Pure TPP, as well as TPP-
i shows a strong endothermal peak at 84 C. This endothermic effect was attributed
to the process of melting. Pure TPP shows the endothermic effect of evaporation
(sublimation) with maxima at about 350 C (curve a), whereas the sample of TPP-i
indicates the exothermal peak of char crosslinking and char formation with maxima
at about 305 C (curve b). Thermal decomposition behaviour of TPP-i is very different
from pure TPP. Instead of the broad evaporation endotherm (TPP) the exothermic
effect becomes predominant. The modification/intercalation of TPP with kaolin creates
an option to decompose TPP-i in solid phase instead of the gaseous degradation of
TPP. Formally, it means the TPP introduced into a layered kaolin structure has the
possibility to char formation with the release of low-molecular phosphorus-containing
fragments and P* radicals. Presumably, this model is more effective than the model
of pure TPP degradation in the inhibition of gaseous phase radical chain reactions
during the polymer combustion.
190
Figure 5.7 Gas chromatogram of the degradation products evolved from TPP
Figure 5.8 Gas chromatogram of the degradation products evolved from TPP-i
191
Handbook of Polymer Blends and Composites
The chromatogram for TPP-i (Figure 5.8) has much less products of degradation than
the one for TPP. It mostly consist of biphenyl, phenol and biphenyl derivative of
phenylphosphine (BDP) [17] (Table 5.5).
The difference between two of these chromatograms is the presence of fused aromatic rings
derivatives in the degradation products of pure TPP: naphthalene, phenanthrene, etc.
These products might be formed through additional reactions of benzene ring condensation
in the gaseous phase. However, the condensation products are absent in the chromatogram
of TPP-i pyrolysis (Figure 5.8). As might be expected, the condensation reactions of phenyl
radicals forming via the P-Ph linkage rapture reaction occur in the iter/intra layered structure
of intercalated TPP. That is why it gives more char residue than the pure TPP.
Finally, it can be noted that the pyrolysis of TPP-i gives an opportunity to form charring layer
onto a polymers surface. It may be an extra possibility to protect polymers against the fire.
192
Table 5.5 Peak assignment of the characteristic peaks on the pyrograms
of TPP and TPP-i
Peak no. Mass Products T PP TPP-i
1 94 + +
Phenol
2 128 + +
Naphthalene
3 154 + +
Biphenyl
4 170 + -
Phenyl ether
5, 6, 7 168 + -
Methylbiphenyl
8 168 + +
Dibenzofuran
9 180 + -
Methylfluorene
193
Handbook of Polymer Blends and Composites
10 176 + -
Diphenylphosphine
11 178 + -
Phenanthrene
12, 15 204 + -
Phenylnaphthalene
13 192 + -
Methylphenanthrene
14 230 + +
Terphenyl
16 260 - +
BDP
194
5.3.4.3 Combustion Tests
Figure 5.12 demonstrates the data of LOI of PS compositions. It is clearly seen that the
formulation of PS/7% TPP-i has the highest value of LOI of 30.0 in comparison with LOI
of 23.0 for PS OMC15A (montmorillonite)-nanocomposite and LOI of 18.0 for PS.
Figure 5.13 shows the char yield values for PS compositions.
A comparison of LOI values for PS and Nylon 6 compositions is presented in Figure 5.14.
The effect of LOI increase is not so pronounced for Nylons as for PS compositions: Nylon
6 - 23.0, Nylon 6-OMC15A-nanocomposite (5%) - 27.5, Nylon 6 + 10% of TPP(i) - 26.3
(Figure 5.14).
(c)
(b)
(a)
(a)
(b)
(c)
195
Handbook of Polymer Blends and Composites
(c)
(b)
(a)
(a)
(b)
(c)
(b)
(c)
(a)
(a)
(b)
(c)
196
New Approaches to Reduce Plastic Combustibility
References
197
Handbook of Polymer Blends and Composites
16. S. V. Levchik, G. Camino, L. Costa and G. F. Levchik, Fire and Materials, 1995,
19, 1.
17. H. Horacek and R. Grabner, Polymer Degradation and Stability, 1996, 54, 2-3, 205.
18. S.V. Levchik, L. Costa, G. Camino, Polymer Degradation and Stability, 1992, 36,
3, 229.
22. L. Costa, G. Camino and M. P. Luda di Cortemiglia in Fire and Polymers, Ed., G.
L. Nelson, ACS Symposium Series No.425, ACS, Washington DC, 1990, 211.
25. G. Camino and S. Lomakin in Fire Retardant Materials, Ed., A. R. Horrocks and
D. Price, CRC Publishers, Boca Ratan, FL, USA, 2001, Chapter 10.
26. P. Haake and C. E. Diebert, Journal of the American Chemical Society, 1971, 93,
6931.
27. Y. Tsuchiya and K. Sumi, Journal of Polymer Science, 1969, A-1, 7, 3151.
198
New Approaches to Reduce Plastic Combustibility
39. V. Babrauskas and R. Peacock, Fire Safety Journal, 1992, 19, 255.
199
6
Fibre Reinforced Plastic Composites for
Biomedical Applications
M. Jayabalan
6.1 Introduction
Polymeric composites are the preferred materials for many industrial and general applications
owing to their improved properties namely strength and stiffness (high strength to weight
ratio and high stiffness to weight ratio) which could not be obtained with conventional
materials. Polymers are extensively used for various biomedical applications. Uses of
polymeric implants in orthopaedic surgery have also increased over the last decade. Ultra
high molecular weight polyethylene (UHMWPE) has been used as both knee joint and hip
prostheses while polypropylene and polytetrafluoroethylene have been used as finger joint
prostheses. However catastrophic failures of clinical implants have been reported during
long-term implantation, which are mainly due to the fatigue failure. Composites having
exceptional resistance to fatigue loading have potential applications as orthopaedic and
dental implants. Composites intended for medical applications incorporate carbon,
polyethylene, glass or polyamide as a fibre component, while polysulphone, polyether
ether ketone (PEEK), polymethyl methacrylate or polyethylene is used as a matrix
component. The long-term success of these composites as biomedical implants is largely
dependent on the complete incorporation of the implant by the host tissue and continued
biomechanical performance and the ability to withstand the biodegradation for intended
duration. Therefore the success of these implants depends on the material aspects, engineering
design, material response to the host tissue and host tissue response to the material.
Considering the material aspects and engineering design of the composites, the long-term
performance of polymeric composites in physiological environments associated with
biomechanical (cyclic) stress, is largely influenced by the viscoelastic nature of the polymer
and discrete interfacial region at the matrix - filler interface.
This chapter deals with various aspects of design, development and evaluation of various
fibre reinforced plastic composites (FRPC) for biomedical applications.
The development of synthetic materials for the application as implants in humans involves
matching the mechanical properties of the specific body parts. The polymeric composites
201
Handbook of Polymer Blends and Composites
intended for the replacement and/or augmentation of hard tissue have to be tailor-made
to match the tensile, compressive and bending properties. The most important body part
in the skeletal system is compact bone, which often requires replacement and/or
augmentation due to accidents or to congenital defect. The different compact bones
associated with the locomotor system are the femur, tibia, fibula, humerus, radius and
ulna. Yamada [1] has dealt extensively with the evaluation of the strength of various
biological materials. Ko [2] has reported that the ultimate tensile strength of specimens
from various bones is greatest in the radius bone (15.2 kg/mm2) and then in descending
order, in the ulna (15.1 kg/mm2), fibula (14.9 kg/mm2), tibia (14.3 kg/mm2), humerus
(12.5 kg/mm2) and femur (12.4 kg/mm2). Tsuda [3] and Ko [2] have reported that the
ultimate bending strength and ultimate tensile strength of compact bone is generally
greater in smaller bones than it is in larger bones. Similarly the specimens from small
bones generally have a greater ultimate percentage elongation than the specimens from
the other bones [2]. On the other hand Yokoo [4] has reported that the large bones have
greater ultimate compressive strength than the small ones. Ko [2] has also observed that
the wet femoral and humeral specimens have somewhat lower elastic modulus (stiffness)
than the specimens from the other bones.
The mechanical properties of a particular compact bone for example the femur, vary in
different parts and on different sides. The ultimate tensile strength and compressive strength
of wet compact bone from the middle of the femoral shaft is a little greater than that of
bone from the upper or lower part of the shaft. Bone from the middle side of the femoral
shaft also has greater tensile strength than that from other sides of the shaft [1]. However
there is no significant difference in ultimate compressive strength with respect to sides [4].
The mechanical properties of the components of teeth for example enamel dentin, crown,
root and periodontal membrane are different from each other. The mechanical properties
(vikers hardness) were evaluated by determining the vikers hardness number using the
diamond pyramid hardness test method. The abrasive coefficient (0.009) and vikers hardness
number (123) of wet enamel of teeth have been reported by Kubo [5] and Nose [6]. Kastura
[7] has reported the compressive properties of dentin (ultimate compressive strength 14.2
kg/mm2, ultimate percentage contraction 4.17 and modulus of elasticity 520). Hardness of
dentin (vikers hardness number 30) was also determined [8]. Kastura [7] has also reported
the compressive properties of dental crown (ultimate compressive strength 14.7 kg/mm2,
ultimate percentage contraction 2.72) and root (ultimate compressive strength 17.6 kg/
mm2, ultimate percentage contraction 3.38). Yoshimatsu [9] has reported the average
ultimate extractive (tensile) strength of periodontal membrane is 0.12 kg/mm2.
The mechanical properties of human skeletal muscle for example rectus abdominus muscle
and fascia from thigh and leg were determined by Katake [10, 11]. Ultimate tensile
strength is greatest in the sartorius muscle and least in the rectus femoris and gastrocnemius
202
Fibre Reinforced Plastic Composites for Biomedical Applications
muscle. Takigawa [12] has reported the tensile properties of calcaneal tendinous tissue
(ultimate tensile strength 5.6 kg/mm2, ultimate percentage elongation 9.9). Yoshimatsu
[13] has reported the tensile properties of ascending aorta (ultimate tensile strength 76 g/
mm2, ultimate percentage elongation 1068) and descending aorta (ultimate tensile
strength 102 g/mm2, ultimate percentage elongation 74).
Using these observations, it is essential to optimise the design of the polymeric composites
to critically match the mechanical properties of the body part that they will be replace,
for ensured clinical success.
Materials implanted in soft or hard tissue always elicit a response. An ideal implant
performs as if it were equivalent to the host tissue. The related consequences with nonideal
implants are foreign body reactions [14, 15]. There are three types of foreign body
reactions associated with an implant:
2. foreign body reaction associated with the chemical characteristics of the material;
Jayabalan and Rathinam [15] have reported that the larger and long-lived biocompatible
implants are associated with normal foreign body reaction and chronic inflammatory
reaction.
Carbon has been known for a long time to be a well-tolerated material in implant surgery.
Carbon is biocompatible. However conflicting reports have also appeared on
biocompatibility of carbon fibres and carbon fibre reinforced composites. The normal
tissue growth around and encasement of the individual fibres integrated the carbon fibres.
No significant foreign body reaction was observed. However, integration of the fibres
into the tissue is not as good when the fibres are coated with epoxy resin [16]. The
mechanical function has been taken over by the formation of a connective tissue structure
resembling a ligament. On the other hand, progressive rupturing of fibres occurs with
implants of pure carbon fibres within the course of several months of implantation.
Vitreous carbon and glazed carbon fibre reinforced carbon (CFRC) were considered
suitable for human implantation with partial success. Carbon particle dissemination
with carbon fibre destruction and exposure as a result of preoperative prosthesis drilling
203
Handbook of Polymer Blends and Composites
were considered the likely causes for the failure of this implant. Carbon fibre microparticles
are absorbed either by macrophages or by foreign body giant-cells and are distributed
throughout the tissue via the lymphatic system. Dahhan and others [17] studied the
performance of carbon fibre bundles, implanted in dogs and rabbits after resecting the
anterior cruciate ligament of the knee. They observed slight inflammatory reaction and
collagen tissue in growth.
Morris and others [18] has investigated the tissue response of the carbon fibres (8 m).
They observed that the carbon fibres induced significantly more tissue ingrowth than
polypropylene mesh at six to twelve months postoperatively. The predominant tissues
associated with carbon fibres and polypropylene mesh were dense connective tissue and
fat, respectively. However the studies by Morris and others [18] reveal no fragmentation
of the implants and implant debris in the regional lymph nodes. Belangero [19] has
studied the inflammatory reaction of rat striated muscle to the CFRC particles and
suggested that the longer carbon fibre fragments of CFRC may be responsible for the
more intensive inflammatory reaction. Christel and others [20] has evaluated the long-
term response (two years) of cortical bone to various types of CFRC mainly for bone
contact and bone ingrowth. CFRC coated with calcium phosphate was found to speed
up the bone formation compared to pyrolytic carbon or SiC coatings. The low modulus
of elasticity of the CFRC materials could be responsible for quicker bone contact compared
to a much stiffer material like sintered aluminum oxide. Adams and others [21] by
implanting specimens in holes drilled in rat femora studied the compatibility of the CFRC
with bone. For a period of up to eight weeks, a thin layer of fibrous tissue bridged the
gap between bone and implant but this tissue mineralises and by 10 weeks, bone can be
observed adjacent to the implant, giving firm fixation.
Kenner and others [22] observed the normal stages of bone modelling and remodelling
following the implantation of carbon plugs. The histological study revealed bone in
direct contact with carbon. The implant has bonded with lamellar bone at the end of
four months of post implantation. There is little or no evidence of fibrous bone. The
lamellar bone was made up of large number of osteons many of which were oriented
proximal-distally. Thieme and others [23] has studied the hard tissue compatibility of
solid-mass glass-carbon compounds in bones. Weber and Rettig have reported that the
abrasion of carbon fibre is low, and the abraded particles do not produce any severe
inflammatory reaction [24]. Wright [25] has investigated the damage on the articulating
surfaces of retrieved tibial components of total knee prostheses of carbon fibre-reinforced
ultra-high molecular weight polyethylene and reported the presence of fragments of carbon
fibre in many of the samples from the areas around the prostheses. Gray discoloration in
synovial tissue adjacent to carbon fibre-reinforced polyethylene total joint implants has
been observed in clinical situations. However the synovial discoloration from shed carbon
fibrils does not appear to present a significant clinical problem. Groth and Shilling [26]
204
Fibre Reinforced Plastic Composites for Biomedical Applications
also evaluated the pathologic response of the synovium to this material. Polyethylene
debris was found to produce a significant synovial reaction with histiocytes and foreign
body giant cells. This was found in both the carbon fibre-reinforced polyethylene and in
those cases without carbon-reinforced components. This reaction is also seen in revisions
of total hips and total knees in which standard polyethylene components have been
implanted. In contrast with this, carbon particles produced only a minimal reaction in
the synovial tissue. The cellular response with the direct contact cell culture evaluation
to the carbon fibre reinforced PEEK and polysulphone composites were negligible [27].
Fungal attack of polymeric material is also one of the serious issues as far as the long-
term shelf life is concerned. Jayabalan and Shunmugakumar [28] have studied the effect
of hydrophilicity and hydrophobicity of polyurethanes based on hexa methylene
diisocyanate/bicyclohexyl methane diisocyanate, polytetramethylene oxide glycol
(PTMG)/polypropylene oxide glycol (PPG)/polyethylene glycol (PEG) and trimethylol
propane on the fungal growth of Aspergillus niger and Penicillium sp. They observed
that fungal growth was not observed with hydrophobic polyurethanes.
The major problem in the success of the orthopaedic implant is the mismatch of modulus
of elasticity (stiffness) between bone and implant. Ceramic and metals are 10 to 40 times
stiffer than the cortical bone. During the load sharing between the implant and bone, the
amount of stress carried by each of them is directly related to their stiffness that is known
as stress shielding or stress protection. Under such circumstances the bone becomes
inadequately loaded leading to atrophy with subsequent thinning and increased porosity
[30]. Therefore it is essential to match the bone and implant stiffness to decrease the
detrimental effect of stress protection.
Arthroplasties are successful only for short-term to medium-term since the permanent healing
is restricted because of possible loosening of the prostheses. Even clinically silent loosening
205
Handbook of Polymer Blends and Composites
can lead to failures, either by progressive degeneration of the bones at the supporting points
of the prosthesis or through breakage of the stem of the prostheses. However the attempts to
overcome such difficulties by changing the shape of the prostheses, were not successful.
Therefore the design of the material employed in the construction of the prosthesis is very
important. Composite materials, which can be very strong while having a low modulus of
elasticity, have potential to be made into isoelastic hip prostheses. The modulus of elasticity
of these composites has to match this property of the proximal femur.
Brakeman and Huttinger [33] developed (1) a high strength isotropic carbon, (2) a silicon
carbide/carbon composite (SiC/C) and (3) CFRC. They investigated the effects of size
(length/to thickness ratio of samples) and processing parameters (final heat treatment
temperature, fibre volume content and fibre orientation) on the mechanical properties of
these materials. Flexural strength of the first two materials, which are both isotropic,
increases with decreasing l/d ratio, whereas the CFRC behaves contrarily. In the latter
case, failure at low l/d ratios is caused by shear failure. The fatigue strength of the isotropic
206
Fibre Reinforced Plastic Composites for Biomedical Applications
carbon and the SiC/C is between 70 and 80% of the static strength, even in Ringers
solution. The Youngs modulus of the isotropic carbon is comparable to that of femoral
bone (see Figure 6.1). This figure shows the Youngs modulus and fatigue strength of
some carbon materials in comparison to those of femoral bone, alumina and cobalt base
alloy. The studies suggest that both isotropic materials are applicable for sliding parts
like ball and socket joint. The CFRC possesses high fatigue strength (approximately 400
MN/m2) with a moderate Youngs modulus (approximately 150 x 103 MN/m2) suggesting
a possible application for shafts and pins.
Christel and others [20] have also reported the development of a CFRC hip prosthesis by
embedding carbon fibres either with dense pyrolytic carbon or silicon carbide (SiC) matrix
through chemical vapour infiltration. The fatigue tests showed no fatigue damage and
also showed better stress transfer as compared to a metal prosthesis having the same
design. The data of static strength of the CFRC composite revealed that the fatigue
loading (7.5 x 106 cycles) has no significant influence on the static strength.
Figure 6.1 Fatigue strength and Youngs modulus of selected biomaterials. A: CoCr
Mo alloy; B: Femur corticalis; C: Isotropic carbon; D: SiC/C; E: CFRC; F: Alumina.
(Reprinted from H. Bruckmann and K.J. Huttinger, Biomaterials, Volume 1, 1980,
with permission from Elsevier Science)
207
Handbook of Polymer Blends and Composites
The production of an endoprothesis with different working parts [34, 24] was also carried
out using carbon fibres. Kenner and others [22] studied static fatigue behaviour of glassy
carbon and observed that the glassy carbon bars aged in vivo for five months did not undergo
statistically significant weakening. However glassy carbon does undergo static fatigue when
aged in a simulated biological environment. However, the conditions necessary to cause this
failure are extreme. Kenner concluded that the CFRC composites have mechanical strength
sufficient to withstand the cyclic fatigue loading in the physiological environment.
Wolter and others [35] have studied the long-term implantation of a CFRC hip joint
endoprosthesis in foxhounds. A hip joint endoprosthesis, comprising a CFRC stem,
aluminum oxide head, and polyethylene acetabular cup, was studied in vivo trials in ten
foxhounds after implantation with bone cement. Observations over six months and one
year demonstrated that CFRC exhibits adequate strength for use as a stem material under
the observed conditions.
Plates currently in use for the management of bone fracture made of metal present with
various problems namely atrophy, delayed hypersensitivity reactions and bacterial infection
(Staphylococcus epidermidis). FRPC composites are being developed as alternatives to
metals for structural orthopaedic implant applications. FRPC composite fracture behaviour
and environmental interactions are distinctly different from those that occur in metals.
CFRPC has the advantage over metals of being virtually free from corrosion and fatigue,
reasonably radiolucent and able to meet a wide range of mechanical requirements. Iizuka
[36] has manufactured CFRPC plates and compared the performance with 316L stainless
steel plates on rabbit tibiofibular bones. The results indicate the influence of CFRPC plate
upon cortical bone was milder than that of stainless steel plate, with an adequate stabilising
force for the repair of fractured rabbit tibiofibular bones. Nelham [37] has developed
prosthetics and orthotics using CFRPC. The prosthetic application related to the construction
of the Chailey Harness that was used to mount externally powered upper limb prostheses
to the torso of limb deficient children damaged by the drug Thalidomide. The application
to orthotics was related to Hip-Knee-Ankle-Orthoses (HKAO) as worn by severely
handicapped children with spina bifida. A shapable CFRPC/aluminium alloy hybrid
composite bar was developed as a potential orthopaedic implant.
Sclippa and Piekarski [38] have studied the uses of carbon fibre and glass fibre reinforced
polyethylene for possible orthopaedic hip joint cups. The studies reveal appreciable
208
Fibre Reinforced Plastic Composites for Biomedical Applications
increase of tensile strength and elastic modulus with the addition of carbon and glass
fibres. But these composites cannot be used for fixation devices such as plates and screws
for large bones or for the replacement of larger load-carrying bones in the body since the
strength of these composites are not sufficient for these devices.
Busanelli [39] has reported massive wear in carbon fibre-reinforced polyethylene (Poly Two;
Zimmer Inc., Warsaw, Indiana) as an insert in a knee prosthesis. The study also confirms that
Poly Two is not more advantageous than that of unreinforced ultra-high molecular weight
polyethylene in terms of wear. However, Halcomb and Bardos [40] have reported their clinical
experience of carbon fibre-reinforced UHMWPE for total knee replacement prostheses. The
composite has significant increases in compressive and tensile strength, elastic modulus and
fatigue strength. Moreover they declared that these composites undergo less wear. These
composites are more resistant to creep than the unreinforced ultra high molecular weight
polyethylene (UHMWPE) (Figure 6.2). The carbon fibre-reinforced polyethylene undergoes
less percentage deformation than the unreinforced UHMWPE even under high stress.
Figure 6.2 Competitive creep data for Poly-Two material and UHMWPE
(Reprinted from F.J. Halcomb and D. Bardes, Transactions of ASAIO, 1981, 27, 364,
with permission from Blackwell Science, Inc.)
209
Handbook of Polymer Blends and Composites
Polysulphone and polyether ether ketones have been considered as the matrix material
for carbon fibre reinforced composite implant materials [27]. Claes [41] has prepared
carbon fibre reinforced polysulphone as plates, screws and spinal segmental replacement
implants. Short carbon fibre reinforced composites could potentially replace some of the
metal alloys used in orthopaedic implants. Stange and Mittelmeier [42] have produced
semi-rigid plates made of thermoplastic polyethersulphone (PES), reinforced with 20%
short carbon fibres. Biomechanical examinations revealed less loss of mechanical function
of the stabilised bones due to elastic osteosynthesis. During more pronounced exercise
loading, a reversible springiness of the fracture results, which might stimulate callus
formation and improved stability. Carbon fibre-reinforced epoxy bone plates secured
with carbon fibre reinforced polysulphone expanding rivets were produced by Sell and
others for torsional failure [43]. However, the carbon fibre expandable rivets failed at a
greater torsional moment.
The extensive use of PMMA as a bone and dental cement materials has provided much
information relating to processing and performance. The disadvantage of such cements
is their mechanical behaviour, revealed by low crack propagation resistance, impact
strength, fatigue resistance and reduced fracture toughness. Acrylic bone cement is
significantly weaker and of lower modulus of elasticity than compact bone. It is also
weaker in tension than in compression. Many attempts have been made to resolve these
problems either by modifying the PMMA chemical structure via copolymerisation or
incorporating reinforcing additives. The latter is of great importance, because it can lead
to the preparation of composite materials with considerably improved performance.
Besides reinforcement, the incorporation of additives, such as fibres, results in better
processing characteristics, e.g., it lowers the polymerisation exotherm [45].
210
Fibre Reinforced Plastic Composites for Biomedical Applications
Saha and Pal [46] also attempted to improve the mechanical properties of PMMA by
reinforcing it with metal wires, and graphite and aramid fibres. Small addition of these
fibres (1-2% by weight for carbon and up to 6% for aramid) improved the mechanical
properties significantly. Due to the improved mechanical properties of fibre reinforced
bone cement, its clinical use may reduce the incidence of cement fracture and thus loosening
of the prosthesis. Saha and Pal [47] found improved mechanical properties of machine-
mixed, commercially available carbon-fibre-reinforced bone cement and a significant
reduction of maximum exothermic temperature rise during polymerisation by use of the
fibre reinforcement. Saha and Pal [48] also reported strain-rate dependence of the
compressive properties of normal and carbon-fibre-reinforced bone cement (1 wt.%).
The modulus of elasticity, the proportional limit, and the compressive strength of the
normal bone cement showed highly significant positive correlation with the strain rate.
Therefore the strain-rate sensitivity of carbon fibre-reinforced bone cement should conform
to that of bone in order to match the elastic properties for all different strain rates.
Wolter and others [49] have also studied the carbon fibre-reinforced bone cement with
optimised mechanical properties. The fatigue strength of the carbon fibre reinforced
PMMA bone cement could be increased by 17% by oxidation treatment with nitric acid
[50]. This increase, however, is less than that observed in the case of other fibre-reinforced
composites, which is due to the pronounced polymerisation contraction of PMMA.
In order to provide a fixation vehicle between a polymeric composite femoral hip prosthesis
and bone tissue, Marcolongo and others [51] fabricated bioactive glass fibres and glass
fibre composites. After immersion in protein-free simulated body fluid for 10 days, they
observed the development of a calcium phosphate layer (specifically, partially crystallised,
calcium-deficient carbonated hydroxyapatite) on the surface of the glass fibre as well as
on the surface of the glass fibre of the composite. The investigation has suggested the
onset of bioactivity reactions in vivo.
Deng and Shalaby [52] have evaluated the physical properties of UHMWPE fibre/
UHMWPE matrix composites. It was found that the tensile strength and modulus, and
creep resistance were significantly increased after incorporating UHMWPE fibres into a
UHMWPE matrix. The longitudinal tensile strength of the resulting self-reinforced
composite increased with fibre content, according to the law of mixtures. The transverse
strength did not change for fibre content of up to 7%. The double-notch impact strength
of the composites was higher than plain UHMWPE. There was no difference in wear
211
Handbook of Polymer Blends and Composites
properties between the composites and plain UHMWPE. Both unidirectional and cross-
ply laminates were much more creep resistant than plain UHMWPE. With tensile loading
and unloading at 37 C with 1 MPa, the crossply laminate had a much higher creep
resistance than the plain UHMWPE (see Figure 6.3). They reported that the self-reinforced
UHMWPE composites may be good candidates for load-bearing biomedical applications.
Figure 6.3 Tensile creep and recovery results for plain UHMWPE and the self-
reinforced cross ply composite at 37 C and 1 MPa.
(Reprinted from M. Deng and W. Shalaby, Biomaterials, 1997, 18, 9, 645, with
permission from Elsevier Science)
212
Fibre Reinforced Plastic Composites for Biomedical Applications
that of the matrix as the fibre content decreases. In general, the modulus of elasticity and
fibre content have a linear relationship [53].
Pourdeyhimi and Wagner [54] have investigated the fracture behaviour of PMMA bone
cement reinforced with short UHMWPE (Spectra 900) fibres. Linear elastic and non-
linear elastic fracture mechanics results indicate that a significant reinforcing effect is
obtained at fibre contents as low as 1% by weight, but beyond that concentration a
plateau value is reached and the fracture toughness becomes insensitive to fibre content.
However the flexural strength and modulus are apparently not improved by the
incorporation of polyethylene fibres in the acrylic cement, probably because of the presence
of voids, the poor mixing practice and the weakness of the fibre/matrix interfacial bond.
Andreopoulos and others [45] have studied the effect of some finishing agents of Tekmilon
ultra high modulus polyethylene fibres reinforced in PMMA. The composites having
polyethylene fibres treated with poly(hydroxy ethyl methacrylate) (PHEMA) showed
higher tensile strength than that having untreated and PMMA treated fibres. This is due
to the wetting of polyethylene fibres by PHEMA coating and increased fibre arrangements
by polar forces between the fibre and the matrix.
Polyester fibres also have been used for orthopaedic applications. Nair and others [55]
have reported the physical (molecular weight data, thermal properties, surface properties
and leachants) and mechanical properties (tensile properties and burst strength) of
polyester fibre for biomedical applications. Nair and others [56] have studied the effect
of repeated gamma radiation sterilisation on the physical and mechanical properties. It
was found that sterilisation with a single dose of 2.5 Mrad affected the polyester yarn
and fibre due to the degradation at the amorphous region. Sterilisation with higher or
multiple doses degraded the polyester at the crystalline region leading to the decreased
crystallinity, breaking load and molecular weight.
The composites based on PHEMA matrix and polyester fibres were developed by
Migliaresi and Nicolais [57]. The mechanical performance of a uniaxially oriented
composite has been adjusted to match the behaviour of a human tendon. A laminate
composite has been modelled for use in plastic surgery. Loosening remains an impediment
to the long-term success of total hip replacements despite numerous improvements in the
materials used. In cemented prostheses, fatigue and fracture of bone cement have been
implicated in the failure of these devices. A new materials self-reinforced PMMA (SRC-
PMMA) composite has been developed by Wright and others [58]. The composite consists
of high strength, highly oriented PMMA fibres embedded in a matrix of PMMA. Three
different weaves of SRC-PMMA were evaluated in bending and fracture toughness in
213
Handbook of Polymer Blends and Composites
air, after immersion for 30 days in saline at 37 C, and after gamma irradiation followed
by immersion. Bending modulus and strength decreased with gamma irradiation followed
by saline immersion. The effect of saline immersion alone on bending strength and modulus
was negligible. Saline immersion and gamma irradiation followed by saline immersion
was shown to have little or no effect on the fracture toughness of woven SRC-PMMA.
Optimally incorporated SRC-PMMA absorbs the same amount of water as bone cement.
The studies reveal that SRC-PMMA is a promising material equal to or better than bone
cement in all tests performed.
Particulate filler based dental or bone cement was routinely used in clinical applications.
These cements are based on the formulation consisting acrylic polymeric powder, acrylic
monomer, particulate filler, crosslinking agent and free radical initiator. The tensile strength
and fracture toughness of these composites is lower or somewhat higher than that of
glassy polymer [59]. Fatigue resistance in dental composites is generally correlated with
the fracture toughness [60, 61]. The relatively low values of toughness and fracture
resistance are partially responsible for relatively short service lives (1-4 years) of dental
composites. A similar problem may also exist for bone cement.
FRPC are being developed for dental applications, such as prosthodontic frameworks,
splints and orthodontic retainers. Fibre-reinforced composite materials can be used to make
metal-free prostheses with excellent aesthetic qualities. Generally, the prevalent mode of
failure for all FRPC investigated was brittle failure under flexure loading. Novel appliance
designs for dentistry are feasible with FRPC with the combination of good flexure properties,
formability, and translucency. One method of reinforcing denture base material is to use
FRPC composite. Different types of fibres, such as glass, carbon/graphite, aramid, and
ultrahigh-modulus polyethylene (UHMP) fibres have been tested for this purpose. The
fibre concentration and its adhesion to polymer matrix influence the transverse strength of
the fibre composite [62]. Adams and others [21] has evaluated CFRC for potential
application as endo osseous dental implant. In a unidirectional composite, high strengths
(1200 MN/m2 longitudinal flexural strength) and high modulus (140 GN/m2 flexural
modulus) may be obtained with interlaminar shear strength of 18 MN/m2. Alternatively,
layers of fibres may be laid in two directions to give more isotropic properties.
The acrylic resin denture base is used more than the usual metal denture base because of
the low cost, simple process to make and easy rebasing (process of refitting the denture by
214
Fibre Reinforced Plastic Composites for Biomedical Applications
replacing the denture base material without changing the occlusal relationship of the teeth).
But, due to the low strength of the resin, the upper resin denture base has to be made three
to four times as thick as the metal denture base. Attempt has been made by some investigators
to make the upper complete resin denture base thinner while maintaining the strength of
the resin denture base. Shimozato and others [63] has studied the effects of thickness and
number of carbon cloth sheets on physical properties of reinforced denture base PMMA
specimen. Miyairi and others has used carbon fibre reinforcement to make the palatal area
of the denture base thinner [64]. The resin denture base reinforced by the carbon fibre
cloth for a thickness of 0.7 mm showed improved (by 10% or more) stiffness and strength
as compared with those of the ordinary resin denture bases.
Production of properly oriented fibres that are well centred within the resin is technically
difficult. Such composites yield less predictable property improvements in comparison
with that of randomly dispersed fibres. DeBoer and others [65] has studied the effect of
carbon fibre orientation on the fatigue resistance and bending properties of two denture
base acrylic resins. Carbon fibres placed perpendicular to the direction of applied stress
produced the most favourable combination of increased resistance to bending and to flexural
fatigue. Kilfoil and others [66] has observed that the resins resistance to fracture is increased
if the carbon fibres are added to a dental composite resin. However, the addition of carbon
fibres to a dental composite resin actually decreased the flexure strength. Glass fibres added
to temporary crown and bridge material (Proviso-Blend, Blendax) by 5% to the weight
increased the tensile strength of the acrylic resin by 38% [67]. The increase was even
greater when carbon fibres were added (50%), but the resin discoloured to a grayish shade.
The amount of improvement in tensile strength depended on the type and geometry of the
fibre and of the fibre concentration. Yazdanie and Mahood [68] have investigated the
transverse strength of carbon fibre acrylic resin composite.
Larson and others [69] compared the moduli of elasticity of three provisional crown and
fixed partial denture resins reinforced with carbon graphite fibres. The mean moduli of
elasticity of all the resins increased significantly with fibre incorporation. Water storage
did not have a significant effect on the moduli of elasticity of the resins tested. The
flexural properties of PMMA reinforced with carbon/graphite (C/G) fibres with three
different surface treatments were investigated by Ekstrand and others [70]. The fibres
used were (1) commercially available fibres, (2) cleaned fibres, and (3) cleaned and epoxy-
sized fibres. Fracture stress and flexural modulus decreased when commercial fibres
were used as a reinforcing material. Composites with cleaned and sised fibres gave only
minor differences in flexural properties after dry and wet storage. After water storage a
substantial part of the cleaned fibres adhered to the matrix material. The adhesion capacity
of the other fibres was reduced since the water absorption caused separation of fibre and
matrix. Shimozato and others [71] has studied the surface treatment of the carbon cloth
and its effect on the bending and tensile strength of the denture base PMMA resins
215
Handbook of Polymer Blends and Composites
reinforced with carbon fibre cloth. Viguie and others [72] have studied the effect of fibre
reinforcements on the mechanical properties (flexural strengths) of epoxy/carbon
composite for use in dentistry.
Kasuga and others [73] have reported the preparation and evaluation of carbon fibre
reinforced fluid type resin denture. They have improved the mechanical strength of self-
cured acrylic resin by application of carbon fibre cloth as reinforcement by a simple
prepreg method. Conventional bridge frameworks of gold alloys are expensive. Bjork
[74] has developed implant (titanium)-fixed, dental bridges from carbon/graphite fibre
reinforced PMMA and reported that these bridges functioned satisfactorily during clinical
use. Ruyter [75] has developed the carbon/graphite fibre reinforced PMMA suitable for
implant-fixed dental bridges.
Continuous woven silanised E-glass, fibres and woven silica fibres were used to reinforce
heat-cured and auto polymerised denture base polymers [78]. The ultimate transverse
strength of the E-glass fibre-reinforced test specimens decreased 14% and that of the
silica fibre-reinforced test specimens decreased 36% after 48 weeks of storage in water
[78]. The effect of fibre reinforcement of autopolymerising PMMA was investigated
[79]. The impact strength of continuous E-glass fibre-PMMA composite was determined
by Vallittu [62]. Rectangular test specimens (n = 10, per group) were modified by
216
Fibre Reinforced Plastic Composites for Biomedical Applications
Karmaker and others [80] have prepared two types of uniaxially oriented long S-2-glass and
E glass fibre reinforced composites using polycarbonate (PC) and BIS-GMA-based copolymers
as a matrix for use in various dental appliances. Mechanical properties (flexural fatigue and
single-lap shear strength) of the small cross-sectional composite strips of S-2 glass composites
were superior to those used previously in clinical studies. The flexural moments in the elastic
region of these composites are much higher than the composites S-2 (60)PU, E(50)-PETG,
E(50)PC and E(40)-PA12 (Figure 6.4). The PC, PU, PETG and PA12 refer to polycarbonate,
Figure 6.4 Linear flexural moment versus angular deflection curve for different composites.
(Reprinted from A.C. Karmaker, A.T. DiBenedetto and A.J. Goldberg, Journal of
Biomaterials Applications, 1997, 11, 3, 318, with permission from Technomic Publishing)
217
Handbook of Polymer Blends and Composites
The polymerisation shrinkage and void formation in acrylic-based dental composites are
the major drawbacks of dental materials. The effect of polymerisation shrinkage of PMMA
coating of fibre roving on the transverse strength of an acrylic resin-glass fibre composite
was evaluated by Vallittu [82]. The fracture resistance of the specimens was statistically
different (p < 0.001) among treatments, and the polymerisation shrinkage of the PMMA-
MMA mixture also varied. The ratio of the PMMA-MMA mixture that caused only minor
polymerisation shrinkage of PMMA in the glass fibre roving improved the fracture resistance
of the test specimens and seemed to increase the amount of PMMA inside the fibre roving.
Investigations were carried out to establish the causes and effects of void space formation
in acrylic- continuous glass fibre composite material and to clarify the effect of polymerisation
time of acrylic resin on the transverse strength of heat-cured acrylic resin test specimens
[83]. The results suggested that the void space inside the test specimens is caused by a lack
of the adsorbed monomer liquid in the fibre bundle before polymerisation. The data showed
that the transverse strengths of test specimens (n = 240, total) subjected to polymerisation
of different time spans did not vary significantly (P > 0.05).
The release of residual monomer MMA can lead to adverse physiological reactions.
Therefore the release of residual monomer has to be controlled. Miettinen [84] has studied
the release of residual MMA into water from heat-cured and chemically-cured test
specimens of continuous glass fibre-PMMA composite. Release of residual MMA from
heat-cured test specimens with glass fibre reinforcement was significantly higher than
that from unreinforced test specimens (P = 0.003), while in chemically-cured test specimens
with and without glass fibre reinforcement the amount of MMA released did not differ
(P = 0.501). However the test specimens made from chemically-cured PMMA released
more residual MMA than specimens made from heat-cured PMMA. The use of glass
fibre reinforcement in heat-cured denture PMMA statistically increases the release of
residual MMA from the material.
Fibre reinforced polycarbonate composites have been developed for dental applications.
Jayabalan and others [85] have studied the physical changes in low molecular weight
218
Fibre Reinforced Plastic Composites for Biomedical Applications
219
Handbook of Polymer Blends and Composites
Figure 6.5 Plot of Youngs modulus versus loading of plasma treated and untreated fibres.
(Reproduced from K.W.M. Davy, S. Parker, M. Braden, I.M. Ward and H. Ladizesley,
Biomaterials, 1992, 13, 1, 17, with permission from Elsevier Science)
The biocompatibility of vitreous carbon and CFRC as middle ear implants was investigated
by Blayney and others [90]. Carbon-carbon middle ear prosthesis was also developed by
Podoshin and others [91]. CFRC was used as an alloplastic material (a graft of an inert
plastic material) for reconstructing the ossicular chain. The good anatomic and hearing
results of the clinical trial suggested the use of this prosthesis as an alternative alloplastic
material to be used in tympanoplasties.
Due to the biocompatibility, carbon fibre reinforced polysulfone composite material has
some advantages when compared to metals, which makes it favourable particularly for
implants in tumour surgery. The radiolucency of the material avoids problems caused by
the reflection of x-rays, using metal implants. This special property allows the exact
calculation of postoperative radiation doses of tumour patients. Simultaneously the
structures behind the implants are not hidden. All implants can be machined during the
operation to adapt them to the individual anatomical situation.
220
Fibre Reinforced Plastic Composites for Biomedical Applications
woven prostheses used as arterial graft and for soft tissue abdominal aortic aneurysm
repair. Burns and others [94] have reported formation of fibrosarcoma in femoral artery
with Teflon/Dacron woven prostheses. The material-induced tumorigenicity may be due
to solid-state tumorigenisis (sarcoma) or to degraded products (carcinoma). Therefore the
implants for long-term use must be developed from biostable polymers, which are resistant
to biodegradation, leaching of additives and to the complexation of degraded molecules
with biological molecules. Anterior tumour removal, cord decompression and spinal
stabilisation gain in significance in surgical treatment of vertebral tumours. An implant
system, consisting of a basket as vertebral body replacement, plates and screws, was
developed using carbon fibre reinforced polysulphone [95]. This system allows performing
individually shaped, stable and short-distance spine fusions from an anterior approach.
Moreover its radiolucence facilitates postoperative care and irradiation.
The involvement of fabrics and fibres in the soft PHEMA matrix also provides an
alternative of making the ultra-thin membranes to overcome the problem of easily being
torn during handling. Membranes of PHEMA-based composites were manufactured by
adding various kinds of weaved and knitted fabrics and fibres into a deionised water
solution of hydroxy ethyl methacrylate (HEMA) monomer, cross-linker and initiator,
and followed by polymerisation under ultraviolet radiation [96]. By varying the amount
of initial water addition, the dimensional change of PHEMA matrix from the newly
fabricated state to the eventually swollen state could be adjusted to reduce the swellability
mismatch with the fabrics. Mechanical properties namely yielding strength, maximum
strength, Youngs modulus and elongation at break of the fibre-reinforced PHEMA
composites are evidently improved depending on the mechanical characteristics of
additives applied. In addition, some of these membranes also exhibit an improvement in
water transmission rate. Morris and others [18] have investigated the use of carbon
fibres for repair of abdominal-wall defects in rats. Jayabalan [97] has developed
polyurethane potting compounds for embedding regenerated cellulose (Cuprophan)
hollow fibre membranes in dialyser housing components for the fabrication of
haemodialysis. The integrity of the polyurethane-hollow fibre composite was determined.
The potting compound based on polyether polyol as cross-linker has appreciable stability
in thermal ageing and in autoclaving and gamma radiation sterilisation conditions. The
potting compound based on cycloaliphatic diisocyanate, castor oil and polyether polyol
is a candidate potting compound for the fabrication of Cuprophan hollow fibre membrane-
based haemodialyser.
Polyurethane composites were also developed for soft tissue implants. Jayabalan [98]
has developed biostable polyurethane for soft tissue implants. The polyurethane based
on the biuret of hexamethylene diisocyanate, polytetramethylene oxide glycol and
trimethylol propane is tissue compatible and resistant to in vivo biodegradation. This
new polyurethane can be used for mammary prostheses, indwelling catheter, etc.
221
Handbook of Polymer Blends and Composites
Gershon and others [99] have developed elastomeric composite using polyurethane
Lycra fibres and polyurethane Pellathane matrix. The composite prostheses are made
by a filament-winding technique comprising Lycra elastomeric fibres embedded in an
elastomeric matrix of mostly Pellethane [99]. Milka [100] has studied the tension
properties of coracoclavicular ligament harvested from cadavers and samples of carbon
fibre prosthesis. They found the superiority of carbon fibre ligamentous prosthesis
over natural ligament in respect to the elastic properties.
6.7 Conclusion
The implant failure in clinical situations is a very serious issue. The failure of the
implant may be due to the effect of material failure or due to adverse host reactions
or due to the combination of both. The failure due to the material has to be eliminated
by proper selection of polymer matrix and fibre and also by careful design of the
composite suitable to the required application. One of the routes of failure of these
implants due to adverse host reactions is by the bacterial invasion. Bacteria namely
Staphylococcus epidermidis enters the body during surgery or as a result of subsequent
infectious diseases. Once the bacteria reach the implant, it becomes adhered through
hydrophobic interaction and produces a slime covering under which it lives,
inaccessible to antibiotics. Subsequently, infection starts, resulting in weakening of
the material-tissue linkage and trauma. Ultimately the patient has to undergo removal
of the infected implant and receive a new implant. Another challenge in orthopaedic
implants is the development of percutaneous implants. Skeletally attached limb
prostheses and orthopaedic fixations are some implants encountering failures. These
implant failures are attributed to the epithelial migration and absence of an epithelial
seal that result from the bacterial invasion. In the absence of an epithelial seal, the
implant-tissue interface is weak, resulting in loosening and infection at the interface.
Biomaterial surfaces resistant to bacterial adhesion are necessary for long-term
orthopaedic implants.
222
Fibre Reinforced Plastic Composites for Biomedical Applications
Acknowledgement
The author thanks the Director, Sree Citra Tirunal Institute for Medical Sciences and
Technology, Thiruvananthapuram for the permission granted to publish this chapter.
The author also acknowledges the assistance of Mr. Vinoy Thomas in the preparation of
this article.
References
10. K. Katake, Journal of Kyoto Prefectual University of Medicine, 1961, 69, 463.
11. K. Katake, Journal of Kyoto Prefectual University of Medicine, 1961, 69, 484.
12. M. Takigawa, Journal of Kyoto Prefectual University of Medicine, 1953, 53, 915.
13. N. Yoshimatsu, Journal of Kyoto Prefectual University of Medicine, 1958, 64, 553.
15. M. Jayabalan and K. Rathinam, Clinical Materials, 1992, 11, 1-4, 179.
16. D. Wolter, Aktuelle Probleme Chirurgie und Orthopadie, 1983, 26, 28.
223
Handbook of Polymer Blends and Composites
19. W. D. Belangero, G. Koberle and W.A. Hadler, Brazilian Journal of Medical and
Biological Research, 1993, 26, 8, 819.
24. U. Weber and H. Rettig, Zeitschrift fur Orthopadie und ihre Grenzgebiete, 1979,
117, 268.
224
Fibre Reinforced Plastic Composites for Biomedical Applications
38. E. Sclippa and K. Piekarski, Journal of Biomedical Materials Research, 1973, 7, 59.
42. J. Stange and H. Mittelmeier, Biomedical Technology (Berlin), 1989, 34, 143.
47. S. Saha and S. Pal, Journal of Biomedical Materials Research, 1986, 20, 6, 817.
48. S. Saha and S. Pal, Journal of Biomedical Materials Research, 1983, 17, 6, 1041.
50. T. Hopf, S. Buttner and W. Brill, Zeitschrift fur Orthopaedics, 1989, 127, 248.
225
Handbook of Polymer Blends and Composites
59. P. W. R. Beaumont and R. J. Young, Journal of Materials Science, 1975, 10, 8, 1334.
67. J. Setz, W. Lin, Schweizer Monatsschrift fur Zahnmedizin, 1989, 99, 544.
226
Fibre Reinforced Plastic Composites for Biomedical Applications
68. N. Yazdanie and M. Mahood, Journal of Prosthetic Dentistry, 1985, 54, 543.
73. H. Kasuga, H. Sato and N. Nakabayashi, Shika Rikogaku Zasshi, 1980, 21, 64.
79. P. K. Vallittu and K. Narva, International Journal of Prosthodontics, 1997, 10, 142.
227
Handbook of Polymer Blends and Composites
96. C-D. Young, J-R. Wu and T-L. Tsou, Biomaterials, 1998, 19, 19, 1745.
97. M. Jayabalan and P. P. Lizymol, Polymer Degradation and Stability, 1997, 58, 3,
251.
98. M. Jayabalan, inventor; Sree Chitra Tirunal Institute for Medical Sciences &
Technology, assignee; Indian Patent 182366, 1995.
100. S. Milka, Chir Narzadow Ruchu Ortopedia Polska, 1993, 58, 453.
228
7
Composite Materials in the Nuclear and Space
Industries: Specific Applications
H. W. Bonin, V. T. Bui, D. Y. S. J. Pag and I. Miedema
7.1 Introduction
Polymers or polymer-based composites are scarcely used in the nuclear industry due to the
vulnerability of polymeric materials to different types of radiation such as gamma rays, high
energy electron, thermal neutron. However, those polymers whose backbone is built up with
aromatic rings such as epoxy, poly(ether ether ketone) (PEEK), show a good resistance to
high doses of radiation up to 10 MGy without undergoing appreciable degradation of their
mechanical properties. Considering this fact, a special application of composites based on
epoxy or PEEK is proposed for radioactive materials from the nuclear industry.
In the nuclear industry, the harsh environment to which the materials are submitted in
nuclear fission and fusion reactors appears as a strong deterrent to the use of polymer-
based composite materials, as most polymers have limited resistance to high
temperatures and intense radiation fields. However, the recent development of better
polymer adhesives, such as PEEK, has opened the doors for the increased uses of
composites for nuclear-related applications. There is another link between the nuclear
industry and polymer-based composites: the radiation processing of the polymers and
the composites [1, 2]. However this constitutes a separate case which is not specifically
addressed in the present text.
Several applications are identified where polymer-based composite materials could play
important roles in the nuclear industry, advantageously replacing metals and alloys which
often have only limited resistance, as neutron absorption considerations result in few
possible choices for the alloy compositions. Generally, polymers and the fibres used in
the fabrication of composite materials are made of atoms which have low neutron
absorption cross sections: beryllium, carbon, hydrogen, nitrogen, oxygen, among others.
For example, neutron economy is a primary factor in the design of CANDU heavy-water
moderated nuclear reactors, and a few zirconium-based alloys have been used or are
presently used for the fabrication of components such as the fuel cladding, the pressure
tubes and the calandria tubes (zirconium alloy tube surrounding the pressure tube). As
demonstrated by the rupture of a pressure tube in the Pickering 2 reactor several years
ago, these zirconium alloys have limited resistance to the hydridation phenomenon and,
until better materials are found, it is now agreed that CANDU reactors need to have
229
Handbook of Polymer Blends and Composites
their pressure tubes changed once in their lifetime. On the other hand, polymer-based
composite materials display the required qualities for such applications within power
reactor cores in terms of mechanical strength and resistance to corrosion and hydridation,
but no such composites have been designed offering suitable resistance to the harsh
radiation environment in a nuclear reactor core. Recent research and development has
however produced new composite materials that are more and more resistant to radiation.
These are several applications within the nuclear industry, and the space industry [3]
as well, for composite materials in environments characterised by sustained exposures
to radiation fields less intense than those found in the core of nuclear reactors. In the
space industry, the light weight of polymer-based composites allows their use to replace
metals and alloys on board spacecraft such as the space shuttles and the space stations.
In the nuclear industry, composites can be used to make tools intended for remote
handling radioactive materials, or for joints in flow systems where the fluid may be
contaminated by radioisotopes. But the primary applications of polymer-based
composite materials is in radioactive waste management, which is the area that will be
more thoroughly covered later.
7.2.1 General
Photons (gammas and X-rays) ionise atoms through three main processes [5]. The
photoelectric effect is mainly predominant for low energy photons (<0.05 MeV) and
230
Composite Materials in the Nuclear and Space Industries: Specific Applications
amounts to a total transfer of the photons energy to one of the electrons from the inner
shells (K or L mainly). As a result, this electron is ejected from the atom which is now
ionised, and a subsequent rearrangement of the remaining electrons within the electronic
shells causes the emission of x-rays and Auger electrons. The second mechanism is the
Compton scattering effect, also predominant for lower photon energies (0.1-10 MeV).
This amounts to a collision between the incident photon and one of the atoms electrons,
preferably those located on the valence shells and which is ejected from the atom. As a
result, the photon survives the collision, but with a reduced energy and in a direction
usually different than the incident direction. Both the scattered photon and the stripped
electron may have sufficient energies to ionise several more atoms.
Finally, the last interaction process is through the pair production mechanism, possible
only if the photons energy is equal to or greater than 1.022 MeV, i.e., twice the electrons
rest mass energy. This phenomenon occurs in the vicinity of a heavy nucleus and amounts
to the disappearance of the incident photon leading to the creation of an electron-positron
pair. Both particles are emitted in opposite directions and subsequently interact by ionising
more atoms. The interplay of all three interaction mechanisms results in a fairly uniform
photon energy deposition within the irradiated material, such as organic polymers in
which the Compton scattering effect is predominant for photon energies below 2.5 MeV.
Fast (1-2 MeV) and thermal (0.0253 eV) neutrons interact with atoms in a more
complicated fashion. Since their electrical charge is neutral, they are not technically ionising
radiation, but in practice they are, albeit indirectly. The ionisation is actually produced
by the recoil charged particles, mostly protons, from inelastic and elastic collisions between
the neutrons and light nuclei through a knockout process. Since the ionising power from
energetic charged ions such as protons is much larger than that of electrons, higher
damage is caused (indirectly) from the incident neutron. In addition, both fast and thermal
neutrons can be absorbed into the nuclei of atoms that are transmuted into (usually)
radioactive isotopes through the nuclear reaction. These radioisotopes then undergo
radioactive decay accompanied by the emission of beta and gamma particles. Depending
on the atoms present in the irradiated material, the induced radioactivity may remain
long after the end of the irradiation. However, for most of the polymers of concern here,
the majority of the atoms present, such as carbon, oxygen, nitrogen and hydrogen, have
either a low neutron absorption cross section or the activation products are short lived.
Consequently, the activation process is a major concern only when impurities (mostly
metallic) are present in significant quantities. The overall effect of the neutrons interacting
with matter can be summarised by the high ionisation powers of the recoil protons and
ions concentrating damages along specific tracks and in more localised sites.
231
Handbook of Polymer Blends and Composites
The level of damage caused by radiation depends not only on the quantity of the energy
deposited in the material (total dose absorbed) but also on how concentrated the energy
deposition is, creating free-radical pairs along the path taken by the energised particle or
photon [6]. Furthermore, irradiation conditions such as temperature and environment
(presence or not of oxygen or moisture) have an impact on the extent of damage. The
destructive potential of radiation is assessed in terms of linear energy transfer (LET)
which is the amount of energy deposited per unit length of the ionisation track, expressed
as dE/dx [7]. Polymer containing aromatic rings have displayed the highest radiation
resistance as the rings absorb and dissipate energy without bond disruption [8].
Radiation-induced chemically active species can react in different ways resulting in structural
and molecular weight changes. The first change is crosslinking, which increases the molecular
weight. This phenomenon can eventually result in the formation of an important network
which radically alters transition temperatures, chemical and mechanical properties of the
materials. The second change is chain scission, where free radicals recombine with low
molecular weight molecules, which causes a decrease in the overall molecular weight of the
material. It is not always clear as to which one will predominate; again, it is a matter of the
nature of the epoxy system and the irradiation conditions previously mentioned.
The third change is the creation of low molecular weight products resulting from chain
scission and followed by the abstraction and recombination reactions. These products are
usually gases such as H2, CO2, CO and CH4, but some polymers with particular structural
groups may release other gases like SO2, in the case of polysulphones, or HCl for
polyvinylchloride [9]. Such small products impact on the overall molecular weight and
create small gas pockets in the polymer which reduce mechanical properties. Furthermore,
they plasticise the polymer matrix at elevated temperatures and embrittle it at low
temperatures [10]. The last induced chemical change is the actual structural modifications
by the appearance of new bonds creating new chemical groups or the elimination of existing
ones. Such changes can alter both chemical and physical properties of the polymer.
The following is a simplified summary of the radiation interaction mechanisms. The first
step is to produce ionisation and excitation, both producing free radicals. The free radicals
will interact with the polymer, leading to crosslinking and chain scission:
232
Composite Materials in the Nuclear and Space Industries: Specific Applications
Pn + RA > X (crosslinking)
The presence of oxygen during irradiation is detrimental to the polymer. It acts as a free
radical scavenger, interfering with crosslinking. Peroxides are also formed in the process,
causing polymer chain breakdown into more free radicals. The latter will in turn produce
more peroxides, thus developing a vicious circle, with the end result of polymer
degradation.
When measuring the effect of radiation on a specific material, not only the energy and
the type of radiation must be considered but also the intensity of the emitting source.
Those parameters along with the distance from the source and the radiation absorbing
properties of the irradiated medium are used to measure the absorbed dose. The SI unit
for radiation intensity is the Becquerel (Bq) or 1 disintegration/s, the energy of the particles
is expressed in eV or MeV and the absorption properties are characterised with absorption
coefficients. The Gray (Gy) and the rad are the most commonly used dose units to measure
the amount of energy transferred by radiation to an irradiated material [9].
This section will deal with the effects of radiation on cured epoxies and epoxy/carbon
composites. Since epoxies can be made from numerous combinations, a baseline system
must be chosen. Based on a detailed review of works done on epoxies and epoxy/carbon
composites subjected to radiation, two systems are predominant: tetraglycidyl-
diaminodiphenylmethane (TGDDM) or diglycidyl-ether of bis-phenol A (DGEBPA) (cured
with either an aliphatic or aromatic agent). Most of the epoxy/carbon composites were
made using a carbon-fibre cloth or from commercially-prepared systems.
Using gamma and neutron radiation for a dose up to 10 MGy, in air, aromatic epoxies
showed higher resistance than aliphatic ones [8]. This is ascribed to the capability of the
233
Handbook of Polymer Blends and Composites
In practically all cases, 60Co gamma rays and electron beam were used, either in air or
vacuum, at 77 K (liquid nitrogen was used) and/or room temperature. Doses up to 170
MGy have been used, leading to radiation-induced degradation of certain mechanical
properties concerned [12]. Studies involving radiation from neutrons are limited.
Radiation effects have been assessed via the change in mechanical properties at 77 K
and room temperature through tensile and three-point bending (3BP) testing. Properties
such as Youngs modulus (E), tensile stress or strength (F), shear stress or strength (J),
and strain (g) give good indication of whether cross-linking or chain scission
predominated. Moreover, the mode of failure of a composite under stress also indicates
how the fibre/matrix interface was affected by the radiation and how the load transfer
is subsequently achieved.
In general, carbon (graphite) fibres are fairly resistant to radiation in vacuum, while
having variation in strength in the presence of air [13]. After receiving a dose of 50
MGy (from a 0.5 MeV source) in the absence of air, there was no change in strength
[14]. Irradiation with neutrons in air showed comparable results [11]. The presence of
air lead to two competing mechanisms regarding the strength: the displacement of
internal atoms which tends to strengthen the fibres by promoting bonding between
planes, and oxidation of the fibre surface, which tends to weaken them.
The main effect is the creation of free radicals leading to cross-linking and chain scission
competing simultaneously. Based on the collated irradiating conditions previously
mentioned, the effects of radiation on epoxies and epoxy/carbon composites may now
be examined by assessing specific mechanical properties, for doses up to 170 MGy. The
latter is chosen as a baseline reference for the following reasons:
It regroups the majority of the previous works done on epoxies and epoxy/carbon
composites;
It exceeds the total dose of 100 MGy received by a satellite in the Geosynchronous Earth
Orbit over its 30-year life expectancy;
It is well above the expected dose of 64 MGy received by a container for high level waste
(HLW) over its 500-year life expectancy at a calculated dose rate of 14.5 Gy/h [15, 16]; and
234
Composite Materials in the Nuclear and Space Industries: Specific Applications
The Youngs modulus was relatively unaffected. It remained practically constant: for
the epoxy, 8 GPa (77 K) and 4 GPa (room temperature); for the composite, 60 GPa (77
K and room temperature) [12]. Similar behaviour occurred when irradiation was done
under vacuum (with doses up to 3.5 MGy) [17]. Even after annealing in vacuum for 2
h at 180 C (with doses up to 20 MGy), it remained unchanged [18]. This is ascribed
to the radiation insensitiveness of the modulus of both the fibre and matrix materials.
The ultimate flexural strength fairly decreased with increasing dose. In the case of
epoxy, it went from 240 MPa (0 MGy) to 100 MPa (170 MGy), for both temperatures
[12]. For the composite, it went from 900 MPa (0 MGy) to 650 MPa (170 MGy) at
77 K, and remained constant at 900 MPa at room temperature. Post-annealing in
vacuum for 2 h at 180 C also caused similar behaviour for the composite: the strength
went from 800 MPa (0 MGy) to 500 MPa (20 MGy) (at room temperature) [18].
Annealing results in the activation of latent radiation damage, causing a decrease in
the load transfer capacity at the fibre/matrix interface. Thus, the degradation of the
composite is mainly determined by the radiation damage in the matrix rather than at
the fibre/matrix interface [12].
The ultimate strain of the epoxy varied from 3% (0 MGy) to 2.5% (170 MGy) at 77
K, and 6% (0 MGy) to 2.5% (170 MGy) at room temperature [12]. For the composite,
235
Handbook of Polymer Blends and Composites
it remained practically constant at 2% for the same dose range, for both temperatures.
By establishing a relationship between the composite ultimate strain and the matrix
ultimate strain, the dose dependence of the composite stress may be assessed by a
change in the matrix ultimate strain due to irradiation. This dependence is primarily
due to the dose independence of the Youngs modulus of the composite.
The shear stress (also found as the interlaminar shear stress, ILSS) of the composite
decreased from 33 MPa (0 MGy) to 23 MPa (140 MGy) at 77 K, whereas at room
temperature, it remained constant at 27 MPa for the same dose range [20]. This dose
independence of the ILSS may be explained by the strength of the fibre-matrix bond
due to the radiation insensitiveness of the friction force mode which contributes to
the total bond strength in carbon fibre composites. At room temperature, this mode
is much less sensitive to radiation compared with that of the chemical bond mode.
However, at 77 K, the ILSS becomes dose dependent and decreases with increasing
dose at that temperature [21]. The resulting behaviour may be ascribed to a change
in the matrix brittleness due to radiation, where the friction bond mode can only
work when the matrix resin is ductile to a certain extent.
It has been well documented that the effects of radiation, such as cross-linking or chain
scission in polymers, depend on the radiation dose but not on the dose rate nor on the
type of radiation [13]. This being the case, the study of the effects of neutrons is not as
extensive as that of gamma rays or electron beam. However, a comparison between the
effects of both types has been established [22]. As a result of the same energy deposition,
the decrease in an epoxy/carbon composite strength is 1.8 to 2.6 times higher for neutrons
than for gamma rays. This is because the radiation sensitivity of the epoxy is higher for
recoil protons following neutron interaction than for gamma rays. Subsequently, the
decomposition efficiency of the epoxy increases with an increase in the linear energy
transfer in the order gamma rays < recoil protons (from the neutrons), ascribed to a
difference in the microscopic non-homogeneity of energy deposition between these types
of radiation [23].
236
Composite Materials in the Nuclear and Space Industries: Specific Applications
number of high technology applications in the space, automotive, electrical and chemical
industries. PEEK is generally a tough, semi-crystalline thermoplastic. The potential of
PEEK for use in the nuclear and space industries was identified early and research work
was initiated by Yoda in 1983 with the study of the effect of high energy electron beam
radiation on the structure and physical properties of PEEK [24, 25]. Yoda observed that
amorphous PEEK suffered enough defects to be unable to crystallise after receiving doses
of 50 MGy. Reduction in crystallinity and melting temperature were observed upon
recrystallisation after the material had been exposed to an electron beam radiation dose
higher than 10 MGy. He also observed that with a radiation dose up to 50 MGy the
average crystallite size was reduced by 15% while the c-axis of the orthorhombic crystal
cell structure of PEEK remained mildly affected.
In the mid 1980s, Tenny and Slemp from NASA studied a number of polymeric materials
targeted for space applications, including PEEK composites [10]. Materials were exposed
to 1 MeV electrons at a maximum dose rate of 0.5 MGy/h for doses up to 100 MGy. An
increase in Youngs modulus was observed for PEEK/graphite composite. This was
confirmed by Sagusa and others [26] who observed that the elongation at break and the
tensile strength of both amorphous and semi-crystalline PEEK decrease while the Youngs
modulus increases for doses up to 50 MGy.
Hegazy and others [27] investigated the effect of gamma and electron beam irradiation
under vacuum on the gas evolution of several aromatic polymers at doses up to 25 MGy.
The two types of radiation seemed to generate comparable damages to the PEEK molecular
structure. Crystalline PEEK yielded less gas than amorphous PEEK, and displayed better
resistance to gamma radiation due to the presence of crystals. Carbon dioxide and carbon
monoxide evolution showed that the ether and ketone groups were the most affected by
radiation, however the evolution of hydrogen also attested to the occurrence of cross-
linking which is the creation of new intermolecular bonds. It was concluded that aryl
ether ketone linkages give a protective effect against radiation [28]. Crosslinking was
attributed predominantly to the amorphous phase between the crystal lamellae structure,
which increased the glass transition temperature while the melting and crystallisation
temperatures were lowered.
Vaughan and others [29] observed that PEEK crystals are slightly affected at levels up to
66 MGy of electron beam radiation. At higher doses changes are so important that only
a small fraction of crystals can crystallise at 260 MGy. It was proposed that the formation
of cross-links above 260 MGy can occur to such an extent that no melting is possible
and no endotherm can be detected by differential scanning calorimetry (DSC). However,
diffraction patterns and transmission electron microscopy (TEM) still indicate some local
ordering in the form of lamellae that could be associated with the crystalline structure of
PEEK [30].
237
Handbook of Polymer Blends and Composites
Sasuga and Hagiwara proposed a higher resistance to electron beam radiation for
amorphous PEEK than semi-crystalline PEEK based on dynamic relaxation results [31-
32]. Important variations on the relaxation behaviour were observed at the -transition
(-100 C), -transition (40 C), -transition (140 C) and -transition (180 C). It
was suggested that the disintegration of tie-molecules at the amorphous/crystalline
interface would explain the lesser radiation resistance of semi-crystalline PEEK.
Past research has concentrated mostly on the study of the effects of high energy electron
beams and gamma radiation on PEEK in vacuo or in air. Previous work on epoxy and
polyimide composites clearly demonstrated that the mechanical properties of the studied
materials were more sensitive to neutron radiation as opposed to an equivalent physical
dose of gamma radiation [33]. No research has been carried out on the effects of fast
and thermal neutrons on the mechanical properties of PEEK, especially in a reactor
pool environment. Since PEEK is known to be a radiation-resistant polymer, such
information may be valuable to assess its potential for applications in nuclear and
space industries.
7.4.1.1 Experimental
Two industrial grade PEEK resins namely the 450P (molecular weight (MW) 43 000,
molecular weight distribution (MWD) 2.5) and 150P (MW 14 000, MWD 2.5) were
supplied by Victrex USA. They were dried then processed on an Engel 55 injection moulder
to produce tensile test samples which were then annealed up to 200 C for seven hours
to relieve internal stress. Details of the sample processing are reported elsewhere [34].
Some processed samples were selected and submitted to neutron activation analysis (NAA)
to determine the impurity content and also establish the level of induced radiation in the
irradiated material in order to set safety handling procedures.
The tensile test samples were irradiated with a combined gamma, electron, and neutron
radiation flux emitted from the SLOWPOKE-II reactor. The radiation doses ranged
from 0.1 MGy to 3.4 MGy. Samples were wrapped in polyethylene bags to limit water
circulation from the reactor pool and provide standard flow conditions around the
specimens [34].
238
Composite Materials in the Nuclear and Space Industries: Specific Applications
The morphological and thermal properties of non-irradiated and irradiated PEEK samples
were characterised by density measurement, DSC and x-ray diffraction. The tensile
properties of samples were determined at room temperature using an Instron tensile
tester model 4206 equipped with the Series IX Automated Materials Testing System
V.5.25. The tensile tests followed the ASTM D638 standard [34].
The density of the tensile specimens of PEEK before and after irradiation were measured
by immersion at room temperature in distilled water following the ASTM D792-00 standard
[35]. The results are presented in Figure 7.1 for the sample of PEEK 150P. Similar results
were obtained for the PEEK 450P sample. Both PEEK grades indicate a slight increase in
density with dosage. This increase may be caused by a morphological change in the polymer
structure, mostly by some degree of cross-linking occurring in the amorphous phase. It is
also noted that the MW of PEEK had some effect on the change of density of irradiated
samples. The Tg, the melting temperature (Tm) and the crystallisation temperature (Tc),
obtained from DSC measurements for both PEEK samples, remained practically unaffected
by irradiation at the studied doses. This is likely to indicate that structural changes induced
by radiation were too small to produce detectable variations in the macroscopic thermal
properties of irradiated material. Figure 7.2 illustrates the Tg values of PEEK 150P as a
function of dose. The crystallinity of irradiated samples was determined by three different
techniques: density, DSC, and x-ray diffraction. The results shown in Figure 7.3 for PEEK
150P indicate no appreciable variations detected outside of experimental uncertainties.
239
Handbook of Polymer Blends and Composites
Figure 7.2 Glass transition temperature as a function of dose for PEEK 150P
This stability of the crystallinity in both PEEK samples confirms that the morphological
changes only occurred in the amorphous phase or at the interfaces between the amorphous
and crystalline phases.
The Youngs modulus and tensile strength of non-irradiated and irradiated PEEK 150P
samples are presented in Figures 7.4 and 7.5, respectively. Similar results were also obtained
for the PEEK 450P grade. The results of non-irradiated specimens are comparable with
those from literature [34]. A slight increase in both Youngs modulus and tensile strength
of irradiated samples probably correspond to some degree of crosslinking in the
240
Composite Materials in the Nuclear and Space Industries: Specific Applications
Adhesives based on epoxy resins are widely used for bonding metals, concrete, and a
whole range of nonporous materials [36]. This is due to their ease of application and the
very high strengths of the joints that they form. The use of epoxy adhesives in radiation-
241
Handbook of Polymer Blends and Composites
intense environments requires previous knowledge of how radiation can affect the cohesive
strength of the adhesive as well as the adhesive strength of the joint. In a recent work
[37], we reported the results of the radiation effects on aluminum-epoxy adhesive joints.
7.4.2.1 Experimental
Two different types of epoxy adhesives were used: the first one was a fast-curing adhesive
from Cole-Parmer, and the second one was a long-curing adhesive from Devcon, which
contained many additives. The preparation of the adhesive joints and the testing procedure
followed rigorously the ASTM D897 standard [38]. The radiation processing was similar
to that used with PEEK samples as described above. The studied radiation doses were up
to about 6.0 MGy.
The results obtained indicate that the cohesive properties as well as the adhesive strength of
the aluminum-epoxy joints made from both types of adhesives, are significantly affected by
the radiation produced from the SLOWPOKE-2 reactor, which consists of high-energy electron,
gamma, and thermal neutron radiation. In both cases, an initial reduction of the tensile
strength is observed, followed by a recovery for the Devcon adhesive, or a stabilisation for
the Cole-Parmer adhesive, as irradiation progressed. At the maximum of the studied dose,
the adhesive strength of the joints diminished by about 50% of the original values. This large
radiation effect on the adhesive strength is believed to be caused by the sensitivity to radiation
of the tie molecules on the interface between aluminum surface and epoxy resin.
In a recent work, we have shown that cured epoxy can be toughened by a great extent by
incorporating a polyurethane (PU) phase, of about 20% in weight, via the interpenetrating
network method [39]. It is now interesting to know how the presence of the PU phase
could modify the adhesive properties of cured epoxy and then how the irradiation could
affect the aluminum-epoxy based interpenetrating polymer network (IPN) joint.
7.4.3.1 Experimental
Epoxy resin (DGEBPA) with an average MW of about 1000, and hardener, were supplied
by Dow Chemicals Ltd., Canada. The PU phase was formed by reacting castor oil with
242
Composite Materials in the Nuclear and Space Industries: Specific Applications
tolylene diisocyanate (TDI) where the NCO:OH ratio was kept constant at 1.50. The
preparation of the epoxy/PU IPN followed a sequential method. Details of the synthesis
method are reported elsewhere [40].
Preliminary results [40] show that the tensile strength of the adhesive joint between
aluminum and toughened epoxy (or epoxy/PU at 17:3:15 IPN) was larger than that of
cured epoxy. This synergistic effect on the adhesive properties seems to be similar to
that observed on the bulk tensile and impact properties of the epoxy/PU IPN itself,
such as previously reported and interpreted [39]. The main cause for that effect was
attributed to the mutual enhancement of both epoxy and PU phases via their interlocking
by means of crosslinking. The irradiation of the joints by the same method as described
in Section 7.4.2.1 up to a dose of about 0.5 MGy did not appreciably degrade the
adhesive strength of the joints.
The problem of the disposal of radioactive wastes is separated into several components,
either from the radiation intensity viewpoint, and/or the retrievability aspect.
Radioactive wastes are usually categorised into high level, intermediate level and low
level activity wastes. The approach to the storage and ultimate disposal of the wastes
is determined by the activity of the main radioisotopes present in the wastes which is
the main source of hazard.
HLW are mostly made of spent fission reactor fuel and the discarded effluents from
spent fuel reprocessing which contain most of the fission products. Other components
making up the high level radioactive wastes include discarded radioisotopic sources (such
as 60Co from irradiators) and highly neutron irradiated or contaminated materials. The
problems for this type of radioactive waste rise both from the intense radiation fields
emitted by these materials and the duration of the emission, since several radioisotopes
take centuries and even millenia to decay to safe levels. The usual approach consists of
letting the spent nuclear fuel decay for at least one year in large water-filled pools at the
nuclear power plant, then place the spent fuel and other high level wastes in leak proof
containers that are placed at the disposal site for long-term storage.
243
Handbook of Polymer Blends and Composites
Then comes the question of retrievability. One school of thought proposes that this dangerous
material be kept isolated from the biosphere forever, in leak-tight containers, placed deep
underground in sites provided with multiple engineered barriers intended to prevent the
underground water from seeping into the container and leaching out the radioisotopes,
preventing transport of them back to the biosphere, for all the time necessary for the
radioisotopes to decay back to stable elements. Appropriate burial sites could be geological
formations which are known for their stability, such as the so-called plutons within the
Canadian Shield, these granitic rock formations being essentially monoliths that have survived
intact the many geological upheavals in the last several billions of years. On the other hand,
another school of thought considers that the spent nuclear fuel still contains valuable isotopes,
such as fissile uranium and plutonium which can provide additional electrical energy through
fuel recycling, and several fission products which can become, once separated, valuable
resources in research, industry and nuclear medicine particularly. In this case, surface or
shallow storage is proposed, for which the leak-proof container would be placed inside a
concrete silo, again with multiple engineered barriers, but with continuing monitoring
provided. The latter question poses the problem of the capability of the present and future
societies to provide adequate monitoring for the several centuries in the future.
As for the intermediate and low level wastes, the problem is much simpler than for the
other wastes since the radiation fields are much less than for the high level wastes, and
the necessary storage times are also less, although in some cases the storage facility must
be conceived to last several centuries. The disposal approaches foreseen here are again,
underground sites, deep or shallow, and surface in concrete silos. As for the HLW, there
is a need for leak-proof containers. The methods explained above are not the only ones
conceivable to solve the radioactive waste management problem. With nuclear reactors,
both fission and, in particular, fusion reactors, the high neutron fluxes produced could
be used to transmute the long-lived radioisotopes into short-lived ones which could decay
rapidly into stable species. Other disposal methods are proposed in which the wastes
could be transported into outer space by spacecrafts, the most appropriate method being
to send the wastes into the Sun where the material would be destroyed by the Suns high
temperature and neutron flux. This solution is obviously too risky and too expensive
today, but it is conceivable that it may become quite feasible within the 21st century, as
the technological development needed is well within our reach.
The container proposed by Atomic Energy of Canada Limited (AECL) [15] for the deep
underground disposal of spent nuclear fuel foresees the use of metals such as copper or
244
Composite Materials in the Nuclear and Space Industries: Specific Applications
titanium-based alloys. The same is also foreseen for the containers for the other high,
intermediate and low level radioactive wastes. However, in view of the rapid development
of radiation resistant polymer-based composites, the authors initiated research several
years ago on the application of such advanced materials as a solution to the radioactive
waste management problem. At first, the dimensions of the AECL proposed metal
container were used, such that the container would contain 72 CANDU reactor spent
fuel bundles: the height is 2.246 m and the diameter is 0.633 m, with a thickness of
0.00635 m for a titanium alloy wall. These dimensions were retained for the case studies
carried out by the authors on the application of polymer-based composite materials as
replacements for the titanium alloy. Only the thickness of the wall was changed in order
to meet mechanical strength constraints.
Several combinations of polymer matrix and reinforcement fibre were investigated: the
polymers selected were polystyrene (PS), polymethyl methacrylate (PMMA), epoxy and
PEEK, and the reinforcing fibre retained was boron fibre in either a 50% or a 70% (mass
percent) mixture. Even though PS and mostly PMMA are not highly radiation resistant
polymers, they are used here as reference materials for the purpose of comparison in our
study. In its proposal, AECL had chosen glass beads as the filling material inside the
container which is intended to permit heat transfer from the decaying CANDU fuel
bundles to the environment. As the studies show, glass beads only offer a very limited
radiation attenuation between the radiation emitting nuclides in the fuel and the container
wall. The investigation carried out by the authors has considered the use of compacted
(or even sintered) thorium dioxide powder to replace the glass beads and has demonstrated
that this material offers excellent protection to the container walls against the radiation
from the spent fuel.
The mechanical stress applied to the container is assessed on the basis of a 13 MPa
constant external mechanical pressure, which represents the conditions of deep
underground storage as assessed in the AECL proposal [15]. The calculations aimed at
determining the minimal wall thickness that would resist such external pressures for an
empty container, and consisted of a four-step procedure, the first being based on solving
the usual axial and hoop stress equations for the conditions stated above; in the second
step, a reversal of external pressure and compressive properties was executed, using
Poissons ratio for the various materials, in the third step, the total strength of the
composite material was found by a volume-weighted average of the strength of the fibre
and that of the polymer, and finally, in the fourth step, the pressure calculations accounted
for the load applied to the container during transportation of the container containing
the 72 CANDU fuel bundles and the filling material to the repository vault. When thorium
dioxide is used as the filler, the total weight that must be supported by the container is
1880 kg. The results of these calculations appear in Table 7.1.
245
Handbook of Polymer Blends and Composites
A key factor for the applicability of polymer-based composite materials for the
fabrication of a suitable container for the long-term storage and/or disposal of spent
nuclear fuel and radioactive wastes is the resistance of these materials to the radiation
emitted by the wastes. It was most important to simulate on computer, the container
being irradiated over prolonged periods of time. The MicroShield Version 3.0 [41] in
order to determine the maximum dose rate (at mid-plane of the container) from the 72
spent CANDU fuel bundles which were assumed to have been placed in the container
after a ten-year cool down period in the fuel storage pool at the nuclear power generation
station following discharge from the reactor core.
For these calculations, the most important radioisotopes in terms of activities were
included in the source term of the simulation model, as determined from depletion
246
Composite Materials in the Nuclear and Space Industries: Specific Applications
calculations carried for a standard 37-rod Bruce A fuel bundle type. The fuel bundle
in these calculations has been submitted during its 228.2 day residence time in the
CANDU reactor to a thermal neutron flux of 1.26 x 1014 neutrons cm-2 s-1, for an
accumulated burn-up of 685 GJ kg-1 (initial U), or 7928 MW-day per tonne initial U.
The activities of these radioisotopes were thus taken from [42] and the actinides having
an activity exceeding 10-5 Ci kg-1 after ten year decay were included in the simulation
model along with the fission products for which the activity was more than10-3 Ci kg-1.
The MicroShield simulation was done for two filling materials: glass beads and thorium
dioxide. Typically, with glass beads as the filler, dose rates of 11 Gy h-1 were calculated
in the container wall, and only 0.05 Gy h-1 were calculated when thorium dioxide was
used as the filling material. Calculations with air as filling material were also done and
yielded a dose rate of 19 Gy h-1. Tables 7.2, 7.3 and 7.4 present the details of these
simulations, which were carried out using conservative assumptions for the various
options possible, such as the representation of the build-up factor. An additional
simulation was performed in which the natural radioactivity for thorium was accounted
for and the effects on the dose rate in the container wall were negligible indeed. The
negative doses absorbed in the container wall are due to uncertainties with the
MicroShield calculations, mostly coming from the build-up effects.
Samples of the four types of polymers were irradiated in the pool of the SLOWPOKE-2
nuclear research reactor at Royal Military College of Canada in order to determine how
they would be capable of sustaining doses comparable to those determined in the previous
section. The dose rate from a mixed radiation field (fast and thermal neutrons, gamma
Table 7.2 Calculated dose rates in container wall: titanium and copper
Container Filler material Dose rate Dose in Absorbed dose
material inside wall outside wall rate wall
Gy h-1 Gy h-1 Gy h-1
Titanium None 14.5 15.2 -0.66
Titanium Glass beads 11.2 7.14 4.02
Titanium ThO2 0.0517 0.0573 -0.00565
Copper None 14.5 4.98 9.5
Copper Glass beads 11.2 1.83 9.34
Copper ThO2 0.0517 0.014 0.0379
247
Handbook of Polymer Blends and Composites
radiation and x-rays, energetic electrons and protons) was determined in a separate study
to be 3.7 1.0 x 104 Gy h-l at half reactor power [43, 44]. The irradiation times in the
SLOWPOKE-2 reactor pool varied from 8 to 80 hours for dog bone shaped polymer
samples and various comparative tests were done on both non-irradiated and irradiated
samples [45-49]. The doses accumulated within the samples during their irradiation in
the reactor pool correspond to a 50 year storage time in the underground vault when the
248
Composite Materials in the Nuclear and Space Industries: Specific Applications
Table 7.4 Calculated dose rates in container wall: epoxy and PEEK
Container Filler material Dose rate Dose rate Absorbed dose
material inside wall outside wall in wall
Gy h-1 Gy h-1 Gy h-1
EPOXY -
None 14.5 19.1 -4.58
50% boron
EPOXY -
Glass beads 11.2 10.5 -0.67
50% boron
EPOXY -
ThO 2 0.0517 0.0871 -0.0354
50% boron
EPOXY -
None 14.5 19.3 -4.77
70% boron
EPOXY -
Glass beads 11.2 11.0 -0.13
70% boron
EPOXY -
ThO2 0.0517 0.0924 -0.407
70% boron
PEEK -
None 14.5 19.0 -4.46
50% boron
PEEK -
Glass beads 11.2 11.2 0.16
50% boron
PEEK -
ThO2 0.0517 0.0923 -0.0406
50% boron
PEEK -
None 14.5 19.1 -4.65
70% boron
PEEK -
Glass beads 11.2 11.3 -0.18
70% boron
PEEK -
ThO2 0.0517 0.0957 -0.0440
70% boron
filling material is glass beads, and to 10,700 years storage time when thorium dioxide is
used, if the initial source strength activity is kept constant. However, when radioactive
decay of the actinides and the fission products is accounted for, the corresponding storage
times are 60 years and infinity, respectively, for the glass beads and the thorium dioxide
filling materials.
249
Handbook of Polymer Blends and Composites
While the PS and the PMMA displayed low resistance to radiations, the epoxy actually
increased its tensile strength from 34 MPa to 44 MPa after 8 hours of irradiation, due to a
predominant cross-linking process. However, after 120 hours irradiation, the tensile strength
had decreased significantly due to chain scission being the dominant phenomenon. The
various grades of PEEK tested displayed much better resistance to radiation, with little
change or even increase observed in the tensile strength after irradiations as high as 83.6 h.
These values have been observed for samples in contact with water during their irradiation,
which represents an extreme case since the presence of highly oxygenated substances promote
chain scission during irradiation, thus decreasing the resistance to radiation.
Consideration must be given to heat transfer within the container since high level
radioactive wastes, the spent fuel in particular, continue to produce heat as they undergo
radioactive decay. According to the AECL study, a maximum temperature at the
container wall must be set as not to exceed 100 C in order to protect the integrity of
the environment. In the case of polymer-based composites for the container wall, such
a temperature maximum also applies for an additional reason: the Tg of the polymer
must not be exceeded, as the mechanical properties of the material would be adversely
affected. Both experimental and numerical data were obtained [50], from measurements
on a scaled-down model of the container, and using a computer simulation model.
Table 7.5 shows the results on the numerical model, both for glass beads and thorium
dioxide used as filling materials. The numerical model allowed for sensitivity analysis,
thus accounting for the uncertainties in the various heat transfer coefficients used for
the model. Again these results show that thorium dioxide is significantly more desirable
as a filling material, since its better heat transfer properties allow much lower
temperatures at the container wall, a major point in favour of using polymer-based
composites for the fabrication of the container.
The results of the various analyses carried out have been estimated with a conservative
approach as to their accuracy. The MicroShield-based simulations confidence has been
evaluated as 20%, while the results from the various irradiation tests in the
SLOWPOKE-2 reactor pool are estimated with a 5% accuracy or better. The dose-to-
flux conversion factor remains a difficult parameter to evaluate, and a conservative
analysis yielded a 28% uncertainty on the dose values. As for the heat transfer analysis,
a 22% uncertainty has been assessed on the results, mostly due to the confidence in the
heat transfer coefficients.
The work done so far clearly demonstrates that at least two of the polymer-based
composite materials studied are promising candidates to replace metals in the
250
Composite Materials in the Nuclear and Space Industries: Specific Applications
251
Handbook of Polymer Blends and Composites
fabrication of the radioactive waste storage containers: the epoxy and the PEEK.
Shielding the container wall against radiation has proven an easy task with the use of
thorium dioxide as filling material. Since this material has little usage in the industry
in general and it is discarded as a residue from uranium refining, its cost has to be
relatively low, even in the sintered ceramic form. Obviously, additional investigation
is needed to fully determine the appropriateness of polymer-based composites as a
suitable material for the fabrication of the radioactive waste storage containers.
Presently this work continues at RMC, and two versions of the container are being
studied, one for HLW, and the other for intermediate and low level wastes. Polymer-
based composite materials may be selected for the fabrication of radioactive waste
long-term storage only if it is fully demonstrated that this material will retain its
integrity over such long periods of time. This means that further testing and validation
must be done, not only to assess the resistance to radiations, but also the resistance
to chemical attack even by weak chemical agents such as slightly acidic or basic
ground water. For this, suitable time acceleration experiments must be designed and
implemented to test the various materials, and suitable interpretation methods of the
results must be designed and validated.
Polymer-based composite materials are a rather novel invention and the applications
of such materials are only in their beginning phase. Lately, an increasing number of
polymers have been conceived displaying suitable resistance to radiation, such that
tools or systems for applications in environments characterised by radiation fields
can be designed with polymer-based composites for their fabrication. While the
resistance to radiation and high temperatures still precludes polymer-based composites
from uses inside the core of fission or fusion nuclear reactors, the fast evolution in
the development of such composites makes it conceivable that in a near future such
applications will become a reality. In the mean time, these novel materials are at a
point such that some of them already appear suitable for applications within the
nuclear industry and the space industry, at least where the radiation fields are moderate
or low. In the present chapter, the authors present an overview of their studies on
radioactive waste storage containers as a example of a possible application. Much
research and development remain to be carried out, but some specific superior
properties of polymer-based composite materials for applications in the nuclear and
space industries (with respect to other classes of materials) indeed justify the research
and development efforts.
252
Composite Materials in the Nuclear and Space Industries: Specific Applications
APPENDIX
Materials
The PEEK/Graphite composite used in this study was obtained from Applied Fiber Systems,
Clearwater, FL, USA. This material consists of woven continuous graphite fibre in a 2 x
2 twill in a matrix of Victrex grade 150 PEEK. This polymer has a reported Mn and
MWD of 14 000 and 2.5, respectively. The graphite, grade HTA G30-500, was obtained
in strand form from Toho Besfight Carbon Fibers. The specimens were obtained from
the manufacturer in nominal 2.5 x 7.5 x 0.3 cm coupons. These samples were cut from
a 45 x 45 cm panel produced by stacking plies of woven fabric pre-impregnated with
PEEK resin powder and compression moulded into a consolidated sheet. The coupons
would become the samples to be irradiated.
Experimental
During the irradiation sequences, the composite samples were placed beside the
SLOWPOKE-2 reactor core in the reactor pool, at 20 C, in a specially designed support.
The highest dose for the samples was 107 Gy (or 10 MGy). The radiation effects were
evaluated using flexural testing, and wide angle X-ray scattering (WAXS). The flexural
tests were conducted on an Instron Model 4204 in accordance with ASTM D790 [51],
using a 16:1 span-to-depth ratio, at room temperature. WAXS measurements were
performed on a Scintag Diffractometer with a q-q geometry goniometer coupled with
DMS2000 Diffraction Management System software from which crystallite sizes of the
polymeric matrix could be determined.
Table 7.6 summarises the flexural strength, flexural modulus and crystallite size of the
composite samples as a function of irradiation dose. The standard deviations were obtained
from results of five specimens for each dose. It is expected that the carbon fibre is
unaffected by the radiation because of the relatively small cross-section of the neutron.
The main effects to be observed would be in the polymeric matrix, the PEEK resin. The
changes in flexural strength and modulus both remain within experimental errors. Upon
radiation exposure polymers tend to display a molecular reorganisation, such as
253
Handbook of Polymer Blends and Composites
crosslinking and chain scission, which leads to changes in the mechanical properties
among others. However, because the matrix has such a small contribution to the expressed
mechanical properties in a composite , the changes are not seen. Alternatively, the results
obtained do show that even at a high dose of 10 MGy, the traditionally vulnerable
interfacial adhesion between the fibres and matrix seems to resist the mixed field radiation
well. The WAXS results show an almost 55% drop in crystallite size at the highest dose,
this leads to speculation that the crystals are being split, leaving behind significantly
smaller crystalline regions within the matrix. This can also be qualitatively observed on
the scattering intensities from the WAXS patterns.
254
Composite Materials in the Nuclear and Space Industries: Specific Applications
References
255
Handbook of Polymer Blends and Composites
19. C. E. Kladunde, R. R. Coltman Jr., Journal of Nuclear Materials, 1983, 117, 345.
27. E-S. A. Hegazy, T. Sasuga, M. Nishii and T. Seguchi, Polymer, 1992, 33, 14,
2897.
28. E-S. A. Hegazy, T. Sasuga, M. Nishii and T. Seguchi, Polymer, 1992, 33, 14,
2904.
256
Composite Materials in the Nuclear and Space Industries: Specific Applications
35. ASTM D792-A, Standard Test Methods for Density and Specific Gravity
(Relative Density) of Plastics by Displacement, 2000.
36. H. Lee and N. Neville, Epoxy Resins, Their Applications and Technology, Mc-
GrawHill, New York, 1957.
38. ASTM D897, Standard Test Method for Tensile Properties of Adhesive Bonds,
2000.
39. M-P. Raymond and V. T. Bui, Journal of Applied Polymer Science, 1998, 70, 9,
1649.
41. MicroShield: Version 3.0 Users Manual, Grove Engineering Inc., Olney, MD,
USA, 1996.
257
Handbook of Polymer Blends and Composites
45. P. E. Poirier, Effet des Radiations sur les Epoxies pour Composites Polymriques,
Document #RMC-CCE-CM417-96- 16, Royal Military College of Canada, 1995.
46. H. W. Bonin, V. T. Bui and E. Poirier, Presented at the 16th Annual Conference of
the Canadian Nuclear Society, Saskatoon, Saskatchewan, Canada, June 1995,
Volume I.
49. D. J. Y. S. Page, H. W. Bonin, V. T. Bui and P. Bates, Presented at the 18th Annual
Conference of the Canadian Nuclear Society, Toronto, Ontario, Canada, 1997.
51. ASTM D790-00, Standard Test Method for Flexural Properties of Unreinforced
and Reinforced Plastics and Electrical Insulating Materials, 2000.
258
8
Advanced Composites for Offshore Developments
Mamdouh M. Salama
Composite materials offer several advantages for marine construction because of their
low density, corrosion resistance, and excellent fatigue performance. In addition, the
use of composites allows for greater design flexibility by tailoring the properties to
meet specific design requirements, thus promoting better system-oriented solutions.
On a one-to-one replacement basis composite components are often more expensive
than their steel counterpart. However, on a performance equated basis, the economic
incentive to use composite components can often be demonstrated based on their
capability to reduce system and life cycle costs. In recognition of these potential
advantages, use of composites has shown phenomenal growth in the development
activities of several offshore structural systems, particularly those that are water depth
sensitive such as mooring and risers. This chapter summarises the status of composite
development for offshore use, presents financial incentives to develop these composite
applications, identifies the main challenges facing the introduction of composites into
service and discusses steps being taken to address these issues. In addition, the market
potential for these composite systems is estimated. Information on composite materials
that are candidates for use in the offshore industry are also discussed. This information
includes discussion of material properties, fabrication methods, design philosophy, safety
requirements and regulation issues.
8.1 Introduction
The deepwater successes of several operators, particularly in the Gulf of Mexico (USA)
and in the Campos basin (Brazil), have created great interest in exploration and development
of oil and gas reserves in deepwater. This interest now extends to deepwater basins in the
North Sea, West Africa, and the South China Sea where deepwater developments are
currently underway. Figure 8.1 shows the historical development of both exploration and
development as a function of water depth. It demonstrates the continued oil industry interest
in deepwater, even at time of uncertain and low oil prices. While the equivalent of over 30
billion barrels of oil has been found in deepwater (>500 m), less than a third of the discovered
deepwater reserves has been developed. Deepwater developments are, however, unforgiving:
physically, technically and economically. There is no room for failure in any area: reservoir
259
Handbook of Polymer Blends and Composites
Figure 8.1 Historical water depth exploration and development of hydrocarbon fields
As part of the oil industrys efforts to reduce the life-cycle costs of deepwater developments
and to improve reliability, considerable attention is being devoted to the evaluation and
application of innovative and cost-competitive composite materials. Although on a one-on-
one basis, components made of composites will most likely be more expensive than an identical
steel counterpart; on a performance equated basis, the economic incentive of composite
components can be demonstrated by their impact on reducing system and full cycle costs.
Composites offer many potential advantages based on their low density, corrosion resistance,
and excellent fatigue performance. In addition, the use of composites permits greater design
flexibility for tailoring properties to meet specific design requirements, thus promoting better
system-oriented, cost-effective solutions. Also, new innovations are being developed to embed
fibre optics and electrical conductors into the composite part to monitor manufacturing
quality, structural integrity, service loads, and operational conditions from remote locations.
260
Advanced Composites for Offshore Developments
Applications of advanced composites can often draw on the knowledge and extensive databases
developed by the defense and aerospace industries. Any successful development must include
extensive economic analysis to ensure that the results are aligned with the needs of the oil
industry, particularly deepwater development. This chapter presents information on offshore
development and on composite materials that are candidates for use in the offshore industry.
Discussion of material properties, fabrication methods, design philosophy, safety requirements
and regulation issues are also presented. The current applications of composites in the marine
and offshore industries are reviewed. Future applications of composites in the offshore industry
and the status of their development are discussed. The financial incentives for developing
these applications are highlighted. In addition, estimates of the market potential for these
composite systems for offshore applications are presented.
261
Handbook of Polymer Blends and Composites
of a floating structure that is moored in place by catenary, taut or tensioned mooring lines.
These concepts include tension leg platform (TLP), SPAR (a large cylindrical buoy that
carries the deck and is anchored to the seabed by cables), and floating production units
with and without storage (FPS and FPSO) as shown in Figure 8.2. The TLP and Spar
concepts are considered the most attractive for deepwater development because they allow
for easy access to the wells. A typical well construction involves a set of concentric tubes
that extends from the reservoir to the seabed. The inner most tube is called production
tubing in which the oil or gas flows. The outer sets of pipes are called casings and are
cemented to each other and to the drilled hole. The casings terminate at the seabed. The
tubing extends from the seabed to the surface. In order to ensure against tubing leaks to
the sea, the tubing is surrounded by either one or two pipes that extend the casings from
the seabed to the surface as shown in Figure 8.3. These pipes are called production risers.
During drilling, the drill pipe is surrounded by a pipe that is called a drilling riser. The
drilling riser allows the drilling fluids containing drill cuttings to be transported from the
seabed to the surface. The diameter of the production risers is in the range of 0.2 to 0.3 m
and the drilling riser diameter is in the range of 0.5 to 0.6 m. The diameter of the production
tubing is in the range of 7.5 to 12.5 cm. All these pipes are designed to withstand the
internal pressure that may vary between 25 to 70 MPa; the external pressure that depends
on the water depth, and tension load is generally 1.5 to 2.0 times their weight in water.
This riser tension will be supported by the floating structure. As the water depth increases,
the weight and the tension requirement of the risers will increase. Similar to risers, the
weight and load imposed on the platform by the mooring system will also increase as the
262
Advanced Composites for Offshore Developments
water depth increases. This makes both risers and mooring lines suitable candidates for the
application of lightweight materials such as composites.
The primary purpose of these floating structures is to support the topside payload
which includes the weight of the accommodation module, helideck, production and
process facilities, and drilling equipment. In addition, the hull of these structures must
support the deck structure and the tension loads associated with the production and
drilling risers and any mooring system. A reduction in topside weight will reduce the
size, weight and cost of the supporting structure (deck and hull) with the savings premium
dependent on the type and size of the structure and design parameters including
environmental loads. There is a significant economic incentive to reduce the topside
weight through the use of composites. The ranges of payload premiums, that are being
considered by the industry as a basis for assessing the cost effectiveness of weight reduction
options, are $9 to 16/kg of payload for TLPs, $7 to 11/kg for semi-submersibles, $2 to
7/kg for ships, and $2 to 4/kg for SPARS [1]. Figure 8.4 shows the impact of payload
reduction on the cost of the different systems and on the cumulative cost of a TLP. While
the riser tension for a TLP is supported by the TLP hull, the riser tension of the SPAR is
supported by external buoyancy tanks. Therefore, the premium for the riser tension for
a SPAR will be similar to that for a TLP, i.e., $9-16/kg.
263
Handbook of Polymer Blends and Composites
Taking into account the weight savings resulting from using composite risers, several
case studies for deepwater development showed that the saving resulting from using
composite risers instead of steel risers rated for 25 MPa pressure is 5% for 1000 m water
depth and 21% for 3000 m water depth. When the design pressure increases to 50 MPa,
the saving is 10% in 1000 m water depth and 28% in 3000 m [2]. These economic
advantages take only weight into consideration. However, in addition to the weight
savings, composite risers provide excellent thermal insulation that is necessary to maintain
production by preventing plugging due to paraffin, asphaltene and hydrate depositions
as shown in Figure 8.5. Figure 8.6 shows a comparison between the steady state oil
temperature profile from perforations to topside for steel, steel with 0.6 cm insulation,
and composite risers [3]. The figure shows that only composites will allow the oil to
reach the topside, at 1500 m above the seabed, while its temperature is higher than the
wax deposition temperature. If steel is to be used in this case, it would require a minimum
of 2.5 cm insulation, which greatly increases the cost of a steel riser.
Figure 8.5 Examples of paraffin and hydrates that can cause offshore flow
assurance problems
264
Advanced Composites for Offshore Developments
Figure 8.6 Steady state oil temperature profile from perforations to topside for bare
steel, steel with 0.6 cm rubber insulation, and composite risers
265
Handbook of Polymer Blends and Composites
Because of the potential savings from reducing the weight of offshore structures, extensive
effort has been devoted to expanding the use of composite materials. In this chapter, the
definition of composite materials is limited to fibre-reinforced polymer (FRP) materials.
Applications where metal matrix composites and particulate composites may be used are
not addressed. FRP materials consist of small diameter fibres of high strength and modulus
embedded in a matrix with bonded interfaces which permit the fibres and matrix to form
a new material which captures the desirable characteristics of each. The fibres are the main
load-carrying member while the matrix maintains the fibres in the preferred orientation,
transfers the load onto the fibres and protects the fibres from the surrounding environment.
The most common fibres are glass, carbon and aramid (aromatic polyamide fibres,
commonly DuPonts Kevlar). Other fibres such as silicon carbide, boron and aluminum
oxide may have superior properties, but are normally not used in offshore applications
because of higher cost. Polymeric matrix materials can be thermoset or thermoplastic.
Fibres are incorporated in the matrix in long continuous lengths or are sometimes utilised
as short discontinuous fibres. Composite components are formed by stacking several lamina
(single plies) to build up a composite structural laminate. The properties of a composite
material depend on the properties of the two main constituents: the fibres and matrix. A
summary of characteristic fibre and matrix properties is given in the following section
8.3.1 Fibres
A large number of fibres with different physical and chemical properties are
commercially available. They can be obtained in several different physical forms
266
Advanced Composites for Offshore Developments
including tow, ribbon, continuous woven mats and chopped strand mats. Fibre tow is
the most common form used for high performance reinforcement. The fibres most
likely to be used in the offshore industry are E and S glass, carbon and aramid. Table
8.1 provides a qualitative comparison of these fibres and Table 8.2 summarises some
of the physical properties. Other fibres such as polyamide (PA), polyester and
polyethylene (PE) are being used in pure fibre form (without matrix) for mooring ropes,
but they are not expected to be competitive in the composite form due to their
significantly lower modulus.
8.3.2 Resin
267
Handbook of Polymer Blends and Composites
268
Advanced Composites for Offshore Developments
8.4.1 Processes
269
Handbook of Polymer Blends and Composites
used to fabricate grating and strands for TLP tendons. Helical braiding and a
combination of filament winding and pultrusion are being used to manufacture
continuous spoolable pipes. The RIM process is ideally suited for fabrication of large
structural configurations such as beams, shells, and large structural boxes. The challenge
that currently faces both the composite industry and the oil companies is the lack of
available manufacturing capacity to fabricate the large quantities (millions of kg)
required for components such as risers or tendons for a single offshore development
within a reasonable schedule.
8.4.2 Inspection
Many non-destructive testing (NDT) techniques have been successfully applied for
inspection of composites. Table 8.6 lists some of these methods and their application in
detecting different types of defects that can be found in composite laminates. Electronic
shearography, thermography, and Mobile Automated Scanner (MAUS) has been proven
suitable for inspection of large composite areas [5, 6]. MAUS is a portable NDT system
that was developed by McDonnell Douglas. The system incorporates multi-mode
inspection options that include ultrasonic P/E depth (time of flight), ultrasonic pulse
echo amplitude, ultrasonic resonance, and eddy current. For continuous pipe production,
axial tomography and both pulse-echo and through transmission ultrasonics have been
shown to be very reliable. In the case of ultrasonics, proper contact between the transducer
and the pipe surface must be assured by a smooth outer surface of the pipe. Low frequency
ultrasonics, i.e., the Impedance Plane Method (IPM) was shown to be very successful for
inspection of adhesively bonded composite pipe joints [7].
270
Advanced Composites for Offshore Developments
271
Handbook of Polymer Blends and Composites
Unlike structures fabricated from metal plates, a composite component achieves its form
entirely at the time of fabrication. Although many NDT techniques have been successfully
applied for inspection of composite components as previously discussed, the application
of these techniques for inspecting offshore composite components such as risers can be
limited because of the presence of internal and external liners. Therefore, the overall
integrity of a composite structure is very dependent on the successful implementation of
a comprehensive quality assurance programme (QAP), particularly when weight
limitations and economic incentives require the use of lower safety factors. The basic
elements of an acceptable QAP should include the following [3]:
1. Inspection and testing of all raw materials (fibres, resins, adhesives, and so on).
7. Proof load testing of actual components. As an example, all riser joints shall be
subjected to pressure test with axial load end effect at 1.5 times the maximum operating
pressure for 15 minutes.
Destructive testing of at least one joint should also be considered as an important element
of the quality control plan. The destructive testing is used not only to confirm design and
manufacturing, but also to develop monitoring parameters such as stiffness, local
displacement and acoustic emission signature for use to verify the quality of composite
risers during proof load testing.
Unlike other commonly used offshore materials such as steel that are homogeneous and
isotropic (fibre directionality with uniform properties in all directions, independent of the
direction of applied load) continua, fibre reinforced composites are microscopically
272
Advanced Composites for Offshore Developments
11
F =
+
11 > 0
11
F = 11 11 < 0
-
11
F = ( 22 )2 + ( 12 )2 22 > 0
+
22 12
2
( 22 + 22 2
+ 12 )
2 4
F = 22 < 0
12
2
( 22 - 22 2
+ 12 )
2 4
F =
-
22 < 0
22
273
Handbook of Polymer Blends and Composites
The values of X+11, X-11 , X+22 , X-22 , and X12 are the strength properties of the composite
lamina under axial tension and compression, transverse tension and compression, and shear,
respectively. Table 8.7 provides typical values for these parameters for AS4 carbon fibre/
epoxy and S2 glass/epoxy composites at room temperature with 60% fibre volume fraction.
The value of the failure function F (F < 1.0) is set using the acceptable safety factor.
The selection of a safety factor for composites depends on many parameters such as material
type, duration of load, failure mode, rigor of design, impact of failure, feasibility of in-
service inspection, quality of fabrication, and accuracy of the materials data. As an example,
the safety factor depends on whether the designer is using short-term strength data or
long-term strength data. The reduction of composite strength with time can be accounted
for by using the stress rupture curves. An example of these curves for carbon fibre composites
is shown in Figure 8.8 (Salama and coworkers [12]). These curves were developed as part
of a composite production riser project. The x-axis represents the load duration and the y-
axis provided the ratio between the long-term and short-term strength values. The different
curves are for different failure probabilities that can be used as different characteristic
values. The load duration ultimate strength can be confusing to metal designers who are
accustomed to a fixed strength value for metals and a single safety factor such as 1.5 based
on yield strength or 2.0 based on ultimate strength. Developing equivalent safety factors
for composites is critical to advancing their application within the offshore industry that
currently lacks experienced composite designers.
274
Advanced Composites for Offshore Developments
Figure 8.8 Normalised stress rupture data for carbon/epoxy composite pipes at
different probability of failure (Pf)
components such as risers. Table 8.8 provides recommended safety factors for composite
risers (Salama and coworkers [13]). The A-basis strength is the strength value above
which at least 99% of the test results are expected to fall, with a confidence of 95%. For
composites, three strength values are required under the proper environmental condition,
e.g., temperature and moisture. These values are the tensile and compressive strengths of
the fibre and the interlaminar shear strength of the composite laminate. Table 8.9 provides
an example of typical mean, and A-basis strength values for carbon fibre and carbon
fibre/epoxy composites.
275
Handbook of Polymer Blends and Composites
Instead of imposing a safety factor on the ultimate strength parameter, one can specify
allowable strength values. The rigorous approach for establishing acceptable allowable
strength values is based on the use of structural reliability calculations (Salama and
coworkers [3]). This approach involves three steps. The first step is to select the acceptable
annual probability of failure. The probability can be found by calculating it for existing
structures, like steel risers in service today, i.e., a calibration to past, acceptable practice.
The annual probability can also be defined as an absolute value, e.g., 10-5, if such a level
has been established for the project in which the riser will be used. The comparative
approach is generally preferred. The second step is to define all loads in terms of their
respective annual extreme value distributions. The type of distribution and the distribution
parameters have to be known. The third step is to define the resistance or strength, also
in terms of its distribution and variation. Once these steps are defined the probability of
failure is given by the following convolution integral over the load distribution (S) and
the resistance (strength) distribution (R) which is schematically shown in Figure 8.9:
[
PF = P R S < 0 =] f R( r )f S ( s )drds = f R ( r )f S ( s )drds = F ( s )f
R S ( s )ds
R S <0
The partial safety factors f and m for load and material, respectively, describe the required
distance between a characteristic load value and a characteristic resistance value in
order to achieve a required safety against failure. The characteristic values can be typically
defined as the mean plus or minus two standard deviations. Different sets of f and m
276
Advanced Composites for Offshore Developments
Unlike metals when failure predominantly occurs as a result of crack development and
growth, composites fail because of multiple damage mechanisms. These damage mechanisms
include fibre fracture, matrix cleavage, matrix shear, matrix plasticity, fibre/matrix
debonding, fibre microbuckling, and interlaminar debonding. Damage mechanisms of
composites are not only impacted by thermal and mechanical loads, but they are also
influenced by environmental sensitivity such as water absorption. Sometimes, damage of
composites can result in a redistribution of stresses and, thus, increases the strength. However,
the type, extent, and location of damage determine whether load is relieved or intensified.
A critical design criterion for composites is impact resistance. Impact damage of metal
components, that may be caused by impact with another object during transportation,
storage, or installation, can be visually obvious and, thus, would be easy for an operator
to make a decision on whether to use the component or not. For composites, it would be
difficult to recognise this damage, particularly for high pressure components such as
277
Handbook of Polymer Blends and Composites
risers that have external liners. Therefore, an impact criterion must be specified for
composite components. Currently, there are no clear guidelines for impact load caused
by dropped objects for metal components. The only exception is the requirement of 250
kJ that was imposed by Norweigan Petroleum Directorate (NPD) on the design of titanium
drilling risers for the Heidrun TLP (Salama and coworkers [13]). This high value is
based on the predicted energy imposed by a casing joint falling in the water. This criterion
should be examined very carefully because it can result in a major cost increase of
composite risers unless innovative protection options are utilised. For composite
production risers, it is suggested that an impact criterion of 5 kJ is appropriate. This
impact energy corresponds to the situation of a riser joint falling from a truck.
Since resin cracking will occur at a much lower stress level than fibre cracking, internal
liners are used for high pressure components to allow cost effective designs based on
fibre failure. Both elastomeric and metal liners are being considered. The liners must be
selected to be compatible with the fluids in contact with it. For drilling and workover
composite risers, mechanical damage to the internal elastomeric liner is of concern. Since
damage to the liner of a composite pipe can result in leaks, the application of metallic
liners is being considered.
8.7 Durability
Since the fire resistance of composites presents a significant technical issue that has limited
the use of composites in many applications, major efforts have been directed to address
this concern. For surface ships and submarines, fire threat is ranked as one of the top
concerns among technical issues that included cost, acceptance process, environmental
effects/disposal, structural analysis/joints, and large scale validation (Sorathia and
coworkers [14]). In 1993, several US government agencies created the Interagency Working
Group on Fire and Materials (IWGFM) to implement a coordinated national research
effort on fire and thermal behavior of polymeric materials and composites, and to develop
materials with improved performance.
278
Advanced Composites for Offshore Developments
minimising structural damage. Most phenolic resins evolve water during curing, yielding
high void content and weaker matrix, and, thus, limiting their use for primary structural
applications. However, efforts are continuing to improve their performance. The poor
structural performance of phenolics can be overcome by modifying the phenolic resin
with siloxane. This resin system is now being used by Ameron Fibre glass Company for
fabrication of pipes for marine applications (Folkers [16]).
The use of fire barriers to protect composites can be necessary in many applications. The
results of a study by the US Naval Surface Warfare Center/Carderock Division (NSWC/
CD), on more than 20 fire barriers for composite materials, have shown that glass mats,
intumescent mat, and a water-based intumescent/ceramic coating are the most effective
(Sorathia and Beck, [14]). The cost of these fire barriers can be reduced and their durability
increased if they are fabricated in-situ during composites processing. Using any of these
fire barriers, a glass reinforced vinyl ester composite system met the ignitability and peak
heat release requirements of MIL-STD-2031 [17] at all heat fluxes.
Fibre glass composite pipes have been used in firewater systems of many offshore
structures. For wet systems, where water is always in the pipe, fire protection is, generally,
not required. However, most companies specify fire protection for joints, fittings, and
flanges. For dry systems, fire protection is always required.
Studies on the effect of seawater on composites by Bradley and coworkers [18] showed
that the interfacial shear strength of seawater saturated E-glass epoxy composites
with good sizing was decreased by 22%. When improper sizing was used, the
interfacial shear strength was decreased by about 50%. This confirms the importance
of selecting the proper sizing. The ply stresses at which interfacial debonding and
transverse cracking of E-glass/epoxy composites occur were reduced by 15% after
ageing, by immersion in seawater to saturation. For pure hoop (ply laid onto a mandrel
at a 90 degree angle) tubes, the burst results of moisture saturated filament wound
E-glass/epoxy tubes was 20% less than the burst results of the virgin tubes. On the
279
Handbook of Polymer Blends and Composites
other hand, for biaxial E-glass/epoxy tubes [903/+25/903], the burst results of moisture
saturated tubes was only 5% less than the burst results of the virgin tubes. Bradleys
results also showed that the impact of moisture on carbon fibre/epoxy composites is
less than that on fibre glass composites.
8.8 Joining
280
Advanced Composites for Offshore Developments
combination of adhesive bonding and mechanical traps (Baldwin and coworkers [19]).
In selecting the joining option, both the level of complexity of fabrication of the
mechanical interlocking system and the long-term integrity of the adhesive bond must
be considered. For joints that rely only on the interlocking systems, a mechanism such
as a lip seal and o-rings must be provided between the metal part and the composite
tube for sealing internal fluids. Mechanical locking systems include reverse taper/curve,
pins and hoop grooves. External sleeves/compression rings are sometimes considered
as an additional feature to increase resistance to fibre pull-off loads and to reduce
bending at the tip of the metal termination.
8.9 Repair
Currently, there are very few codes with any relevance to offshore composite systems.
Therefore, the pragmatic approach to evaluate composite systems is to apply a
combination of codes for metal systems and codes for composite structures together
with sound engineering judgement. Codes for composites can be used for material
related aspects. However, most composite codes are written for specific applications.
When using these codes for other applications, the safety philosophy behind each specific
code has to be carefully evaluated. It also has to be checked whether assumptions
about loading conditions, stress states, etc. that are implicit are also relevant to the
new systems. Table 8.11 gives a list of some relevant codes for composite components.
Although these codes may not be entirely applicable for the composite system under
study, much useful information can be extracted from these documents. However, for
composite systems to be extensively used, relevant industry guidelines and codes must
be developed. This gap is currently being addressed by several industry initiatives such
as the FRP Design guideline and the composite riser design guideline being developed
by Det Norske Veritas (DNV).
281
Handbook of Polymer Blends and Composites
One of the first structural applications for composites was in the automotive industry
where fibre glass was introduced in the 1950s for sports car bodies. Fibre glass pipe
and tanks have been used in the onshore oil industry for almost 45 years (Oswald
[27]). Fibre reinforced composites have been used in many marine applications including
canoes, speedboats, coastal and ocean-going yachts, work boats, lifeboats, naval vessels,
high performance crafts, and underwater vehicles (Dow and Bird [28]). The reasons
composites are popular in marine applications include weight reduction, increased
diving depth, flexibility in sculpturing structural geometries, acoustic signature
reduction, reduction in magnetic and electrical signatures, and reduced maintenance.
The most significant application of composites in the marine industry has been the
construction of mine countermeasure vessels (MCMV). The world first GRP
minesweeper was launched by the British Navy in 1972. The reasons for using GRP
were their low magnetic signature, lower cost, improved durability and better shock
resistance than laminated wood. Currently, GRP minesweepers are being used by the
navies of at least 15 nations around the world. The hull of many of these vessels is
sandwich construction incorporating a rigid polyvinyl chloride (PVC) foam core and
282
Advanced Composites for Offshore Developments
Fibre glass composites, and to a lesser extent carbon, Kevlar, and Nomex composites, have
also been used by the offshore oil industry in a variety of applications. Some specific examples
for the offshore application of fibre glass composites include fire water piping, seawater
piping, storage vessels, grating, fire and blast walls, cable trays, mud mats, and subsea
wellheads (Williams [29]). Almost all TLPs in the Gulf of Mexico starting with Mars
have used GRP pipe for the firewater ring main and glass/phenolic resin for the gratings.
Figure 8.10 gives an example of fibre glass pipes and grating on an offshore structure. It
should be noted that only during the last decade has the fibre glass industry started to
seriously consider new technologies such as improved resins, reliable connections, selectively
using carbon fibres, implementing better quality control and NDT techniques, publishing
better design and installation guidelines, and using improved fire resistance formulations
and coatings. Therefore, past negative experiences with fibre glass composites should not
be used as a basis for judging current materials. As an example, in the past many operators
had very disappointing experiences with leaks of fibre glass pipe joints, but recent
applications of GRP pipes on platforms such as Heidrun and Ram-Powell have shown that
the leak rate of several thousands of field joint was less than 0.1% (Salama [1]).
As subsea developments move to deeper waters, the use of remote operating vehicles
(ROV) and autonomous underwater vehicles (AUV) will expand for installation, retrieval
and repair functions. The efficient use of ROV requires the development of lightweight
equipment and tools. Components for which composites can offer major advantages
Figure 8.10 An example of fibre glass pipe and grating on an offshore structure
283
Handbook of Polymer Blends and Composites
include: manifold support foundation, mud mats, subsea protection covers, ROV
manipulators, ROV skids, tools, control pod housing, vessels, accumulators, manifolds,
and guide cones. Most AUV use a pressure vessel to house their energy source. Composites
are being considered for these vessels because for every kilogram saved, an extra kilogram
of battery payload can be added (Griffiths [30]).
The application of composites for load bearing floors, bulkheads, helidecks and living quarters
can take advantage of the technology that has been developed for design and manufacture of
composite bridges. Both sandwich and pultruded composites are being considered as a
substitute for steel flooring which is currently fabricated from plates welded on I or U shaped
beams, and then laid on the primary beams of the deck. Coflexip developed composite flooring
that is made by juxtaposition of composite beams that are stiff enough to be laid directly on
the primary deck beams. The low cost pultrusion process is used to produce these beams that
are formed with profiles to allow interlocking and equipment installation. Composite beams
can be built with more strength and more fire resistance than steel, which could have special
applications in rig floors and hull support (Massot [31]).
Composites have also been used for many high-pressure vessel applications. These vessels
are manufactured from glass and carbon fibre composites with a thermoplastic liner
(Salama [1]). Composite pressure vessels have been developed and qualified for use as
mud gas separators. Composite accumulator vessels have been used for production-riser
tensioning systems on several TLP.
These successful composite experiences coupled with the need to reduce the lifecycle costs
of deepwater development has motivated the industry to examine the application of
composites in several new critical systems. The following sections discuss these applications.
So far, the main emphasis on new applications of advanced composites for the offshore
industry has focused on high pressure pipe (above 7 Pa pressure) which are either discrete
(6 to 24 m long) for use as drilling and production risers, choke and kill lines, tubing,
casing, and pipe or continuous (many thousands of m long) for use as coiled tubing (Sas-
Jaworsky and Williams [32]) and flowlines (Melve and coworkers [33]). Continuous
pipes are of relatively small diameters (< 12.5 cm) and are designed to be spooled.
Spoolable composite pipes have been proposed for use as coiled tubing, velocity strings
(a smaller diameter pipe that is inserted in the tubing of a well to allow the produced
fluids to flow faster), capillary tubes, subsea pipeline clean out lines, subsea flow lines
and subsea control lines. High pressure composite coiled tubing is attracting major
attention because it provides enabling capabilities to work over, log and complete highly
deviated wells as well as provide operational improvements for a wide variety of other
284
Advanced Composites for Offshore Developments
oil field applications. There also are other applications under consideration for composites
which do not fall under the high pressure tubular classification such as TLP tendons.
One proposed tendon concept uses a strand assembled of many continuous small diameter
carbon fibre rods or a belt laminate constructed using carbon fibre rods. The tether, like
coiled tubing, would be designed to be spoolable to improve the economics of installation.
Non-bonded flexible risers are critical elements in floating systems for deepwater offshore
oil and gas production. A non-bonded flexible pipe is designed as a multi-layer conduit
consisting of thermoplastic layers for sealing the conveyed fluid from the external
environment and metallic layers of carbon steel to provide axial and hoop structural
reinforcements. In deeper water where longer length flexible pipes will be used, the weight
of the carbon steel reinforcements becomes a critical factor in the design of a flexible pipe.
The large tension load induced by the pipe weight could lead to unacceptable stress levels
in the pipe structure and excessively large deck and installation loads. As the pipe stress
level increases, larger cross sectional areas of the steel members are required, further
increasing the weight of the pipe. To reduce the pipe unit weight, while retaining the required
strength level, composites have been proposed and used as armor wires (reinforcing wires)
to replace the steel axial reinforcement (Fischer and Salama [34]). The shapes of the
composite wires are flat and rectangular similar to those used for steel armour wires. The
first application of composites in a non-bonded flexible pipe was carried out by Coflexip
Stena Offshore using armor wires made of glass fibre reinforced epoxy (Figure 8.11). The
lighter weight flexible pipe allows smaller less expensive vessels to be used for installation.
285
Handbook of Polymer Blends and Composites
Fibrespar estimates that the cost of a spoolable composite pipe will be competitive
with the acquisition cost of a similar API 17J flexible pipe on a per-mere basis. However,
the life cycle cost is expected to be reduced because the spoolable composite tubular is
corrosion resistant, lighter and easier to deploy, has longer fatigue life, and has the
286
Advanced Composites for Offshore Developments
potential to reduce overall riser system complexities and costs. Figure 8.13 shows a
10.3 MPa rated 10 cm ID Fibrespar flowline with a metal connector. Hydril has also
developed spoolable composite tubulars in the 1-inch to 6-inch diameter range for
deepwater applications. Figure 8.14 shows a medium pressure (5.2 MPa) spoolable
composite pipe placed on a truck.
Figure 8.13 High pressure spoolable composite pipe section (courtesy of Fibrespar)
Figure 8.14 A 5.2 MPa Rated, 7.5 cm hydril spoolable composite pipe
(Courtesy of Hydril)
In addition to these spoolable composite tubulars that are made with thermoset resins, a
new class of spoolable composite tubulars made with fibre-reinforced thermoplastics is
beginning to be used in the oil and gas industry. For the manufacture of this reinforced
thermoplastic pipe, pre-formed fibre-reinforced tapes having a compatible thermoplastic
matrix are utilised. Layers of these tapes are helically wound around a thermoplastic
tubular. Usually, pairs of tape layers are applied having opposite wind-angles for each
287
Handbook of Polymer Blends and Composites
layer to ensure torque balance of the fabricated pipe. Following application of the tape
layers, an outer layer of thermoplastic pipe is co-extruded on top of the tape-wrapped
inner pipe. The layers of this sandwich-type construction are usually bonded together by
thermal or other means either as the layering proceeds or following the co-extrusion
step. A reel of pipe having intermediate connectors every 12 m is shown in Figure 8.15.
Extensive effort has been devoted to qualification of composite risers that are
categorised as top tensioned, stiff pressurised, composite pipes such as production
risers, drilling risers, casing, tubing, choke and kill lines, and mud transfer lines.
Design and qualification of composite risers have been the subject of three recent
288
Advanced Composites for Offshore Developments
joint industry studies. Composite production and low pressure drilling risers have
been developed by a joint-industry projects that were initiated in 1995 and are partially
funded by the US Department of Commerce National Institute of Standards and
Technology (NIST) Advanced Technology Program (ATP). High pressure composite
risers are being developed as part of a joint industry project organised by Norske
Conoco A/S and Kvaerner Oilfield Products (NCAS/KOP) and is partially funded by
EU Thermie. Table 8.12 provides a summary of the basic specification and the
performance of NIST ATP composite production riser and NCAS/KOP drilling riser
joints. Details on these projects have been discussed by Salama and coworkers [12,
289
Handbook of Polymer Blends and Composites
Figure 8.16 Composite drilling riser joint schematic and test specimen [3]
13]. Figure 8.16 shows a schematic of the main features of the composite drilling
riser joint and a prototype joint that has been used for testing. Figure 8.17 shows a
production riser prototype joint that has been used for testing. The joint was
instrumented with conventional strain gauges and fibre optic sensors that were
embedded in the composite laminate during fabrication. The wires shown in the
photograph are for the leads of the sensors.
290
Advanced Composites for Offshore Developments
3. Acceptability of the different elements in the design strategy to avoid these failure
modes.
5. Acceptability of the basis for selecting materials allowables and safety factors.
For drilling and workover risers (a pipe connecting the platform to the seabed allowing
workers to enter the well to perform maintenance), the internal liner will not only be in
contact with chemicals, it will also be in contact with running tools such as drill bits,
milling tools, fishing overshots (a tool used to grab and lift a broken part from the well),
coring tools, under reamers, packers, safety valves, etc. Tear damage of the elastomeric
liner by running tools is probable. Therefore, an elastomeric internal liner alone may not
be sufficient to ensure pressure integrity because of the high probability of its tearing
damage. Considerations should also be given to delamination of the inner liner from the
composite reinforcement as a potential failure mode by collapse due to permeation of
291
Handbook of Polymer Blends and Composites
high pressure gas. The bond between liner and reinforcement in these systems is weak
and will become weaker on exposure to hydrocarbons. Another important damage
mechanism that must be accounted for in design and verification of composite risers is
external damage from mishandling and impact by dropped objects.
The most critical component in the design of composite risers is the metal-composite
interface (MCI). The integration of composite and metal fittings is a challenging aspect
of composite structural design. The differences in material properties between an
orthotropic filament wound composite and an isotropic metal end fitting impose special
requirements on the selection of the manufacturing process and design. Numerous
attachment methods have been demonstrated, with a wide range and variety of load-
carrying capabilities. These attachments rely on two basic approaches to provide the
load transfer between the composite tube and the metallic joint, adhesive bonding and
mechanical interlocking. Because the long-term integrity of the adhesive bond is difficult
to achieve, they are not recommended to be used alone for composite risers. Because of
the complexity of a mechanical interlocking system with respect to inspection, verification
must take into account both the consistency of its fabrication and the reliability of its
design/analysis. Since composite riser joints are located above the bottom taper or flex
joints, bending loads are small, i.e., the primary loading is pressure plus axial loads.
Although many NDT techniques have been successfully applied for inspection of composite
components, the application of these techniques for inspecting composite risers can be limited
because of the presence of internal and external liners. Therefore, the overall integrity of the
composite riser must heavily rely on the successful implementation of a comprehensive QA
programme (QAP). Review, validation and monitor of the QAP should represent a critical
element of the composite riser verification program.
8.12.3 Tethers
Tethers used on most of existing TLPs have been made of steel pipe sections to maximise
payload and facilitate installation and inspection. In deepwater TLPs, two key parameters
governing the tubular tendon size are maximum heave/pitch natural periods of the platform
and hydrostatic collapse pressure of the tendons. These two parameters in turn dictate the
minimum solid sectional area required for adequate tendon stiffness and the minimum t/D
(wall thickness/diameter) ratio to resist the hydrostatic pressure. Carbon fibre composites
possess several attractive properties such as lightweight, high specific stiffness, and excellent
corrosion and fatigue resistance that make them attractive as lightweight carbon fibre
composite cable that can be used as TLP tethers (Salama [4, 36, 37]). The concept considered
is based on the use of parallel strands as shown in Figure 8.18. Approximately 50 mm in
diameter, they are formed of lightly twisted bundles of small diameter (3-6 mm) pultruded
carbon fibre composite rods. The TLP tether body will be formed by several strands whose
292
Advanced Composites for Offshore Developments
number of strands is determined by the strength and axial stiffness requirements. This type
of tether can be spooled for transportation and installation offshore. The proposed concept
provides an attractive approach to the standardisation of the tether system, thus eliminating
uncertainty and minimising project specific qualification requirements. The feasibility of
this concept can be credited to the steady decrease in the cost of carbon fibres, particularly
the heavy tow grade. The composite rods can be fabricated from these large tow, low cost
carbon fibre in epoxy or vinylester matrix. The strength and stiffness of rods fabricated
from these fibres depends on the fibre volume fraction. Figure 8.19 gives values obtained
293
Handbook of Polymer Blends and Composites
by testing rods supplied by different pultruders and fabricated of several of these low
cost fibres. The small diameter rods are manufactured using the well established
pultrusion process. The strands can be manufactured using typical wire rope stranding
methods. Assembly of the strands into a tether can be achieved using umbilical
fabrication facility. Using the umbilical manufacturing approach, the tether cross section
can be constructed from a number of strands. The strands are free to move individually
in the length direction, allowing individual adjustment and hence a better distribution
of the axial loads. Terminations using resin-potted cones have been extensively used in
the wire rope industry, and resin terminations have been proven to be successful also
for terminating composite strands. While for the wire rope application the termination
is required to develop the full strength of the rope product, this is not be an issue of
concern for TLP tethers. The tether design is driven more by stiffness than by strength
and, therefore, a high load transfer in the termination is not required. The results of
tensile fatigue tests on 54 mm strands with resin potted terminations confirmed the
adequacy of this termination concept because its fatigue life far exceeded the currently
acceptable steel tethers as shown in Figure 8.20. In order to allow flexibility of
fabrication, assembly and adjustment, the current plan is to terminate the 5 cm strands
individually and then assemble the terminations of the different strands into a single
tether termination.
Figure 8.20 Comparison between fatigue strength of girth welded steel pipes and
composite rods and strands
294
Advanced Composites for Offshore Developments
Figure 8.21 Installed cost of steel and composite tethers in millions of dollars for a
small TLP in the Gulf of Mexico
For TLPs with conventional steel tubular tethers, the tether loads are monitored using
complex load cells or electromagnetic devices. These cells utilise strain gauges to measure
the compression in a material supporting the tether top connector. The load cells are of
a complex construction, and are hence expensive and can be unreliable. Fibre-optical
sensor technology has been developed extensively in the past few years, and different
optical sensing methods are now being used in several applications offshore, e.g., a
downhole measurement system (pressure/temperature), and smoke detectors. Carbon
fibre composite rods can be fabricated with embedded optical fibre. Measurements of
the strain in the rod to an accuracy of about 1 m strain by the application of Bragg
grating technology have been demonstrated.
Several studies on TLP designs for application in the North Sea and the Gulf of
Mexico were conducted to assess the potential benefits of composite tethers. One of
these studies included two TLP configurations for installation in water depth from
800 to 3,000 meters. Figure 8.21 provides an example of the cost data generated in
this study for a Mini TLP in the Gulf of Mexico for steel risers and composite and
steel tethers [38].
The interest in the development of deepwater reserves is expected to increase. Figure 8.22
provides the expected number of required deepwater floating units to meet the known
development plans for the next few years (Gulbrandsen and coworkers [2]). Over 100
km of top tensioned production risers will be required for these units. Figure 8.23 shows
295
Handbook of Polymer Blends and Composites
Figure 8.23 Potential production riser market for composites in the next five years
the potential market for composites risers. If current developments and qualifications of
composite risers are successful, one would expect that the demand of composite risers
will be between one to two million kilograms within the next five years. Currently, there
is no composite manufacturing capacity to meet this demand. However, both Kvaerner
and ABB Vetco are developing plans to develop such manufacturing capacity.
296
Advanced Composites for Offshore Developments
Figure 8.24 Potential tether market for composites in the next five years
It is projected that four TLP will be required for the next five years. The total length
for the tethers of these TLP will exceed 50 km. The expected demand of composite
tethers, as shown in Figure 8.24, will also be between two to four million kilograms
within the next five years. While the demand for composite risers will require new
manufacturing facilities, the demand for composite tethers can be accommodated
using existing capacity.
The expanded application of composites in the oil industry continues to face technical,
financial and emotional barriers that must be overcome to allow the full potential of
composites to be realised. Solutions to these barriers involve developing a more
comprehensive design and manufacturing data base to allow competent risk
assessment, establishment of reliable cost structure for both components and systems,
and educating offshore contractors and suppliers. This effort should create sufficient
confidence to expand the use of these new materials. While regulatory agencies have
demonstrated receptiveness to new technology, the value of developing industry
accepted design codes for specific composite components couldnt be overemphasised.
This gap for composite risers is being addressed by projects that are sponsored by
several offshore operators and suppliers. In addition to the technical issues, it must,
however, not be overlooked that the key to the successful development and
commercialisation of advanced composite for the offshore oil industry is project
implementation within a value chain partnership including owner, designer, material
supplier, fabricator, regulator, and system integrator. Such a partnership is illustrated
in Figure 8.25.
297
Handbook of Polymer Blends and Composites
References
7. A. Boye Hansen and B. Asdal, Revue del IFP, 1995, 50, 1, 113.
298
Advanced Composites for Offshore Developments
17. MIL-STD-2031, Fire and Toxicity Test Methods and Qualification Procedure for
Composite Material Systems Used in Hull, Machinery, and Structural
Applications Inside Naval Submarine, 1991.
18. W. L. Bradley and T. S. Grant, Journal of Materials Science, 1995, 30, 5537.
20. F. Barnes, Carbon Fibre Composites for Strengthening and Repair, Davenport
Management Limited, UK, 1997.
21. BS 4994 Specification for Design and Construction of Vessels and Tanks in
Reinforced Plastics, 1987.
22. BS 7159 Code of Practice for Design and Construction of Glass Reinforced
Plastics (GRP) Piping Systems for Individual Plants or Sites, 1989.
23. BS 6464 Specification for Reinforced Plastics Pipes Fittings and Joints for Process
Plants, 1984.
299
Handbook of Polymer Blends and Composites
26. DIN 18820 Glass Fibre Reinforced Unsaturated Polyester (GF-UP) and Phenacrylic
(GF-PHA) Resin Structural Composites Structure, Fabrication and Properties.
300
9
Functional Polymer Composites
Many materials can change their shape, colour, form, phase, electric properties, magnetic
properties, optical properties, or other physical characteristics in response to stimuli
from the environment or a control system. Functional materials allow the combination
of the functions of two or more components into a single material. Designers can
harness the functional materials ability to react automatically to an environmental
change or take specific action in response to a control signal. Examples include a
component that senses when a room gets too warm and opens a ventilation skylight or
a material that converts vibrations directly into an electrical signal for monitoring
rotating machines.
301
Handbook of Polymer Blends and Composites
With some exceptions, functional materials markets are young and remain largely
unexplored. Many existing applications are relative simple and derivative. Examples are
conductive composites, which are used for electromagnetic interference (EMI) shielding.
When an electrically conducting phase is dispersed in sufficient quantity in a polymeric
resin, a conductive composite is formed. The unique properties of such composites make
them technologically superior to or more cost effective than alternative materials. EMI
shielded enclosures are mainly designed to protect against radiation. The largest market
in this area is housings for business machines. In addition, conductive composites can be
used to shield individual electronic components, which are either particularly noisy or
particularly sensitive to outside electromagnetic noise. In the EMI shielding market,
conductive composites compete with a number of other technologies based on surface
coating. These methods include vacuum metallising, sputter coating, zinc arc spraying,
nickel-based paints, and electroless metal coatings.
302
Functional Polymer Composites
background for the understanding of the physical properties in such materials. The
arrangement of the filler particles and the resulting mechanical, electrical, or
dielectric response can be described either by percolation or by effective medium
theories. In the following sections the different functional effects occurring in
composites will be discussed.
Polymer composites consist of at least two phases, a filler material and a polymer host
matrix. The filler can be any kind of material, a metal, a ceramic, or a polymer with
different morphologies. It can be an isotropic particle, a flake, and a fibre or woven
fabric. Functionality depends critically on the content of the filler material and its
connectivity or percolation within the matrix.
9.2.1 Connectivity
303
Handbook of Polymer Blends and Composites
Figure 9.1 Ten connectivity patterns for a diphasic solid. Arrows are used to indicate
the connected directions (Reproduced with permission from D. S. McLachlan [19],
published by The American Ceramic Society, 1990)
For very low filler fractions, the mean distance between conducting particles is large and
the polymer matrix determines most of the composites properties. As an example, a
polymer filled by an electrically conducting material would show for low filler content a
conductivity close to the value of the polymer (about 10-15 -1 cm-1, Figure 9.2). When a
sufficient amount of filler is loaded, the filler particles get closer and form linkages,
which result in an initial conducting path through the whole material. The corresponding
filler content is called the percolation threshold. In this concentration range, the
conductivity can change drastically by several orders of magnitude for small variations
of the filler content. Finally, at high filler contents, the increasing number of conducting
paths forms a three-dimensional network. In this range the conductivity is high and less
sensitive to small changes in volume fraction.
Percolation theory is, for example, able to give a good phenomenological description of
the conductivity of a system near to a metal-insulator transition. Excellent introductions
are given by Zallen [9] and Stauffer [10]. In model percolation studies one starts with a
large, preferably infinite regular periodic lattice. A site of this lattice is then assigned at
304
Functional Polymer Composites
random with a probability P. All sites are bonded to adjacent sites by a conductive bond.
At a critical probability Pc, an infinite conducting cluster is first formed in the infinite
lattice. This means that at an initial percolating path through the material the conductivity
of the lattice changes from insulating to a finite value. Above Pc, the conductivity is
found to increase as:
( p ) ( P Pc )n (1)
with a parameter n between 1.65 and 2 for three-dimensional lattices [8, 11, 12, 13]. Pc
is called the critical or percolation threshold.
Percolation theory is often limited to the cases in which either the insulating medium has
zero conductance or the conducting filler has zero resistance. For real systems, where the
resistivities are always finite values, percolation theory becomes valid with the use of
scaling factors [14]. Critical volume fractions vc have to be calculated by using filling
factors [15]. For random packing in three dimensions, vc is about 16%.
The shape of the filler can influence the percolation behaviour. Fibres show, for example,
a percolation threshold of only several percent. Additionally, geometry and dimension of
the regarded composite sample, as for example, the distance of external electrodes have
305
Handbook of Polymer Blends and Composites
There are two special cases for which effective medium theories exist, called the symmetric
and the asymmetric cases. The symmetric case assumes that a random mixture of spherical
(or ellipsoidal) particles of two or more components fills all space. When one component
is a perfect insulator, the symmetric medium theory contains a conductor-insulator
transition. In the asymmetric case, the surface of the particles of one component is always
completely covered by the other component. This means either a 0-3 or 3-0 connectivity.
This case does not contain a transition. In order to describe the conductivity of conductor-
insulator composites, McLachlan proposed a generalised effective media (GEM) equation
[17, 18]. It is derived as an interpolation between the symmetric and asymmetric effective
media theories and has the mathematical form of a percolation equation within the
appropriate limits. The GEM has been successfully applied to describe the piezoresistive
behaviour of Fe3O4-filled flexible epoxy [19]. A change in the resistivity of five orders of
magnitude, from 1010 to 105 cm, could be described by the GEM equation in good
agreement with the experimental data.
At very low filler content, well below the percolation threshold, the mean distance between
conducting filler particles is large and no conducting paths through the composite are
established. The mean separation distance of neighbouring filler particles is so large that
tunnelling from particle to particle cannot take place. In this case, the composite
conductance is determined by the conductivity of the polymer host matrix. Hence, the
306
Functional Polymer Composites
filler content has only a small effect on the electrical conductivity of the entire composite
(initial plateau in Figure 9.2).
At high loading with filler material many particles are in close contact. The conduction of
charge carriers occurs through the continuous structure of the chain of filler particles in the
polymer matrix. The filler material and its microscopic contacts to adjacent filler particles
mainly determine the conductivity. The resistance of the contact spots is the leading
contribution to the overall composite resistivity, particularly for good conducting filler
materials. Two effects contribute to the contact resistance, the constriction resistance of a
contact spot and the tunnelling resistance between the particles [20]. Therefore the contact
spots are the region of largest resistance and thus of highest electrical losses and may be the
source of large temperature development during current flow through the composite.
The constriction resistance, Rc, at a circular contact area with radius a, is given by Equation
(2) which is already well known from the study of macroscopic electrical contacts [21]:
T
Rc = (2)
2a
Here T the bulk resistivity of the filler particles. The constriction resistance at the contact
spots dominates over the bulk resistance of the filler particle as long as the contact spot
radius, a, is much smaller than the filler particle radius. The spot radius may, of course,
depend on the force with which the particles are pressed together.
The large conductivity changes observed, e.g., in thermistor composites and piezoresistive
materials can be explained by modifications in tunnelling current and/or variations of
the spot area due to shifting mechanical properties.
9.3 Thermistors
Figure 9.3 shows the temperature dependence of the heat flow in differential scanning
calorimetry (DSC), the resistivity, and the relative linear expansion as measured by
dilatometry for a composite consisting of 50 volume % titanium diboride (TiB2) in high
307
Handbook of Polymer Blends and Composites
Figure 9.3 Results of the heat flow in DSC (a), the resistivity (b) and the relative linear
expansion (c) of 50 vol.% TiB2 in HDPE are shown as a function of temperature
density polyethylene (HDPE) [31]. The composite material has an initial resistivity of
about 0.01 cm. Far below the melting temperature the resistivity increases only gradually
with increasing temperature. But upon approaching the melting temperature of the matrix,
which is determined by DSC as 133 C (Figure 9.3a), a huge step-like increase of more
than eight orders of magnitude is observed (Figure 9.3b). Also the thermal expansion
(Figure 9.3c) exhibits a stronger increase in the same temperature range.
Well below the melting temperature of the polymer, the conducting filler particles are in
a state of close packing with intimate contact to their neighbours, forming conducting
paths throughout the composite. During heating, the polymer expands much more then
the filler particles themselves. The contact pressure between adjacent filler particles is
308
Functional Polymer Composites
reduced leading to a moderate resistivity increase for T<120 C, as shown in Figure 9.3b.
This effect is strongly enhanced when approaching the melting temperature of the polymer
matrix. Then the filler particles and the conducting paths are obviously interrupted due
to the enhanced thermal expansion of the polymer. This leads to the huge jump in the
resistivity. Similar results have been obtained for carbon black dispersed in polyethylene,
which support this view [32].
Active heating of the thermistor by an electrical current can also cause the transition from
the conducting to the insulating state. The filler size then has a strong influence on the time
until the transition is performed [33]. Most of the resistance of the PTC material in the
conducting state originates from constriction resistance at the contact points. Therefore,
the contact points are the dominant sources of heat in the material. The thermal conductivity
of most of the conducting filler materials (1-102 W/mK) is at least one order of magnitude
higher than that of the polymer matrix (< 0.3 W/mK). Consequently, one can show from
the heat equation that the generated heat dissipates first, within 110 ms, into the filler
particles [33]. Then, with a time delay of 0.20.5 ms, the matrix is heated as well. Therefore
the temperature distribution inside the filler particles will be quite homogenous, except in
the close vicinity of the contact spots, where temperatures up to 2726 C can be expected
[34]. A reduction of the particle size leads to a larger number of contacts and smaller
volumes of polymer matrix material embedded within the network of metallic filler particles.
Due to a reduced mean distance of the matrix material to the next neighbouring particle
contacts, where the energy is generated, the heating of a composite with smaller filler
particles should occur faster. Together with the properties of the polymer matrix this is
important for nonlinear changes of the resistivity on temperature increase.
The transition from low resistivity to high resistivity upon an applied current can be
utilised for applications as current limitation. Such limiters are presently available as
309
Handbook of Polymer Blends and Composites
small elements for the protection of circuit boards at rated currents between fractions of
an ampere to several amperes [4]. As a three-phase current limiting module [37], they
are used in low voltage distribution systems for ratings up to 63 A [37].
Similar changes in resistivity can be achieved if stress is applied externally. This can lead
to a change in conductivity through a change in the distribution of the filler particles,
especially at low filler contents in soft polymers, due to their low Youngs moduli. At
higher filler content, i.e., for more rigid composites, it is more likely to modify the number
and efficiency of the electrical contacts by the application of external forces. This effect
has been demonstrated by Carmona and others [38] for flexible composites filled with
carbon black or graphite fibres. The initial resistivity of about 103 cm is reduced by a
factor of two under an applied hydrostatic or uniaxial pressure of 100 MPa and 1.5
MPa for Araldite epoxy resin (Ciba-Geigy) and Rhodorsil (Rhodia) silicone, respectively.
Larger resistivity changes have been observed by Yoshikawa and others [39] using Fe3O4
or SnO2 in a flexible epoxy matrix. Under an uniaxial pressure of 20 MPa the resistivity
of a composite containing 22 % SnO2:Sb (particle size < 0.1 m) is reduced from about
108 down to 106 cm. In the case of Fe3O4 (particle size: 0.2 m), the maximum reduction
is observed for a filler content of 47 volume %. The resistivity of 106 cm is reduced
down to 103 cm by an applied pressure of 10-12 MPa.
Larger resistivity changes can be achieved by using a hard oxide-free filler material. Figure
9.4 shows the resistivity as a function of uniaxial pressure for different filler contents of
TiB2 (particle size: 100 200 m) in Powersil 250 Q (Wacker Silicones) silicone rubber.
The composite was compounded using a Brabender Plasticorder at 60 C. The mixture
was then pressed to form a plate of 46 mm thickness and vulcanised at 180 C for 20
minutes. For the highest and lowest filler content, the pressure dependence under an uniaxial
pressure of 10 MPa is relatively small (three orders of magnitude). Between 25 and 35
volume %, however, the resistivities change by six to eight orders of magnitude. This
means that for intermediate filler contents not only an applied pressure improves the
particleparticle contact, but also additional conducting paths are established.
In a similar way, extensional strain can change the resistivity, as reported for nitrile-
rubber filled with carbon fibres. The specific resistivities of the initial state were between
0.3 and 2 cm. A strain of 40% resulted in a reduction of by a factor of 8 to 10 [40].
310
Functional Polymer Composites
Figure 9.4 Resistivity as a function of uniaxial pressure for 25, 30, 35 and 40 vol. %
of TiB2 in silicone rubber (Powersil 250 Q)
Figure 9.5 Mechanical switch with a piezoresistive composite for arc suppression
phenoxy polymer filled by nanometer-sized carbon particles [41]. The measured forces
can be in the range of 4.4 N to 4.4 kN.
Power switches are another area of possible applications, where the resistive composite
can have the function of an arc inhibitor. A simple example is shown in Figure 9.5 [42].
311
Handbook of Polymer Blends and Composites
The switch consists of an anchor as one electrode and a flexible, conducting polymer
below the counter electrode. In the closed position of the switch the anchor presses onto
the composite and reduces its resistance to a low level. When the mechanical force is
released, the material expands due to elastic forces. It becomes insulating, interrupts and
limits the current. Subsequently, the anchor can open the contact without arcing.
Tension sensitive material may also be used as current limiter and arc inhibitor in a switch
[42] as shown in Figure 9.6. Normally the material is highly conducting. When the piston
of a switch is withdrawn from the closed position (dashed) to the open position, the current
is commutated to the parallel path with the elastic polymer composite. The current leads
to and from the polymer composite are very close together. Electrodynamic repulsing forces
between the current leads expand the polymer in the case of a sufficient high current,
which has to be switched off. The commutation and the increase of the polymer resistance
serve for an arc-free switching off. A further advantage of this arrangement is that the
contacts of the switch are solid metal and can carry a high nominal current. The polymer
composite has to carry the current only in the case of switching off.
The increase in resistance due to the expansion of the polymer matrix is utilised for
chemical sensors as well. Polymer matrices loaded with conductive filler near the
312
Functional Polymer Composites
The general time-dependent behaviour of a composite sensor is shown in Figure 9.7 for the
example of a 100 m thick carbon black/polyurethane composite upon the exposure to
toluene and acetone [43]. While the response time and resistance ratio greatly depend on
the composite and the solvent, the responses seem to possess similar features. A small
quick increase in resistance is noted in region 1, which is probably due to surface adsorption
and field fluctuations. In region 2, the diffusion of the solvent through the sensor film can
be traced. It is accompanied by a steady increase in resistance due to the changing number
of conducting paths through the composite. In order to get a small response time, this
region has to be minimised. In region 3 the increase in resistance is quite dramatic
313
Handbook of Polymer Blends and Composites
corresponding to the breaking of the last remaining filaments. After this critical swelling
occurs, a smaller resistivity increase is observed in region 4, which may be further swelling.
When finally the samples are removed from the solvent environment, the resistance rapidly
decreases. After a quick decline, the resistance gradually decreases back to its initial values.
Response has been proven to solvents such as acetone, carbon tetrachloride, cyclohexane,
diethyl ether, ethanol and pentane [43]. The resistivity change can exceed seven orders of
magnitude. A certain degree of selectivity can be obtained for polyethylene-, polyurethane-
and polyvinyl alcohol-based systems. However, the reproducibility of the final resistivity
changes is not yet sufficient for industrial application.
Very interesting and useful effects can be achieved by using nonlinear conducting filler
materials. The electrical conductivity of the filler may depend in a nonlinear way on the
applied electrical field. Ceramic varistor (variable resistor) devices are made from doped
zinc oxide. The ohmic j(E) behaviour at low voltage changes to a highly nonlinear
relationship for electrical fields higher than the breakdown field EB:
J ~ E (3)
For ceramic zinc oxide varistors the nonlinearity coefficient is usually larger than 40
[44]. Polymer composites with a substantial varistor characteristic can be prepared by
sintering spray-dried, doped ZnO varistor powder and mixing it into a polymer. The
polycrystalline filler particles act as varistors due to their typical grain boundary structure.
Such varistor composites containing 20 to 50 volume % of varistor filler can have
nonlinearity coefficients up to 30. Figure 9.8 shows a comparison of the j(E)
characteristic of an epoxy/varistor composite and a ceramic varistor material.
The ceramic was sintered under the same conditions as the varistor filler. In the case of
the composite, the leakage currents at low field are reduced by 1/8 - 1/3 and the breakdown
field EB is increased by a factor of about 1.5, compared to the bulk ceramic. The increase
of EB can be understood by the longer current path in the composite, while the reduced
contact area explains the shift in the leakage current. For a dense packing of the powder
in a face centered cubic structure (FCC) structure, the geometrical path length is increased
by 2 (see inset of Figure 9.8). In a close-packed hexagonal closed-packing (HCP) structure
it would be a factor of 8 / 3 = 1.63. This is close to the observed value. Together with the
high B-values, it indicates that the varistor characteristic is dominated by the grain
boundaries of the filler and not by the interfaces between the filler particles. Consequently,
the characteristic can also be changed by the sintering conditions of the filler [45]. Using
fillers with large grains, breakdown fields down to 200 V/mm can be achieved [46].
314
Functional Polymer Composites
The most important area of application is field control at cable terminations. The effect
is based on the increase of the conductivity in those regions, where the electric field is
enhanced. Terminations using material with a relative small nonlinearity have now been
commercial products for several decades [47]. Zinc oxide filler material with a stronger
non-linearity, as described before, has been proposed recently for such applications [48].
The varistor composite is able to consume quite some energy so that it can be applied as
a surge arrester in medium or high voltage systems [49]. Due to the ease of processing
and high flexibility in shaping compared to ceramic devices, this material offers a big
potential for new applications and cost reductions.
Piezoelectricity is the linear relationship between electric and mechanical variables. Electric
polarisation is proportional to mechanical stress in the direct piezoelectric effect. The
thermodynamically related converse piezoelectric effect relates mechanical strain to the
applied electric field. Lead zirconate titanate (PZT) is the most prominent piezoelectric
material and widely used, e.g., as wake-up buzzers in radio clocks.
315
Handbook of Polymer Blends and Composites
S = sT + dE (4)
where S is the strain (vector of relative elongation), T is the stress (vector of force per
unit area), E is the electric field, s is the tensor of elastic compliance and d is the tensor
of piezoelectric coefficients.
Likewise, for the inverse effect of generating an electric field by stressing the material, we
have the following relationship:
E = gT + ( T )1 D (5)
where g is the tensor of voltage coefficients, D is the displacement vector, and T is the
tensor containing dielectric coefficients at zero tension. Commonly the following
mnemonic rule to reduce the double index notation in the component and matrix form is
used [50]:
Hence, for a freestanding piezoelectric device with applied electric field in the z-direction,
the displacement is given by:
s 3 = d 33E3 (6)
s1 = d 31E3 (7)
dh = d 33 + 2d 31 (8)
Typical values for PbZr0.52Ti0.48O3 are d33 = 223 x 10-12C/N and g33 = 39.5 x 10-14C/N. A
high d constant is desirable for actuator materials that are intended to develop motion or
vibration, while a high g constant is wanted for sensor applications.
A useful figure of merit for hydrophone applications is the product of the hydrostatic
strain coefficient dh and the hydrostatic voltage coefficient g h. Hydrophones are
underwater listening devices, which respond to hydrostatic pressure waves. Among the
applications of hydrophones are sonar systems for submarines, off-shore oil platforms,
geophysical prospecting equipment, fish finders and earthquake monitors.
316
Functional Polymer Composites
It is worthwhile to point out that the dh gh product of 0-3, 1-3, 3-1, 3-3, 3-2 piezoelectric
composites (see Figure 9.1) is usually much larger that that one of single phase PZT, with an
increasing dhgh in this order [51]. This occurs because in composites the planar coupling,
which contributes negatively to dh,, is interrupted by the elastically compliant polymer. For
good PZT material the dh value is only in the order of 45 x 10-12 C/N, because d33 and d31 are
opposite in sign. For piezoelectric composites dh can be 60 to 70 x 10-12 C/N [52]. Hence,
piezoelectric polymer composites are of particular interest for hydrophone applications.
In addition, as for other functional composites, a larger variety in shaping and hence the
chance for new types of application is possible. Depending on the filler material or the
processing route one can achieve different types of connectivity.
Some relative early attempts to fabricate 0-3 composites consisting of a polymer matrix
loaded with ceramic particles of PZT were made by Furukawa and others [53]. A 200
m thick PZT-epoxy film has been prepared using ceramic grains with diameter range
between 0.2 and 2 mm up to a volume loading of 23%. For a poling field of 12 MV/m a
piezoelectric coefficient d31 = 0.55 x 10-12C/ N was found.
Pardo and others [54] have also reported on these composites. They found that a mixed
connectivity composite was obtained if the ceramic grain size was larger than the thickness,
regardless of the volume fraction of the ceramic. The d33 coefficient was found to be
higher for the highest ceramic volume fraction (43%) and grain to thickness ratio (3.4)
used and equal to d33 = 165 x 10-12 C/ N.
A series of 0-3 type composites called piezo rubbers (PR303/PR307) have been developed
by NTK Technical Ceramics Dev. from NGK Spark Plug Co. [55, 56, 57]. They consist
essentially of very fine lead titanate particles dispersed in a chloroprene polymer matrix.
With such a material good transmitter/receiver pulse responses were achieved with dhgh
products in the range of 1 to 5 x 10-12 Pa-1.
Hanner and co-workers have done research on a practical way of preparing a piezoelectric
paint [58]. Two polymers, an acrylic and a polyurethane were utilised and loaded with
60 to 70% of PZT or lead titanate ceramic. The polymers were dissolved in a suitable
dispersing agent to which the powder was then added. Films were prepared by casting
the mixtures onto a brass plate and cured in a vacuum oven. Resulting film thicknesses
were 200 to 500 m. The best results were achieved for the lead titanate-acrylic composite.
The piezoelectric constants were dh = 32 x 10-12 C/ N, gh = 67 mV m/ N and dhgh = 2.15
x 10 -12 Pa-1. Piezoelectric 0-3 composites have been proposed for applications as
hydrophones [59], vibration sensors [60] or active control of sound [61].
317
Handbook of Polymer Blends and Composites
The fabrication of piezoelectric composites with 1-3 connectivity has been the subject of
numerous studies. Early methods, however, have been very time consuming and the
resulting composites were limited in scale and size. For tape cast composites [52] thin
PZT films are stacked with spacers and infiltrated with an epoxy. After curing, the stack
is diced in a direction perpendicular to the plane of PZT layers to form slits or trenches.
Finally the stack is refilled with epoxy and one gets a 1-3 composite. In a similar way
honeycomb structures can be infiltrated, diced and refilled.
The recently developed processing for direct spinning of ceramic and piezoelectric fibres
[62, 63, 64] offers much easier processing routes for 1-3 composites. The fibre possesses
a dense microstructure with grain sizes of 2 to 8 m. A small amount of microporosity
(0.3 m in diameter) is visible at the fracture surface. Such fibres can be combined to a
yarn of about 200 filaments. In order to prepare a 1-3 composite they may be woven
through a honeycomb support structure, heat treated and fired, before they are
impregnated with a polymer [52].
PZT-silicone rubber 1-3 composites have been used, e.g., as acceleration sensors which
had twice as much output voltage as that of a ceramic PZT transducer of a similar design
[65]. This was observed in a wide range of frequency, acceleration and added load. In
addition, it has been proposed to use resistively shunted piezoelectric fibres for passive
vibration damping [66]. When a finite resistor shunts a piezoelectric element, deformation
would cause a current through the resistor, dissipating energy through joule heating. For
composites the idea would be to incorporate resistive fibres or particles in parallel to the
piezoelectric fibre or to coat them with a resistive layer. A theoretical study showed that
a peak damping loss factor of 12 % should attainable at a piezoceramic fibre volume
content of 30% [66]. The highest loss factors were obtained in this study for 3-3 coupling.
318
Functional Polymer Composites
A new method for preparing 3-3 connectivity composites has been developed by Gururaja
and co-workers [69]. It consists of preparing a PbTiO3 ceramic using a co-precipitation
method, and then forming pellets using a binder, which is burnt out during a firing
process. The voids were subsequently filled with an epoxy to form composites with filler
loadings of about 70%. The composites were poled both conventionally and by a corona
method, which provided similar results. A measurement of the (002) and (200) x-ray
diffraction peaks before and after poling showed an apparent saturation of the poling.
The piezoelectric hydrostatic figure of merit dhgh was 1.75 x 10-12Pa-1.
The polymer consists of a polyelectrolyte, which possesses ionisable groups on its backbone.
First investigations of ionic chemomechanical deformations have been made on polyacrylic
acid (PAA) and polyvinyl acetate (PVA) [70, 71, 72, 73]. First electrochemical transduction
was reported for the PVA-PAA co-polymer system [74]. It can generate mechanical tension
comparable to those typical for animal muscles in response to chemical stimuli [75].
where M+ is the counter ion (H+, Li+, Na+ or others). The structure of Flemion is similar:
319
Handbook of Polymer Blends and Composites
In order to chemically electrode thin films of these materials, metal ions (platinum, gold
or others) are dispersed throughout the hydrophilic parts of the polymer, and are
subsequently reduced to the corresponding zero-valance metal atoms. This results in the
formation of dendritic type electrodes (Figure 9.9). Finally the IPMC films are plated
with metal and contacted.
One property of the IPMC is the ability to absorb large amounts of polar solvents, e.g.,
water. If the interstitial space of these polyelectrolyte networks is filled with ion (M+)
containing liquid (water-based), the electrophoretic migration of such ions inside the
structure due to an imposed electric field can cause the macromolecular network to
deform accordingly. For Nafion it is also assumed that electro-osmosis plays an
important role, where each cation drags some water in with itself. In particular, by
replacing Na+ by Li+ the efficiency is dramatically reduced, while the osmotic pressure
should be essentially unaltered.
320
Functional Polymer Composites
Once the shape has been recovered there is no change in shape when the specimen is
cooled to below Mf. The shape memory can only be reactivated by deforming the
martensitic phase again. This is the one-way shape memory effect. A reversible effect can
be achieved by applying an external force, which deforms the material in the martensitic
phase and creates the one-way effect all over again. The reason is that the martensite has
a much lower strength than the high temperature phase.
321
Handbook of Polymer Blends and Composites
Figure 9.10 Principle plot of a property change, e.g., electrical resistivity versus
martensitic transformation occurring in a shape memory alloy including the
characteristic transformation temperatures Ms, Mf, As, Af.
six or seven training cycles, microstresses become inbuilt in the parent phase. These
stresses bias the normal martensitic transformation to the extent that a preferred variant
of martensite is formed on cooling. Accordingly the transformation shape change of this
single variant of martensite gives rise to a spontaneous deformation of the specimen on
cooling. When the material is heated the original shape is recovered due to the one-way
effect. A comparison of one-way and two-way effects is given in Figure 9.11.
Among the advantages of SMA actuators compared to other types of actuators are the
very large deformations (>5%) and stresses (>500 MPa) that can be generated. Typical
applications of SMA are fasteners, connectors, flexible robot structures, movable flaps
for ventilation or lightning and actuators replacing motors, where small space and quite
operation is required.
SMA in the form of thin wires can be integrated in fibre reinforced polymer composites,
yielding adaptive composites. They can actively change their shape, stiffness, internal
stress level and natural frequencies of vibration either due to a current applied to the
SMA wires or due to a change of the ambient temperature. Most functional properties
of SMA composites are directly related to the following mechanism on the level of the
SMA wires. Prestrained martensitic SMA wires operate upon heating against the elastic
322
Functional Polymer Composites
Figure 9.11 Comparison of the one-way (left) and two-way (right) shape memory effect
stiffness of the host matrix, biasing the strain recovery of the SMA wires. The result is
that recovery stresses are generated gradually by the SMA in the composite during heating
and strain recovery of the wires is delayed. After overcoming a temperature hysteresis,
the reverse process occurs during cooling. There is not yet much literature on the
integration of SMA into composites. However, a number of experimental studies have
been devoted to active strain energy tuning and active modal modification in order to
modify the vibrational characteristics [81, 82, 83].
An example is given in Figure 9.12. It shows the natural vibration frequency upon change
of temperature for an epoxy composite containing 5 volume % NiTi fibres, which had
been pre-strained by 2.5% [82]. The temperature increase is achieved by heating the
SMA wires by an electrical current. One observes an increase in the natural vibration
frequency by about 50%. The change in frequency occurs over a temperature range of
less than 20 C, which means that only small electrical power is needed to actuated this
effect.
A variety of shape memory plastics have also been developed. Polynorbornene has excellent
shape memory recovery properties at temperatures of 35 C. It is available under the
trade name Norsorex made by Zeon Chemicals Inc. However, drawbacks are its super
high molecular weight, which makes it hard to process, and that the shape memory
transition temperature can not easily be changed, both hindering commercial applications.
So far they do not play an important role.
323
Handbook of Polymer Blends and Composites
Figure 9.12 Effect of 5 vol.% NiTi fibres with 2.5% pre-strain embedded in an epoxy
matrix; closed circles: heating; open circles: cooling (Reproduced with permission
from J-E. Bideaux [82], published by Materials Research Society, 1997)
The dielectric constant of composites depends largely on the polymer matrix but to
some extent also on the filler. For increasing permittivity p of the particle, however,
the electric field E p inside the particles decreases, i.e., the field is expelled from the
filler. This affects in particular the transfer of nonlinearity from nonlinear particles
to the composite. A simple nonlinear effective medium approach has been developed
to study this effect [84]. As an example, consider particles with a permittivity which
increases in increasing electric field p (E p) = 1+(E p /E 0 -1) 2 , with an arbitrary
normalisation constant E0. Figure 9.13 shows the increase of the resulting effective
permittivity as a function of the external field E a for different aspect ratios of the
filler particles. The upper three lines show the permittivities within the particles, p,
and the lower lines the effective composite permittivity, . For the latter, a saturation
effect at high E a is clearly seen. Even if p rises up to about 100, the effective
permittivity stays below 10.
Composites filled with nonlinear dielectric particles have been studied experimentally
using barium titanate [84]. Barium sulphate exhibits an increase of around the Curie
temperature (the critical temperature above which the ferroelectric behaviour of
materials disappears (coupling of electrical dipoles is destroyed by the thermal
324
Functional Polymer Composites
Figure 9.13 Effective nonlinear permittivities for various aspect ratios. The filling
factor for all curves is q = 0.3. The nonlinear permittivity of the filler material is
chosen to be p(Ep) = 1+(Ep/E0 1)2 for Ep > E0
movement)). The room temperature value of is 580 at 120 Hz. Around 130 C it
increases up to 730. For higher temperatures the decreases again. Figure 9.14 shows
how this nonlinearity is transferred to the composite properties. The temperature
dependence of at 50 Hz is plotted for different filling fractions of coarse barium
titanate (545 m) in Araldite CY 184. For 42 volume %, increases slightly from
18.6 to 25 between 30 C and 120 C, but between 120 C and 155 C it reaches a
value of 39. Of course, lower filler fractions result in smaller dielectric constants, but
also the high-temperature increase is smaller and shifted to higher temperatures.
325
Handbook of Polymer Blends and Composites
The highest field stress is obviously located at the interface between the insulator and the
inner electrode. One main objective is to reduce the dielectric stress in the insulating
material and at the interfaces between the insulation and the electrodes in order to increase
the lifetime and reliability of the system. By the use of an insulating material, which
increases its dielectric constant for increasing electric field, the field stress could be
considerably reduced at the inner electrode. In Figure 9.15 the calculated field distribution
is depicted for E2.
326
Functional Polymer Composites
Figure 9.15 The radial field distribution for E = const (solid line) and E E2
(dashed line)
the electric field is reduced at the inner conductor, where the temperature is highest, i.e.,
where the dielectric stress is highest.
For other geometries the problems are similar. In cable terminations, additional AC losses
occur especially in the area where the shielding electrode is cut off. Temperature-dependent
dielectric stress relief should lead to better control of the field in this crucial area as well.
Nonlinear optical properties of polymer composites or blends are quite attractive for
large volume applications as smart windows, mirrors or liquid crystal displays (LCD).
Control of heat and light transmission can improve comfort and safety. Many automobile
companies have begun to offer electrochromic mirrors. The mirrors will automatically
dim in response to glare from the headlights of trailing vehicles. Control of interior
heating of automobiles via solar transmission is already important. It will become even
more important in future with the appearance of electrical vehicles when the power
demand of air-conditioning units has to be limited.
327
Handbook of Polymer Blends and Composites
Domestic applications are transparent insulation materials (TIM) and solar collectors. TIM,
which can be attached to the dark coloured, light absorbing outside walls of houses, permit
simple conversion of the incident sunlight to heat. The collected heat might be conducted
directly through the wall to the buildings interior and can be used as a partial substitute for
conventional heating sources. Major heat losses to the environment are avoided due to the
insulating properties of the TIM. In this way, up to 80% of the heating energy may be saved.
However, just like solar collectors, such systems need to be protected against overheating by
reducing the incident light intensity, in particular at high temperatures in summer. For this
purpose a plastic film with a temperature dependent transmission would be ideally suited.
The light transmission of such films with different thickness is shown in Figure 9.16. Up
to 4050 C the transmittance is high, about 0.80.9. Between 50 and 60 C the film
328
Functional Polymer Composites
becomes cloudy. Upon further temperature increase the transmittance is strongly reduced
and can be as low as 0.1 for 2.3 mm films. The highly interesting aspect of this system
and most unusual for polymer blends is that the phase separation is rapid and completely
reversible below the switching point, even at temperatures as low as 30 C.
For display applications a new route to produce polarisation dependent colour filters
has been studied by Dirix and others [86]. The filter is based on high-density polyethylene
filled with pure silver nanoparticles. Flexible films containing pearl-necklace-type arrays
of the metal nanoparticles are obtained by stretching. The aspect ratio achieved has been
25. Such films exhibit a colour that is strongly dependent on the polarisation direction of
the incident light. The films exhibit bright colours and appear either yellow or red
depending on the polarisation direction. The main advantage is that, compared to classical
LCD, the top polariser (analyser) and colour filter can be replaced by one single film
with polarisation dependent colour resulting in cheaper and brighter displays.
9.12 Conclusions
An overview has been given on the major physical properties of functional polymer
composites. It has been shown that a large variety of effects are feasible by the right
choice of filler material, polymer matrix, and connectivity of the filler and processing. In
all cases the ease of shaping the functional material to complex forms and cheap
manufacturing routes are main drivers for the development. It can be expected that due
to these advantages more and more classical single-phase materials will be replaced by
functional polymer composites.
Acknowledgement
The authors wish to thank Felix Greuter, and Claus Schler for many helpful discussions,
and Ruzica Loitzl, Andreas Garbin and Markus Zehnder for experimental support.
References
329
Handbook of Polymer Blends and Composites
7. T. Kobayashi and H. Endo, NEC Research & Development, 1987, 86, 81.
9. R. Zallen, The Physics of Amorphous Solids, John Wiley & Sons, New York,
USA, 1983, Chapter 4.
10. D. Stauffer, Introduction to Percolation Theory, Taylor & Francis, London, UK, 1985.
21. R. Holm and E. Holm, Electrical Contacts; Theory and Application, 4th Edition,
Springer Verlag, New York, USA, 1967.
330
Functional Polymer Composites
22. G. Pearson, inventor; Bell Telephone Laboratories, Inc., assignee; US 2,258,958, 1941.
23. E. Frydman, inventor; Automatic Telephone & Electric Company Ltd., assignee;
UK 604 695 I 718 14S, 1948.
25. K. Ohe and Y. Natio, Japanese Journal of Applied Physics, 1971, 10, 99.
30. T. R. Shrout, D. Moffatt and W. Huebner, Journal of Materials Science, 1991, 26,
1, 145.
32. M. B. Heaney, Journal of Applied Physics Letters, 1996, 69, 17, 2602.
33. R. Strmpler, G. Maidorn and J. Rhyner, Journal of Applied Physics, 1997, 81,
10, 6786.
38. F. Carmona, R. Canet and P. Delhaes, Journal of Applied Physics, 1987, 61, 2550.
331
Handbook of Polymer Blends and Composites
40. P. K. Pramanik, D. Khastagir and T.N. Saha, Journal of Materials Science, 1993,
28, 13, 3539.
47. L-G. Virsberg and P.H. Ware, IEEE Transactions on Power Applications and
Systems, 1967, PAS-86, 9, 1124.
50. K. F. Etzold in The Electrical Engineering Handbook, Ed., R. C. Dorf, CRC Press
Boca Raton, FL, USA, 1993, 1087.
53. T. Furukawa, K. Ishida and E. Fukada, Journal of Applied Physics, 1979, 50, 7, 4904.
54. L. Pardo, J. Mendiola and C. Alemany, Journal of Applied Physics, 1988, 64, 10, 5092.
55. H. Banno and S. Saito, Japanese Journal of Applied Physics, 1983, 22,
supplement 2, 67.
332
Functional Polymer Composites
60. S. Egusa and N. Iwasawa, Journal of Materials Science, 1993, 28, 6, 1667.
62. K-C. Chen, H. Zheng and J. D. MacKenzie, inventors; The Regents of the
University of California, assignee; US 5,072,035, 1991.
63. W. Glaubitt, D. Sporn and R. Jahn, Journal of Sol-Gel Science and Technology,
1994, 22, 525.
333
Handbook of Polymer Blends and Composites
72. W. Kuhn, B. Hargitay, A. Katchalsky and H. Eisenberg, Nature, 1950, 165, 514.
80. Shape Memory Effects in Alloys, Ed., J. Perkins, Plenum Press, New York, USA,
1975.
81. C. A. Rogers, C. Liang and D. K. Barker, U.S. Army Research Office Workshop
on Smart Materials, Structures and Mathematical Issues, Virginia Polytechnic
Institute and State University, Blacksburg, VA, 1988, 39.
82. J-E. Bidaux, J-A. E. Mnson and R. Gotthardt, Materials Research Society
Symposium Proceedings, 1997, 459, 107.
86. Y. Dirix, C. Bastiaansen, W. Caseri and P. Smith, Advanced Materials, 1999, 11,
3, 223.
334
10
Conducting Polymer Composites
S. J. Dahman
10.1 Introduction
Electrostatic discharge (ESD) [1] is an issue that has been a focus of attention to those in
the electronics industry for well over a decade [2]. ESD is the rapid, spontaneous transfer
of electrostatic charge induced by a high electrostatic field. Common observable ESD
events include lightning and the occasional zap felt when reaching for a door knob. ESD
in these forms is easily seen and quite noticeable. However, it is ESD that is not readily
apparent that is more of a hazard to the electronics industry. For instance, the typical
human threshold of ESD awareness begins around 4000 volts and a discharge of less
than 250 volts can be harmful to a device.
ESD causes both catastrophic and latent failures in sensitive electronic devices. This
leads to higher manufacturing costs, lower yields, decreased reliability, and reduced profits.
It is estimated that losses caused by ESD are costing the electronics industry billions of
US dollars annually [3, 4]. This is in the order of 8-33% of production or 5% of total
sales [5, 6]. Despite a major effort to rein in losses, the continuing trend for smaller and
faster electronic devices will cause ESD to remain a major concern. It is essential, therefore,
that ESD be effectively controlled.
Contributing to the problem of ESD control is that plastics have become the materials of
choice in the manufacturing and packaging of electronic devices. This is a result of
economic benefits and greater design freedom over metals. However the typical plastics
of choice in these applications are insulators, which raises the likelihood of an ESD
event. In order to provide ESD protection, a degree of static dissipation needs to be
imparted to the plastic part.
Materials can be classified according to their electrical performance. For instance, materials
are defined as metals, semiconductors, or insulators based on their conductivity as shown
in Figure 10.1 [7, 8]. The basic unit of conductivity, , is the Siemen per centimeter (S/cm).
Metals have the highest conductivities (>10 0 S/cm). Insulators have the lowest
conductivities below 10 -7 S/cm. Semiconductors are materials with intermediate
conductivities that range between 100 S/cm and 10-7 S/cm. Electrical conductivity spans
one of the largest ranges in the measurable properties of materials. For example, some
335
Handbook of Polymer Blends and Composites
L
= (1)
R A
336
Conducting Polymer Composites
It is within the dissipative range (105-1012 ohm/sq) that ESD is most effectively controlled.
Charges will not be generated and will not remain localised on the surface of the part.
Additionally, within this range, the material is slightly conductive which provides a resistive
path to ground. Therefore, both conditions for proper ESD control are met [11]:
337
Handbook of Polymer Blends and Composites
moulded or extruded without these drawbacks and have been utilised successfully in a
number of ESD control applications. In fact, these represent one of the fastest growing
classes of specialty thermoplastics.
Filled static dissipative thermoplastics have traditionally been prepared by the compounding
of additives such as hygroscopic agents or conductive fillers such as carbon black, carbon
fibres, metallic particles. into the polymer matrix. Filled polymers have a long history and
the widest use in electronic devices. This is attributed to the aforementioned ease of
processing as well as a wide range of electrical properties available with the fillers.
This is not to say that conventionally filled polymers for ESD control do not have
disadvantages. Their inherent weakness is that they are inhomogeneous, multi-phase systems.
This usually leads to reduced properties such as strength and toughness.
There are also shortcomings with each type of additive. For example, hygroscopic agents
are by nature dependent on moisture for proper functioning, and therefore, fail in low
humidity environments. Additionally, hygroscopic agents are migratory which leads to
permanence and contamination issues. The agents are easily washed or wiped off and
performance degrades over time. Finally, these additives can only achieve the antistatic
portion of the dissipative range (see Figure 10.3). In this range, a part will resist triboelectric
charge generation (generation of charge by two materials coming into contact and then
separating), but may not be able to safely drain a charge to ground fast enough.
Systems containing conductive fillers such as carbon black also have drawbacks. Such
systems are percolative which form conductive pathways above a critical concentration of
additive (the percolation threshold). The percolation threshold varies with additive type
and polymer matrix. These types of systems can become quite conductive. However, high
surface conductivity leads to uncontrolled discharge. For many ESD applications, the optimal
surface resistivity range is between 106 to 109 ohms/square (see Figure 10.3) [12].
It is often difficult to provide a reproducible surface resistivity in the optimal ESD range for
percolative systems. The compounds are usually either too conductive or not conductive
enough since the insulator to conductor transition occurs over a very narrow range of additive
concentration. Figure 10.4 shows this behavior for a typical carbon black compound.
Inherently conductive polymers (ICP), represent another potential class of conductive additives
for ESD control. These are a different class of materials than filled polymers, which are simply
physical mixtures of polymers and additives. ICP are polymers that are conductors by themselves.
The repeat units of several families of ICP are shown in Figure 10.5. The fundamental
characteristic of ICP is an extensive -conjugated system in the main chain. This is formed from
the overlap of carbon pz orbitals and alternating carbon-carbon bond lengths or nitrogen pz
orbitals and C6 rings. Examples of conjugation paths are shown in Figure 10.6.
338
Conducting Polymer Composites
ICP are typically insulators or weak semiconductors in the natural state. They are
transformed to the conductive state when complexes are formed between the chains and
guest species. This process has been termed doping in analogy to conventional
semiconductor doping. Doping creates the charge carriers and the mobility necessary for
electrical conduction. An example structure of a doped ICP is given in Figure 10.7.
To date the use of ICP in ESD thermoplastic applications has been nearly non-existent
due to [13]:
339
Handbook of Polymer Blends and Composites
340
Conducting Polymer Composites
Polyacetylene
Polypyrrole
Polyaniline
A-
H H H H H H
C C C C C C
C C C C C C C
H H H H H H H
Figure 10.7 Doping of polyacetylene with the charge on the polymer backbone is
delocalised over several repeat units
341
Handbook of Polymer Blends and Composites
The problems encountered during processing are of prime concern since conducting
polymers will never be more than a laboratory curiosity if useful articles with
commercial application cannot be produced.
ICP were initially regarded as intractable materials (infusible and insoluble). Although
-conjugated structures and doping lead to conducting polymers, they also give rise
to increased chain stiffness and chain interactions leading to intractability. Several
ingenious methods have been devised to overcome the intractability. The first is to
simply polymerise directly into the final shape [14]. The next was the use of precursor
polymers [15]. Others include substituted derivatives and copolymers [16]. Direct
synthesis and precursors are typically impractical from a commercial viewpoint while
substituents and copolymers gain processability at the expense of conductive
performance.
Even with these advances in processability, doping has historically taken place after
an article has been shaped into its final form such as a film, fibre, or moulded article.
Due to the high intractability, it has been generally accepted that doped ICP cannot
be processed from common nonpolar organic solvents or from melts. Since ICP are
commonly doped by exposure to liquid or vapour phase dopants, the doping step is
diffusion limited. Therefore, the time needed for the dopant to completely saturate
the object and reach the final desired conductivity is usually inappropriate for mass
production [17]. The problem is compounded if the ICP is to be part of a multi-
component system. Other methods of doping or system formulation are needed to
make use of ICP as additives for ESD applications.
Recently, several methods have been devised to improve the utility of ICP. These
advances centre on the use of ICP as additives into polymer matrixes. The advantages
of these composites are to control the conductivity into the optimal ESD protection
range, improve mechanical properties, and at lower cost compared to past ICP
alternatives. Additionally, processing is sufficiently enhanced so that thermoplastic
compounds can be produced on existing equipment. These developments include the
use of ICP as coatings onto various substrates and the use of novel doping schemes
[19-21].
342
Conducting Polymer Composites
ICP/carbon black (ICP/CB) structures are a new class of conductive additives for
thermoplastics and thermosets. These materials are essentially ICP coatings on carbon
blacks prepared via in situ polymerisation and deposition of ICP onto a CB substrate
[20]. A simple structure of an ICP/CB is represented in Figure 10.8.
Figure 10.8 Representation of a typical ICP/CB structure showing the ICP coating on a
carbon black substrate
ICP/CB composites are very flexible conductive additives that offer excellent tuning
opportunities. For example, tuning prospects include the variation of the type and
amount of conductive polymer, the associated dopant, and the grade of CB. This
inherent control offered by ICP/CB additives coupled with improved thermal stability
allow ESD control materials based on a variety of commercially available
thermoplastics such as polypropylene, high impact polystyrene, Nylons, polyesters,
or acrylonitrile-butadiene-styrene (ABS). A fine conductive ICP/CB percolative
343
Handbook of Polymer Blends and Composites
structure is formed within the matrix resin and eliminates the possibility of
macroscopic regions with a sharp gradient of conductivity, i.e., hot spots.
The ICP/CB additive results in a greatly reduced surface area and pore volume
compared to that of the original CB. As a result, improved mechanical and melt-flow
properties are obtained. It is also possible to tailor the polarity of ICP/CB additives
to make them more compatible with the chosen thermoplastic [23]. Therefore, the
mechanical properties of the resin can be further improved over traditional CB
compounds.
10.2.1 Stability
344
Conducting Polymer Composites
The ICP/CB additives are much more stable at 300 C in air and therefore show superior
thermal stability over doped polyaniline. One of the possible reasons for such enhanced
stability is the multiple interactions between the conductive polymer backbone and the
carbon black substrate. It was also found that there was no significant change in the electrical
properties of ICP/CB after storage under ambient conditions for over 9 months [23]. The
shelf life of several tested additives made earlier is at least in the range of 3 years.
Another check of thermal stability is to monitor the conductivity of composites that contain
ICP/CB additives during long-term ageing at elevated temperatures. Figure 10.10 shows
the effect that long-term aging at elevated temperature has on the conductivity of finished
parts that are produced from Nylon compounds. In both systems, the conductivity has not
changed over a time period of over 4000 hours while at a temperature 100 C. Similar
results with excellent long-term stability were measured between 80 and 130 C [26].
In addition to thermal stability, the chemical stability of ICP additives must be addressed.
Both doped polyaniline and polypyrrole are usually considered as the most
environmentally stable ICP [27]. However, none of these polymers will be resistant to
treatment by a strong base. For instance, doped polyaniline with organic acids will be
345
Handbook of Polymer Blends and Composites
undoped with a loss of conductivity and irreversible polymer degradation will occur
[28]. On the other hand, the ICP/CB compounds have good chemical stability as shown
in Table 10.1. Table 10.1 shows that no significant change in conductivity was seen for
an ICP/CB-polypropylene compound when exposed to acid and only a slight change was
measured for that exposed to strong base.
One of the major advantages of the ICP/CB composites over traditional additives for
ESD applications is their flexibility with respect to the surface area, surface energy,
and electrical properties. This versatility is achieved as a result of the in situ synthesis
and deposition of ICP onto carbon black substrates. In short, this can be accomplished
by changing the carbon black substrate, the type of ICP chosen, the total coating level,
and the dopant system. Thus, the additive can be tailored to be compatible with a
variety of commercial plastics and an ESD material with the optimal electrical and
structural properties.
One concern with compounds based on carbon is the difficulty in processing due to
the increase in viscosity as a result of high surface area. This is related to the interaction
of a higher surface area additive with the polymer matrix. Deposition of ICP onto CB
reduces surface area. For instance, the surface area of ICP/CB additive containing 30
wt% of doped polyaniline is about three times less than the surface area of uncoated
Ketjenblack EC as shown in Figure 10.11.
The reduction in surface areas should provide easier processing of the composites.
Indeed, compounds incorporating ICP/CB were found to process much easier than
compounds containing standard CB. Figure 10.12 demonstrates this that even at higher
concentrations, the ICP/CB additive shows improved melt flow behaviour compared
to high structure CB.
346
Conducting Polymer Composites
Figure 10.11 Surface area as a function of coating level for polyaniline on high
structure carbon black
Figure 10.12 Melt flow behaviour of conductive Nylon 6 compounds: (O) 5 wt.%
high structure carbon black and () 7.5 wt% of ICP-C/CB2.
347
Handbook of Polymer Blends and Composites
ICP/CB additives are also versatile with respect to their surface properties. This makes
it possible to tailor the polarity of ICP/CB additives for greater compatibility with
the resin system of choice. Table 10.2 shows the effect that different ICP coating
systems have on the surface properties of two different types of carbon blacks, CB1
and CB2. It is observed that the surface polarity can be tuned depending on the ICP
system applied. For example, the polarity of CB1 was found to become about 8
times less polar by the addition of ICP-A. In contrast, the polarity of CB2 was increased
by about fifteen times for ICP-B and about one hundred times for ICP-C.
This can be important in the selection of the most suitable ICP/CB additive for a
certain polymer matrix. Increased compatibility tends to lead to enhanced mechanical
properties. Table 10.3 gives the surface properties for two different polymers: silicone
and polyamide. By comparing the surface parameters of Table 10.2 to Table 10.3, it
is expected that ICP-B/CB2 is a good for silicone while ICP-C/CB2 should be chosen
for polyamide.
Table 10.2 Surface properties of some carbon blacks and ICP/CB additives
Sample Average Water Total Surface Percent Surface
Contact Angle Energy Polarity
(degrees) (mJ/m2) (%)
CB1 114 15.0 0.53
ICP-A/CB1 115 18.0 0.07
CB2 114 15.6 0.28
ICP-B/CB2 97.2 24.9 4.14
ICP-C/CB2 67.6 40.2 27.5
348
Conducting Polymer Composites
As discussed previously, perhaps the major drawback with traditional conductive additives
is the lack of control over the electrical properties. With traditional additives, the steepness
of the percolation curve is the cause of the problem. It indicates that the conductivity
changes by several orders of magnitude over a very small range of filler content.
Furthermore, experience has shown that at these content levels the electrical properties
are highly sensitive to processing conditions such as the mixing regime, processing
temperature, the speed of extrusion and cooling, etc. As a result, the material is likely to
contain non-uniformity and hot or cold spots (macroscopic regions of high or low
concentrations of conductive additive) will result during compounding and processing.
The discharging of static electricity will be too fast or too slow from these regions and
ESD protection may not be adequate for highly sensitive electrical components.
The surface resistivity of potential ICP/CB Nylon ESD control composites are given in Table
10.5. This demonstrates that the optimal ESD range can be obtained with ICP/CB additives
349
Handbook of Polymer Blends and Composites
and that the performance can even be customised within the range. In many cases, the desired
levels of surface resistivity can be maintained to within an order of magnitude.
The excellent ESD properties of ICP/CB composites displayed in Table 10.5 are the result
of shallower percolation curves. Typical percolation curves for ICP/CB composites are
shown in Figure 10.13. The tuning capability of ICP/CB additives leads to compounds
having tailored electrical properties with shallower percolation curves. This leads to much
less sensitivity due to processing conditions or fluctuations in the amount of filler content.
The result is a more robust and reproducible ESD control compound. It is also observed in
Figure 10.8 that not only are the percolation curves shallower but also that the saturation
levels are variable. This allows even more precise tailoring of an ESD control material.
350
Conducting Polymer Composites
The observed onset of saturation of the percolation curves indicates that at higher ICP/CB
concentrations (beyond the percolation threshold region), a well established and
interconnected network is formed. Such a high content of the additive and its uniform
distribution within the matrix polymer will eliminate the existence of non-homogeneous
conductive regions or hot spots within the material. This intrinsic property of the
conductive network is a key to creating proper ESD control materials. On the other
hand, in the CB-filled materials, the existence of conductivity fluctuations is predictable
due to fact that to keep the resistivity at the ESD range, it is necessary to use an exact
amount of a certain CB. At such filler contents (the middle of the percolation threshold)
the formation of the conductive network is not yet fully completed. Thus, there are
separated islands of CB agglomerates surrounded with insulating plastic. Incompatibility
of CB in certain resins might also contribute into formation of undispersed macroscopic
conductive islands. Thus there are macroscopic regions that may be large enough to
cause an ESD event.
10.2.5 Purity
The emission of volatile components, such as water, monomers, oligomers, and reactor
residues is known as outgassing and is a potential problem concerning the contamination
of electronic component surfaces [29]. Therefore, it is highly desirable to keep the total
emissions as low as possible, especially for materials to be used in specialised electronics
industries. The outgassing of some ICP/CB compounds are shown in Table 10.6. Both
compounds show very low levels of outgassing and are well within the specification
limits required for materials suitable for spacecraft applications.
Table 10.6 Outgassing for ICP/CB in Nylon and PP tested to ASTM E-595
Spec. Limits ICP/CB-2 in Nylon ICP/CB-5 in PP
TML % 1.0 0.54 0.32
CVCM % 0.10 0.01 0.07
TML: Total Mass Loss; CVCM: Collected Volatile Condensable Material
Also important is the potential for leachable contaminants. Table 10.7 shows the
concentrations of some leachable ions (chloride and sulphate) found in an ICP/CB-
polypropylene resin compared to a standard CB in the same resin. Typically, chlorides
and sulphates are the most important to those applications requiring high purity materials.
The results indicate that the ICP/CB-resin compound has a significantly higher purity
351
Handbook of Polymer Blends and Composites
than standard CB. It is over four times cleaner with respect to chloride and over twice as
clean with respect to sulphate as a standard CB compound.
As seen in previous sections, the flexibility of ICP/CB additives allows for the tailoring of
desired properties such as surface resistivity into the optimal ESD control range. Such
tuning can be accomplished by varying: the type of conductive polymer, coating level,
the nature of the dopant system, or the grade of carbon black. These differences were
also observed to cause changes in surface properties, specifically polarity.
Double percolation phenomenon has been reported previously for two-phase polymer
blends loaded with carbon black [30-32]. When conductive carbon black was
preferentially localised at the phase boundary, very low percolation thresholds were
observed [30]. The ability to adjust the polarity of ICP/CB additives should allow for the
opportunity of tuning for localisation in a variety of immiscible blends.
Solvent extraction indicates that the more polar ICP-C/CB2 is preferentially located in
the Nylon phase as anticipated from Table 10.3. In this case, when Nylon 6/polypropylene
blends containing ICP/CB were ultrasonically mixed in formic acid (formic acid is a
solvent for Nylon 6 and a non-solvent for polypropylene) the entire ICP/CB additive was
washed away with the soluble Nylon phase. The PP phase remaining after extraction
was white in colour indicating that all the additive is located within the Nylon phase.
Also, differential scanning calorimetry (DSC) tests on the PP phase indicated efficient
Nylon phase extraction from the conductive compounds. There was no detectable peak
corresponding to melting point of Nylon 6 present in the DSC spectra.
352
Conducting Polymer Composites
Figure 10.14 Volume resistivity of polyamide (PA)/PP blends with 4 wt% of ICP/CB
versus ICP coating level (PA/PP = 1/1)
It is expected that an ICP/CB with intermediate polarity would tend to be localised at the
phase boundary of plastic blend [30]. By varying the conductive polymer to CB ratio, it
may be possible to fine tune the polarity of ICP/CB additive for multi-phase blends and
achieve lower percolation thresholds due to selective localisation at the phase boundary.
This behaviour is observed in Figure 10.14 as a drop in volume resistivity at a coating level
centred around 5 wt%. This occurred at a concentration of 4 wt% of ICP/CB additive as
compared to more typical percolation thresholds of 7-12% as seen in Figure 10.13. Thus,
the tuning ability of ICP/CB materials and multi-phase blends can be used to obtain double.
The substituted derivative method has been the most successful approach in rendering
ICP more capable of being processed into useful articles. Typically, flexible side groups
such as alkyl or alkoxy chains are attached to the backbone. When the chains are
sufficiently long, usually greater than 3-4 carbon atoms, the polymers become solution
and/or melt processable. Conducting polymers that have been altered by this method
include polythiophene, poly (phenylene vinylene), and polypyrrole as shown in
Figure 10.15 [33-34, 16].
353
Handbook of Polymer Blends and Composites
Unfortunately, conducting polymers with chemically anchored side chains typically exhibit
lower conductivities than linear unsubstituted conducting polymers. Additionally,
substitution generally does not facilitate processing the doped polymer. For instance,
doped poly (3-octylthiophene) (P3OT), crosslinks and undergoes thermal dedoping during
melt processing and becomes insoluble if solution processing [18].
Unfortunately, more attention has been paid to the processing of conducting polymers
with functionalised dopants from solution [35-38]. However, many ESD applications make
use of thermoplastics. The advantages of thermoplastic processing are the ease of producing
moulded or extruded parts as well as a lack of solvent removal and recovery concerns.
Even with the use of functionalised dopants, ICP typically have mechanical properties
that are unacceptable for most ESD applications. Polymer blend processing would be a
354
Conducting Polymer Composites
(a)
(b)
Figure 10.16 Some functionalised protonic acids for polyaniline: (a) camphor
sulphonic acid (CSA), and (b) dodecylbenzene sulphonic acid (DBSA)
For a substance to exhibit electrical conductivity, the existence and mobility of charge
carriers is required. In order to maximize conductivity, it is implied that charge carrier
density and mobility should have the highest possible level where e is the charge per
carrier, n is the number of carriers, and is the mobility according to [39]:
=en (2)
In conducting polymers, the doping process creates a generous number of charge carriers
either by adding electrons to or, more typically, by removing electrons from the conjugated
backbone.
In a polymer solid, since single chains would not be expected to stretch all the way
across a manufactured article, the bulk three-dimensional conductivity not only depends
on charge transport along the polymer chain, but also on charge hopping from polymer
chain to polymer chain. Transport along a backbone is referred to as intrachain transport
while hopping between chains is called interchain carrier transfer. Additionally, if
isolated conductive domains exist in an insulative matrix another type of transfer
mechanism is required: interphase carrier transfer. A possible path illustrating all the
types of transport for a charge carrier travelling through a conducting polymer material
is shown in Figure 10.17.
355
Handbook of Polymer Blends and Composites
Figure 10.17 Potential conduction path of a charge carrier travelling through an ICP-
polymer blend: (A) intrachain transport, (B) interchain transfer, and (C) interphase transfer.
From this it can be deduced that carrier mobility is the limiting factor for conductivity since
there are a large number of carriers introduced into the conjugated system by the doping
process and the complicated resistance to mobility found in three-dimensional articles. Since
conducting polymers contain conjugated systems and doping delocalizes the charge across a
length of the chain, mobility is usually less for interchain transfer than intrachain transport.
Mobility is significantly less for interphase transfer as the charge pass tunnel through an
insulative region or a phase boundary. This implies that in order to obtain the higher charge
carrier mobilities, isolated conductive phases must be reduced, the resistance to interchain
carrier transfer decreased, and intrachain carrier transport length increased.
In the ICP-polymer blends, the conduction paths for a charge carrier travelling through
a specimen as described above are possible. In order to achieve good electrical performance,
it is imperative that the individual contributions to the overall resistivity be minimized or
optimized. For optimal control in ICP-polymer blends, this means that the conducting
polymer should form a co-continuous phase throughout the blend. In this manner, the
interphase contribution, as shown in Figure 10.17, to the overall resistance can be
eliminated. If a co-continuous phase is not achieved, behaviour similar to traditional
filled systems is to be expected.
Blends with functionalised doped polyaniline have been prepared from solution [40].
The developed morphology was significant different from that seen in standard filled
polymer systems. A typical morphology for filled systems can be represented as shown in
Figure 10.18.
356
Conducting Polymer Composites
357
Handbook of Polymer Blends and Composites
from conductive additives with higher aspect ratios is greater at any given concentration.
Therefore, the conducting polymer blends which have much higher aspect ratios than
spherical particles will form conduction pathways at much lower concentrations [41].
Recent advances have been made that render ICP complexes quite capable of forming
co-continuous phases in polymer blends within the melt. For instance, using a mixture of
functionalised dopants was found to be a simple method to improve the melt processability
of ICP while retaining relatively high levels of conductivity [19, 21]. The fundamental
difference between these complexes and the additives that were previously discussed is
that there is now no longer any need for another medium, i.e., CB, to assist dispersion
and conductive pathway formation. Therefore, these compounds can be considered as
polymer-polymer blends as opposed to filled systems. Very importantly these blends can
be manufactured directly from the melt rather than from solution as has been studied
extensively in the past.
While not strictly within classical definitions, these types of ICP-polymer blends may be
considered as interpenetrating networks. These polymers are not synthesised within one
another nor are they chemically crosslinked. However, conducting polymers usually have
a strong self-association potential [42] and the network may be considered as physically
crosslinked. The potential advantages of co-continuous morphologies can be discerned
by comparing Figures 10.18 and 10.19. In contrast to percolative types of pathways
present in filled particulate compounds, ICP-polymer blends that form fine interpenetrating
networks are expected to provide improved ESD control at lower conducting polymer
content. In turn this leads to lower cost, improved mechanical properties (especially
impact values), and the possibility for a certain degree of colourability (other than black).
358
Conducting Polymer Composites
and notched impact, are very similar to those of the matrix polymer. Particulate fillers
with a low degree of aspect ratio and reinforcement efficiency such as CB tend to increase
the strength and modulus slightly as shown. However, impact properties especially notched
impact typically falls off significantly. An advantage of the interpenetrating network is
to retain the impact properties of the base resin. Another item to note from Table 10.8,
is that the ICP composites can be coloured to a degree.
The electrical properties from different lots of composites that contain carbon black can
vary extensively because of the steep percolation curve. Consequently, wide ranges must
be listed for materials that attempt to provide ESD control. On the other hand, the
electrical performance of ICP composites is more reproducible and usually falls within
one order of magnitude. As a result, the surface resistivity can be finely tuned into the
optimal ESD protection range.
Figure 10.20 is a curve showing the conductivity as the content of the ICP complex
increases in an ICP-polypropylene composite. An extremely smooth onset of conductivity
is present at notably low ICP content. This behaviour is typical of complex network
formation [40].
359
Handbook of Polymer Blends and Composites
density increases with additional ICP complex. This performance is compared against
typical percolation behaviour in Figure 10.10.
Also shown in Figure 10.21 is that the ICP-polymer composites can be tuned to immediate
conductivity versus conductive fillers such as CB. This tuning ability coupled with a smoother
conductivity curve allows for optimal ESD protection. This is shown in Figure 10.22 for
an ICP-PP compound. In this compound, the surface resistivity is remarkably continuous,
even linear, through the optimal range (106-109 ohm/sq) making it ideal for ESD protection.
360
Conducting Polymer Composites
361
Handbook of Polymer Blends and Composites
10.3.4 Processing
Composites which contain interpenetrating ICP networks are not any more difficult to
process than the matrix resin. The melt flow behaviour of an interpenetrating ICP-PP is
given in Figure 10.23 and compared to the neat PP used as the matrix. As seen, the melt
flow of the ICP material is very similar to the matrix resin. This is different from most
particulate fillers that tend to increase the melt flow of the compound compared to the
base resin. This is especially a problem in carbon black composites that usually have
very high relative viscosities due to the high surface area.
362
Conducting Polymer Composites
Table 10.9 Comparison of surface resistivities before and after blow moulding
for ICP interpenetrating networks and carbon black in HDPE
ICP Carbon Black
7
Before 10 108
After 107 Dead
which occurred during this blow moulding operation was enough to disrupt the
percolation in CB composites while the interpenetrating network material maintained
its electrical properties. This is to be expected if a strong network has been formed.
When polymer chains or networks are oriented, the mean square step length is increased
in the orientation direction. Since the step length in the orientation direction increases, it
is expected that the average distance that a charge travels in the direction of orientation
also increases for a charge that is proceeding along the chain. As a result, the conductivity
can be enhanced in the orientation direction for stretched conducting polymers.
i
[ ] (
= fcn 2 1 ) (3)
The ability to preserve a network upon extension may also be important in certain
elastomeric applications. The maintenance of conductivity in a thermoplastic elastomer
when stretched is shown in Table 10.10.
Derivations have been made that show that the conductivity can even increase with
extension [49]. Although the conductivity did not increase substantially, it is still
remarkable that electrical performance was maintained and slightly increased when
samples were stretched unlike percolative systems. However, the stretch ratios were
363
Handbook of Polymer Blends and Composites
relatively low in this case and as seen in Figure 10.24, the conductivity would not be
expected to improve a great deal.
10.3.5 Stability
The thermal stability interpenetrating network capable ICP complexes is sufficient for
compounding into low temperature thermoplastics such as polyolefins, polystyrenes,
thermoplastic elastomers, etc. The stability of the compounds is very good at room
364
Conducting Polymer Composites
temperature, as no loss of conductivity has been seen. The stability at elevated temperature
is shown in Figure 10.25. While more typical of ICP behaviour than the ICP/CB composites
discussed earlier, the half-life at 80 C is more on the order of days and weeks rather
than the hours or even minutes that are usually reported for ICP [13]. This gives some
limited usage of these ICP compounds at elevated temperatures up to 80 C especially if
the desire is to remain within the optimal 106 to 109 ohm/square range. Additional
capability can be gained as the temperature is lowered.
10.3.6 Purity
Compounds containing ICP have the potential for high purity. Since they are based on
polymers, volatiles and leaching are expected to be low. Also, sloughing which is typically
found with particulate additives such as CB will be non-existent. However, due to the
current techniques that are used in the production of interpenetrating network capable
ICP complexes, compounds that contain these additives will typically have higher leachable
sulphate contents than most high purity conductive composites. If high purity to leachable
sulphate ions is not an issue, these compounds compare favourably to traditional high
purity compound as sloughing of conductive particles is not observed and outgassing to
conventional materials is similar or better.
365
Handbook of Polymer Blends and Composites
10.4 Conclusions
New classes of conductive compounds have been developed based on ICP. These classes
include ICP-coated fillers and interpenetrating network formable complexes. Both classes
display the unique capability of adjusting electrical performance into the optimal ESD
control range.
ICP-coated additives are manufactured via in situ polymerisation of ICPs onto various
substrates, typically carbon black. These composites are extremely thermally stable and
display useful tuning characteristics. This ability allows the additive to be tailored to the
required level of ESD protection. Furthermore, this versatility allows the control of
processing, mechanical performance, and purity to be optimised for the desired
thermoplastic.
Even more recently, ESD control compounds have been developed based on ICP-
polymer blends [48]. These materials exhibit excellent ESD properties without
degradation of physical properties or processing capability. These systems form fine
network morphologies (with the appearance of interpenetrating networks) as opposed
to the morphologies typical of percolative filled compounds. The result is excellent
ESD control with retention of the mechanical properties and flow characteristics of
the matrix resin. These composites may also attain a degree of colour other than
black. Further work is underway to improve the stability and purity of these types of
compounds.
References
3. L. Brown and D. Burns, Electronic Packaging & Production, 1990, April, 50.
366
Conducting Polymer Composites
11. J. Kanarek and W. Tan, ESDiscovery 98 Tutorial, Santa Clara, CA, USA, 1998.
13. L. W. Shacklette, C. C. Han and M. H. Luly, Synthetic Metals, 1993, 57, 1, 3532.
14. R. V. Gregory, W. C. Kimbrell and H. H. Kuhn, Synthetic Metals, 1989, 28, C823.
16. R. L. Elsenbaumer, K-Y. Jen, and R. Oboodi, Synthetic Metals, 1986, 15, 169.
22. S. J. Dahman and J. K. Avlyanov, Proceedings of SPE ANTECH 98, Atlanta, GA,
USA, 1998, Volume II, 1313.
367
Handbook of Polymer Blends and Composites
25. H. Kuhn and A. Child in Handbook of Conducting Polymers, 2nd Edition, Eds.,
T. A. Skotheim, R. L. Elsenbaumer and J. R. Reynolds, Marcel Dekker, New
York, 1998, 993.
26. S. J. Dahman and J. K. Avlyanov, Proceedings of SPE ANTEC 99, New York,
NY, USA, 1999, Volume II, 1521.
27. A. D. Child and H. H. Kuhn, Synthetic Metals, 1997, 84, 1-3, 141.
35. Y. Cao, P. Smith and A. J. Heeger, Synthetic Metals, 1992, 48, 91.
39. G. H. Wannier, Statistical Physics, John Wiley, New York, NY, USA, 1966.
40. C. Y. Yang, Y. Cao, P. Smith and A. J. Heeger, Synthetic Metals, 1993, 53, 3, 293.
368
Conducting Polymer Composites
41. R. Zallen, The Physics of Amorphous Solids, John Wiley, New York, NY, USA,
1983.
46. S. Tokito, P. Smith and A.J. Heeger, Polymer, 1991, 32, 3, 464.
48. Press Release, RTP Company, Winona, MN, USA, August 2000.
369
11
Recycling of Automotive Composites
Anand K. Kulshreshtha
11.1 Introduction
Up until 1970, petrol was thought to be abundant and cheap. In oil producing
countries of the Middle East, petrol was used for cleaning the roads because water
was scarce. The first oil embargo enforced by the Organisation of Petroleum Exporting
Countries (OPEC) changed all that and countries around the world experienced the
first oil shock or energy crisis in 1973. This led to a series of oil conservation measures
by various industries. The automobile industry faced this crisis by substituting metal
in cars with plastics both thermoset as well as thermoplastics. The former were used
as they were available as BMC and SMC which could be fabricated into car frames
using RIM, RTM and other fabrication techniques. In thermoplastics many varieties
of polymers and polymer blends were experimented with. Feasibility of making all
composite cars was shown by producing demonstration models at intervals. The
end-of-life cars faced difficulties in disposal: metal could be almost completely
recovered from scrapped cars but plastics were not. Thermosets were difficult to
recycle because they are non-melting and thermoplastics could not be properly
segregated or identified due to their multitude. Assuming 12 years as an average life
of car between production and recycling and non-availability of information on this
topic due to intense competition between manufacturers and resulting confidentiality,
it has been difficult to prepare an overview such as the present chapter. The readers
ought to realise that due to this time gap, information about recycling of cars is
available on models which were produced at least 12 years ago. The chapter highlights
the problems encountered in recycling mainly of parts due to contamination,
identification, cost of landfill, toxicity due to combustion, low value shredder waste,
inhomogenity, absence of supply chain from manufacturer to user to shredder, lack
of parts integration, degradation, painted surface, etc. The most convenient to recycle
parts are large ones such as, bumper, intake manifold, hatchback, etc. Also, recycling
has been done for material recovery (if possible) otherwise for energy recovery
(burning as fuel). As a result of these experiences car manufacturers today are wiser
and more organised than a decade earlier. Today, the trend is to use polypropylene or
its copolymers either in car interiors or outside, which will reduce the need of
identification when the car is recycled in future.
371
Handbook of Polymer Blends and Composites
The legislation to put recycling in place is now in effect in many advanced countries.
Responsibility for recycling now lies with car manufacturers who have created an
infrastructure for shredding, dismantling and recycling of used cars. This chapter discusses
the recycling strategies adopted by car manufacturers in many countries.
372
Recycling of Automotive Composites
springs/driveshafts have not reached early expectations and volumes have been very low.
Almost all vehicles from Ferrari to Ford now feature 40% glass reinforced Nylon injection
moulded throttle or clutch pedals, while some vehicles use the same materials for the pedal
box assemblies. Prototypes have been tested for a composite brake pedal.
The use of thermosets and plastic materials is increasing rapidly in vehicle construction,
particularly in Europe. Almost all external trim now uses plastic materials, particularly
PP, PC and PU for bumpers and panels.
The use of fibre reinforced composites is growing (Table 11.1). In 1966 America
absorbed about 42% of the world composite sales with transportation taking about
25% of the global composites market. In studies released by Owens-Corning Fiberglass,
the average usage of glass fibre composites in European passenger cars and trucks was
about 1.9 kg in 1975.
This grew to 11.5 kg in the 1991, representing a European volume of 220,000 tonnes.
373
Handbook of Polymer Blends and Composites
By far the largest material in this segment is polyester SMC which has shown global growth
in the automotive sector. The largest market is the USA, where SMC usage grew from
8,000 tonnes in 1970 to about 120,000 tonnes in 1995 (Table 11.9). Composite body
panels have been successfully made at all volumes on both sports cars and family vehicles.
Bonded composite applications are being implemented much more widely either as bonded
sub-assemblies which are then mechanically fastened to the main structure (Citroen
Evasion tailgate, Alfa GTV hood) or to bond composite panels to metal structures (Ford
Transit roof).
The use of composite materials is slowly growing as issues of cost, cycle time, impact
modelling, quality, fit and finish, painting, waste and recycling are gradually being overcome.
374
Recycling of Automotive Composites
The upper and rear spoilers of cars are made by resin transfer moulding (RTM) process
using low shrink resin grades. Advantages of RTM are one piece moulding and injection
moulding of high volumes of highly aligned glass fibre reinforcement with low viscosity
polyester resins quickly (see Chapter 4.2). Other competing materials include:
Granada and Ford are among the main producers of composite-rich concept
demonstration cars or test models made around 1980, which illustrate the viability of
composites in car fabrication. Table 11.2 illustrates this.
Typical examples of demonstration cars made by using composites are listed in Table 11.2.
Among those concept cars, the largest percentage of advanced composite materials
(ACM) is used in the Ford test car. This test car used CFRP for making the body,
frame, doors, etc., and was about 656 kg lighter (about 33%) than the 1978 model
LTD test car. As a result of the light weight, the mileage of the LTD test car increased
to 10.4 km/l, about 44% better than that of the 1978 model (7.2 km/l). About 160
parts were made of CFRP.
The body, entirely made of carbon fibre reinforced unsaturated polyester was about 115
kg (61%) lighter than that made of steel. At the same time, the original 83 pressed parts
become 10 parts, making it possible to increase the productivity. In addition, frame and
doors were all made of CFRP. The parts contained more than 60% CFRP including
body, door, bumper, front end, hood, and deck lid. They contributed to a 40% reduction
in the weight of a car compared with a steel car. The door was made of CFRP, including
an internal window frame made of high-modulus carbon fibre and a door guard beam
made of a carbon fibre/glass fibre reinforced extrusion-moulded part. The hood and
deck lid were made of SMC, each resulting in a reduction in weight of more than 10 kg.
375
Handbook of Polymer Blends and Composites
An example of the materials composition of the interior trim is given in Table 11.3.
376
Recycling of Automotive Composites
Plastic materials in the interior are used in the main areas: the dashboard, pillar and
quarter trimming and paddings.
Moulded plastics are mainly used for the complete dashboard including the instrument panel
housing, and to cover the metal of the door and rear quarter panels in the other areas. Plastic
foams are also moulded and covered with other trim materials to achieve a soft, warm feel.
A simple dashboard, e.g., for basic-models, is rigid and composed of injection moulded
PP the use of which still has technical problems. These problems include flammability,
poor low temperature impact strength, poor gloss, poor heat resistance and poor
printability. In top range models and luxury cars, the dashboard is a combination of a
reinforced ABS as support, PVC skin for a warm feel and PU foam as a filler. The most
innovative dashboard consists of PU foam with a PU skin.
The principal disadvantage of PVC coverings is the fogging effect, that is the release of
plasticisers and other additives from PVC by the high temperatures, that can exist if the car
is exposed to direct sunlight and then deposit on the windows, resulting in a foggy view.
377
Handbook of Polymer Blends and Composites
Dashboard and pillar and quarter trimmings are put together because the materials choice
for the dashboard covering is in many cases the same as for the other trimmings, because
of stylistic arguments. Therefore, the importance of the single-material solution for the
dashboard is very much enhanced by the stylish coupling of dashboard covering and the
other trimmings.
11.2.4.4 Padding
PU foam is extensively used in seat cushioning. It is possible to change the mix proportions
during the foaming process, creating areas of different hardness within a single cushion.
New plastic appliances are not expected for the interior, as the interior seems to be a
market in which there is no further scope for development. Yet, some inter-polymer
substitution might take place. If this would be the case, PP and/or PU might be used in
higher volumes at the expense of ABS and PVC.
About 15% of the materials in body works and exterior trimmings are plastics. The
main advantages of plastic polymer materials as potential car body materials are that in
general they are easily and accurately formed into extremely complex shapes, lighter in
weight, sufficiently flexible to withstand minor impacts, without permanent damage,
and virtually free from corrosion problems.
The main disadvantages of these materials in body applications are that without some
form of reinforcement they have poor stiffness; that their long term chemical stability,
especially under ultra-violet radiation, is in doubt in some cases; that they call for a
completely different approach to body assembly and finishing, since they cannot be welded
to metal nor (in general) be painted with the same systems.
To overcome the problem of lack of stiffness and thus broaden their range of
applications, most polymers can be reinforced by the addition of high strength fibre,
invariably glass except in some very high strength, high cost applications where carbon
or Kevlar (or a mixture of the two) reinforcement is used. The reinforcement can take
the form of continuous fibres, providing high stiffness, or short lengths of fibre which
confer less stiffness but are easier to introduce into automatic manufacturing systems
with high rate of component production. Figure 11.2 shows the fabrication of car
body or underbody by two different composite manufacturing techniques: filament
winding or high speed RTM.
378
Recycling of Automotive Composites
Another problem which needs to be solved if the plastic composite panels are to be widely
used, is the manner in which they are mounted to the sub-structure (no motor industry
body engineer is prepared seriously to contemplate a composite single structure for large
scale production). Here the balance of opinion favours mechanical fastening, either by
bolting or by some form of clip system. Adhesives are a possible alternative but one with
which engineers are clearly less happy, as they are reluctant to abandon an assembly
technique about which they know a great deal (spot welding) in favour of the widespread
use of adhesives which might present long-term problems of joint deterioration. As a result,
structural adhesives are still relatively little used in car body construction.
Major polymer appliances in the body are bumpers, wheel coverings, front and rear-
ends, and add-on, such as spoilers, lateral protections, and wheel caps. New components
are introduced, such as heat- and sound-insulating plates, side-impact absorbers, and
underbody coverings. All these mostly compose of plastics. Increasingly plastic body
panel are introduced, mainly fenders, bonnets and tailgates, and front and rear ends.
Thermoplastics are, as yet, not used extensively in body panels. Research continues and
the prospect of major weight savings could still bring about a change in established
motor industry attitudes. The balance of industry opinion seems to favour a material
which can be injection moulded to achieve high production rates, in a process analogous
to sheet steel pressing, even at the expense of high initial tooling costs. Some promising
results along these lines have been achieved with PP, PPO, and with polymer alloys
mixtures of different materials. One well known example is the BMW Z1, that has an
all-polymer body, consisting of engineering plastics. In fact, BMW Z1 was a test model,
e.g., for thermoplastic body panels. But other manufacturers also work on thermoset
body panels, e.g., Renault in the new Clio and the successor to the Renault 25.
379
Handbook of Polymer Blends and Composites
Three composite materials have been used in vehicle body production: epoxy resin
reinforced with glass fibre (GFRP), SMC/BMC, and recently, RIM/PU.
The oldest composite material is generally known as glass reinforced plastic (GRP), in
which continuous glass fibres, usually in the form of woven mat are laid into a mould
which is then filled with epoxy resins. Already in the early 1950s all GRP bodies appeared
in low volume, specialty cars, e.g., GM Corvette. This approach is best suited to low
rates of production although Lotus has developed its Vacuum Assisted Resin Injection
(VARI) system to achieve higher rates. GRP does have the advantage of being suited to
the manufacture of extremely large single components; car bodies made of this material
can if necessary be made in just two (upper and lower) sections.
Glass fibre reinforced polyesters (SMC, BMC) are the second most successful
composite material. Differences between the various polyester materials arise from
the exact mixture of polyester type and other chemicals for specific functions, and, in
the type of glass fibres used. SMC is composed of polyester sheet in which continuous
fibres are laid, which can be pressed into shape after a curing period. SMC has
extremely high stiffness and is suitable for large horizontal panels such as roof sections;
specific additives are required to give finished pressings a class A surface finish. BMC
has short fibres and therefore is a little less stiff, but is more easy to process. Some
BMC can even be injection moulded. Well known examples of polyester applications
are the BMC tailgates in the FIAT Tipo (injection moulded BMC), the compressed
BMC tailgates of Citroen BX and ZX, and the SMC body of Renault Espace.
11.2.4.7 Bumpers
In bumper applications a wide number of plastics are used, e.g., PP/EPDM, PC/PBT,
PPO, PBT, reinforced polyurethane (PUR), SMC.
Bumpers fall into two categories, hard and soft bumpers. Soft bumpers are generally
integrated into the body design and therefore have to be able to absorb collision energy.
380
Recycling of Automotive Composites
They are to be found on the more expensive models. Hard bumpers are generally attached
to the chassis, and they function to transfer collision energy to the chassis as there is less
need to absorb the energy.
All bumpers consist of three functional elements, that might be integrated or not: the
cover, or fascia, which is the visible part, the reinforcing beams, giving structure to the
bumper, and the energy absorbing systems, which might be a physical system, or a foam
filling behind the cover.
In the 1970s, largely as a result of pressure from consumer organisations and insurance
companies a federal law was passed in the USA demanding that no permanent damage
must be done to a car involved in a collision of up to 5 mph. Similar safety standards
were adopted in Europe where the EC standard stated a speed of up to 4 kmph. As a
reaction to this legislation automotive companies in the USA developed a bumper
consisting of heavy steel beam attached to the longitudinal runners of the chassis by
hydraulic shock absorbers. The space between bumper and bodywork is filled with flexible
PU panels. European manufacturers adopted this principle for models intended for export
to the USA. For the European market they experimented with a great many systems
some of which were based on the principle of energy absorption by means of plastic
foam or a large number of metal or plastic ribs.
381
Handbook of Polymer Blends and Composites
The complicated aerodynamic design of Fords successor to the Cortina, the Sierra, together
with the large production runs anticipated, led to the companys decision to use an injection
moulded thermoplastic. The aim was to fabricate a bumper made from an impact resistant
moulding that did not need painting, was capable of being coloured with suitable pigments,
and at the same time was able to meet all of the relevant weather resistance specification.
The companys R&D department investigated a number of options and a polymer alloy
of PC and PBT was identified as a suitable candidate material.
As a result of these developments the use of polymers in the fabrication of bumpers has
increased dramatically over the last three decades. This point is illustrated by an
examination of bumper material composition by fascia, reinforcing beam and energy
absorber subsections (Table 11.4).
The need for weight savings in suspension systems is particularly acute which is likely to
lead to the increased use of high strength composite materials for suspension members
(springs) and for wheel disks. Currently, laminated springs have been constructed from
thermoset, glass fibre reinforced materials, mainly for application in commercial (heavy
weight) trucks.
11.2.4.9 Transmission
Major mechanical components are made from cast iron and cast aluminium alloy. Markets
already exist for a small number of plastic items such as windscreen washer bottles, brake
fluid reservoirs and coolant expansion tanks. Recent applications in more demanding areas
include radiator header tanks, sump oil pans, valve gear covers, and inlet manifolds. Engines
including plastic camshafts and pistons are at the R&D stage. Some research has been also
been conducted into (lined) plastic composite cylinder blocks in a search for low weight
and ease of manufacture, but these give rise to severe acoustic problems.
382
Recycling of Automotive Composites
Great interest is being shown in the use of plastic composite materials for the production
of inlet materials (Figure 11.1) as this is a relatively cool application, especially if the
engine is crossflow the inlet and exhaust of opposite sides of the cylinder head. Such
materials can be manufactured to very close tolerances with excellent internal surface
finishes an important consideration in any inlet manifold while low thermal
conductivity means there is less tendency for fuel to condense on to the manifold walls,
though this is clearly not a factor when port-type fuel injection is used.
Plastic composite materials are increasingly used for the valve gear on top of the engine.
Some interest is also being shown in oil sumps of similar material. Composites are very
well suited for the valve gear cover, now that grades of material are readily available which
withstand the temperatures involved and which also resist chemical attack by oil. Covers
can be accurately moulded with integral components for systems such as crankcase
ventilation. Noise attenuation properties can be excellent, and the component can be made
to look attractive. It therefore seems likely that plastic composite materials will become
virtually standard for this application in the near future. Its use for sumps (oil pans) is less
certain since most current engine designs depend on heat transfer to atmosphere through
the sump as the principal if not the only means of oil cooling. The use of a composite
sump would, therefore, probably involve the fitting of a specific oil cooling radiator.
Plastic composites are already widely used for other casing components, such as the
timing case cover (or the toothed belt drive guard). Here again the noise attenuation
properties of the material are potentially valuable, both to improve cabin interior
refinement and to meet external noise regulations, thus more development to refine this
application is likely.
The weight saving potential is excellent with an 80% reduction in the weight of a
composite for cars and a massive 95 kg saved on a typical heavy good vehicle spring.
383
Handbook of Polymer Blends and Composites
The use of a hybrid combination of glass and carbon reinforcement provides wide design
flexibility in terms of spring rate alternatives and variable spring rate possibilities. By careful
design the composite construction can be made to fail safe, with the spring splitting along the
plane of maximum shear stress to form two separate spring both capable of bearing load.
Fibre reinforced plastics posses a number of features which enhance the operational
suitability of composite springs compared to conventional materials:
Plastic waste from cars is part of the shredder waste. In Western Europe, about 1,300
kilo tonnes of plastic waste from cars was generated annually in 1990 (Table 11.5). This
grew to 1,900 kilo tonnes in Europe by the end of 2000.
Total shredder waste in the EC can be estimated, assuming that shredding and separation
of metals from 13 million cars weighing 1,000 kg each, results in 25% shredder waste, or
3,250 kilo tonnes. The composition of shredder waste is largely unknown. Research showed
that 52% of the shredder waste could not be traced back to any material. This undefined
fraction accounts for 75% of the polymers, 90% of the rubbers and 50% of the glass.
384
Recycling of Automotive Composites
Most of the currently discarded cars, were constructed 10-15 years ago and their
development started 4 to 8 years before that date. Thus, changes in vehicle design can
only become effective after a period of at least 10 to 15 years.
The production of a car takes place on an assembly line. In vehicle production integration of
parts also reduces the number of assembly steps, e.g., two shells of glass reinforced polymer,
instead of welding some 15-20% parts for a hatchback, reduces the number of assembly
steps dramatically. Composites offer the opportunity to integrate parts into units with a
reduced number of assembly steps. When recycling, some spare parts are removed for repair
and reuse. However, the disassembly is uneconomic and takes high cost time (Table 11.6).
385
Handbook of Polymer Blends and Composites
Figure 11.3 Shows the entire chain of operations which take place in car purchase,
dismantling, shredding and finally, in recycling
The disassembly of parts and selling them in the second-hand market is a profitable
business. Of course, the income depends on the state of the vehicle: partly damaged
cars wrecks, e.g., from road accidents, are worth much more than worn out, 15 year
old vehicles. Also, luxury cars are worth much more than economy cars.
386
Recycling of Automotive Composites
11.3.4 Logistics
BMW, Volkswagen and Mercedes-Benz use the existing new part distribution system to
collect used parts, such as bumpers in the case of Mercedes-Benz and Volkswagen and
engines and starter motors in the case of BMW. Renault plans to use the dealer network
to collect recyclable parts from old cars.
A related problem is the unattractive weight to volume ratio of plastics. This makes
storage and transport expensive. However, a few examples of small, mobile grinders at
the dismantling site have been found to operate satisfactorily in practice. Two examples
in which these grinders have been used are GE Plastics and KBD. GE Plastics co-operated
with Kotrac, a Dutch transportation firm that collected bumpers at car wreckers in the
Benelux and Germany. KBD, a British consultancy firm, has organised a collection
structure for plastic parts at about 60 dismantling sites. In both cases a mobile grinder
was used to decrease the volume of the collected plastic parts.
Identification takes place via a code on the component. It provides information on the
materials used (whether thermoplastic, thermosetting, or elastomeric materials; whether
pure blended or copolymeric materials), or material type, e.g., PA6 or PA66, or filling
and reinforcing materials, or percentage of filling and reinforcing materials. It does not
provide information on plasticisers (as the ISO 1043 [1] does) nor any other additives.
Plastic waste from cars and shredder waste differ in quality. Therefore, several ways to
treat plastic waste from cars and shredder waste have been developed. Four separate
streams of plastic waste from cars can be identified:
Well defined plastics from specific parts, such as HDPE from fuel tanks, PP from
battery cases and fluids containers, for which plastics substituting reuse can be applied.
Easily separated plastic parts for which high-quality reuse is not possible without
further information. With improved identification and the aid of a databank these
parts can also be reused in plastics substituting applications.
For mixed plastics and plastics combined with other materials only low-quality re-
use is possible.
387
Handbook of Polymer Blends and Composites
11.3.4.3 Disposal
Plastics in vehicles reaching the end of their useful lifetimes need to be disposed off in an
environmentally sound manner. Currently, a scrap automobile recycling industry existing
in North America, Japan and Europe that purchases or collects discarded vehicle and
uses them as a source of spare parts first, and then ferrous and non-ferrous metals. In the
USA, the non-metallic fractions obtained after the metal has been recovered used to be
landfilled.
Approximately 60-75% of the plastics purchased end up in the final assembled products.
The remainder becomes internally generated waste, usually called prompt scrap by the
recycling industry.
1. Prevent formation of waste during design, reuse of parts composition and materials
at high quality level (separation of relatively uncontaminated materials for re-use in
the same application).
2. Reuse of materials at a low quality level (applies to materials that are not separately
collected, or spoiled with dirt and reused in less demanding applications).
5. Landfill.
For automotive thermosets (urethanes and SMC), even prompt scrap (scrap generated
during processing ) is difficult to recycle economically. In some cases, a market exists for
the material. For example, clean, flexible PU foam may be sold to manufacturers of
carpet underlay and padding. The recyclers of this waste, chop the foam, combine it
with additional isocyanates and water and allow the mixture to react, thereby producing
cushion-like materials where the waste particles are basically flexible fillers glued together
by an adhesive. The value of the scrap in this case is related to the price of raw materials
that are saved and to the market value of the product. At the present time, urethane
scrap demand is high and virtually all virgin PU foam scrap is being reused.
388
Recycling of Automotive Composites
Reuse at a high quality level is not possible for all parts and materials. Waste will remain
after disassembly and reuse of parts and materials. The next preferred strategy is to make
use of chemical and thermal technologies. These are those technologies that chemically
break down the polymers, using heat. Examples are incineration with valorisation of heat,
or just for reduction of volume, but also pyrolysis, hydrolysis, gasification, etc. These
technologies can be applied to make re-use of chemical components possible, e.g., pyrolysis,
hydrolysis, gasification, to use the energy content, to reduce the volume of waste or to use
both the energy content cement works or in blast furnaces for production of steel.
Since a great deal of plastics waste is currently contained in shredder waste, many initiatives
worldwide currently aimed at utilising this source of organic material. Techniques under
investigation include:
Pyrolysis of complete fraction to produce fuel gas and liquids/solids capable of being
used as chemical feedstock.
In the case of SMC, 27 companies, including materials suppliers and SMC holders have
formed a group to study SMC parts reclamation and disposal. General Motors has been
a driver by demanding that SMC moulders supplying parts for their new A-van accept
return of the prompt scrap generated at the assembly plant for disposal. This is a welcome
development, since it places some disposal responsibility on the people who know the
chemistry of the materials best.
389
Handbook of Polymer Blends and Composites
The problems of recycling prompt scrap are usually multiplied several fold once the car
body has been shredded because of the mixing and contamination that takes place there.
Shredders produce two fractions with substantial amounts of plastics: the fluff (or air
fraction), which contains lightweight materials with high surface to volume ratios such
as cloth, urethane foam, vinyl upholstery and padding; and the dense media reject. The
dense media reject is essentially PP and ABS left once the ferrous material has been
removed magnetically.
Experiments conducted by Ford Research in the early 1980s have shown that it is possible
to separate urethane foam from the fluff and to hydrolyse it using the Ford process
described later [2].
An alternative approach investigated involves the grinding and reuse of coarsely separated
mixtures as fillers for low grades products. This concept is not unlike the approach
taken by carpet underlay makers with prompt urethane scrap. Generally, the reground
shredder waste is processed with either other virgin thermoplastics or with adhesives
and thermosets that form a matrix holding together the waste filler. Thus, floor tiles and
plastic wood have been made in this manner.
390
Recycling of Automotive Composites
The use of shredder waste as fuel in steel blast furnaces or cement ovens is in a pilot stage.
With high volume thermoset material consumers such as the Ford Motor Company
mandating the use of recyclate in composite products, the thermoset industry has steeped
up efforts in demonstrating the recyclability of their products.
Utilising granulators, hammer mills and sieves, scrap BMC and SMC parts can be
reclaimed as a filler. In this process, the granulator would reduce a part down to the
manageable size required to feed a hammermill pulveriser which then reduces the scrap
to the size of filler particles (30.1 m average), and finally a classifying unit would be
used to separate or refeed to the particles (Figure 11.5). Studies have shown a 9% level
of regrind incorporated into polyester BMC material has no significant effects on physical
properties of the material (Figure 11.1). Plaslok Corporation and Rogers Corporation,
who manufacture phenolic materials, utilise a similar process to regenerate scrap supplied
by their customers in active recycling programmes.
391
Handbook of Polymer Blends and Composites
An element in the pilot project of Peugeot SA is the thermal treatment of shredder waste.
CFF (the largest European shredder company) and VICAT (cement works), who
participate with Peugeot in the project, have developed a process in which shredder
waste can be used as fuel in the production of cement. The process does not produce
ashes, nor any polluting emissions, while replacing ordinary energy feedstock [3]. The
remaining inert granulates are to be processed in for example isolation materials for
construction.
Because of the relative low caloric value of shredder waste (14-15 MJ/kg), shredder
waste can only be used to limited extent as a fuel for cement works. Due to the large
scale of cement production, it was concluded that under appropriate conditions about
50% of shredder waste in Germany could be used as an energy source in cement works.
The reason why Peugeot, CFF and VICAT made another effort to use shredder waste in
cement ovens, despite these generally known constraints, could be that they think they
can rely on a constant composition of shredder waste, due to a controlled dismantling
and shredding process by one single partner in the project.
11.3.5.2 Incineration
The only thermal treatment of waste applied on large scale is the incineration of a mixture
of several types of plastic waste, contaminated and co-mingled with other waste streams,
i.e., municipal solid waste (MSW). The available processes have not specifically been
developed for plastics, but are compromised on a best average practice for MSW, with
large control rooms.
One way to increase the thermal efficiency of incinerators is to separate the plastic waste
from the other waste streams in order to increase the caloric value. If necessary, other
waste with high caloric values can be added. This so-called high efficiency incineration
technology is still under development.
392
Recycling of Automotive Composites
Recycling composite materials that combine thermoset polymer matrices, such as epoxy
resins, unsaturated polyester resins, and polyimides, with glass, polymer, carbon, or metal
fibre reinforcements has been a tough task. The desired characteristics of thermoset
composites, especially strength, toughness, and durability, have made it easier and more
cost effective to dump them into landfills (both industrial scrap and composite products
that have outlived their usefulness). However, the closing of landfills across North America,
increasing environment concerns, and the growing necessity for industries to conserve
resources have worked together to change that.
1. The first process is a secondary recycling method that involves the mechanical
(physical) breakdown of composite materials to produce a flake, chip, or powder
product that is used either as filler or as a reinforcement in new composites or
compounds.
2. The second process is a tertiary recycling method in which the thermoset polymer
composite is reduced to its basic chemical components, or derivatives of them, using
heat applied in varying media, or chemical agents.
With either process, the necessary first step is to reduce scrap thermoset composite products
to small, manageable sizes. This task is difficult, physically and economically, because of
the tough, indestructible characteristics that are the most desirable qualities of composite
products.
The automotive scrap industry uses shredding and chipping machines that can be adapted
to reduce thermoset composite scrap and products to a manageable size.
In deciding which recycling process to use, the properties of the composite scrap and of
the final product must be considered. If, for instance, the values of the filling material
and reinforcing materials in a scrapped thermoset are great, e.g., carbon fibre, and if
they can be reused as fillers or reinforcements the first method might be chosen. On the
other hand, if the value of the thermal energy contained in the scrap thermoset composite
393
Handbook of Polymer Blends and Composites
is high and that energy could be used easily to provide energy for an industrial or
commercial process, the second method may be preferred.
The second method, thermal and chemical composition can be adjusted to produce
valuable, clean reinforcements and fillers which have been shown to be reusable, along
with quantities of chemical produced.
In this case, the grinding of composites is a more expensive because the scrap must be
reduced in size to reinforcing lengths approximately 0.3 to 1 cm or ground to a fine,
powdery substance as a filler.
The addition of small amounts typically 15% regrind into new composites (Figure 11.1)
has led to several successful, but minor, applications.
In method (2) grinding the thermoset scrap or product to a manageable size is the first
step before the compounds are exposed to heat. In this method four approaches have
been examined and have produced varying results:
11.3.6.1 Depolymerisation
An alternative recycling route for clean, well defined thermosets involve reactions that
reverse the polymerisation process to produce starting compounds. For example, Ford
has developed a process (Figure 11.5) to hydrolyse water-blown polyether based PU in a
glycol solution and to purify the resulting polyether polyols.
Economic studies, however, indicate that the sale of the scrap to carpet underlay makers
is usually more attractive than depolymerising. The increasing amount of polyurethane
being manufactured with methylene diisocyanate-based isocyanates (MDI) and with
polyurea backbones, however, creates problems in both the hydrolysis reaction step and
in the purification step. For reinforced RIM (RRIM) urethanes, a hot filtration step is
required after hydrolysis to remove the reinforcing fibres. For these compounds,
transesterification reactions might be more attractive if a use for the product resins
containing fibres can be found.
394
Recycling of Automotive Composites
A key factor in the future success of the GMT route is likely to be the recyclability of these
materials. Much work remains to be done to establish the true recyclability parameters, but
early indications are that GMT parts can be successfully reprocessed by chopping and injection
moulding. Moreover, it also appears that in some circumstances GMT can also be reworked
by the stamping process, thus retaining much of the benefit of the long fibres initially present.
Waste thermosets can be used as fillers or converted into low molecular weight products.
The material recycling of PU is possible, albeit with certain limitations. Flexible PU foam
is usually granulated and processed into rebonded foam (for example for gym mats or
carpet underlay). Waste from PU shoe soles can be used to produce new shoe soles. Rigid
PU foam powder from the disposal of refrigerators and chlorinated fluorocarbon (CFC)
is used as an oil binder or is processed with isocyanates to manufacture building boards.
From the automobile industry, ABS can be collected from radiator grills, instrument
panels, lamp housing etc.
The transportation market is considered to be the key to recycling ABS because of the
large amount of ABS used in this application, and the relatively advanced collection
features. ABS polymers can be remelted and reused several times or properly stabilised
with various antioxidant and thermal stabiliser packages.
Both co-extrusion and co-injection moulding allow the production of plastic products
having a sandwich structure. The skin material may be a virgin resin characterised by
high tensile strength, good scratch resistance and good colour. The core material may
include regrind scrap from multilayer regrind, somewhat degraded material, colour rejects
and off-grade material.
Increase in plastics use is likely to be observed under the hood, in the transmission and in the
body work. In the interior trimming a inter-plastics substitution might take place, favouring
single material solutions (either PP, PU or other solutions) that are easier to recycle.
395
Handbook of Polymer Blends and Composites
Research has been undertaken on Audi (fluoridated) fuel tanks, VAG, BASF and KAUTEX
(a major producer of fuel tanks) have been involved in the project. Consecutive stages in
the recycling process included the removal of non-HDPE parts, the evaporation of fuel
residues and grinding of the tanks into processable flakes. Fuel tanks were processed
into new fuel tanks at least once. The functional requirements on tanks were passed.
Thus recycling of HDPE fuel tanks was shown to be technically feasible.
11.3.7.2 Bumpers
One example is the project by Volkswagen, DSM and Evert Heeren GmbH (a large
independent shredder company) on the recyclability of EPDM-modified PP bumpers.
Bumpers are collected at the Heeren plant (where all the dismantling operations of Volkswagen
take place), from scrap cars delivered there and via the dealer network of VAG. After grinding,
DSM upgrades the material by compounding it to specifications for bumper applications. If
there is no demand for paintability the percentage of recycled material can be as high as 70-
80%. More than 7,000 bumpers were collected and recycled in this way till 1992. Recycled
bumpers were used experimentally on the Volkswagen Polo.
The recyclability of PP bumpers was also studied at Hoechst. Technical problems with the
recycling of PP from bumper applications exist with regard to UV degradation and paintability.
To date, the only large-scale recycling of automotive plastic waste is polypropylene (PP)
from battery cases. Recycling of the lead in batteries by specialised firms generates a
396
Recycling of Automotive Composites
stream of homogeneous PP. Taken together with relatively easy recycling of PP, this made
it economical for specialised plastics recycling firms to enter the business. PP is regranulated
and used to produce for example flower pots. In Europe some 10 companies are currently
involved in the recycling of this PP stream. Recently also, metal recovery firms have gone
into the PP battery case recycling, e.g., Metallgesellschaft. These firms compete in a fully
developed market of recycled PP from battery cases.
Recycling experiments on (thermosetting) GRP have been carried out. It has been shown
that after grinding and/or milling, they can be fed into SMC or BMC production,
substituting for the fibre or filler material to a maximum of about 30% by weight. The
technical characteristics of the product are:
5% reduction in weight
In the above automotive sector application recycled GRP is in non-class-A surface parts.
Hence the material can be applied as a reinforcement for thermoplastics or in construction
applications, e.g., road foundations.
ERCOM (founded November 1990) is a joint venture of DSM, BSAF, Vetrotex, Owens
Corning Fiberglass and some German plastics processors. It collects, grinds and processes
GRP from production scrap and from used products. Manducher (manufacturer of plastic
components, France) has demonstrated a capability for reprocessing SMC into new products.
397
Handbook of Polymer Blends and Composites
Bayer has developed a system to reuse PU/RIM not only from production scrap but also
from used car parts. After washing and granulation, the flakes are subjected to heat and
high pressure in a mould, resulting in a rigid form. Parts produced by this process have
about 50% of the tensile strength and 15% of elasticity of their original parts. Application
of the reuse materials are thought to be in wheel caps, body coverings, battery tops.
Flexible PU foams can also be re-used for new PUR parts by reprocessing flocks of
grinded PUR soft foams with some virgin PUR into rigid foam applications. Bayer says
this process is technically more secure than hydrolysis of PU foams because the output of
hydrolysis of PUR foams, is a mixture of polyolefins. To produce PU foams pure
polyolefins are needed.
The recent introduction of thermosetting polyester skins on some large volume vans, has
generated more interest in recovering these and other thermosetting materials. One large
automobile manufacturer has concentrated on pyrolysis of glass reinforced panels, which
results in the production of a mixture of combustible gases, oily material, calcium
carbonate, glass and carbon char. The gases are burned in the pyrolysis furnace and
supply all the heat needed for the process, leaving the oil for other fuel uses.
ICI in the Eureka/RECAP project works on the hydrolytic and glycolytic depolymerisation
of PU-based plastics to yield reactive polyol liquids. Sources of
PU scrap are from both moulding and assembly operations in vehicle production and
also disassembly of vehicles prior to shredding. ICIs main interest is in establishing
tolerance limits of contamination by other plastics and textiles for depolymerisation and
the reaction of recovered polyols to new products.
The aim is to fabricate PU products from recovered reactive liquids for reuse within the
automobile industry.
At Ford Werke AG in Cologne, 1.5 tonnes of PU-foam are collected daily as production
waste from the cutting of seats and cushions. With alcoholysis and special catalysts, it is
possible to regenerate a homogeneous mixture of polyols. Contaminants of polyamides
and polyesters are also dissolved. The polyol mixture can be reacted again with isocyanates
to produce PU-hard foam, partly to be used in Ford Scorpio body-parts. Other applications
398
Recycling of Automotive Composites
still have to be found. Separate collection is a prerequisite for this process; therefore,
only production waste can be treated.
With high volume thermoset material consumers such as the Ford Motor Company
mandating the use of recyclate in composite products, the thermoset industry has stepped
up efforts in demonstrating the recyclability of their products.
Work has so far focussed on processing a component made from a polyester SMC.
The FRP feed unit is a hopper at the base of which is a centre-less corkscrew type screw
feeder turning with a large clearance within a feed tube. The feed hopper is sealed to
prevent the back flow of hot gas through the feed material. A compressed air bleed into
the feed tube helps to prevent the crushed scrap from bridging in the hopper and provides
a positive airflow into the fluidised bed preventing the feed tube overheating.
The fibre and filler recovery is achieved using a cyclone (Figure 11.6). Particulate material
is usually extracted from the base of a cyclone by a rotary valve, however, in this case,
such a device is avoided to prevent damage to the recovered fibres. Instead, an air bleed
from the base of the cyclone is used to compress a high fibre fraction collection bin
followed by a small cyclone to trap fine filler particles.
The recovered fibres are mixed with a substantial proportion of fillers in the collection
bin. Washing of the fibres to remove the loosely bound contamination has shown that
399
Handbook of Polymer Blends and Composites
the fibres are contaminated with up to 60% fillers by weight. Microscopy of the washed
fibres shows that there is very little contamination attached to the fibres. This indicates
that the thermal processing effectively removes the polymer and leaves the fibres
substantially clean [5].
Open-cell rigid urethane fluff (RUF) panels produced from binding finely shredded PU
foam from old car seat cushions with polyisocyanate have been shown to be strong
candidates as substrates or spacers for vacuum insulation panels, and thus as replacements
for fumed silica, glass fibres, or open-cell foams in this application. Since the recovered
PU foam is compressed to four to five times its original moulded density, all of the
recycled foam from an old car can essentially be utilised in vacuum panels that are inserted
into one newly manufactured refrigerator (Figure 11.7).
A large number of companies in the vehicle production and disposal cycle are already
undertaking research and technology development R&T initiatives aimed at providing
long term solutions to the automotive plastics waste management problem. The EC should
develop a consistent legislative and regulatory frame-work for optimum waste disposal
400
Recycling of Automotive Composites
Figure 11.7 Composite car floor pan produced by high speed resin transfer moulding [6]
in the long-term. Co-ordination of R&T efforts would concentrate on the three themes
that need to be addressed in order to achieve an optimum vehicle disposal system for the
long term:
A consistent demand for the recycled material - this requires confidence in the quality
and the continuity of quality of secondary materials and the stimulation of a broad
range of applications.
1) Identify those areas of design which, if altered, are likely to have the greatest impact
on waste volumes in the shortest time period.
2) Stimulate research to set standards for the quality of supply of recovered plastics
from cars.
401
Handbook of Polymer Blends and Composites
The variation around the average trends can be rather large, e.g., the use of PP is
much higher in Renault 19 (being 40% by weight) and Nissan Primera (being 50%)
than average (about 15%). Other differences are in use of unsaturated thermoset
polyesters (SMC/BMC). Some manufacturers, such as Citroen, Renault and FIAT,
use SMC/BMC in relatively large quantities whereas some have tended to phase out
SMC/BMC, e.g., Nedcar, BMW, or looked for substitutes such as aluminium,
e.g., Audi.
A. Well-defined plastics from specific parts, mainly PE and PP from fuel-tanks, battery-
cases, and fluids containers.
B. Easily separated plastic parts for which high-quality reuse without further information.
With an improved identification aid of a databank these parts can also be reused in
plastics substituting applications.
C. Mixed and combined plastics with other materials. Only low-quality re-use is possible.
402
Recycling of Automotive Composites
From the plastics in Stream A at least be 2 kilotonnes had to be recycled by the year
2000; in stream B at least 12 kilotonnes; in stream C 5 kilotonnes.
This means that 19 of the expected 55 kilotonnes of plastic waste from cars by now
will have to be reused. This leaves 36 kilotonnes for landfill or recycling.
The development of recycling technologies is mainly aimed at the separation and treatment
of shredder waste, which consists of four fractions:
i. The ultra-light fraction, consisting of PU foam. ICI and Enichem developed a glycolysis
process for this fraction, put together with the other industrial scrap. The resulting
polyols could be used for production of rigid PU foams (see Section 11.3.6.1).
iii. The heavy fraction, consisting of materials like ABS, PVC, PA, etc. The light and the
heavy fraction acting as a filler for more general purposes, like floor tiles, containers,
etc.
iv. The ultra-heavy fraction, consisting of cables (PVC and copper), SMC, BMC, glass,
metals and elastomers.
In the year 2000, the quantity of shredder waste should not have surpassed that in 1986,
that is: 100 kilotonnes; 90% of this shredder waste should be incinerated by the year
2000, and 10% should be reused in an useful application. The implication is that there
will be no landfilling of un-processable waste from scrap cars.
Ford, foresees a gradual but slow adoption of new polymer derived components over the
next 25-30 years. This will in part be due to Fords policy of only using parts with which
403
Handbook of Polymer Blends and Composites
it is 110% sure. In addition to the design and production developments required, two
other factors were identified as slowing the adoption of new components; the need to
improve the quality and consistency of batch produced resins, and Fords continued
investment in metal pressing tools which will not readily be scrapped in favour of less
tried technology.
Without a radical change in car design, the plastics content of the average car is expected to
level-out at 150 kg or 16-8% (Figure 11.8). The type of radical change thought capable of
another wave of polymer application is the introduction of a manufacturing system based on
a modular space-frame design, which will increase the attachment of polymer body parts,
such as wheelbase, roof, interior, dashboard and under the bonnet parts, etc. However, it
should be stressed that a change of this magnitude could take up to 20 years to be realised
and a further 10-12 years to affect the waste stream. As a result the plastics content of scrap
vehicles is unlikely to increase dramatically over the next two or three decades.
In the meantime within the current design paradigm, new polymer derived components
will continue to be used primarily in niche markets where changes in fashion require
short lead-times and production flexibility, e.g., Ford Cosworth, and only expand to
larger volume production lines, e.g., trucks and vans, and finally into main car models
once Ford are fully satisfied with their quality and realibility.
Figure 11.8 Growth rates for the mass fraction of plastics in various vehicles from the
one manufacturer
404
Recycling of Automotive Composites
Projections (based upon industry data) suggest that there will be an overall increase in plastics
waste from approximately 100 kilotonnes in 1990 to 650 kilotonnes by the year 2010, with
an increase in all types of polymers with the exception of PVC, phenolic and acrylic.
In the near future shredder waste might be classified as hazardous waste. This implies
that cost for landfill will rise dramatically, from $30-95 for municipal waste, to $240-
480 for hazardous waste, i.e., 5 to 8 times more expensive.
The Fachgruppe and Ruckkmontage, Germany states that about 2 million vehicles were
being discarded every year [4], of which 1.5 million are cars. The recent data estimate
2.4 million cars being discarded. The number of discarded cars should have increased to
2.8 million by the year 2000. About 5% of this number is exported. Parts and components
of about 10% of the car wrecks are disassembled and sold.
The increasing content of plastic materials in cars, and the increasing number of discarded
cars will result in a massive increase of plastics waste from scrapped cars, from 90
kilotonnes in 1984 to 375 kilotonnes in 2005. These data seem to be an underestimate
as it is reported that in 1989 in Germany about 530 kilotonnes of cars (based on 2
million vehicle disposals by BMW Volkswagen). These data are also used for later years.
By the end of 1993 a nationwide take-back system was established. BMW said it might
in the future impose a recycling goal for plastics in cars. Based on a fleet average, 25% of
the plastic content should consist of recycled polymers.
References
3. A. Weber, Plastics & Rubber Composites Processing & Applications, 1991, 16, 3, 143.
405
Handbook of Polymer Blends and Composites
Bibliography
T. Gomi, New Developments in Advanced Composites for the 90s, Toray Research
Centre, Tokyo, Japan, 1986.
406
Contributors
V. Antonucci
Department of Materials and Production Engineering
P. le Tecchio 80
80125 Napoli
Italy
H.W. Bonin
Department of Chemistry and Chemical Engineering
Royal Military College of Canada
P.O. Box 1 7000
Station Forces
Kingston
Ontario
Canada
K7K 7B4
V.T. Bui
Department of Chemistry and Chemical Engineering
Royal Military College of Canada
P.O. Box 1 7000
Station Forces
Kingston
Ontario
Canada
K7K 7B4
Sam J. Dahman
RTP Company
580 East Front St
Winona
MN 55987
USA
415
Handbook of Polymer Blends and Composites
M. D. Gilchrist
Department of Mechanical Engineering
National University of Ireland at Dublin (NUI-D)
Belfield
Dublin 4
Ireland.
M. Giordano
Institute for Composite Materials Technology CNR, Italy
P. le Tecchio 80
80125 Napoli
Italy
J. Glatz-Reichembach
ABB Corporate Research Ltd.
CH-5405 Baden-Dttwil
Switzerland
Hidemitsu Hojo
Industrial Engineering and Management Department
Nihon University
1-2-1 izumi-cho
Narashino-shi
Chiba-ken 275-0006
Japan
M Jayabalan
Sree Chitra Tirunal Institute for Medical Sciences and Technology
Polymer Division
Biomedical Technology Wing
Thiruvananthapuram - 695 012
India
A.K. Kulshereshtha
Indian Petrochemical Corporation Ltd
Vadodara 391346
India
S.M. Lomakin
Institute of the Biochemical Physics
Russian Academy of Sciences
119991
Kosygin 4
Russia
416
Contributors
I. Miedema
Department of Chemistry and Chemical Engineering
Royal Military College of Canada
P.O. Box 1 7000
Station Forces
Kingston
Ontario
Canada
K7K 7B4
L. Nicolais
Department of Materials and Production Engineering
P. le Tecchio 80
80125 Napoli
Italy
D.Y.S.J. Page
Department of Chemistry and Chemical Engineering
Royal Military College of Canada
P.O. Box 1 7000
Station Forces
Kingston
Ontario
Canada
K7K 7B4
Giuseppe R. Palmese
University of Delaware Center for Composite Materials
Newark
DE 19716
USA
L.V. Ruban
Institute of the Biochemical Physics
Russian Academy of Sciences
119991
Kosygin 4
Russia
Mamdouh M. Salama
Conoco Inc.,
Ponca City
Oklahoma
USA
417
Handbook of Polymer Blends and Composites
N.G. Shilkina
Institute of the Biochemical Physics
Russian Academy of Sciences
119991
Kosygin 4
Russia
R. Strmpler
ABB Corporate Research Ltd.
CH-5405 Baden-Dttwil
Switzerland
S.V. Usachev
Institute of the Biochemical Physics
Russian Academy of Sciences
119991
Kosygin 4
Russia
Cornelia Vasile
Romanian Academy
P.Poni Institute of Macromolecular Chemistry
41A Gr. Ghica Voda Alley
RO 6600 Iasi
Romania
G.E. Zaikov
Institute of the Biochemical Physics
Russian Academy of Sciences
119991
Kosygin 4
Russia
418
Terminology
1
Terminology
There are some controversial definitions of the terms used in the field of polymer mixtures,
therefore their meaning has to be clarified. Most terms are based on the thermodynamics
and include the analysis of the phase diagrams of polymer systems as the basis of
differentiation, see Figure 1.1.
Utracki defines the polymer blend as a mixture of at least two polymers or copolymers
comprising more than 2 wt% of each macromolecular component. Depending on the
sign of the free energy of mixing, the components of the blends are miscible and immiscible
[1, 2, 3, 4].
When significant interactions exist between the two initial components, Gm becomes
negative and the blend is miscible. However, the miscibility is not as complete as in the
case of small molecules. Nevertheless, this nanometric phase separation cannot be easily
1
Handbook of Polymer Blends and Composites
Figures 1.1a and 1.1b, show plots of Gibbs free energy of mixing (Gm) for blends
with differing miscibility, which lead to correspondingly various morphologies.
Figure 1.1c is a schematic representation of a phase diagram for a system exhibiting
both LCST and UCST
2
Terminology
The complete miscibility phenomenon is rare for polymers. For example it is present in
the case of polyphenylene oxide and polystyrene (Noryl from GEP or Vestoblend from
Hls) which is a miscible blend of amorphous polymers or polyvinylchloride (PVC) and
polycaprolactone (PCL) which are miscible in the amorphous state. For concentrations
of PCL higher than 50%, PCL crystallises. A biphasic blend is observed mixing a crystalline
phase of PCL and an amorphous phase of miscible PVC/PCL.
Partially miscible polymer blends are a subclass of polymer blends including those blends
that exhibit a window of miscibility, i.e., they are miscible only at certain concentrations
and temperatures, see Figure 1.1b.
Miscibility has been also defined in morphological terms of homogeneous domain size.
The strict definition of miscibility requires a single phase over the entire composition
range at a particular temperature [5].
The miscibility defined in the terms of the equilibrium thermodynamic (Gm = 0) must
be considered only within the range of independent variables (temperature T, pressure P,
molecular weight, chain structure, etc). The rate at which the thermodynamic equilibrium
can be achieved depends on the driving force, i.e., the polymer-polymer interaction
parameter, and the resisting rheological forces, i.e., the diffusivity. Only detailed studies
of the tendency of these blends to mix or to phase separate over a long period of time can
answer the question of their true thermodynamic miscibility. In thermodynamics
miscibility means single phase down to molecular level. In a pragmatic sense miscibility
means that the system appears to be homogeneous in the type of test applied in the study,
i.e., it is defined in terms of degree of dispersion. In the literature one finds conflicting
reports on miscibility of a given polymer/polymer blend. In a global sense, polymer/
polymer miscibility does not exist, observed miscibility is always limited to a miscibility
window also for a range of independent variables namely, composition, molecular
parameters (as molecular conformation and configuration, molecular weight, molecular
weight distribution), temperature, pressure and others. More than 1600 of these
miscibility windows have been identified for two, three or four component blends.
However, immiscibility dominates the polymer blends technology [1-4, 6].
For the blends with a miscibility window the miscibility is obtained only for a limited
domain of concentrations. Outside the miscibility window interactions are not strong
enough to prevent the demixing of the blend. Nevertheless, no phase consists of a pure
3
Handbook of Polymer Blends and Composites
polymer, each one must be considered as a miscible blend of major component with a
small amount of a minor component. In this case two glass transition temperatures Tg
are observed, but each one is different from the values reported for the virgin polymers.
As an example, terpolymer acrylonitrile (AN)/butadiene/styrene (ABS) and polycarbonate
(PC) with strong interactions between acrylonitrile and PC give a miscible alloy for a
certain range of concentrations.
Miscible or not, most polyolefin (PO) blends are compatible, with the enhancement of
their physical performance responsible for the predominance of blending in PO industry.
The compatibilisation must not only ensure a certain morphology and improvement in
performance. The compatibilised blends must be stable and reproducible, insensitive to
forming stresses and repeated processing. Compatibilisation prevents the phase separation.
The compatibilisation does not mean that immiscible mixtures become miscible at
thermodynamic equilibrium of the system. The immiscible blends become more
transparent by adding a block copolymer (compatibilising agents) due to the reduce of
the interfacial tension that facilitates dispersion therefore the decrease of domain size,
that stabilises morphology against high stress and strain processing and enhance adhesion
between the phases in the solid state. By compatibilisation systems with acceptable
properties can be obtained from mixtures of immiscible polymeric components. As results
improved mechanical properties results.
4
Terminology
compatibiliser should migrate to the interface, reducing the interfacial tension coefficient
and the size of the phase domains (a finer dispersion of the minor component), as well as
creating improved adhesion in the solid state. Intensified phase interactions and a
controlled phase morphology lead to improvements in mechanical properties of the blends.
An additional function of compatibilisers is stabilisation of the blend morphology against
coalescence and agglomeration of the dispersed particles that can take place during the
following processing and forming steps. Different groups of compatibilisers are known
and used for such applications. As an example the thermoplastic elastomer Kraton (Shell)
is a typical compatibilising agent. This compatibilising agent is a copolymer consisting
of a central elastomeric block (polybutadiene, polyisoprene, hydrogenated styrene-
butadiene rubbers) with two lateral glassy polystyrene blocks. It is used to compatibilise
and to improve the impact properties of PO/polystyrene (PS) and PO/
polyethyleneterephthalate (PET) blends. The elastomeric block is miscible with the olefinic
component and the glassy styrenic block is miscible with PS or with PET.
It is important to know that miscibility per se may not be required or even desired in
optimising the properties of a given blend. However what is important is that the miscibility-
immiscibility boundary also delineates the regions of compatibility Figure 1.3(a).
5
Handbook of Polymer Blends and Composites
Incompatible blends. This is the most frequent case because polymers have not often
good affinity for each other. Differences in crystallisation increase repulsive interactions.
This results in a very coarse morphology. The interphase is very thin corresponding to a
very low adhesion between the components. The blend is characterised by two Tg with
values corresponding to those of the virgin polymers and by one or two melting
temperatures (Tm) if crystalline polymers are present.
Usually miscibility (in fact the compatibility) is assessed against the method or technique
of study [7]. Often compatibility is judged by analysing the Tg temperature of the blends
6
Terminology
The term pseudo-compatibility is also used to differentiate between real miscibility and
compatibility [8]. Compatibility being found only for a narrow composition range.
Rtzsch and Haudel [10] define the alloys using interface properties. If no new phases
occur, i.e., no miscibility in the interfaces is demonstrable then the term of blend is
used, if an interface (a new phase) is present it is denoted as an alloy. A polymer alloy
is also a specific sub-class of polymer blend. They could be subdivided into two categories:
1) those in which the compatibilisation leads to very fine dispersion so the molded part
will show neither streaking not excessive weld-line weakening, and 2) those where some
compatibiliser is added in order to facilitate the formation of the desired morphology is
a subsequent processing step.
The term compound will be used as generic term for blends, alloys, filled and reinforced
polymers.
A special case of polymer blends are semi- and full interpenetrating polymer (SIPN and
IPN) networks and other systems that are formed from two or many polymers during
the chemical reactions. In IPN both components form a continuous phase and at least
one is synthesised or crosslinked in the presence of the other [11, 12]. Such networks
interdisperse two immiscible polymers down to a fine scale of phase separation, and
7
Handbook of Polymer Blends and Composites
usually use crosslinking to stabilise this morphology. This can produce a remarkable
synergism of properties.
An engineering polymer blend is a polymer blend or alloy either containing or having the
properties of the engineering polymer.
Thermoplastic Elastomers (TPE) are materials that possess, at normal temperatures, the
characteristic resilience and recovery from the extension of crosslinked elastomers and
exhibit plastic flow at elevated temperatures. They can be fabricated by the usual
techniques such as blow moulding, extrusion, injection moulding, etc. This effect is
associated with certain interchain secondary valence forces of attraction, which have the
effect of typical conventional covalent crosslinks, but at elevated temperatures, the
secondary bonds dissociates and the polymer exhibits thermoplastic behaviour.
Block Copolymers (BC) and Graft Copolymers In a block or graft copolymer, several
incompatible components are linked each other by chemical bonds, and the bulk of the
copolymers form a stable structure even if the two phases are separated. Usually they
show the character or function of each component independently.
Techniques used to retard or eliminate demixing of the polymer blends are called
compatibilisation procedures [13]. They are:
2. Techniques rely on slow diffusion rates by mixing high molecular weight polymers
and co-crystallisation;
8
Terminology
9
Handbook of Polymer Blends and Composites
because of its structure, stabilises a diffuse interface between pure polyethylene and
polystyrene phases, acting more as solubilising agent for the homopolymers than an
anchoring agent. In all cases, however, enhanced strength was observed in the blends,
particularly in the polystyrene rich composition region where the ductility was also
improved. Factors that favour compatibilisation are: an increased polarity, decrease
of molecular weight, specific group attractions, presence of block and graft copolymers
10
Terminology
Figure 1.4 (a) Ideal configuration of a block copolymer at the interface between
polymer phases A and B; (b) Formation of an interphase between them
Pseudoplastic is a rheological term used to describe material which has a yield stress
above which flow occurs and where the flow rate increases sharply with further increases
in stress. Printers ink and paper (cellulose) pulp are pseudoplastic in behaviour.
Bioplastics are biodegradable materials with high water vapour permeability, a good
oxygen barrier, not electrostatically chargeable and usually environmentally friendly
materials. As an example they are obtained by blending thermoplastic starch (TPS), with
biodegradable polymers like polylactic acid (PLA) [20] or poly(vinyl alcohol) PVA and
plastifying agents (glycerol and its derivatives, sorbitol and its derivatives, etc). Bioplastics
can be processed with the existing plastic processing industrial production lines [21, 22,
23, 24]. Bioplastics reduced the waste amounts in an ecological manner by closing the
natural material cycles [25]. Commercial biodegradable plastics, used for packaging and
agricultural mulch films, are Mater-bi AF10H, prepared from a blend of starch (60%)
and natural additives with modified PVA (40%) and plasticisers [26, 27].
11
Handbook of Polymer Blends and Composites
Blood compatibility means that a foreign surface placed in contact with blood should
not induce adverse responses such as thrombosis, destruction of the blood cellular
components, alteration of the plasma proteins, damage to adjacent tissue and toxic or
allergic reactions [31].
Special terms are used for highly interacting polymer systems, such as water-soluble
ones: An interpolymer complex is a chemical compound obtained by mixing solutions as
a result of the interaction of two or more compatible polymers. It can occur both between
two oppositely charged polyelectrolytes due to an ionic bond and through hydrogen
bonding between a polyacid and a proton-acceptor groups of a noncharged polymer
[32] or two components having electron donor-electron acceptor groups [33-39]. An
interpolymeric complex is characterised by a decomplexation temperature at which phase
separation occurs. Interpolymeric complex formation depends on the association-
dissociation equilibrium.
The complex coacervation means a spontaneous liquid/liquid phase separation that occurs
when oppositely charged colloids are mixed as a results of electrostatic attractions.
Polymers which do not dissociate into ions, form solutions of polymeric non-electrolytes
or neutral polymers. Polymers which undergo electrolytic dissociation when dissolved
form solution of polymeric electrolytes or polyelectrolytes. A polyelectrolyte is a
macromolecule in which a substantial portion of the constitutional units have ionisable
or ionic groups, or both [40]
Hydrocolloids are colloids which can interact with water to form new textures and perform
specific functions [41]. Polysaccharides such as starches, carrageenan, pectin, xanthan,
arabic gum, agar gum, locust beam gum, guar gum, carboxymethyl cellulose (CMC),
gellan, konjac mannan and alginates are known from the literature as hydrocolloids.
12
Terminology
the nature of the liquid and the polymer, as well as on the crosslink density of the latter.
A gel with strong intermolecular bonds between its structural elements (gelatine gel) is a
thermodynamically unstable system, the state of equilibrium of which corresponds to
the formation of two phases. Such gels form only because the energy of interaction of the
polymer chains with each other is higher than the energy of their interaction with the
molecules of the low molecular liquid. The latter must be a poor solvent for polymer. If
the liquid is very polar, it cannot disturb the interaction between polymer chain fragments
of low polarity. If the liquid is nonpolar or of low polarity crosslinks appear as a result of
interaction between the polar groups of polymer. Since the affinity of polymer for the
liquid is low, the system should separate into phases, but this separation takes time. They
often have a biphasic structure induced by segregative interactions (thermodynamic
incompatibility) between the constituent polymers [44]. Systems phase separation occurs
in the pre-gel solution state or during cooling [45] or one component gels first from a
single phase solution and the other then forms a second (dispersed) phase by gelling
within the pores of the original (continuous) network [45-48]. The ultrastructure of the
resulting co-gels can vary widely depending on the time-temperature course of gelation
in relation to the rate of the segregation and network formation [49].
The term hydrogel is used to describe materials that are hydrophilic in nature and exhibit
the characteristic structure of a gel [43].
The gelation process is the beginning of phase separation. It could induce the differences
between phase relationships and pre-gel solution state. In the early state of gelation after
the onset of intermolecular association but before development of a continuous network
there will be a massive reduction in the number of species free to move independently
which will reduce the relative importance of entropy of the mixing and hence promote
further segregation therefore the assumption of complete de-mixing could be not entirely
unrealistic (especially when the rate of gelation is slow enough) [48].
The overall physical properties of such co-gels depends on the relative deformation [12, 49].
Syneresis. The process of gradual phase separation of a gel and segregation of the low
molecular weight liquid is called syneresis. The seepage or release of water from gels,
particularly on storage is also known as syneresis. Syneresis in food applications may be
avoided by choosing a suitable chemically stabiliser.
The transition of a nonfluid gel into a fluid solution is called melting of the gels.
13
Handbook of Polymer Blends and Composites
linear amylose molecules are attracted to each other and form bundles of parallel
polysaccharide chains by the formation of hydrogen bonds between hydroxyl groups on
neighbouring molecules [50].
References
1. L.A. Utracki, Polymer Alloys and Blends, Hanser Publishers, Munich, Germany,
1989.
3. A. Ajji and L.A. Utracki, Progress in Rubber and Plastics Technology, 1997, 13,
3, 153.
4. A. Ajji and L.A. Utracki, Polymer Engineering and Science, 1996, 36, 12, 1574.
5. M.M. Coleman, J.E. Graf and P.C. Painter, Specific Interactions and the
Miscibility of Polymer Blends: Practical Guides for Predicting & Designing
Miscible Polymer Mixtures, Technomic Publishing Company Inc., Lancaster, PA,
USA, 1991.
8. C. Vasile and I.A. Schneider, European Polymer Journal, 1973, 9, 10, 1063.
11. M.A. Winnik, Polymer Engineering and Science, 1984, 24, 2, 87.
12. J.A. Manson and L.H. Sperling, Polymer Blends and Composites, Plenum Press,
New York, 1976.
14
Terminology
14. R.D. Deanin and M.A. Manion in Handbook of Polyolefins, Second Edition, Ed.,
C. Vasile, Marcel Dekker Inc., New York, NY, USA, 2000, Chapter 23, p.633.
25. J. Lrcks, Polymer Degradation and Stability, 1998, 59, 1/3, 245.
26. T. Iwanami and T. Demura, Japanese Journal of Polymer Science and Technology,
1993, 50, 767.
28. Polymers of Biological and Biomedical Significance, Eds., S.W. Shalaby, Y. Ikada,
R. Langer and J. Williams, ACS, Washington, DC, USA, 1994.
15
Handbook of Polymer Blends and Composites
30. B.K. Gale, R.S. Besser, J. Brazzle, I. Papautski and A.B. Frazier, Packaging for
Biomedical Analysis Systems, Proceedings of the 198th Meeting of the
Electrochemical Society: Microfabricated Systems and MEMS V, Phoenix, AZ,
USA, 2000.
31. M. Gheorghiu, G. Popa and O.C. Mungiu, Journal of Bioactive and Compatible
Polymers, 1991, 6, 2, 164.
33. M.C. Piton and A. Natansohn, Macromolecules, 1995, 28, 1197, 1598, 1605;
35. U. Epple and H.A. Schneider, Thermochimica Acta, 1987, 112, 123.
36. U. Epple and H.A. Schneider, Thermochimica Acta, 1990, 160, 103.
37. M. Bolsinger and H.A. Schneider, Makromolekulare Chemie, 1994, 195, 2683
38. M. Bolsinger and H.A. Schneider, Journal of Thermal Analysis, 1998, 51, 643.
39. M. Bolsinger and H.A. Schneider, Journal of Thermal Analysis, 1998, 52, 115.
40. A.D. Jenkins, P. Kratochvil, R.F.T. Stepto and U.W. Suter, Pure and Applied
Chemistry, 1996, 68, 12, 2287.
43. R.J. LaPorte, Hydrophilic Polymer Coatings for Medical Devices: Structura/
Properties, Development, Manufacture and Applications, Technomic Publishing
Company, Lancaster, PA, USA, 1997.
44. V.V. Suchkov, V. Ya Grinberg, V.B. Tolstoguzov, Carbohydrate Polymers, 1981, 1, 39.
45. I.S. Chronakis, S. Kasapis and R. Abeysekera, Carbohydrate Polymers, 1996, 29,
137.
46. A.H. Clark, R.K. Richardson, G. Robinson, S.B. Ross-Murphy and A.C. Weaver,
Progress in Food and Nutrition Science, 1982, 6, 149.
16
Terminology
47. A.H. Clark, R.K. Richardson, S.B. Ross-Murphy and J.M. Stubbs,
Macromolecules, 1983, 16, 1367.
48. S. Kasapis, E.R. Morris, I.T. Norton and A.H. Clark, Carbohydrate Polymers,
1993, 21, 269.
50. Dictionary of Food Starch Terms, National Starch & Chemical website, New
York, NY, USA, www.foodstarch.com.
17
Thermodynamics of Multicomponent Polymer Systems
2
Thermodynamics of Multicomponent Polymer
Systems
The thermodynamics of polymer solutions and blends has the same basic principles as
classical thermodynamics however, it was developed as a new discipline.
The fundamental scientific questions raised by polymer systems and their commercial
significance, as well as their potential for achieving novel materials properties, are among
the reasons for continued interest in study of their structural and thermodynamic
properties. Rational development of polymer blends technology has raised many
interesting scientific challenges. The answer these questions requires progressively higher
levels of sophistication and knowledge about the systems using a thermodynamic theory
or model appropriate for the level of detail to be predicted. The reason for the effort is
that thermodynamic solubility is an exception and if compatible systems can be found,
the probability of obtaining a proprietary compatible blend is high. The thermodynamic
study of polymer systems also stimulated the study of the individual polymers. Some of
parameters, related only to the pure components, have to be specified through appropriate
characterisation or physical properties determination, e.g., pressure-volume-temperature
behaviour, as in recent years experimental observations have demonstrated the sensitivity
of macrophase and microphase boundaries on local polymer architecture. The knowledge
of the polymer-polymer interactions are generally required, too. In many cases, the most
practical means, is to reverse the prediction procedure, i.e., the parameters are deduced
from observation of phase behaviour. A quantitative prediction of phase diagram should
be the most useful for practical purposes.
Because of its attractive and exciting theoretical aspects in solving problems related to
structure, conformation and configuration of macromolecules and also due to the direct
practical implications, thermodynamics of polymer mixing has seen over the years many
steps of developments and it is still a very fascinating domain. Many very specialised and
superb reviews and monographs have been published in a relatively short period of time,
starting with Flory, Huggins and Staverman books [1-3] and their co-workers papers,
who developed the first theory that was commented, contested, revised and again
reconsidered in the last year as demonstrated by Sanchez and Stone [4]. Due to the
reasonable character of the assumptions underlying the Flory-Huggins-Staverman theory
19
Handbook of Polymer Blends and Composites
and mainly due to the remarkable simplicity of the final formulas, this theory up to now
remains widely popular among investigators dealing with experimental data treatment.
During the last ten years important reviews appeared with special reference to polymer
blends. The approaches developed in thermodynamic characterisation of polymer systems
including polymer blends are summarised in many books and reviews [5-30].
Two quite divergent approaches are currently in use for describing or predicting the
phase behaviour of polymer mixtures. One uses association or quasi-chemical models
and is applicable for systems where strong specific interactions, like hydrogen bonding
are involved. The other uses a mean field approach which is appropriate when the
interactions are given by geometric mean of the interactions between the two like pairs.
This last approach leads to endothermic heats of mixing and positive Gibbs free energy
(G) values and always predicts immiscibility in the limit of very high molecular weight
of the components, see Figure 1.1a.
Fundamental studies were started independently by Flory, Huggins and Staverman. The
advancement of the theoretical concepts in the field of thermodynamics of homopolymer
solutions and blends was due to development by relaxing some of the five assumptions
of the Flory-Huggins theory (see subchapter 2.1.2). Further progress in the Flory-Huggins
theory was associated with an account for polydispersity of macromolecules and the
selection of more complicated dependencies of the enthalpy of mixing on concentration
and temperature [31]. The most serious shortcoming of the Flory-Huggins theory is
apparently, a disregard of the volume change occurring at dissolution. The effect was
taken into account in the framework of the molecular theories of polymer solutions
proceeding from their equation-of-state that originate from fundamental studies by
Prigogine and co-workers [32, 33] who demonstrated the possibility of applying the
principles of corresponding states to polymer systems. The refinement of the Prigogine
theory [34-36] enabled to predict the existence of upper and lower critical solution
temperatures, to explain the phenomenon of volume change during mixing. The difficulty
of the theory resides in finding the values of reduction parameters and the accuracy of
their determination [37-39], (because the predicted phase diagram very sensitively depends
on them), which was obviated by new Florys theory [40-45] and lattice liquid theory
[46-50]. The new Flory theory underwent a number of modifications mainly connecting
with regard for inhomogeneity of mixing and some other factors [51-55]. The construction
of phase diagrams Figures 1.1b and 1.1c - was started 40 years ago by Scott, [56],
Tompa [57, 58] and Stockmayer [59] who applied Florys formalism. The theories were
experimentally verified first by Koningsveld and co-workers, in a series of works what is
undoubtedly of first-rate important for physical chemistry of polymers [16, 31, 60-65]
20
Thermodynamics of Multicomponent Polymer Systems
In this chapter the milestones and the progress in thermodynamics of polymer mixing is
discussed, emphasising the aspects and relationships currently used in miscibility/compatibility
studies and the practically applications.
A polymer blend is a special type of polymer solution. All results developed for polymer
solutions, including those for multicomponent mixtures can be applied to polymer blends,
allowing that a polymer is distinguished from a small molecule by some size parameters,
such as the degree of polymerisation.
The main task of the thermodynamics of binary polymer systems is to establish the
theoretical background to find the conditions of mixing and the ranges of mutual solubility
of different polymers depending on their chemical nature and molecular mass. The models
provide for the specification of the size or length of the macromolecule. The mixing of
polymers at elevated temperatures and of amorphous polymers can be considered as
21
Handbook of Polymer Blends and Composites
mixing of two liquids. According to the general principles of thermodynamics, the state
of system can be described by the variation of the Gibbs free energy (G):
G = H- TS (2.1)
The necessary but not sufficient condition for the system stability is that G < 0 which is
fulfilled if enthalpy H < 0 and entropic term TS > 0 or if H >0, |TS| > |H| see
Figure 1.1a and 1.1b.
At first sight, it appears that for two polymers with a high molecular mass, the contribution
of the entropic term is very small, so the change in enthalpy is responsible for the mixing
behaviour of two polymers. However the Russian school showed that contribution of
entropy cannot be neglected at least for several systems [5, 13, 14].
The change of molecular packing upon mixing and its effect upon the thermodynamic
properties of the mixture, have not, as far as we know, received adequate attention, although
several studies are known [78]. The packing affects results because the molecules will
arrange themselves in such a way as to shorten the intermolecular distances, i.e., decrease
the volume, as much as possible in order to maximise intermolecular interactions. The
most favourable arrangement depends on the molecular geometry. The best arrangements
are achieved in crystals, but in liquids mutual molecular orientations are still preserved in
a rudimentary form. Parallel arrangement generates liquid crystals. Their entropy is very
low because their freedom of movement is restricted along their molecular axes.
For over 50 years the lattice models as Flory-Huggins-Staverman theory dominated the
polymer solution thermodynamics. The Flory-Huggins theory, being the original from
which the thermodynamics of polymer systems emanated, in its first version has the
following assumptions [1-7, 27]:
Fluctuation effects are excluded from consideration. That is why, like any mean field
theory, this one does not work when dealing with dilute solutions or with phenomena
occurring in the close vicinity of the critical point.
22
Thermodynamics of Multicomponent Polymer Systems
The application of the rigid lattice model. Within its framework, no account is taken
of distinction in size and shape of monomer units of different homopolymers or in
copolymers.
Polydispersity is ignored.
G m = (RTV / Vr ) A ln A + B ln B + AB A B (2.2)
rA rA
where, R is the universal gas constant, T is the absolute temperature, V is total volume of
the mixture, rA and rB are the numbers of segments (or polymerisation degree) of polymer
A and B, ri = Vi /Vr (Vi is molar volume of component i) A and are volume fractions
of components.
Koningsveld kept these terms in Equation 2.2, but added an interaction function. By
means of this function several improvements of the theory have been accommodated in
Gm expression [79]. For example:
where S and h/T represent the entropic and enthalpic contribution or the parameter, AB
is a free energy parameter comprising an energetic H and an entropic part, S:
23
Handbook of Polymer Blends and Composites
AB = H + S.
2 i
AB (, T) = a i 2i ; with a i = a Tij
i
(2.3)
this relationship being introduced to explain many experimental data as multiple peak
spinodals. Other expressions have been also proposed by Koningsveld and other authors
[80]. They concluded that four factors determine the polymer/polymer miscibility namely:
interacting surface areas of various types of segments, coil dimensions as a function of
temperature, molar mass and concentration, molar mass distribution and free volume.
The simplest form is:
( A , N)
(T, N, ) = + ( A , N) (2.5)
T
In this approach the contribution to the entropy of mixing represented by the first two
terms in the brackets (Equation 2.2), is purely combinatorial and will become negligibly
small as the molecular masses of each component increase, i.e., ri becomes large. However,
the entropy of mixing also has other significant non-combinatorial contributions.
This means that the miscibility becomes increasingly dependent on the interactional free
energy, expressed as a segmental interaction parameter, which will required to be either
very small or negative, if a one-phase system is to be obtained (Gm negative). The AB
parameter is then assumed to be a composite term that includes contribution from
dispersive forces, specific interactions, non-combinatorial entropy effects and, to a lesser
extent compressibility effects. In the recent treatment the AB is considered a binary
24
Thermodynamics of Multicomponent Polymer Systems
= AB (AA + BB)/2
equal with algebraic sum of energies of intermolecular interaction. Hildebrand [83] realised
that the geometrical average seems to be reasonable:
ij = ii jj
It should be realised that the equation of Gm (Equation 2.2) is only a crude approximation
because the distribution of the molecules on the pseudo-crystalline lattice is non-random
and the molecules usually have different size and shapes. A much more correct analysis
replaced the volume fraction , by the surface fraction and the composition independent
parameter , by a concentration dependent function gij, with a character of a free energy
function. All these values result mainly from contact interactions. Various kinds of
interactions have to be considered but they must be understood in local terms not in the
terms of the whole molecule. While the energies of the homo-contacts, ii, may be obtained
easily from the properties of the pure components, the hetero-contacts energy, ij, are
generally derived from the thermodynamic studies of the mixtures themselves, frequently
a very difficult task. Being able to predict the value of ij from the properties of the pure
components would therefore be of great practical importance.
The blends tend to reach equilibrium in two ways. Firstly, it tends to lower its Gibbs free
energy by further phase separation into a state of smaller mutual solubility and the
second effect is the increase of mutual solubility in the case of immiscible blends. This
phenomenon was named thermodynamic ageing [79]. This phenomenon appears because
the chemical potential gradients that will force the system into some degree of mutual
solubility, however slow the change may be that will eventually lead to physical properties
deviating from those of the initial blend. Special measures must be taken if this inevitable
process of thermodynamic ageing is to be prevented. The problems associated with the
25
Handbook of Polymer Blends and Composites
production of polymer blends under far-from equilibrium conditions can be very difficult.
During the products lifetime, its properties drift toward the equilibrium. For poorly
mixed homogeneous polymer blends this may be detrimental although it may even lead
to an improvement of properties. However, the heterogeneous polymer blends, the phase
ripening and deterioration of mechanical properties may lead to disastrous results.
Vitrification of one of the phases or both will usually take place, but reprocessing of the
blend will then start renewed thermodynamic ageing. Crosslinking of one or both phases
might offer a better perspective because a network tends to expel free chains.
When attempting to modify or tailor the properties of a polymer by blending, one quickly
realises it would be highly desirable to have available a technique that provides guidelines
for estimating miscibility or regions (windows) of miscibility. A simple approach that
could provide guidelines rather than a rigorous theoretical treatment, particularly useful
to the industrial polymer scientist/engineer is solubility parameter approach. It is capable
to emphasise trends and not rigorous quantitative results.
The dispersive forces are often approximated from interaction parameters. AB can be
calculated from the relationship:
Equation 2.6 is valid for non-polar substances that means in the presence of only van der
Waals or London interactions, where A and B are solubility parameters of components.
It can be seen that calculated values of AB are always positive. Critical values of interaction
parameter and volume fraction are:
1
( A ) crit =
[1 + (rB / rB)1 / 2 ]
26
Thermodynamics of Multicomponent Polymer Systems
find a miscible polymer pair in absence of any specific interactions. Therefore the
components of the blend would have to have closely matching values of i if miscibility
is to be achieved. Some small assistance might come from the combinatorial entropy but
consideration of the critical conditions for miscibility, namely AB < (AB)crit, shows that
the difference between the solubility parameters of the components is very small indeed.
For low molecular weight compounds, the solubility parameters can be directly calculated
from heat of vapourisation measurements = [E/V]1/2 in (cal/cm3)1/2, where E is the energy
of vapourisation or 2 = CED, cohesive energy density. The polymer solubility parameters
are determined by indirect methods as swelling, viscosity or surface tension measurements in
various solvents or by calculation from group molar attraction constants (Fi):
Fi
=
V
Weight fractions of repeated units are multiplied by their respective individual solubility
parameter contributions. (A - B)2 is considered a miscibility parameter, denoting
miscibility for values 0.1 or lower. To use this parameter to estimate miscibility a bank
(or database) of solubility parameter is necessary connected to a program (Matprop
from SF Technologies) including polymer properties [10, 87].
Coleman, Graf and Painter [10] produced a practical guide to polymer miscibility based
on solubility parameter approach, using the Mac II MATLAB programme, that is used to
select the components of a blend, reducing the time used in searching the potential miscible
polymer systems and indicating the extent of chemical modification of polymers by
incorporation of some functional groups to render them compatible. It allows calculation
of the free energy, phase diagrams and miscibility windows. However due to the errors
involved in such an evaluation it is not recommended for theoretical purposes. Prediction
of compatibility on the basis of solubility parameters using a computer programme known
27
Handbook of Polymer Blends and Composites
Molar attraction constants
as spherical volume of solubility gave good correlation with mechanical properties for
the polycarbonate (PC)/acrylonitrile-butadiene-styrene (ABS) blends [88].
The molar attraction constants tabulated by Small, Barton or van Krevelen, [89-91]
have been compared and selected for unassociated and weakly associated groups, see
Table 2.1, using the same set of model compounds.
28
Solubility parameters Thermodynamics of Multicomponent Polymer Systems
29
Handbook of Polymer Blends and Composites
The errors inherent in the indirect experimental methods used to determine polymer
solubility parameters are too large to be useful in prediction of polymer miscibility.
It has been established on the basis of numerous studies that, for high molecular weight
polymer blends in the absence of favourable intermolecular interactions, miscibility is
only achieved when the solubility parameters of the two polymers are within about
0.1 (cal/cm3)1/2 of one another for rA = rB = 5000. Coleman, Graf and Painter [10]
found that the errors in calculating solubility parameters are approximately 4 times
higher (of 0.4 (cal/cm3)0.5) than mentioned value of 0.1 and standard error in molar
volume determination was ~ 2.8 cm3/mol. Therefore for such a kind of blends, the
solubility parameters cannot be used to predict phase behaviour. For one-phase blends
to be obtained it is desirable to have systems in which favourable specific interaction are
present. The ability to quantify such contributions, however, is less well-established.
Coleman, Graf and Painter [10] simply add an extra term to Equation 2.1 and then scale
this up to accommodate varying strength of different specific interactions. The critical
values of (AB)crit, or solubility parameter difference ()crit change in a particular manner
with the blend composition and type, number and relative strength of specific favourable
intermolecular interactions. Taking into account the errors involved in calculating
solubility parameters to avoid over-interpretation of the results and also considering the
equilibrium constants describing both the self-association and the inter-association of
the two polymers the following values of ()crit have been proposed to be used to discern
significant trends in phase behaviour of some sets of polymer blends, in respect with the
forces involved in mixing process and that can be accommodated if a one-phase blend is
to be obtained as follows:
Weak, moderate and strong hydrogen bonding (or other specific interactions):
()crit = 1.0, 1.5, 2.0 and 3.0 (cal.cm3)1/2, respectively, see Table 2.3.
30
Thermodynamics of Multicomponent Polymer Systems
G m G Hm G FH
m
= +
RT RT RT
a term that takes into account the specific interaction GmH/RT, which has a complicated
composition dependence, given the necessity of taking into account inter- and self-
associations. According to this relationship any immiscible binary system may be
transformed into a compatible one by chemical incorporation into the chain of a sufficient
number of groups of specific interactions through their favorable contribution to dominate
over unfavorable ones,
Hydrogen-bonded groups can be determined by high resolution (~2 cm-1) Fourier Transform
Infrared Spectroscopy (FT-IR) or from the isotherm of adsorption [92]. The samples have
31
Handbook of Polymer Blends and Composites
to be sufficiently thin (usually as films deposited on KBr tablets from diluted solution) to
be within the absorption range where the Beer-Lambert law is obeyed [93]. Non-hydrogen
bonded carbonyl stretching frequency at ~ 1739 cm-1 and hydrogen bonded carbonyl band
is observed some 21 cm-1 lower at 1718 cm-1. Considering the equilibrium constants
describing both the self-association and the inter-association of the two polymers, that
describe the stoichiometry of hydrogen bonds in the blend and, assuming that the system is
single phase, one can readily calculate the fraction of hydrogen bonded groups as a function
of composition. Using this procedure, Painter, Coleman and others [93] compared
experimental and calculated phase diagrams for a great number of systems and have
established some limit of miscibility as: poly(2,6-dimethyl-4-vinyl phenol) (PDMVPh), forms
miscible blends with the homologous series of poly(n-alkylmethacrylates) up to poly(n-
hexyl methacrylates) and also with polyethylene-co-vinyl acetate (EVA) copolymers
containing greater than 40 wt% vinyl acetate, or less than 35 wt% vinyl acetate, two
phase systems are formed. The accentuation of the strength of interassociation over self-
association enhances the favourable contribution from the GmH/RT term and is also a
favourable trend for miscibility.
32
Thermodynamics of Multicomponent Polymer Systems
The specific interactions between electron donor and electron acceptor groups could be
taken into account as a negative contribution to the interaction parameter as: VmEda/
RTVctc a d, where d and a are volume fractions of donor and acceptor components in
the mixture. Vm and Vctc are average molar volumes of the component of the blends and
respectively of the charge transfer complex and Eda is average cohesive energy of the
charge transfer complex usually its value is about ~ 6 kJ/mol [94].
It has been concluded that high molecular mass polymers may be miscible only at the
negative values of AB when the interactions of the A-B type dominate over A-A and B-B
interactions, therefore for specific interactions between two polymers. It is only one
example of atactic polyvinylethylene and cis-1,4-polyisoprene where no specific
interactions are present (only van der Waals) and system is miscible, but AB has a very
low value of 0.0004 [5]. However, some authors found that because of the self-association/
inter-association equilibrium the existence of the groups able to interact is not a necessary
and enough condition for miscibility, because they can give positive values of interaction
parameter when self-association is dominant [95].
However, when A and B are almost equal, the calculated value of AB may be below the
critical value. Other authors considered both AB or arising from various contributions
[96, 97] but the findings are still unable to deal adequately with the presence of miscibility
windows in copolymer blends and they also fail to recognise the part played by structural
factors such as tacticity, sequence distribution in copolymers, and non-combinatorial
contributions. Consequently, following analyses developed by various authors [96-98]
consider the segmental parameters established initially from experimental phase behaviour
measurements are used to predict behaviour of a range of similar systems [98]. The
effects of the local environment or micro-structural differences (as tacticity in PVC-based
blends, on the miscibility behaviour are described by using AB as arising from inter- and
intra-segmental interactions and co-solvent concept in explaining the miscibility in blends
involving copolymers [11, 12], or by Cantow-Schulz approach [76, 77]. The miscibility
maps for various homopolymer/copolymer or copolymer/copolymer blends have been
calculated [10-12] both in binary and ternary blends on the basis of solubility parameter
approach.
To predict the miscibility in solid state the following thermodynamic criteria can be used:
A and B very close to each other, with approximately the same values
w B M 2 B
H mix = w A M A 1 A1 ( A B)2 [ (1 w A )M A A ]
(1 w B)
33
Handbook of Polymer Blends and Composites
where Mi , i and wi are molecular mass of monomeric unit polymer density and weight
fraction, respectively. It was established that for miscible polymers, values of Hmix are
in the range from 10-3 to 10-2 J. In spite of the satisfactory agreement between experimental
data and theoretical calculations some exceptions are found. Heat of mixing approach is
useful for explaining the: miscibility windows of a series of polymer blends as polyamides,
polyesters and copolymers. It is not useful in generation the phase diagrams [99, 100]
Jacobson and others have described how molecular modelling studies can give a rapid
estimation of whether two polymers will form a miscible blend using a method that can
account for specific interaction between polymer segments [101, 102]. The theoretical
foundation of their method relies upon the premise that miscibility is determined by
thermodynamic factors alone, and furthermore that these thermodynamic factors are
dominated by the energetics of local interactions between segments of the polymer chain.
Mainly rigid high-performance polymers were studied. Tiller and Gorella, have extended
and modified the two-segment approach to miscibility prediction of flexible polymers by
a flexiblend algorithm [103]. A computer model was elaborated using high-temperature
molecular dynamics or rotational-isomeric-state Monte-Carlo sampling and also
amorphous cell approach. The FLEXIBLEND method developed by Tiller and Gorella is
useful for initial screening of potentially miscible combination of polymers for a given
application and for understanding changes in phase behaviour resulting from
modifications in chemical structure. It is also very convenient for the polymers with
close solubility parameters. The FLEXIBLEND method is quick, typically requiring only
a couple of central processing unit (CPU) hours but only energetics of local interactions
are considered. The amorphous cell method is more rigorous but requires greater
computing resources, calculations on a polyethylene oxide (PEO)/polypropylene (PP)
blend took about three CPU days on a Silicon Graphics 4D/35 workstation and it allows
investigation of the effects of concentration by varying the volume fractions of the two
components in the mixed cell.
Sanchezs theory predicts lower values for G, therefore predicts a better solubility in
polymer systems than other theories.
34
Thermodynamics of Multicomponent Polymer Systems
Finite compressibility adds a destabilising influence, and often this is origin of phase
separation upon heating of LCST diagrams (see Figure 1.1c). The average mutual distances
among the molecules depend not only on the interplay of attractive and repulsive forces,
but also on the temperature and pressure. Generally the entropy of a system increases
with its volume. The expansion of a condensed system is paid for by a decrease in cohesive
energy. An equilibrium is established, at which the free energy of the system is minimised
and the volume adopts an equilibrium value. With increasing temperature, the entropy
effect becomes more important and the equilibrium shifts to larger volumes. At large
expansions, and especially close to the critical point, the behaviour of most substances is
remarkably similar. Under such conditions, the behaviour of many compounds is described
well by the theory of corresponding states. In the equation-of-state theories, the key term
is the change in volume upon mixing. Excess contact energies increase the volume when
positive and decrease it when negative. The equation-of-state is written in terms of reduced
volume, pressure and temperatures.
Among the new developments the most important are equation-of-state models proposed
for the first time in 1950 by Prigogine and co-workers and extended by many others
[32, 33].
In contrast a hole model has been developed by Sanchez and Lacombe [46, 105] where
free volume is represented as vacant lattice sites. To characterise the stability of the
systems consisting of two polymers of high molecular mass, Sanchez has developed the
theory based on the model of compressible lattice. This theory is capable of describing
the volume changes by mixing and predicts the existence of LCST. It is assumed that
polymer segments are rigid cores and that four or six parameters operating in the theory
can be found from pressure-volume-temperature (PVT) properties of individual
components. On the basis of this theory Sanchez derived the expression for chemical
potentials, phase stability conditions, and interfacial tension. The lattice-fluid equation-
of-state has the following simple closed form:
2 + P + T [ln(1 ) + (1 1 / r) ] = 0 (2.7)
35
Handbook of Polymer Blends and Composites
Where the i are close-packed volume fractions. The P* replaces the interaction parameter
in the Flory-Huggins theory. The P* = when = i = 1.
An examination of these equations shows that Tc increases with r while Pc and c decrease
with increasing r.
The expressions for other thermodynamic functions have been developed but they are
not reproduced here.
This class of theories retained forms similar to the Flory-Huggins theory for the
combinatorial entropy and the interaction energy but adds free volume contribution to
both the entropy and enthalpy of mixing. They predict LCST behaviour even when P*
is constant and independent of temperature by volume contribution. In the case of the
enough strong specific interactions even P* (or ) values deduced from this framework
will be temperature dependent.
A general result of the lattice fluid theory is that differences in equation-of-state properties
of the pure components make a thermodynamically unfavorable entropic contribution
to the chemical potential. The difference between the pure component parameters,
especially T* values tend to destabilise a solution and make it more susceptible to phase
separation. A small change in combinatorial entropy term limits miscibility both at low
and high temperatures so both LCST and UCST may appear.
This theory was widely applied to polymer blends [106-109] and some improvements
have been proposed [110-112].
36
Thermodynamics of Multicomponent Polymer Systems
Another complication of the theory is that the interaction energy is dependent on blend
composition, but for current objectives, it is more important to know the range of
temperature where phase separation occurs than the precise phase diagram shape rather
than to predict or interpret the phase diagram in fine details. However, sometimes this
effect and also the effect of polydispersity on the bimodal curve are necessary to be
known. Theories and computer software that allow for these effects are available.
Two systems are considered very convenient to test the theory EVA-copolymer/chlorinated
PE and polyether sulfone/PEO because their properties are very diverse and they show
very large effect of pressure on the phase separation temperature and a very large volume
change on mixing. LCST of these systems increases with increasing pressure.
Recently Sanchez and Stone [4] developed a generalised lattice fluid model (with z
coordination number) able to explain most of thermodynamic properties of polymer
blends including systems with strong interactions. So, both the old Sanchez-Lancombe
theory and Flory-Huggins theory appear as special cases, all relationships being deduced
in certain conditions. They demonstrated that although the Flory-Huggins model does
37
Handbook of Polymer Blends and Composites
indeed have limitations and drawbacks, it is still much more general than is accepted by
the thermodynamics scientists. They have obtained former Flory-Huggins relations
without ever using the lattice description of a fluid.
Sanchez and Stone [4] took into account the finite compressibility of polymer solutions
that affects the phase stability and chemical potentials in a significant way. The Gibbs
free energy at a T and P constant can be separated into its compressible and incompressible
contributions. Searching for phase stability they concluded that a compressible solution
is always less stable than the corresponding incompressible solution, because the equation
of G contains a term for the compressible contribution that is an unfavorable contribution
to phase stability. This term increases with temperature and at high enough temperatures,
this unfavourable term becomes large enough to overwhelm the positive constant volume
term and a phase instability develops. This phenomenon of thermally induced phase
separation (LCST) is well known in polymer solutions and blends [115]. With increasing
temperature, the interactions are weakened, causing LCST behaviour, while the
noncombinatorial entropy contribution becomes more and more dominant, leading
eventually again to phase stability of the blend or UCST behaviour. This phase behaviour
is experimentally easily accessible only for blends with component having very different
polymerisation degree. The parameter has to be negative at low and high temperatures
and positive in a temperature range between.
The next step was to consider that thermodynamic behaviour depends on local structure.
These effects are generally referred as free volume effects which have been extensively
theoretically treated but not entirely successful from a quantitative point of view.
Attempts to quantify the effect of free volume have proceeded by two ways. The first
uses a cell model, described initially by Prigogine and Simha [32, 33] and later in a
modified form by Flory and co-workers [40-42] which allows the lattice site to vary in
size so as that each site, or cell, can contain a portion of the systems free volume. Patterson
and Robard [116] using the Florys model approximated in the original Flory-Huggins
theory to the sum of two terms, one describing contact or interactional energies and the
second consisting of so-called free volume terms. The former decreases with temperature
through the usual 1/T dependence, while the second increases, see Figures 2.1a and 2.1b.
The sum of the interactional and free volume terms has a parabolic dependence for AB
crossing the line of (AB)crit twice, giving both UCST and LCST behaviour with a region
of miscibility in-between. The LCST behaviour appears usually for hydrogen bonding
materials, where entropy changes due to association (changes in hydrogen bonding) and
it is seems to be critical factor and due to the free volume differences observed in mixtures
of some non-polar polymers or both factors could contribute. Other theories manage the
last effects as compressibility effects as the most common so the phase separation is
entropically driven.
38
Thermodynamics of Multicomponent Polymer Systems
A significant failure of the Flory-Huggins model is its inability to account for the universal
property of polymer solutions to phase separate on heating. The simple Flory-Huggins
theory does not predict LCST behaviour because it was developed as an incompressible
model. For this reason it is useful to use an equation-of-state theory that can predict
LCST behaviour stemming from compressible nature of polymers.
The Flory-Huggins theory is useful if one only wishes to know whether miscibility occurs
or not, but an equation-of-state theory is generally needed to answer more detailed
questions about phase diagram.
The surface in TPx (x denotes concentration) space that separate the stable (metastable)
from thermodynamically unstable region is defined by spinodal condition namely G
=0, because G < 0 inside the spinodal and G > 0 outside the spinodal. On the spinodal
line we have the following conditions:
39
Handbook of Polymer Blends and Composites
Therefore unfavourable entropies drive the phase separation in an LCST system, while
unfavorable energetics drive the phase separation in an UCST system.
All known miscible polymer blends appear to phase separately at elevated temperatures,
unless thermal decomposition intervenes.
The equation-of-state for the lattice fluid modified by this new approach is [4]:
Po / kT + ln(1 ) / (1 1 / r)ln(1 ) / = 0
or (2.10)
1
Po / kT + ln(1 ) / (1 1 / r)[1 + + ........] = 0
2
For
0, the ideal gas law is recovered while setting = 0 we obtain the Sanchez-
Lacombe equation-of-state (Equation 2.7).
Parameters of equation-of state for polymers can be found in [4, 5] and in the Polymer
Handbook [117].
40
Thermodynamics of Multicomponent Polymer Systems
between different segments and also the interaction parameter with AB (1-jS) where S is
volume fraction of solvent. So when S0 the equations for a binary system are obtained
while for S1, the AB0 and system becomes a single phase one. The critical values of
the system do not depend on the solvent nature at the same S. This should means that if
two polymers are not miscible in a given solvent, they will not be miscible in any other
solvent. This conclusion is not in accordance with many experimental data, being
established that the influence of solvent on the phase behaviour is considerable [120-
122]. The influence is manifested both in solution and in solid state (films from various
solvents exhibits various morphologies). A correlation between critical concentration of
mixing and the exponent from well-known Mark-Houwink equation has been also
established. Due to this influence the Scott and Tompa theories have been reconsidered
by Patterson and Koningsveld [64, 119] whose thermodynamic treatment took into
account the role of solvent. According to their theories the spinodal equation is:
ri i 2 rri j ( i + j ) i j + rArB
(2.12)
( S A + S B + A B ) S A B = 0
with: 2S= SA + SB - AB
The other ij are obtained by cyclic interchange of subscripts. where ri are number of
segments in components, i are volume fractions, ij correspond to three energies of the
formation of contacts between segments. Contrary to their significance in Florys theory
where contacts were related to the molecules of the polymer ij (Flory) = riij (Tompa).
Symmetric bimodals and spinodals are obtained when SA = SB and AB are constant.
Higher difference in ij leads to a great broadening of the immiscibility region. At SB = SB
and 23 = 0, the spinodals disappear and therefore the polymers become miscible both in
presence and absence of solvent. If SA SB at AB = 0 or <0, the region of immiscibility
will be present and shifted to lower concentrations. In this case the spinodals have a
closed loop shape and do not intersect the axes of triangle diagrams, therefore binary
mixtures are fully miscible [119] see Figure 2.2. Increasing the quality of the solvent for
polymer produces almost the same effect as decreasing it. Increasing the molecular weight
of one of the polymers enlarges the region of high concentration of the polymer of lower
molecular weight.
When molecular masses of polymers decreased and SA SB, closed loops of immiscibility
rapidly narrow and at low molecular masses disappear. Koningsveld and others [64]
have studied a system, which shows a closed immiscibility loop (high density polyethylene/
isotactic polypropylene (IPP)/diphenyl ether) and pointed out that polymer immiscibility
41
Handbook of Polymer Blends and Composites
Figure 2.2 Schematic representation of ternary diagrams for systems: polymer A /polymer B/
solvent S a) curve a and b: AB = 0.0025; curve a SA= SB; curve b SASB ; curves c) AB =
0, SASB with increasing differences between SA and SB. [119] b) PS/PVME/chloroform
[116, 123] experimental closed region of incomplete miscibility decreased with increasing
temperature; the system is turbid inside the region and clear outside c, d) polymer A/polymer
B/copolymer block copolymer with various composition and molecular weight [124-126a]
and e, f) polymer A/polymer B/polymer C (phenoxy (PH)/aromatic polyamide/polyethylene
terephthalate (PET) and phenoxy resin/PC/polycaprolactone) [95]. The light and shaded
zones represent transparent, turbid and opaque films, respectively
42
Thermodynamics of Multicomponent Polymer Systems
in solution does not necessarily imply incompatibility in the absence of solvent. The
origin of the closed loop is not completely understood.
Another area where the solvent effect could be important is in the segregation of the
blocks within isolated block copolymer macromolecules [127].
For real systems the restriction of miscibility in the region of small concentration is connected
with an inequality of polymer-solvent interactions, whereas in the region of high
concentration the immiscibility is the result of unfavourable polymer-polymer interactions.
The ternary mixtures of polymer homologues, a phase diagram of closed type may be
obtained. Such kind of phase diagrams are also typical for miscible mixture exhibiting
LCST, such as the PS/PVME system Figure 2.2.b. On the basis of AB measurements has
been found that = SA - SB diminishes with decreasing temperature, the immiscibility
region narrows. But for this system AB should increase with temperature according to
the Scott theory [56]. Contrary, Pattersons theory [119] explains the experimental data,
emphasising the role of inequality of ij on the miscibility of two polymers in common
solvent and so the role of solvent in ternary mixtures. Moreover, it was established that
polymers immiscible in solution are immiscible in absence of solvent, too. For the systems
with UCST as oligostyrene and dimethylsiloxane mixture, AB should decrease with
increase in temperature. For such systems, the introduction of solvent should rapidly
decrease AB and change its dependence on temperature from negative to positive.
In the other theory based on the Florys theory of equation-of-state the interaction
parameter was obtained function on solvent nature and polymer concentration and the
43
Handbook of Polymer Blends and Composites
interaction parameters are multiplied with surface fractions [130]. The use of surface
fraction instead of volume fraction counts for the molecular contact probability. This
theory explained how the affinity of each polymer influences the ternary phase diagram.
It was found that if a solvent has a different affinity to the mixture components, the
region of immiscibility of the phase diagram of the blends shifts rapidly towards the
polymer-solvent axis. For a good solvent for both polymers and small values of , the
binodals are symmetric.
The influence of concentration has been studied for PS/PMMA in toluene and
bromobenzene. It has been established that in dilute solutions the does not change,
whereas in a semidilute solution increases with concentration. The linear increase of
AB begins at concentrations exceeding crossover concentration. The extrapolation of
the AB in solution to 1 gives the AB, corresponding to polymer-polymer pair.
The morphology of the polymer blends obtained by solvent casting depends of the solvent,
temperature, time, concentration, etc. This dependence is important in membrane
manufacture.
The thermodynamic behaviour of the ternary systems constituted from three different
polymers has been described using the same approaches as for ternary polymer-polymer-
solvent system. The interaction parameter having the same expression (Equation 2.12).
If crit0, then the ij >0 therefore the ternary polymer system is immiscible. The general
picture of the effect of the third component thermodynamic behaviour of the ternary
mixture was developed on the basis of the Flory theory of equation-of-state taking into
account the constants between surface
N
by the fraction of surface i = (Si/S)i where Si is a
segment surface area and S = S
i=1
i i is total surface area.
44
Thermodynamics of Multicomponent Polymer Systems
a) The addition of the third component to the two immiscible polymers makes all system
miscible if this is miscible with two other examples: PMMA/polyethylene methacrylate
(PEMA)/polyvinylidene fluoride (PVDF); PVDF is miscible both with PMMA and PEMA.
b) By adding a component miscible with one of the component of miscible binary mixture,
the phase separation proceeds as demixing of single phase regime. With increasing
temperature, the miscibility region broadens example PS/PVME/polypropylene oxide
(PPO) system;
c) The following systems are of practical interest: 1) two with LCST and one with
UCST (PS/PVME/PCL, PVC/PC/chlorinated polyvinyl chloride (CPVC), PVC/PMMA/
styrene/acrylonitrile copolymers (SAN), p-methylstyrene-acrylonitrile copolymers (p-
MSAN)/PEMA/PMMA, SAN/PMMA/PVA); 2) three with LCST; 3) two with UCST
and one with LCST (PS/PC/PCL) [5].
Block and graft copolymers display unique properties, in solution as well as in solid
state, which can essentially be attributed to the incompatibility of the dissimilar blocks
in their molecule. Due to their possessing two incompatible blocks within a molecule,
block and graft copolymers exhibit surface activity and colloidal properties [124-126a].
45
Handbook of Polymer Blends and Composites
The transition from immiscibility to miscibility and again to immiscibility with appearance
of miscibility windows has been theoretically assessed by equation-of-state theory, the
theoretical variation of the interaction parameter with temperature and composition
proving this behaviour because in function of copolymer composition (x), the takes
both positive and negative values, see Figure 2.3.
Kammer [147] introduced a new parameter to the generalised that reflects the difference
in the segment dimension, the final equation predicting both LCST and UCST.
Because the aim of this handbook is mainly practical, the expressions for thermodynamic
functions are not given. Several compatibility windows for hopolymer/copolymer mixtures
of potential practical interest are given in Table 2.4.
Figure 2.3 Schematic representation of the variation of the with temperature and
copolymer blends whose miscibility changes with copolymer composition x. Tobs is
temperature where the system changes from immiscible to miscible [5]
46
Compatibility windows
Polymer mixtures
47
Handbook of Polymer Blends and Composites
From the miscibility maps the following values of the segment-segment interaction
parameters have been determined [11, 12], see Table 2.5.
Random copolymers offer a power tool for achieving the desired results since they permit
adjustment of both the intermolecular interaction between the mixture components and
48
Thermodynamics of Multicomponent Polymer Systems
within components. In fact random copolymers are more often found to be miscible
with other polymers than are homopolymers because of this. A thermodynamic treatment
of random copolymers is based on a mean field binary interaction model in which
interaction energies Bij (or P*) between monomer unit pair is considered as a sum of
terms the positive terms reflecting intermolecular interactions and negative terms
intramolecular interactions [148]. A copolymer composition mapping is useful in
establishing the compatibility windows. Formal theories have been proposed for dealing
with inductive effects but no theory of inductive effect itself has appeared and unit sequence
distribution is not yet considered.
A small amount of block copolymer is quite often used as compatibiliser in many polymer
blends for miscibility enhancement between immiscible polymer pairs, leading to increased
film transparency and improved mechanical properties. The transparency is related to
the presence of domains with dimensions less than 1000 . The considerable effect of
the copolymer on the reduction of domain sizes is clearly observable. The effect of such
copolymers depends on their composition and molecular weight as is easily remarked
from diagrams c, d of Figure 2.2 [124-126a] The block copolymers polystyrene
polyisoprene are much effective than PS-PMMA block copolymers in improving
compatibility of corresponding homopolymers.
According to the binary interaction model for a blends of two copolymers composed of
units 1 and 2 and 3 and 4, respectively, the net interaction energy parameter is written as
a sum of at least 6 terms [149]. The interaction parameter is considered a segmental
interaction one. Some information about them can be deduced from the experimental
miscibility maps. The theory was applied for copolymer blends based on
glycidylmethacrylate (GMA)/methylmethacrylate or GMA/S copolymers with S/AN
copolymers or with PPO and tetramethylcarbonate, [120] establishing the maps of
copolymer compositions where miscibility exists. These help to model the GMA use as a
functional monomer in reactive processing.
The microphase boundaries and ordered phase symmetries for diblock copolymer met in
the weak segregation regime were examined by mean field theory [150, 151],
incompressible field theoretical approaches [152] and by fluctuation theory [153-155]
within highly coarse grain. Theoretically has been predicted in a quantitative manner the
interfacial and emulsifying activity of block copolymers A-b-B in immiscible
homopolymers A and B. The authors [150-155] claimed that the compatibilising effect
depends not only on the ratio of molecular weight of block copolymer to that of
homopolymers but also on the composition f = Nc/N (where N is the chain length of
block copolymer and Nc is the chain length of block C) of the block copolymer. A
maximum compatibilising effect can be observed when the block copolymer has the
same composition as that of homopolymer blend.
49
Handbook of Polymer Blends and Composites
fAA(B) are the pair probabilities of AA, AB and BB in a single chain so that fA= fAA + fAB, etc.
The dependence of interaction parameter on composition was found for several systems
as isotopic blends (protonated and deuterated components), but for some systems such
as polyethylbutylene and polymethylbutylene, the dependence was not found [160] either
theoretically nor experimentally (by small angle neutron scattering; SANS).
The best method of estimating the phase state of polymer mixtures consists of determining
their phase diagrams. Unfortunately most studies deal with establishing only the region
of composition having a better compatibility or determination of properties of the blends.
The earlier investigations used very simple method as visually observation of the turbidity
of polymer mixture solutions or films [5, 161]. Other criteria for compatibility have
been developed based on dynamic mechanical measurements (DMA) and differential
scanning calorimetry (DSC). If a binary blend has only a single glass transition temperature
50
Thermodynamics of Multicomponent Polymer Systems
the system is considered compatible, depression of the melting point and the viscosity of
solutions and melts was also used to estimate compatibility. Heat effects of mixing have
been directly related to compatibility and later the role of mixing entropy has been also
established in determining compatibility as being due to the negative values of excess
volume of mixing, but positive values of excess volume of mixing have been also found
(PMMA/PVDF).
Many efforts have been dedicated to the estimation of the thermodynamic interaction
parameter AB using different methods as SANS, melting point depression, heat of mixing,
cloud points, light scattering, osmotic pressure, inverse gas chromatography, FT-IR, etc.
It has to be mentioned that the extrapolation of the results obtained in ternary polymer/
polymer/common solvent system for binary polymer/polymer system must be made very
carefully. To sustain this, it was supposed that in ternary systems two polymers do not
fully mix, i.e., the number of contacts between them is much smaller than it should be
for random mixing, in order to avoid the coil overlapping should lead to their repulsion
and to segregation of the two different macromolecular species.
H A VA 1 1
( 0 ) + ln B / N B (1 / N B 1 / N A ) = AB2A
RVB Tm , A TmA
where Vi and Hi are molar volume of monomeric unit of i-th component and enthalpy
of mixing per segment correspondingly, Tm and Tmo are melting temperature of sample
and of totally crystalline polymer. This equation does not include the dependence of AB
on concentration and temperature. This deficiency can be solved by using inverse gas-
chromatography, but the dependences obtained by this methods depend on the molecular
probe used. The Prigogine-Flory-Patterson theory corrects in part this shortcoming.
To minimise the free volume effects, polymers with similar Tg values should be selected.
During the past two decades, several groups have attempted to incorporate the dependence
of interaction parameter on concentration, pressure, molecular weight, etc., into predictive
microscopic theories as follows:
51
Handbook of Polymer Blends and Composites
equations of state for individual components but is necessary to demonstrate its application
to the mixtures [162-167].
Polymer Reference Interaction Site Model (PRISM) allows certain structural features of
polymers to be incorporated. The well-known Flory-Huggins theory has been remarkably
successful in describing many aspects of polymer mixing. This theory is known, however,
to suffer from several deficiencies that can be traceable: it requires that both components
to have the same volume, a mean field or random mixing approximation is made which
effectively ignores chain connectivity. It does not include packing effects and can not be
used to make quantitative molecular engineering calculations. Recently Curro and
Schweizer developed a new approach for treating these blends that is PRISM theory.
This is based on a continuous space description of a polymer liquid, which includes
chain connectivity and non-random mixing effects in a computational tractable manner
[168-174]. It is considered that each polymer has a certain structure factor that includes
details on polymer structure. By using of a series of coupled integral equations that take
into account the repulsive interactions while the attractive interactions are handled by a
perturbation theory. The equations-of-state obtained are inferior to those obtained by
other theories, but PRISM model has ability to consider polymer blends. Solubility
parameters can be calculated directly from the radial distribution functions, by integrating
the pair-pair energy function. It can be also applied to polymers with different architectures
as block copolymers, star polymers and ring polymers because subtle structural effects
can be systematically investigated both in one polymer component system or mixtures.
Applying the density functional theory thermally induced phase separation both UCST
(even in isotopic blends) and LCST behaviour has been explained [172].
Lattice Cluster Theory (LCT) is an extension of the Flory-Huggins lattice model, but it is
able to distinguish polymer structural details [175-179] because a superior solution is
developed. This includes contributions to thermodynamic properties from packing and
induced local correlations. In LCT, has an energetic and an entropic component and
can be developed in a compressible and an incompressible form. Chain stiffness is
introduced in model by using a bend or flex energy. The advantage of LCT is that it can
be used to investigate in a systematic way the detail effect of structure on polymer blend
miscibility. The differences in the miscibility of different polyolefin (PO) systems have
been explained using this model.
52
Thermodynamics of Multicomponent Polymer Systems
Other models and theoretical treatments proposed are able to describe multicomponent
polymer systems with arbitrary distribution of both polymerisation degree and
composition, based on variational principles [183], molecular dynamic simulation [184],
computer simulation using bond fluctuation model [185-187], surface enrichment [188].
The effects of non-local mixing entropy have been also taken into account [189]. It was
found that these effects are significant near the glass transition of the blend.
References
1. P.J. Flory, Principles of Polymer Chemistry, Cornell University Press, Ithaca, NY,
USA, 1953 and 1981.
3. A.J. Staverman, Recueil des Travaux Chimiques des Pays-Bas, 1941, 60, 76, 640.
4. I.C. Sanchez and M.T. Stone in Polymer Blends, Volume 1, Formulation, Eds., D.R.
Paul and C.B. Bucknall, John Wiley & Sons, New York, NY, USA, 2000, 16.
6. Polymer Blends, Eds., D.R. Paul and J.W. Marlow, Academic Press, New York,
USA, 1978.
8. A. Manson and L.H. Sperling, Polymer Blends and Composites, Plenum Press,
New York, NY, USA, 1976.
9. L.A. Utracki, Polymer Alloys and Blends, Hanser Publishers, Munich, Germany,
1989.
10. M.M. Coleman, J.E. Graf and P.C. Painter, Specific Interactions and the
Miscibility of Polymer Blends: Practical Guides for Predicting & Designing
Miscible Polymer Mixtures, Technomic Publishers Co. Inc., Lancaster, PA, USA,
1991.
53
Handbook of Polymer Blends and Composites
12. J.M.G. Cowie, and J.H. Harris, Polymer, 1992, 33, 4592.
14a. N.V. Kuleznev, Smesi Polymerov. Structura I Svoistva, Himia, Moscow, Russian
Federation, 1980.
18. F.E. Karasz and W.J. MacKnight in Multicomponent Polymer Materials, Eds.,
D.R. Paul and L.H. Sperling, American Chemical Society, Washington, DC, USA,
1986, 67.
19. D.R. Paul in Multicomponent Polymer Materials, Eds., D.R. Paul and L.H.
Sperling, American Chemical Society, Washington, DC, USA, 1986, 3.
20. R.J. Roe and D. Rigby, Advances in Polymer Science, 1987, 82, 103.
21. W.J. MacKnight and F.E. Karasz, Comprehensive Polymer Science, Volume 7,
Ed., G. Allen, Pergamon Press, New York, 1989, 111.
22. D.J. Walsh, Comprehensive Polymer Science, Volume 7, Ed., G. Allen, Pergamon
Press, New York, 1989, 35.
23. M.T. Rtzsch and H. Kehlen, The Frontiers of Macromolecular Science, Eds., T.
Saegysa, T. Higashimura and A. Abe, Blackwell Science Publishers, New York,
NY, USA,1989.
24. M.T. Rtzsch and H. Kehlen, Progress in Polymer Science, 1989, 14, 1.
54
Thermodynamics of Multicomponent Polymer Systems
26. V.P. Privalko and V.V. Novikov, The Science of Heterogeneous Polymers,
Structure and Thermophysical Properties, John Wiley and Sons, Chichester, 1995.
27. S.I. Kuchanov and S.V. Panyukov, Statistical thermodynamics of Copolymers and
their Blends, Chapter 13.4 in Comprehensive Polymer Science, Second
Supplement, Ed., G. Allen, S.L. Aggarwal and S. Russo, Pergamon Press, New
York, 1996, 441.
36. D. Patterson and G. Delmas, Discussions of the Faraday Society, 1970, 98.
39. K. Choi, W.H. Jo and S.L. Hsu, Macromolecules, 1998, 31, 1366.
40. P.J. Flory, R.A. Orwoll and A. Vrij, Journal of the American Chemical Society,
1964, 86, 3507.
41. P.J. Flory, R.A. Orwoll and A. Vrij, Journal of the American Chemical Society,
1964, 86, 3515.
42. P.J. Flory, R.A. Orwoll and A. Vrij, Journal of the American Chemical Society,
1965, 87, 1833.
55
Handbook of Polymer Blends and Composites
44. P.J. Flory, Journal of the American Chemical Society, 1965, 87, 1833.
45. B.E. Eichinger and P.J. Flory, Transactions of the Faraday Society, 1968, 64, 2035
46. I.C. Sanchez and R.H. Lacombe, Journal of Physical Chemistry, 1976, 80, 2352.
47. I.C. Sanchez and R.H. Lacombe, Journal of Physical Chemistry, 1976, 80, 2568.
48. I.C. Sanchez and R.H. Lacombe, Journal of Polymer Science, Polymer Letters
Edition, 1977, 15, 2, 71.
49. I.C. Sanchez and R.H. Lacombe, Macromolecules, 1978, 11, 6, 1145.
56. R.L. Scott, Journal of Chemical Physics, 1949, 17, 268, 279.
62. R. Koningsveld, L.A. Kleintjens and M.H. Onclin, Journal of Polymer Science,
Part B, 1980, 18, 363.
56
Thermodynamics of Multicomponent Polymer Systems
66. Y. Hu, X. Yang, D.T. Wu and J.M. Prausnitz, Macromolecules, 1993, 26, 6817.
73. D.R. Paul and J.W. Barlow, Polymer, 1984, 25, 487.
74. R.P. Kambour, J.T. Bendler and R.C. Bopp, Macromolecules, 1983, 16, 763.
75. G. ten Brinke and F.E. Karasz, Macromolecules, 1984, 17, 815.
76. H.J. Cantow and O. Schulz, Polymer Bulletin, 1986, 15, 449.
77. H.J. Cantow and O. Schulz, Polymer Bulletin, 1986, 15, 539.
78. P. Munk, P. Hattam and Q. Du, Journal of Applied Polymer Science, Applied
Polymer Symposium, 1989, 43, 373.
80. R. Koningsveld and W.J. MacKnight, Polymer International, 1997, 44, 356.
82. F. Scheffold, E. Eiser, A. Budkowski, U. Steiner, J. Klein and L.J. Fetters, Journal
of Chemical Physics, 1996, 104, 8786.
57
Handbook of Polymer Blends and Composites
83. J.H. Hildebrand and R.L. Scott, The Solubility of Non-Electrolytes, Dover
Publishers, New York, 1964.
87. D.J. David, Proceedings of Antec 95, Boston, MA, USA, Volume 2, 2748.
88. V.N.S. Pendyala and S.F. Xavier, Polymer, 1997, 38, 3565.
91. A.F. Barton, CRC Handbook of Solubility Parameter and Other Cohesive
Parameters, Second Edition, Boca Raton, FL, USA, 1991, Chapter 6, 157.
92. W.P. Hsu, A.S. Myerson and T.K. Kwei, International Journal of Polymeric
Materials, 1998, 40, 81.
93. G.J. Pehlert, X. Yang, P.C. Painter and M.M. Coleman, Polymer, 1996, 37, 4763.
94. M.C. Pascu, Gh. Popa, M. Grigoras and C. Vasile, Synthetic Polymer Journal,
1998, 2, 158.
95. E. Espi, M.J. Fernandez and J.J. Iruin, Polymer Engineering and Science, 1994,
34, 1314.
96. A. Wakker and M.A. van Dijk, Polymer Networks and Blends, 1992, 2, 123.
97. D.J. David and T.F. Sincock, Polymer, 1992, 33, 4505.
98. D.R. Paul and J.W. Barlow, Polymer, 1984, 25, 487.
58
Thermodynamics of Multicomponent Polymer Systems
101. G.V. Nelson, S.H. Jacobson and D. J. Gordon, Chemical Design Automation
News, 1992, 7, 39.
102. S.H. Jacobson, D.J. Gordon, G.V. Nelson and A.C. Balazs, Advanced Materials,
1992, 3, 198.
106. T.A. Callaghan and D.R. Paul, Macromolecules, 1993, 26, 2439.
107. T.A. Callaghan and D.R. Paul, Polymer, 1993, 34, 3796.
108. C.K. Kim and D.R. Paul, Polymer, 1992, 33, 1630.
109. C.K. Kim and D.R. Paul, Polymer, 1992, 33, 2089.
110. J.J. Gyu and B. Y. Chan, Journal of Applied Polymer Science, 1998, 70, 1143.
111. K.S. Jeon, K. Char and E. Kim, Polymer Journal, 2000, 32, 1.
115. S.C. Chiu, T.K. Kwei and E.M. Pearce, Journal of Thermal Analysis and
Calorimetry, 2000, 59, 71.
117. J. Cho and I.C. Sanchez, in Polymer Handbook, Fourth Edition, Eds., J.
Brandrup, E.H. Immergut and E.A. Grulke, John Wiley, New York, NY, USA,
1999, VI/591.
59
Handbook of Polymer Blends and Composites
121. Ch. Huglin and A. Dondos, Die Makromolecular Chemie, 1968, 126, 206.
129. R. Kuhn and H.J. Cantow, Die Makromolekulare Chemie, 1969, 122, 65.
130. M.G. Prolongo, R.M. Masegosa and A. Horta, Macromolecules, 1989, 22, 4346.
133. C. Huang, M.O. de la Cruz and P.W. Voorhees, Acta Materialia, 1999, 47, 4449.
134. R. Vukovic, V. Kuresevic, F.E. Segudovic, F.E. Karasz and W.J. MacKnight,
Journal of Applied Polymer Science, 1983, 28, 1379.
135. R. Vukovic, V. Kuresevic, F.E. Segudovic, F.E. Karasz and W.J. MacKnight,
Journal of Applied Polymer Science, 1983, 28, 3079.
136. G. ten Brinke, F.E. Karasz and W.J. MacKnight, Macromolecules, 1983, 16,
1827.
60
Thermodynamics of Multicomponent Polymer Systems
137. G. ten Brinke, F.E. Karasz and W.J. MacKnight, Macromolecules, 1992, 25,
4716.
138. G. ten Brinke, F.E. Karasz and W.J. MacKnight, Macromolecules, 1986, 19,
2274.
139. A.C. Balazs, I.C. Sanchez, I.B. Epstein, F.E. Karasz and W.J. MacKnight,
Macromolecules, 1985, 18, 2188.
140. A.C. Balazs, F.E. Karasz, W.J. MacKnight, H. Ueda and I.C. Sanchez,
Macromolecules, 1985, 18, 2784.
141. J.V. Hunsel, A.C. Balazs, W.J. MacKnight and R. Koningsveld, Macromolecules,
1988, 21, 1528.
142. A.C. Balazs and M.T. Meuse, Macromolecules, 1989, 22, 4260.
143. T. Shiomi, H. Ishimatsu, T. Eguchi and K. Imai, Macromolecules, 1990, 23, 4970.
146. H.W. Kammer, T. Inoue and T. Ougizawa, Polymer, 1989, 30, 888.
148. D.R. Paul, Pure & Applied Chemistry, 1995, 67, 977.
149. P.P. Gan and D.R. Paul, Polymer, 1994, 35, 3513.
153. G.H. Fredrickson and E. Helfand, Journal of Chemical Physics, 1987, 87, 697.
154. G.H. Fredrickson and E. Helfand, Journal of Chemical Physics, 1987, 87, 124.
155. S.A. Brazovskii, Soviet Physics JETP - USSR, 1975, 41, 85.
156. S.H. Kim, W.H. Jo and J. Kim, Macromolecules, 1997, 30, 3910.
61
Handbook of Polymer Blends and Composites
157. S.H. Kim, W.H. Jo and J. Kim, Macromolecules, 1996, 29, 6933.
160. C.C. Lin, S.V. Jonnalagadda, N.P. Balsara, C.C. Han and R. Krishnamoorti,
Macromolecules, 1996, 29, 661.
162. S.H. Huang and M. Radosz, Industrial Engineering Chemistry: Product Research
and Development, 1990, 29, 2284.
163. S.H. Huang and M. Radosz, Industrial Engineering Chemistry: Product Research
and Development, 1991, 30, 1996.
164. S.J. Han, C.J. Gregg and M. Radosz, Industrial Engineering Chemistry: Product
Research and Development, 1997, 36, 5520 45.
165. T. Kraska and K.E. Gubbin, Industrial Engineering Chemistry: Product Research
and Development, 1996, 35, 4727.
167. D.Ghonasgi and W.G. Chapman, Journal of Chemical Physics, 1994, 100, 6633.
168. K.S. Schweiser and J.G. Curro, Advances in Chemical Physics, 1997, 97, 1.
169. A. Yethiraj, J.G. Curro, K.S. Schweizer, J.D. McCoy and K.G. Honnell,
Macromolecules, 1993, 26, 2655.
170. D.S. Pope, I.C. Sanchez, W.J. Koros and G.K. Fleming, Macromolecules, 1991
24, 1779.
171. E.F. David and K.S. Schwetizer, Macromolecules, 1997, 30, 5118.
172. S.K. Nath, J.D. McCoy, J.G. Curro and R.S. Saunders, Journal of Polymer
Science, Part B, Polymer Physics, 1995, 33, 2306.
173. J.G. Curro, Proceedings of Antec 94, San Francisco, CA, USA, 1994, Volume II,
2083.
174. P.A. Tillman, D.R. Rottach, J.D. McCoy, S.J. Plimpton and J.G. Curro, Journal of
Chemical Physics, 1997, 107, 4024.
62
Thermodynamics of Multicomponent Polymer Systems
175. K.F. Freed and J. Dudowicz, Modern Trends in Polymer Science, 1995, 3, 248.
176. K.F. Freed and J. Dudowicz, Macromolecular Symposia, 1994, 78, 29.
178. K.F. Freed, J. Dudowicz and K.W. Forman, Journal of Chemical Physics, 1998,
108, 7881.
179. K. W. Foreman and K.F. Freed, Advances in Chemical Physics, 1998, 103, 335.
180. Y. Song, S.M. Lambert and J.M. Prausnitz, Macromolecules, 1994, 27, 441.
181. T. Hino, Y. Song, and J.M. Prausnitz, Macromolecules, 1995, 28, 5717
182. T. Hino, Y. Song, and J.M. Prausnitz, Macromolecules, 1995, 28, 5725.
183. S.I. Kuchanov and S.V. Panyukov, Journal of Polymer Science, Part B Polymer
Physics, 1998, 36, 937.
184. P.A. Tillman, D.R. Rottach, J.D. McCoy, S.J. Plimpton and J.G. Curro, Journal of
Chemical Physics, 1998, 109, 806.
186. M. Muller and K. Binder, Journal of Computer Aided Material Design, 1998, 4,
137.
187. M. Muller and M. Schick, Journal of Chemical Physics, 1996, 105, 8282.
188. I.G. Kokkinos and M.K. Kosmas, Macromolecules, 1997, 30, 577.
189. I. Ya. Erukhimovich, A.R. Khokhlov, T.A. Vilgis, A. Ramzi and F. Boue,
Computational and Theoretical Polymer Science, 1998, 8, 133.
63
Phase Behaviour
3
Phase Behaviour
Cornelia Vasile
3.1 Introduction
In technical applications one often has to handle multicomponent mixtures and the
accurate knowledge of their phase separation is essential. In most cases of practical interest,
polymers are mixed with many other components, as during the synthesis of copolymers
(two monomers, a solvent, a polymer) or in membrane forming systems, numerous
additives are also required in the processing of various materials. With all applications a
detailed theoretical understanding of the mixing behaviour of such systems is mandatory.
The properties of polymer blends are intimately tied to their morphology, which depends
on the miscibility of the components and the mechanism and kinetics of phase separation.
For some applications, phase-separated morphologies are required for improving
mechanical properties such as impact toughness, while in other instances, a miscible
blend is desired, e.g., for improving processability or extending an expensive material
with a cheaper one. For either a single phase or two-phase blend, the development of the
optimum morphology and properties requires knowledge of the phase diagram and the
characteristics of the phase separation process [1-7].
To determine miscibility, isobaric phase diagrams are usually constructed. Figures 1.1
and 3.1 are represented typical curves for binary polymer systems [1-7]. The coexistence
curves, binodals, separate the region of complete miscibility from that of partial miscibility.
At a given temperature any point situated above the curves with lower critical solution
temperature (LCST) and below curves with upper critical solution temperature (UCST)
(points C and D) corresponds to the concentration and temperatures of an homogeneous
system. Points in the region beneath the curves, e.g., E and F, correspond to a
microheterogeneous two-phase system which separates in time into two clear layers in
equilibrium with each other. The composition of layers can be found by drawing an
65
Handbook of Polymer Blends and Composites
isotherm at each temperature. It will intersect the coexistence curve at two points, the
abscissas of which correspond to the volume fractions of the second component in each
layer. The connecting line is called the tie line. Raising the temperature changes the
concentrations of the components in both layers, bringing them closer together, but each
temperature corresponds to a definite solubility of one component in the other. At the
critical miscibility temperature, the composition of both phases becomes the same, the
interface between them disappears, and an homogeneous solution results. When critical
state is reached, the compositions of the two coexisting phases have became identical
and the tie-line has changed into a point, the critical point.
In most polymer systems the miscibility of the components improves with decreasing
temperature and phase separation occurs by heating and they exhibit a LCST. This
temperature corresponds to a minimum on the coexistence curve. There are two types of
LCST. The first type of LCST, which is more common one, is characteristic of systems in
which strong hydrogen bonds form between the component molecules, promoting mixing
of components. By heating these bonds are broken by thermal motion, worsening the
miscibility of the components.
The second type of LCST is observed in systems whose components are close in chemical
constitution, but differ in the size of their molecules [8].
If for a binary system, the plot of free energy of mixing (Gm) is concave upward
throughout the composition range Figure 1.1a bottom, then the two components are
miscible in all proportions. This is because any point on this curve (Q) has lower free
energy than any two phase system of the same overall composition (Q*). Contrary, the
lower free energy is obtained by phase separation in the case illustrated in Figure 1.1b. In
respect to the positions of the mixture with composition they will be metastable against
separation into phases in the spinodal region and non stable against separation for
composition situated on concave downwards that means under the bimodal curve.
When an uniform mother solution is cooled very slowly, a second phase begins to separate
at a definite temperature which is called the cloud point temperature or the cloud point.
In the homogeneous area, the system is stable, the free energy (Gm) is minimal and can
only be increased by a demixing process. In the heterogeneous area the overall Gibbs
energy of demixed system consisting of the coexistence phases is less than the Gm of the
homogeneous system. This area breaks into two domains. One is called metastable, here
the beginning of phase separation leads to an increase of Gm. Therefore the phase
separation cannot proceeds spontaneously, it is bound to nucleation and growth. The
demixed system consists of a continuous matrix-phase and a non-continuous drop-phase.
66
Phase Behaviour
The second domain located within the two phase area are only touching the homogeneous
region at the critical point is called unstable domain, Gm is decreased during the whole
demixing process. If the phase separation takes place under these conditions, the well-
known spinodal pattern with two co-continuous phases is observed.
Binodals and spinodals provide important information because together with the
characteristics of the phase state, they give data for analysis of the mechanism of phase
separation.
A binodal curve indicates the limits of miscibility. They are sometimes (incorrectly)
identified with the experimental cloud-point curves. Binodals are computed from the
equality of the chemical potential of polymers in two coexisting phases, the general
equation for a binodal curve is:
i = n i (G m / n i ) V , T , n j (3.1)
( A )' ( Ao )' = ( A )'' ( Ao )'' and ( B)' ( Bo )' = ( B)'' ( Bo )'' (3.2)
where:
where i - io are chemical potentials of the i-th component of the mixture, rA and rB are
the numbers of segments (or polymerisation degree) of polymer A and B, ri = Vi /Vr (Vi is
molar volume of component i) A and B are volume fractions of components. In the
general expression (Equation 3.1), n i designates various modes of concentration
evaluations (volume fraction is the next). AB is a free energy interaction parameter
calculated from Equations 2.3-2.8 and 2.11 depending on the type of polymer blend
studied, practical or theoretical purpose and selected theory for study.
Using a computer it is preferable to use a more direct method in which the Gm function
itself is used. The coexisting polymer volume fractions are found by means of numerical
iterative construction of the double tangent to G () curves at different values [1-4]
Normally the calculation of the tie lines connecting the coexisting phases at constant
temperature and pressure is carried out by numerically solving the equations describing
the equality of the chemical potentials i, which is connecting with the Gibbs energy and
the derivatives of the Gibbs energy with respect to the composition variables.
67
Handbook of Polymer Blends and Composites
The region of the metastable states, located between the cloud-point curve and the
spinodal, is of special significance for polymer systems compared to low-molecular ones.
This is because the phase separation in the polymer system proceeds much more slowly
than in low-molecular systems. Consequently, the lifetime of the metastable states may
often far exceed the duration of a thermodynamic experiment and even the polymer
operation time. In such cases, which are typical of concentrated solutions, melts and
blends of polymers, a metastable system is expected to exhibit an equilibrium behaviour.
A more complicated situation is encountered where more than two phases coexist.
For a complete description of the phase diagram, the spinodals and the critical point
must be calculated from the second and third derivatives, respectively:
2 G 3 G
( ) P , T = ( ) P ,T = 0 (3.5)
3A 3A
The spinodal line is given by the equation containing the second derivatives of G. Only
for very simple theoretical equations for G, can the derivatives be easily calculated.
Very complicated dependencies of G on the composition result in the case of polymer
blends and copolymers. The calculation is more complicated for sheared polymer blends,
where G is extended by a term which contains implicit equations and describes the
energy stored in the flowing system, where the analytical representation of the derivatives
is impossible. The extension of ternary blends needs a fitting function [9].
To calculate the phase diagram on the basis of these equations one must know the
concentration (i volume fraction), temperature T and pressure P.
The spinodal condition for binary mixture deduced by lattice fluid theory (LFT) is given
by (see also Equation 2.12):
d2G 1 1 1 *
P * v * 2 TP
= ( + )
( + )=0 (3.6)
d A 2 rA A rB B
2
kT 2
68
Phase Behaviour
1 1
sp = 0.5[ sp + (1 sp )] (3.7)
rA rB
and from LFT for the simplest case of the identical characteristic volumes of components:
1 1
sp = 0.5[ sp + (1 sp )]/ T 2 P * A (3.8)
rA rB
The critical point is the point where the bimodal and spinodal curves are cotangent:
The corresponding values for critical values of concentration and interaction parameter are:
1
( A ) crit = 1/ 2
and ( AB) crit = (1 / 2)[rA1 / 2 + rB1 / 2 ] 0.5 (3.10)
[1 + (rB / rB) ]
The FH theory leads to a reasonable prediction of critical values, but there are significant
deviations between predictions and experimental measurements of other quantities. A
better fitting of calculated and experimental bimodal and spinodal curves has been
obtained by assuming a dependence of AB on concentration and temperature.
According to Lipatov and Nesterov, the phase behaviour all the polymer mixtures can be
classified in five groups [17]:
69
Handbook of Polymer Blends and Composites
1. Systems with specific interactions between components which are miscible blends
having only LCST [Examples: oligodimethylsiloxanes/oligomethylphenylsiloxanes;
poly(aryl ether ether ketone/poly(aryl ethersulfone); polybenzimidazole/polyimides;
1,4-polybutadiene/cis-1,4 polyisoprene; chlorinated polybutadiene/chlorinated
polyethylene (PE); poly(2,6-dimethyl-1,4-phenylene oxide/deuterated polystyrene or
polystyrene ionomer; brominated poly(2,4-dimethyl-1,4-phenylene oxide) (50-
76%Br)/polystyrene (PS) or polymethylmethacrylate (PMMA) or atactic
PMMA(aPMMA); poly(hydroxyether of bisphenol A)/poly--caprolactone;
polyepichlorohydrine/ polymethacrylate (PMA) or polyethylacrylate (PEA), poly-n-
propyl acrylate, PMMA, poly(ethylmethacrylate), poly(n-butyl methacrylate), poly(N-
vinyl-2-pyrrolidone), polyethersulfone/poly(ether imide); poly(ethylene oxide)/
poly(epichlorohydrin) or poly(D)(-)-(3-hydroxybutyrate), or PMMA; PMMA/poly
(-methyl styrene), or chlorinated polypropylene, or poly(vinylchloroacetate);
poly(1,1,1,3,3,3,hexafluoroisopropyl methacrylate)/PMMA or poly methyl ethyl
acrylate (PMEA), poly(tetrahydrofurfurylmethacrylate), or poly(2-fluoroethyl
methacrylate); poly(1-chloroethylmethacrylate)/PMMA, polyethyl methacrylate
(PEMA), poly n-propylmethacrylate (Pn-PMA), polyisopropyl methacrylate (PiPMA),
poly n-butyl-MA (BMA), poly(tetrahydrofurfurylmethacrylate), poly(3-
chloroethylmethacrylate)/poly(tetrahydrofurfurylmethacrylate, poly(2-
iodoethylmethacrylate)/poly(butylene adipate), or poly(2,2dimethyl-1,1,3-propylene
adipate, or poly(hexamethylene sebacate), poly--caprolactone); poly(2,2-
dichloroethylmethacrylate)/PEMA, or polybutyl methacrylate (PBMA), Poly-n-PMA,
PiPMA, poly isoamylmethacrylate); poly(2,2,2,-trichloroethylmethacrylate)/PMMA,
PEMA, Pn-PMA, PiPMA, poly(tetrahydrofurfurylmethacrylate); polyvinyl acetate
(PVAc)/cyclic oligomer bisphenol A polycarbonate, poly(2-bromoethylmethacrylate),
poly(2-chloroethylmethacrylate), PMMA; polyvinyl alcohol (PVA), poly(acrylic acid),
PVAc, polyvinylchloride (PVC)/polybutyl acrylate (PBAc), PBMA PMMA; chlorinated
PVC/poly--caprolactone, PMMA, polyurethane, atactic PMMA; poly(vinyl methyl
ether)/poly(2-chlorostyrene), PEA, polypropyl acrylate, PBAc, poly(hydroxyether of
bisphenol A); polyvinylidene fluoride/PEA, PEMA; poly(vinylfluoride)/poly(vinyl
phenol), poly(p-phenylene benzothiazole), polyamide 66 (PA 66); PS/poly(2-
chlorostyrene); deuterated PS/poly(vinyl methyl ether; (Li, Mg, Zn, Mn)-sulfonated
PS ionomers with sulfonation degree 6.5-9/PA 66; or polypropylene oxide (PPO), PS
containing (1,1,1,3,3,3,-hexafluoro-2-hydroxyisopropyl)--methyl styrene)/PBMA,
biodegradable poly(D,L-lactide)/poly(-caprolactone)] [18].
2. Oligomer or polymer mixtures having UCST and both LCST and UCST [Examples:
hydrogenated oligocyclobutadiene/isotactic PP or isotactic polybutene; poly(ethylene
glycol)/poly(propylene glycol); poly(vinylethylene)/cis-1,4-polyisoprene or
poly(ethylethylene); poly(vinylidene fluoride)/PMMA].
70
Phase Behaviour
3. Systems composed of chemically similar (or non-similar) polymers may have both
LCST and UCST [Examples: polybutadiene (PB)/deuterated polybutadiene or
polyisoprene (Mw = 35 500); protonated polybutadiene/deuterated polybutadiene;
brominated poly(2,4-dimethyl-1,4-phenylene oxide (50-76%Br)/syndiotactic PMMA;
PMMA/deuterated PMMA; poly(vinyl methyl ether)/polyisophthalamide (2-
10 methylene groups); PS/poly(-methylstyrene)].
Figure 3.1 Schematic representation of four types of phase diagrams. a) UCST < LCST,
the region of complete miscibility lies between curves. The higher is molecular weight
of components the narrow will be this region; b) UCST > LCST the region of complete
miscibility is found outside the curve; c) No unlimited miscibility region hourglass
type diagram; d) complex phase diagram
71
Handbook of Polymer Blends and Composites
Only from this summary it is clear that the most common phase diagram type for polymer
mixtures is that with LCST, it being characteristic to the majority of miscible polymer blends.
With increasing temperature, the interactions are weakened, causing LCST behaviour,
while the noncombinatorial entropy contribution becomes more and more dominant,
leading eventually again to phase stability of the blend or UCST behaviour. This phase
behaviour is experimentally easily accessible only for blends with component having
very different polymerisation degree. The parameter has to be negative at low and high
temperatures and positive in a temperature range between see Figure 2.1.
The shape of the phase diagram as well as the width of the miscibility window between
the UCST and LCST will be critical affected by the molecular characteristics of
components. The phase diagrams constructed according spinodal equation of LFT
(Equation 3.6) may change the shape from the simplest one Figure 3.1a to closed
loops Figure 3.1b, hourglass Figure 3.1c, etc depending on the relative mismatch between
characteristic parameters of components.
72
Phase Behaviour
in the literature both from a theoretical and practical perspective. Interest in these materials
stems, in part, from their improved mechanical and processing properties and, in part,
from the current environmental focus on plastics recycling. The novel polymers of
controlled molecular weight, controlled stereoregularity, and comonomer composition,
distribution, are being produced from metallocene catalysts, leading to polymers with
superior mechanical properties as compared to those produced by conventional catalysts.
In this context, blending becomes an invaluable tool for molecularly-engineering resins
with desired set of properties and ease of processing. Due to the interest in these blends,
especially HDPE/IPP and IPP/ethylene propylene rubber (EPR), the theoretical explanation
is necessary to evidence the factors that determine their miscibility/immiscibility (see
Section 3.3).
In qualitative terms, the major factor ensuring stability of a single phase state, i.e., a
negative value of the excess enthalpy will be favoured both by a negative value of the
interaction parameter as well as by higher than unity ratios of the characteristic
parameters v*A/v*B and T*A/T*B.
However, the calculation of phase diagrams for mixtures of more than three components
is very difficult. The calculation becomes more complicated when the dependence of
interaction parameter on composition and temperature are taken into account. Therefore,
a simple method is required with which the phase diagram can be calculated: tie lines
and bimodal, spinodal, critical points and their stability. A simple iteration process for
the determination of the composition of the coexisting phase was proposed by Horst
and Wolf [27, 28].
3.3.1 Structure
The structure of components is the major factor determining phase behaviour. Structural
differences determine the compatibility of somehow similar blends. If PVC is miscible
with polylactones, they are not miscible neither poly(vinylfluoride) nor with
poly(vinylidene fluoride) (PVDF), while poly(vinyl bromide) exhibits an improved
compatibility with polylactone, the behaviour being explained by differences in
electronegativity. PVDF is miscible with many poly(alkyl acrylates) and poly(alkyl
methacrylates).
During the last years many reactive oligomers have been developed, such as oligoesters,
oligoethers and oligoesteracrylates, and these are widely used as mixtures in network
73
Handbook of Polymer Blends and Composites
production by curing. The strong interaction between these oligomers leads to formation
of associates in reaction medium that will influence the reaction mechanism and structure
of cured polymer. Knowledge of thermodynamic behaviour of such systems is useful in
process control and for understanding the morphology and properties of resulting
materials. Examples are the PEG/PPG, polyethylene glycol adipate/PPG mixtures. The
composition and temperature dependence of AB evidences a behaviour with both UCST
and LCST. The appearance of the LCST in such systems can result due to the weakening
of hydrogen bonds, that is induced by increase in temperature. The minimum evidenced
in the dependence of AB on temperature and its negative value show that this system
demixes on heating.
Besides hydrogen bonds, the miscibility of binary polymer mixtures may depend on ion-
ion, ion-dipole and other favourable interactions.
where kB is the Boltzman constant and crit is volume fraction at critical point.
The evaluation of crit for block copolymers showed that this is invariably larger than
that for binary polymer blends of the same composition. In the other words, chain
connectivity makes phase separation more difficult in block copolymers than in the
corresponding blends, although in other respects both systems exhibit similar
characteristics.
74
Phase Behaviour
A key parameter controlling such morphologies is the number of segments of the block
chains. Also the microdomain morphology of block copolymer systems is determined by
the shape of the interface between two phases, i.e., the local curvature of the interface
(local aspect) and the topological continuity of the interface in space (global aspect).
Curved surfaces can be classified into three groups, namely, elliptic, parabolic and
hyperbolic surfaces. The hyperbolic surfaces are the ones that form new, complex,
bicontinuous microdomain morphologies. The size of the grain, the coherent region of
75
Handbook of Polymer Blends and Composites
the microdomains, depends on the conditions of sample preparation and the segregation
power of the constituent polymers in the block copolymers. Three-dimensional continuity
of each microphase at the grain boundary is especially important for cylindrical and
lamellar morphologies because it is expected to affect significantly the rheological and
diffusion properties. The grain boundary is merely a surface with thickness of the order
of the repeat distance of the microdomains. Other kinds of morphologies are shown for
copolymer/homopolymer and copolymer/copolymer mixtures. Sometimes, microphase
separation can be followed by a macrophase separation or inversely microphase separation
follows after macrophase separation.
Since microdomain structures of block copolymers have their repeat distances in the
order of 10 nm, small angle scattering of X-rays (SAXS) or of neutrons (SANS) and
transmission or scanning electron microscopy (TEM and SEM) of ultrafine sections are
the major techniques to characterise the structures. Scattering gives information about
size and symmetry of the structure, whilst microscopy gives information limited to the
local area but related to the structure entity. A mesh structure was shown for a
70/30 wt%/wt% mixture of styrene-butadiene copolymer and PS homopolymer, which
consists of alternating parallel sheets of the phase composed of PS block chains and PS in
the matrix of PB phase, and the sheets of the major component (PS block or PS
homopolymer) are interconnected by catenoidal channels of the same component
traversing through the sheets of the minor component (PB).
Therefore, melts and solutions of block copolymers as well as blends containing their
macromolecules often form thermodynamically stable mesomorphic states, characterised
by an inhomogeneous density distribution on monomer units in space. In the vicinity of
critical points, these mesomorphic states consist of blocks localised within the
microdomains periodically distributed in space. Microdomain formation attracted much
interest both from the theoretical and experimental standpoint [32, 33].
Two extreme limit regimes of the formation of spatially periodic structures are generally
considered. Strong Segregation Regime (SSR) and Weak Segregation Regime (WSR). In
the first regime (SSR) the different type of units of a binary block copolymer present a
strong incompatibility. Microdomains consist mainly of units of the same type, separated
by interfaces that are very thin in comparison with their size. Such systems are situated
far from the critical points of their phase diagrams. The copolymer chains are strongly
extended and gyration radius exceeds that of an unperturbed Gaussian coil.
In the second regime (WSR), the systems is close to critical point and the differences in
concentrations of monomer units inside the domains are small, so no sharp interfaces
exist between them. Under such a regime the size of macromolecule is only slightly
perturbed by volume interactions.
76
Phase Behaviour
For interpenetrating polymer networks (IPN), vapour absorption data and morphological
studies revealed that the overwhelming majority of both simultaneous IPN and sequential
IPN are phase separated systems. The expression of the interaction parameter contains
term depending on network density and crosslink functionality, and in most cases free
energy of mixing was found positive [17].
3.3.2 Tacticity
The miscibility of two polymers depends also on the stereoregularity. Isotactic PMMA
(iPMMA) is not miscible with PVC in the whole range of compositions whereas
syndiotactic PMMA forms miscible mixture when the content of syndiotactic PMMA
does not exceed 60%. Polyethylene oxide (PEO) shows miscibility with atactic and
syndiotactic PMMA and is not miscible with isotactic PMMA. Other examples are cited
in [17]. This is due to the differences in interaction parameters. For example the interaction
parameters for the mixtures of PMMA with polyvinylidene fluoride (PVDF) are equal to
0.13; 0.1 and 0.06 for isotactic, atactic and syndiotactic PMMA, respectively.
3.3.3 Branching
The effect of branching can be attributed to exclusion of some part of the branch from
the interaction with another polymer. In the polyvinylmethyl ether (PVME) in the mixture
with PS the shift of the cloud point curves to the lower temperatures was explained by
kinetic effects of phase separation due to a greater number of entanglements of the chains
as compared with usual branched macromolecules. Mehta and Honnell [36] used an
atomistic skeletal united-atom model to represent HDPE, IPP and EPR and, on the basis
of generalised Florys approach derived an accurate equation-of-state with an
compressibility term and concluded that if the phase separation occurs in the melt, it is
attributable only to architectural differences. The branching destabilises these mixtures
inducing phase separation.
With the increase in molecular weight of components, the composition range over which the
blend is miscible decreases Figure 3.1a. Below a critical molecular weight of ~14,000 g/mol
of polyolefin, the mixtures exist as single phase over the entire composition range [36].
77
Handbook of Polymer Blends and Composites
So they explained practical results obtained with commercial polyolefin blends. This
theory can be extended to -olefin comonomers (1-butene, 1-hexene or 1-octane) in linear
low density PE (LLDPE).
In the PS/PS or PEG/PEG systems, the components differing by average molecular mass,
and consequently by thermal expansion coefficient and having close values of the solubility
parameters, there should observed both LCST and UCST and a bimodal concentration
dependence of AB.
Figure 3.3 schematically shows some examples of phase diagrams with various gaps of
miscibility of binary polymer systems.
McMaster theoretically assessed that when the molecular weight increased, the mutual
solubility of the two polymers decreases. Even at molecular weights as low as 50,000,
differences in thermal expansion coefficients as small as 10% can lead to virtually complete
immiscibility. Small positive values of the interaction energy parameter lead to both
UCST and LCST behaviour. The two critical points have already merged to give skewed
hourglass shaped bimodal and spinodal curves. The temperature where the LCST and
UCST merge is a function of the magnitude difference of the component thermal expansion
coefficients. The simultaneously occurrence of an LCST and a UCST should be rare for
polymer/polymer systems.
3.3.5 Pressure
In mixtures PEA/PVDF exhibiting LCST, the cloud points decrease with increase of
pressure, reach the minimum, and increase again, if the content of the second component
does not exceed 30%. At 40% PVDF, only monotonous increase of cloud point takes
place. LCST increases with increasing pressure in the entire composition range for the
mixtures of copolymers poly(2,6-dimethyl-1,4-phenylene oxide)-poly(styrene-co-
fluorostyrene). The UCST of the oligomeric mixtures of styrene (S)/butadiene (B) and B/
propylene (P), increases with increasing pressure. The effects can be described by the
modified theory of the equation-of-state. As a rule the increase in pressure increases the
temperature of phase separation for the systems with LCST. Mixtures with UCST and
negative heat of mixing increase the miscibility region along with increase in pressure,
whereas for positive heats of mixing this region narrows [38]. Mixtures with LCST and
78
Phase Behaviour
with high molecular mass increases their miscibility with pressure. Both LCST and spinodal
temperature increased with pressure [39, 40].
Figure 3.3 Effects of various factors on the shape of the phase diagrams. The curves
represent the following effects: the thermal expansion coefficient, molecular weight,
and interaction parameter (------------) (1) M1 = M2 = 30 000, 2 = 0.640 x 10-3 K-1,
relative difference of thermal expansion coefficients 10.3%; (2) M1 = M2 = 50 000, 2
= 0.630 x 10-3 K-1, relative difference of thermal expansion coefficients 8.6%; of
molecular weight. ( ) (3) M1 = 30 000; (4) M1 = 80 000; ( ) of interaction
energy parameter = (5) AB 0.01; (6) AB 0.05; (7) AB 0.100
79
Handbook of Polymer Blends and Composites
The casting solutions from common solvent is a very useful method for study and obtaining
polymer blends. Miscible blends in solvent with great difference in the interaction
parameters between each polymer and solvent give narrow region of immiscibility see
Figure 2.2a. Such systems may be two-phase mixtures because during evaporation the
system passes through the region of immiscibility and increase in viscosity causes the
two phase structure of the mixtures to be frozen. In such cases a false UCST can be
detected. Rapid evaporation may lead to the formation of a pseudo-homogeneous system
with false LCST. Typical cases are PMMA/polycarbonate (PC), PS/PMMA and PMMA/
styrene-acrylonitrile (SAN) copolymer blends. The effect of apparent dissolution in the
two-phase region is very important for determining the LCST for films dried from solutions
and having two-phase structure. Thermal treatment also has an important effect. The
pseudo-homogeneous, non-equilibrium systems are more frequently encountered when
system undergoes phase separation with a very low rate. Slow phase separation by heating
may lead to the overestimation of the miscibility region, especially when LCST is situated
near Tg. The effect of low rate of phase separation on the estimation of the cloud point
curves was investigated in detail for the mixtures of bisphenol A polycarbonate with
PMMA [41]. The annealing allows thermodynamic equilibrium to be reached.
Incorporation of a filler changes the shape of phase diagram due to the formation of the
boundary surface layer near the filler interface, with a different composition than bulk
one. As an example incorporation of even 10 wt% filler in PS/PBMA mixture decreases
the thermodynamic stability and for the PMMA/PVAc mixture filled with 10 wt% silanised
fumed silica, the curves of filled mixtures are situated at much lower temperature than
those of unfilled mixtures. Increased filler content affects the asymmetry of the interactions
and also the temperature of phase separation. The phase diagrams characterise the systems
only when they are phase separated. In the region of incomplete phase separation, e.g.,
interfacial layer, only changes in the free energy of mixing and thermodynamic interaction
parameters can be used to characterise the blends. The difference between free volumes
of the components in the surface layer and in the bulk contributes markedly to increases
in thermodynamic stability. The thermodynamic explanation consists of the fact that the
coordination number depends on the conformational state of macromolecules, and it
will be different in the surface layer and in bulk. Therefore the interaction parameter can
be considered as a sum of two parameters characterising the surface layer and the bulk:
80
Phase Behaviour
where is the volume fraction of the mixture in surface layer. The calculations show
that the interaction parameter in the surface layer is considerably lower than in the bulk,
therefore the interface with solid increases the thermodynamic stability of immiscible
polymer pair even in the melt. But not all fillers increase the thermodynamic stability. By
addition of filler (kaolin) to the miscible mixture PVAc/EVA copolymer, the interaction
parameters increase in the surface layers in the whole range of compositions. This indicates
the diminishing thermodynamic stability of the filled mixture. The effect is explained by
independent adsorption of both components at the interface decreasing the interaction
between component macromolecules [42].
The present theories described well enough the dependence of the phase behaviour on
the composition of the blends and the molecular mass of components, but some
particularities of such systems have been found that cannot be explained presently.
The mechanism and dynamics of phase separation in binary polymer blends have long
been the subject for many studies, in an attempt to obtain high performance materials by
controlling the morphology.
As has already been shown there are three regions of phase separation of a binary
polydispersed blend: miscible, metastable and immiscible. Phase separation takes place
in the last two cases but by different mechanisms. In the metastable region between
bimodal and spinodal some forms of activation mechanism must trigger the phase
separation. In the immiscible region such a triggering is not needed, the phase separation
being spontaneous. The metastable mixtures are characterised by a nucleation and growth
mechanism of phase separation, with a finite cooling into the metastable region required
for nucleation. Due to the fluctuation of density within the metastable region spinodal
exhibits diffuse boundary, the need for activation rapidly vanishes as the conditions
move from bimodal to spinodal.
Two mechanisms of phase separation are considered: nucleation and growth (NG) and
spinodal decomposition (SD). Both take place mainly in three steps: early, intermediate
and late step of evolution Figure 3.4.
81
Handbook of Polymer Blends and Composites
Figure 3.4 A schematic representation of the stages of the phase separation mechanisms
NG and SD (concentration patterns or spatial segmental density profile of component
A) [9, 16, 43]. - is wavelength of concentration fluctuation, is domain size
In the first mechanism, the initial evolution of phase separation takes place by localised
fluctuation of concentration (or density) and the mechanism responsible for phase
separation is known as nucleation and growth. Into the spinodal region (enclosed by
spinodal) the concentration fluctuations are delocalised, spontaneous phase separation
takes place known as SD. Due to the large chain dimensions and slow diffusion rates, the
mean field theory describes well enough, the phase separation in polymer blends, but
other theories have been also developed [24]. A two-step phase separation was described
by Hashimoto and others [44, 45].
Under the conditions of gradual temperature decrease and low viscosity, the phase
separation begins in a metastable region, where nuclei with increased concentrations of
one of the blend components appear in the homogeneous medium. Of these,
thermodynamically favourable are only nuclei having sufficiently large dimensions
(because the chemical potential is size dependent) whose formation requires a longer
time. If the viscosity is sufficiently high, the system will remain spatially homogeneous
with decreasing temperature until entering into the region of thermodynamic instability
82
Phase Behaviour
The transition of the system from one phase to two-phase state takes place with
amplification of fluctuations in the composition with development of the microregions
of a new phase. The diffusion flows of components in the metastable region are directed
towards diminishing fluctuation. These fluctuations are considered as critical nuclei. If
these exceed critical values the second step will be developed. The process is similar to
the appearance, growth and coalescence of the micro-regions of the evolved new phase
by condensation or crystallisation.
The metastable regions are located on the convex part of the G versus curve
Figure 1.1a, within the phase separation region, where: (2G/2)P,T>0 or ((2G/
2) V>v((2G/v)2P,T where v is the specific volume of the mixture and is
compressibility. When phase separation occurs, the system across the phase diagram
from b1 to b2, consumes an energy, which is the activation energy required for nucleation.
This activation energy depends on the interface energy required for nucleus formation
which is the product of interfacial tension coefficient () and surface of nucleus (S).
These values characterise the nucleation step. The nuclei or droplets with critical size
grow by diffusion of macromolecules into the nucleated domains. The rate of growth
can be approximated by the Ostwald ripening equation:
or
d t1/ n c (3.14)
where nc 3 is coarsening exponent, d and Vd are the droplet diameter and volume,
respectively; at points b1 or b2 - Figure 1.1b, are the equilibrium concentrations. Vm is
molar volume of the dispersed phase and Dt is diffusion coefficient. The diffusion step of
the droplet growth is followed by the coalescence coarsening determined by the interphase
83
Handbook of Polymer Blends and Composites
The theory of SD of homogeneous system was developed by Cahn and Hillard [46-50],
at the initial stage of the process where the concentration fluctuations appear. These
fluctuations can be considered as a set of sinusoidal waves with a fixed length (). The
wavelengths represent the dimensions of the structures formed during the course of SD,
when both the composition and size depend on time. For example, in the case of the PS/
PVME system, the probability distribution function for concentration versus time was
determined by digital image analysis and it changes from sharp Gaussian to bimodal
[51]. The three stages identified in SD are: diffusion, liquid flow and coalescence
accompanied by morphology changes. The coarsening flow mechanism is evident in the
last stage, where the process is in part similar with the phenomenon of creation of infinite
clusters and could be treated by the percolation theory [52]. The diffusion stage follows
the Ostwald equations. The kinetics of the initial stage of SD can be easily estimated by
solving the diffusion equation [46-50]:
where M is mobility constant and K is the energy gradient term. The coefficient 2 can
be identified with diffusion coefficient D. The solution of Equation 3.15 is:
84
Phase Behaviour
spinodal region D>0 and R<0, so that concentration fluctuations are rapidly damped. At
the spinodal q=qm and Rm = D = 0 and a plot of Rm = R(T) or D = D(T) can be used to
define the spinodal temperature Ts.
The SD was computer simulated using Monte Carlo, Runge-Kutta methods reproducing
the observed patterns of phase separation in both regions [61]. Numerical simulation of
the region between NG and SD postulated a dynamic spinodal with a transient percolating
structure. The simulation also revealed that in the early stage of phase separation the
coarsening exponent (nc) is not necessarily equal with 3, this value corresponds mainly
for particle growth (coarsening) in stationary fluid whilst it decreases in the flow presence
so the equation is:
(d / d 0 )n c = 1 + k c t (3.18)
where d is average droplet diameter nc varies from 5/2 to 3/2 depending on the flow type.
kc the coarsening rate constant strongly increases with volume fraction of the dispersed
phase.
Knowing the characteristic features of SD, the mathematical modelling of the phase
separation in the region of unstable state is important. At a definite composition of the
system, there can be found temperature corresponding to the maximum rate of formation
of interconnected micro-regions of phase separated. The development of modulated
structures by SD corresponds to a short period of development of interacting nuclei.
During the later stages of SD, the important role belongs to hydrodynamic initiated
85
Handbook of Polymer Blends and Composites
The most useful methods of study of the phase separation are based on time-resolved
scattering (light, neutrons, X-ray reflectivity, static secondary ion mass spectroscopy or
other irradiation sources, atomic force microscopy (AFM)) techniques, etc. Therefore it
is a direct correlation between the scattering variable and thermodynamic quantities
such as interaction parameter, concentration fluctuation, etc., usually expressed in terms
of the ratios /s or T/Ts. This correlation is very important because it establishes that
the Cahn-Hillard relationship provide a good and easy method of determining the
molecular parameters of polymeric materials.
Linearised theories developed by Cahn and others described the early stages of SD and
predict the exponential growth of intensity of light scattering, the appearance and growth
of maximum of light scattering at a constant angle with increasing time of decomposition
[46-50]. New theories and experimental investigations revealed that the amplitude of
composition fluctuations (maximum of light scattering intensity) is shifted with time to
smaller angles (intermediate stage) and then amplitude reaches its equilibrium value.
Interfacial profile and heterogeneity length are continuously developed [17, 24].
Time and compositional wavenumber resolved light scattering and nuclear magnetic
resonance (NMR) spin-lattice relaxation times during the decomposition time permit
discrimination between SD and NG mechanism [62]. NMR experiments on PS/PMVE
indicate that the homogeneous phase as well as the phase separated quenched blend
contain short scale inhomogeneities. Segment orientation on phase separation has been
also treated [63].
A numerical study of the long-term dynamics of phase separation during SD was made
by solving nonlinear diffusion equations [64]. A particular formalisms for SD in the
mixtures containing liquid crystalline polymers have been developed [65]. The SD
mechanism has also been studied in presence of block copolymers [66, 67] and
compression [68, 69]
86
Phase Behaviour
The role of various factors influencing the kinetics of phase separation has been also
established as mainly on temperature then on tacticity (PS/isotactic PVME and atactic
PVME, PC/iPMMA and aPMMA), addition of a diblock or triblock [74] copolymer (PS/
PB/PS-b-PB), presence of a solid filler, shear applied.
87
Handbook of Polymer Blends and Composites
A special case of polymer blends are semi- and full IPN and other systems that are formed
from two or many polymers during the chemical reactions. In these systems, during their
formation, at a definite stage of conversion due to arising thermodynamic incompatibility
of constituent network fragments, a process of phase separation proceeds accompanying
chemical reactions. A high degree of mechanical entanglements in IPN hinders phase
separation processes, which are always incomplete. The regular phase-separated structures
in the final polymerised (or cured) materials are interpreted as a result of structure
formation by the SD mechanism and is confirmed by light scattering experiments. A
statistical theory of microphase separation in IPN was developed by Binder and Frisch
[77]. The free energy of mixing strictly depends on the method of IPN formation
(sequential or simultaneously). In both cases the expression contains the following
contributions: elastic energy and entropy of crosslinking of two networks, the energy of
interaction between monomeric units of both networks and free energy of entanglements.
The estimation of the limits of stability of one-phase state of various IPN (spinodals)
shows that its stability is higher in the region of small amount of one of the networks.
The theory of SD in the case of chemical quenching is valid, if curing time is smaller than
relaxation time of the system. This leads to the formation of thermodynamic non-
equilibrium structures which depend on the prehistory of the system and essentially
88
Phase Behaviour
affect the IPN properties. [78-82]. A comprehensive review on the reaction-induced phase
separation was published by Williams, Rozenberg and Pascault [81].
Most of the modifiers exhibit an UCST behaviour (epoxies cured with diamines and
toughened with carboxyl-terminated polybutadiene-acrylonitrile copolymers), i.e., the
miscibility increases with increasing temperature. As reaction proceeds the modifier
becomes less miscible at Tr. The LCST behaviour is also observed for modifiers as
polyethersulfones. A change in temperature shifts the miscibility gap mainly in the pre-
gel region, therefore most the phase separation takes place before gelation, although the
continuation of the phase separation process after gelation (or even vitrification) of the
thermosetting polymer has been observed.
Some chemical groups are continuously transformed into different chemical groups during
polymerisation, the interaction parameter between the components may vary throughout
the reaction and interaction energy may be a function of conversion. Thus may increase
89
Handbook of Polymer Blends and Composites
Figure 3.5 Evolution of the miscibility gap with conversion for modified thermosetting
polymers showing a UCST (__________) and LCST (------------) behaviour. cp, cloud point
conversion, - conversion [81, 82]
90
Phase Behaviour
Two alternative ways are used to represent the reaction induced phase separation in a
polycondensation step: time-temperature-transformation and conversion-temperature
transformation diagrams. However a triangular diagram is more useful to describe the
miscibility curves in the case of a chain polymerisation, at a constant temperature.
Of course the phase separation mechanism depends on the position of the system on the
phase diagram. It is very difficult to plot the phase diagrams for such systems. The phase
diagram can be obtained only in a simple case when the IPN is obtained from a linear
polymer and a bifunctional polymer that form a three-dimensional network (crosslinked
PS /linear PBMA, crosslinked polyurethanes (PU)/linear PBMA). Under crosslinking
reaction conditions both composition and thermodynamic compatibility continuously
change. Only at very beginning the changes in composition can be considered almost
constant and the Cahns theory may be applied, because the phase separation occurs
during 200-300 s.
The morphology and properties of SIPN and IPN are determined by the kinetics of their
formation and by processes of microphase separation accompanying the reaction
proceeding by NG mechanism or due to SD mechanism, these aspects being
interconnected. Three cases are identified:
1. If one network is formed much faster than the other, the initial stage of IPN formation
proceeds practically in the liquid state of the components of the second network. The
phase separation takes place relatively fast and the evolution of the swollen first
network in components of the second network is possible. This case is typical of IPN
formation by sequential curing and is characterised by a high degree of phase
separation or segregation.
3. The third case is the most frequent, the reaction rates of components are comparable
(simultaneously IPN) so microphase separation begins simultaneously with the
chemical reactions before gel point. The process is stopped after reaching gel-point
at a rather high degree of crosslinking. The two-phase structure with incomplete
phase separation appears as a result of these two overlapping processes dependent
on both reaction rates and kinetics of phase separation. The phase separation in
IPN is a two-stage process. The initially pure SD mechanism becomes more complex
due to simultaneous processes of phase separation according to the NG mechanism.
In some cases one mechanism is replaced by the other. These processes are very
complicated thus pure mechanisms of NG or SD can hardly occur in their pure
91
Handbook of Polymer Blends and Composites
If the hardener has a low reactivity, small and large spherical domains are formed, i.e.,
bimodal structure. Small domains are distributed among large domains. Due to the
competition between phase separation and chemical reaction, two-phase morphology
can be controlled, as well as physical and mechanical properties and it is possible to
keep the spinodal structures at different stages of SD [87]. For example the
interconnecting structures generated by SD were frozen in PS/poly(-chlorostyrene)
system by irradiation and in PB/PI system by crosslinking reaction [88, 89]. The period
of modulated structure of gel formation varies with the reaction time and conversion
degree. It was shown that the phase separation in IPN strongly depends on composition.
For IPN based PU, it was found that at small amount of initial monomer in network
matrix, more intense interpenetration takes place and therefore smaller segregation
degrees are observed. An apparent compatibility due to non-equilibrium frozen state
may arise which may be considered as forced compatibility [90, 91]. Forced
compatibility should lead to an additional contribution to the free energy of the system.
The contribution of elastic energy is small at the onset of phase separation and increases
with increase in composition difference between phase separated regions. As a result,
the system is stabilised when the thermodynamic driving force for phase separation is
balanced by elastic forces from entanglements of network fragments. The stability
limit of such a polymer system is characterised by a curve below the chemical spinodal.
For the system with UCST, the real spinodal is situated below the chemical spinodal.
For some IPN, the real spinodal is so remote from chemical spinodal that the unstable
state can not be reached.
The formation of multiphase structures for the given rate of cure, modelling the effect
of reaction, was studied by numerically solving of non-linear diffusion equation with a
time-dependent free energy term for a series of values characterising the depth of cure.
Diffusion equations admit analytical solutions that enable the characterisation of the
system above and below the spinodal [92].
Phase separation in chemically reacting polymer blends is a new field of research [93].
92
Phase Behaviour
The methods used to study miscibility of polymers may be classified as I) phase equilibria
methods, II) evaluation of interaction parameter and III) indirect compatibility tests (differential
scanning calorimetry (DSC), differential thermal analysis (DTA), differential thermo-
mechanical analysis (DTM), NMR, infrared (IR) spectroscopy, optical or light microscopy,
SEM, TEM, AFM). The applicability of each group of methods depends on domain size
which may vary from atomic level (10-4 m) to large dimensions (102 m). The decrease in
sensitivity of several groups of methods is AFM > TEM > SEM > optical microscopy and
WAXS > SAXS > SANS > light scattering. Dielectric, mechanical, thermal and IR spectroscopy
measurements have close limits of detection of 10-2 - 10-3 (10-4) m [24].
Phase separation in solution leads to cloud point determination from which cloud point
curves are constructed. Cloud point curves should follow spinodal for binary systems.
The spinodal can be determined in a series of quenching experiments (heating/cooling
rate recommended of 0.1 C/min). The test is very important for aqueous systems with
strong interactions where complexation/decomplexation can be studied (see Chapter 6).
Phase separation can also be appreciated from film clarity, display of properties which
are at least a weighted average of the components, Tg, densitometry, etc.
Light scattering methods are rigorous extended turbidity studies mainly by pulse induced
critical scattering. The problem is the time to reach equilibrium in the polymer system, because
the temperature is limited by degradation and self-diffusion coefficient of macromolecules is
of order of magnitude of 10-15 m2/s. The phase separation is very slow and may never reach
completion. By centrifugation an acceleration of the phase can be achieved.
93
Handbook of Polymer Blends and Composites
Melting point depression can be used for interaction parameter determination (see Chapter 2).
IGC [17, 94, 95] which is mainly recommended for polymer-polymer-solvent systems,
because the interaction of each polymer with the solvent determines the phase behaviour.
For example PS-PVME gave clear a solution when mixed with benzene, toluene or
tetrachloroethylene but a cloudy solution with chloroform, dichloromethane or
trichloroethylene. The phase diagram determined by IGC is presented in Figure 2.2b.
IGC has been somewhat neglected in recent years due to scatter in data and because of
solvent (chemical probe) dependence found with Bij. Phase equilibria provide another
source of interaction data in polymer solutions. However, they are useful mainly in the
vicinity of the temperature of demixing. The most dependable data are from the
measurement of the melting point depression by DSC, but this method is applicable only
for blends where one of the polymers is crystalline.
A large number of methods are available for determining the existence of multiple phases
in polymer systems such as: multiple transition temperatures determined by volume -
temperature measurements, modulus - temperature measurements, DCS, DTA, phase
separation in common solvent, phase contrast optical and electronic microscopy, SAXS,
SANS [96-99], positron annihilation lifetime spectroscopy (PALS) technique that gives
information on an angstrom scale etc., PALS uses a POSITRONFIT and the MELT
program to evaluate the spectra [99, 100].
Thermodynamic interactions between polymer chains are often studied by SANS with
one component of a binary blend labelled with deuterium [101]. A surprising effect is
the dependence of thermodynamic behaviour on isotopic substitution. SANS in principle,
yields the interaction coefficient of polymer blends in a similar procedure to the light
scattering of polymer solutions.
The variability of morphology with the processing conditions, the phase ripening on
ageing are two characteristics of heterogeneous blends which limit the penetration of
blends into the market. However, from the practical point of view the hetero-phased
systems are advantageous; the dispersed phase improves the behaviour of the brittle
94
Phase Behaviour
polymer or has a reinforcing effect. One of the most important problems facing the
polymer blend industry is stabilisation of the desired properties, i.e., the elimination of
residual stresses and stabilisation of the process-generated morphology and conditions
of normal use of the manufactured part. The way to obtain such systems is by stabilisation
of the phases by compatibilisation or alloying.
Addition of a small quantity of a third component that is either miscible with both
phases (co-solvent) or a copolymer with one part which is miscible with one component
of the blend and the second part which miscible with another component (0.5-2 wt%);
usually block type less frequently a graft one has been used.
Addition of a large quantity (35 wt%) of a core-shell compatibiliser that behaves like
a multipurpose compatibiliser-cum-impact modifier and.
Co-solvent can also be used to generate compatible blends. Frequently two immiscible
polymers will form a true solution in a common solvent. After the removal (freezing-
drying) the interfacial area is so large that even very weak polymer-polymer interactions
will sufficiently stabilise such a pseudo-homogeneous system. This method works
particularly well in systems with weak hydrogen bonding. Another application of the co-
solvent principle is an addition of co-miscible oligomeric plasticisers
95
Handbook of Polymer Blends and Composites
References
2. R. Koningsveld and A.J. Staverman, Journal of Polymer Science, Part A-2, 1968,
6, 305.
3. R. Koningsveld and A.J. Staverman, Journal of Polymer Science, Part A-2, 1968,
6, 325.
4. R. Koningsveld and A.J. Staverman, Journal of Polymer Science, Part A-2, 1968,
6, 349.
6. Polymer Blends and Mixtures, Eds., D.J. Walsh, J.S. Huggins and A.
Maconnachie, M. Nijhoff, Boston, MA, USA, 1985, 89.
12. I.C. Sanchez and R.H. Lacombe, Journal of Physical Chemistry, 1976, 80, 2352.
13. I.C. Sanchez and R.H. Lacombe, Journal of Physical Chemistry, 1976, 80, 2568.
14. I.C. Sanchez and R.H. Lacombe, Journal of Polymer Science: Polymer Letters
Edition, 1977, 15, 2, 71.
96
Phase Behaviour
15. I.C. Sanchez and R.H. Lacombe, Macromolecules, 1978, 11, 6, 1145.
16. I.C. Sanchez and R.H. Lacombe, Journal of Macromolecular Science B, 1980, 17,
3, 565.
17. Y.S. Lipatov and A.E. Nesterov, Thermodynamics of Polymer Blends, Technomic
Publishing Co. Inc., 1997, Lancaster, PA, USA.
18. J.C. Meredith and E.J. Amis, Macromolecular Chemistry and Physics, 2000, 201,
6, 733.
20. C.H. Lee, H. Saito and T.T. Inoue, Macromolecules, 1995, 28, 8096.
21. C.H. Lee, H. Saito, G. Goizueta and T.T. Inoue, Macromolecules, 1996, 29, 12,
4274
23. R. Koningsveld and L.A. Kleintjens, Journal of Polymer Science 1977, C-61, 221.
24. L.A. Utracki, Polymer Alloys and Blends, Hanser Publishers, Munich, Germany,
1989.
25. M.M. Coleman, J.E. Graf and P.C. Painter, Specific Interactions and the
Miscibility of Polymer Blends. Practical Guides for Predicting & Designing
Miscible Polymer Mixtures, Technomic Publishing Company Inc., Lancaster, PA,
USA, 1991.
27. R. Horst and B.A. Wolf, Journal of Chemical Physics, 1995, 103, 3782.
28. R. Horst and B.A. Wolf, Macromolecular Theory and Simulations, 1995, 4, 449.
30. S.I. Kuchanov and S.V. Panyukov in Comprehensive Polymer Science, Second
Supplement, Ed., G. Allen, Pergamon Press, New York, NY, USA,1996, Chapter
13, 441.
97
Handbook of Polymer Blends and Composites
31. V.P. Privalko and V.V. Novikov, The Science of Heterogeneous Polymers,
Structure and Thermophysical Properties, John Wiley and Sons, Chichester, UK,
1995, 106.
36. S.D. Mehta and K.G. Honnell, Proceedings of ANTEC 97, Toronto, Canada,
1997, Volume 2, 2648.
38. M. Beiner, G. Fytas, G. Meier, S.K. Kumar, Physics Review Letters, 1998, 81,
594.
39. D.J. Lohse, M. Rabeony, R.T. Garner, S.J. Han, W.W. Graessley and K.B. Migler,
ASC Polymeric Materials Science and Engineering, 1998, 78, 87.
40. M. Rabeony, D.J. Lohse, R.T. Garner, S.J. Han, W.W. Graessley and K.B. Migler
Macromolecules, 1998, 31, 19, 6511.
41. M. Nishimoto, H. Keskkula and D.R. Paul, Polymer, 1991, 32, 2, 272.
42. A.E. Nesterov and Y.S. Lipatov, Polymer, 1999, 40, 5, 1347.
46. J.W. Cahn and J.E. Hillard, Journal of Chemical Physics, 1958, 28, 258.
47. J.W. Cahn and J.E. Hillard, Journal of Chemical Physics, 1959, 30, 1121.
98
Phase Behaviour
48. J.W. Cahn and J.E. Hillard, Journal of Chemical Physics, 1959, 31, 688.
49. J.W. Cahn and J.E. Hillard, Journal of Chemical Physics, 1965, 42, 93.
50. J.W. Cahn and J.E. Hillard, Acta Metallurgica, 1962, 9, 795.
51. H. Tanaka and T. Nishi, Physics Review Letters, 1987, 59, 692.
53. C. Varea and A. Rodledo, Journal of Chemical Physics, 1981, 75, 5080.
54. M. Metiu, K. Kitahara and J. Ross, Journal of Chemical Physics, 1976, 65, 383.
58. P.G. de Gennes, Journal of Physics Letters (France), 1977, 38, 441.
59. P.G. de Gennes, Scaling Concept in Polymer Physics, Cornell University Press,
New York, NY, USA, 1979.
61. R. Petschek and H. Metiu, Journal of Chemical Physics, 1983, 79, 3443.
62. N. Parisel, F. Kempkes, C. Cirman, C. Picot and G. Weill, Polymer, 1998, 39, 2,
291.
64. E.V. Prostomolotova, I.Y. Erukhimovich and L.I. Manevich, Polymer Science,
Series A, 1997, 39, 6, 682.
65. A.J. Liu and G. H. Fredrickson, Macromolecules, 1996, 29, 24, 8000.
67. L. Kielhore and M. Muthukumar, Journal of Chemical Physics, 1999, 110, 4079.
99
Handbook of Polymer Blends and Composites
69. H. Warkins, G.D. Brown and V.S. Ramachandra Rao, Macromolecules, 1999, 32,
7737.
70. E.V. Prostomolotova and I. Ya. Erukhimovich, Polymer Science, Series A, 1998,
40, 5, 419.
72. T. Nishi, T.T. Wang and T.K. Kwei, Macromolecules, 1975, 8, 227.
73. T. Kyu and M. Saldanha, Journal of Polymer Science, Physics Edition, 1990, 28,
97.
75. T.T. Inoue, T. Ougizawa and K. Miyasaka in Current Topics in Polymer Science,
Eds., R.M. Ottenbrite, L.A. Utracki, and S. Inoue, Hanser Verlag, Munich,
Germany, 1987.
77. K. Binder and H.L. Frisch, Journal of Chemical Physics, 1984, 81, 2126.
81. R.J.J. Williams, B.A. Rozenberg and J.P. Pascault, Advances in Polymer Science,
1997, 128, 97.
82. Y.S. Lipatov, T.T. Alekseeva and V.V. Shilov, Polymer Networks & Blends, 1991,
1, 129.
100
Phase Behaviour
86. J.K. Yeo, L.H. Sperling and D.A. Thomas in Polymer Compatibility and
Incompatibility, Ed., K. Solc, Harwood, 1982.
90. I. Sakurada, A. Nakajima and H. Aoki, Journal of Polymer Science 1959, 35,
507.
91. Y.S. Lipatov, L.M. Sergeeva, A.E. Nesterov and L.V. Karabanova, Reports of the
Academy of Science - USSR, 1975, 220, 637.
92. S.A. Shaginyan and L. I. Manevich, Polymer Science, Series A, 1997, 39, 908.
94. D.D. Deshpande, D. Patterson, H.P. Schreiber and C.S. Su, Macromolecules,
1974, 7, 530.
96. W.W. Maurer, F.S. Bates, T.P. Lodge, K. Almdal, K. Mortensen and G.H.
Fredrickson, Journal of Chemical Physics, 1998, 108, 2989.
100. C. Wstlund and F.H.J. Maurer, Macromolecules, 1997, 30, 19, 5870.
101
Interface (Interphase) in Demixed Polymer Systems
4
Interface (Interphase) in Demixed Polymer
Systems
Cornelia Vasile
The main properties of the interphase are thickness (dinter), interfacial tension (AB),
concentration profile (i (x)), etc.
The boundary or interfacial layers possess an effective thickness, beyond which the
deviations of local properties from their bulk values become negligible. In polymeric
materials the thickness of interfacial layer (or region) may be rather high due to the
chain structure and consequently will impart a specific behaviour to the polymers unlike
low molecular substances.
Surface (interfacial) tension is the reversible work required to create a unit surface
(interfacial) area or excess free energy of a system at constant temperature (T), pressure
(P) and composition (n):
G
=
A T , P , n
103
Handbook of Polymer Blends and Composites
where is the surface (interfacial) tension, G is the Gibbs free energy of the system and A
is the surface (interfacial) area. It is a function of the distribution of quasicomponent
density.
= d + p
where d is dispersive and p is polar component of surface tension. The polar component
in its turn can be separated into electron donor and electron acceptor components.
The interfacial tension is related to the surface tension and the polarity of the two
contiguous phases by the equation:
4 dA dB 4 Ap BP
AB = A + B (4.1)
dA + dB Ap + Bp
104
Interface (Interphase) in Demixed Polymer Systems
The harmonic-mean, e.g., Equation 4.1, is much satisfactory as geometric mean, e.g.,
Equation 4.2 [10, 11], equation valid if reaction does not takes place between polymers.
= o(1-T/Tc)11/9
At ordinary temperatures -d/dT for most polymers is practically constant. For polymer
pairs that exhibit upper critical solution temperature (UCST), the interfacial tension
decreases with increasing temperature and vanishes at the UCST. This to the contrary of
polymer pairs having a lower critical solution temperature (LCST), the interfacial tension
vanishes at the LCST and increases with increasing temperature in the immiscible region.
The change in entropy of an interfacial layer can by found from a dependence of interfacial
tension on temperature and composition of the mixture considered in thermodynamics
of surface phenomena.
The surface tension varies also with density according to the Macleod equation:
= o m
( P )/M )
o = ( i i where P i is the parachor of the repeat unit and Mi is molecular
weight. The parachor is a group additive quantity and its values are tabulated [12, 13].
m is the Macleods exponent and for most polymers has a value of 3.0-4.5.
The effect of molecular weight is usually ignored for polymers except for oligomers.
105
Handbook of Polymer Blends and Composites
The properties of the interfacial layer are dependent on the conditions of preparation of
the mixtures and are controlled by thermodynamic and colloidal chemical factors. The
condition G<0 also describes the process of spontaneous emulsification of one polymer
in the medium of another, that leads to the formation of a stable colloid systems. In this
case thermodynamic parameters of the system are changed due to dispersion of one
polymer in another but also because of formation of an interface between two polymers.
By dispersion, parts of the dispersion medium and the dispersed phase move into the
interfacial layer (interphase). The interfacial adhesion, structure and properties of the
interphase are very important characteristics of the immiscible blends. Interactions between
polymers result in the formation of interfaces, i.e., in the formation at least one new
polymer phase. It is reasonable to consider the interfacial region as an interphase, a
third phase in the immiscible blends with its own characteristic properties. The interphase
is a region of interdiffusion of the two types of macromolecules.
The segmental mobility within the interphase is lower, as shown by increased glass
transition temperature [14]. The relationship between glass transition temperature (Tg)
and interphase thickness is non-monotonic, as are the packing density, surface tension
and the free mixing energy.
where g and r are isobaric volumetric thermal expansion coefficients in the glassy and
rubbery region, respectively.
There are only a few data available which show the equilibrium structure and equilibrium
profile at the interface between the two polymers. In that case it is important to study the
kinetics of the interphase formation. De Gennes [16] and Harden [17] considered the
formation of a diffuse interface for 1 and using the scaling concept, obtained the
dependence of the interphase width on time t1/2 at short times and as t1/4 at longer times
106
Interface (Interphase) in Demixed Polymer Systems
and that the first dependence is characteristic to miscible mixtures and the second to
immiscible mixtures. These studies led to the analysis of the critical fluctuation of
concentration near the critical point, spinodal decomposition, dynamics of chain motion,
the coil-globule transition, etc.
The interfacial tension is a measure of material compatibility and it plays a role to failure
because the mixed units in the interphase may contribute to the properties in measurable
ways. In a composite material, the interface acts to transmit a stress from one phase to
another phase. The efficiency of this stress transfer is dependent on the nature of interface.
The interphase region may be a non-negligible fraction in a finely dispersed system.
The formation of the interphase distinguishes the properties of blends from properties of
solution. After all, the rheological and mechanical properties of polymer blends are
determined by compatibility of the components and by the degree of microphase separation
that implies the formation of an interphase. For example, the blends of polycarbonate
(PC) with acrylonitrile-butadiene-styrene (ABS) are commercial products that do not use
any compatibiliser component. The useful properties generated, especially at a copolymer
content of ~ 25 wt% are explained by developing a strong interface [18].
The theory of interfacial tension and composition and structure of an interfacial layer in
polymer mixtures has been developed by many authors. Helfand, Tagami and Sapse [19-
28] used the approximation of mean field theory predicting that the thickness of the
interfacial (dinter) layer is ~ 1 nm and depends on AB. Roe [29] established that the
interfacial tension and thickness of interfacial layer depend on the chain length and on
= - crit. Nose [30] showed the change in the polymer coil dimensions at the interface.
Sanchez [31-34] used gradient theory and compressible lattice theory to describe the
properties of interfacial layer. Other theories [16, 17, 35] have applied the scaling concept
to this purpose. New steps in understanding the polymer-polymer interphase have been
made by the Wool and co-workers [36-40], who used either methods of molecular
dynamics [41] or considered the fractal structure of polymer interfaces [38]. When
diffusion occurs at an interface, the concentration profile varies smoothly as a function
of the one-dimensional depth. However, when the diffusion process is viewed in the two-
or three-dimension diffusion field, it exhibits a rough nature. The random nature of
diffusion permits the formation of complex structures with fractal characteristics. The
authors analysed gradient percolation using computer simulation for the metal-polymer
interface and the fractal structure of a polymer-polymer interface was simulated by
reptation diffusion. A definitive statement regarding the validity of the reptation model
will be made by development of accurate simulation of dense systems that depends on
the current state of computers associated with performance methods for determination
of concentration profile [42].
107
Handbook of Polymer Blends and Composites
Such studies are important for the following cases: welding of polymers in the melt,
lamination of composites, coalescence of powder and pellet resins, drying of latex paints,
internal weldlines formed during injection and compression molding, tack of uncured
elastomers and a multitude of processing operations where interdiffusion is required to
develop full strength at an interface. The fractal analysis compares well with experimental
image examination of the interface.
d int er =
( A B )V
RT[ln sA + r AB ( sB)2 ]
b(T * / T)1 / 2
or d i nter = (2 / 3 3) 1/ 2
AB [1 1.32T / T*]1 / 2
108
Interface (Interphase) in Demixed Polymer Systems
b = m1/2 (<R>/M1/2).
The concentration (as molar fraction) of component in the interphase (xs) can be obtained
from temperature dependence of interfacial tension of phase (1) and (2):
(d/dT) + (d/dT)(1)
xs
1/2 (d/dT)(2) + (d/dT)(1)
It can easily be seen that the direct proportionality of the interfacial layer thickness and
b are inversely proportional with AB, that means that the interpenetration is realised
only on a very small distance which is much less than macromolecular dimensions.
The difference between a molecule near a surface and in bulk is that, a polymer molecule
tries to get away from a surface because, near the surface it loses conformational entropy
(all conformations which will cross the boundary are disallowed). At the interface, the
interaction between statistical segments of polymers A and B are expressed by the
interaction parameter (Florys) AB [48, 49]. They obtained the following relations for:
interfacial thickness (dinter) tension coefficient (AB) and density profile across the interface
(or interfacial composition profile) () (see Figure 4.1):
109
Handbook of Polymer Blends and Composites
AB = bo kBT(AB/6)1/2
1/ 2
B(x) = {1/2 + 1/2 tan h [6 1 / 2 ( x / d ) ]} with y = exp{6 AB (x/dinter)}
AB int er
where i = A or B, b is a lattice parameter and x/d is the reduced distance across the
interface, kB is Boltzman constant. Therefore interfacial tension can be determined from
solubility parameters values.
In binary blends the interphase thickness is inversely proportional to the interfacial tension
coefficient; AB 0.1 is a quite a large driving force to broaden the interface. The reciprocity
between dinter and AB is not to be expected to hold in compatibilised blends.
+ B b B 1 A b A B b B
2
1 / 2 A b A
AB = k B (o AB) +
2 6 A b A + B b B
where o is the density of pure polymer in units of monomer segments per unit volume.
The effective length of monomeric units b, is chosen in such a way that mean square
distance between chain ends is zb2, b is Kuhns statistical segment length, b = C1/2 bo
with bo real mean of monomer length. The real mean-square end-to-end distance for
long chains is frequently written as: C Z bo2 and Z is degree of polymerisation, C is
Florys characteristic ratio and kB is the Boltzman constant. Kuhns statistical segment
length for vinyl polymers has been written as a function of structural characteristics
as [47]:
bvinyl = 2a[sin(/2)]/(1+cos)2
where (a) is the bond length, is the tetrahedral bond angle and represents the steric
hindrance to internal rotations, = <R2>/<R2>f, <R2>, being the average square end-to-
end distance and <R2>f is the average square end-to-end distance of the chain in the
freely rotating state. The average square end-to-end distance is usually obtained by solution
measurements. The b and bvinyl have the same physical significance, but they have different
values because each of them is obtained from another type of measurement.
110
Interface (Interphase) in Demixed Polymer Systems
This theory led to the following conclusions: the surface free energy is proportional to
1AB
/2
, the chain-ends of both polymers concentrate at the interface and any low molecular
weight third component is repulsed to the interface and the interfacial tension coefficient
increases with molecular weight to an asymptotic value:
AB = AB - aoMn-2/3
An effect of this dependence that should take into account is the change in the
concentration of polymeric homologues of low molecular weight in the interfacial region
because they are much mobile and will lead to decrease in interfacial tension and to
promotion of dispersion when the system is heated or subjected to mechanical stress.
The thickness of the interphase is comparable with the dimension of the monomeric unit
therefore only limited diffusion of chain segments is possible in the interphase region.
The equation for dinter has been corrected to account the number of segments for each in
the chains rA and rB:
Later refinements by Helfand and Sapse [26-28] taking into account interaction energy
density (B) gave the relationship:
[ ]
2RT 1/ 2
d int er = A / 6 ( R 2A / M A )1 / 2 + Bi / 6 ( R 2B / M Bi )1 / 2
Bij
Where R 2B is the mean square unperturbed end-to-end chain distance and Mi is the
molecular weight. Interfacial tension is expressed as:
1/ 2
RTBij
( ) ( )
1/ 2 1/ 2
AB = A / 6 R 2A / M A + Bi / 6 R 2B / M Bi x
2
2
( ) ( )
1/ 2 1/ 2
2
A / 6 R A / M A Bi / 6 R 2B / M Bi
1
1 + 2
3
( ) / 6( R )
1/ 2
2
1/ 2
2
A / 6 R A / M A + Bi B / M Bi
Later, Broseta and others [50] using an asymptotic approximation to the square gradient
theory have treated the situation with finite molecular weight and due to the entropic
111
Handbook of Polymer Blends and Composites
gain obtained a broad interface and reduced interfacial tension. This correction can
increase interphase thickness with 10%30%. It was also demonstrated that there is an
entropic increase when small chains segregate to the interface because they can lower
interfacial tension and increase the interfacial thickness.
A model was proposed to fit neutron reflectivity data that accounts for surface roughness
and for a diffuse interface between the polymer layers modelled as either an error function
or hyperbolic tangent shaped profile (see Figure 4.1 and Equations 4.3-4.6), according
to theory [51]. The interfacial thickness calculated graphically from the width of a tangent
drawn to the inflection of the interfacial profile corresponds with that evaluated by
Helfands theory see Figure 4.1. In the case of the polypropylene oxide/styrene-acrylonitrile
copolymer) PPO/SAN copolymer, a 10 interfacial thickness was found independently
on copolymer composition, but various annealing conditions can lead to the increase up
to 60-65 as copolymer composition decreased from 40 to 25 wt% acrylonitrile (AN).
A decrease toward the miscibility limit is recorded for several systems.
The theory correctly predicts an interphase thickness for immiscible blends of about 1-4 nm.
Roe [29] and Joanny and Leibler [52] developed a quasi-crystalline model valid under
the conditions of high immiscibility (near the phase separation point) correcting Helfands
relationship. The basic concept is that the nature of the interphase region is substantially
dependent on the degree of immiscibility.
The described classical thermodynamic theories treat the interphase as a black box but
do not describe how the material is distributed within this new phase. In this sense
information is limited even now.
The theory developed by Sanchez and Lacombe [31-33] is based on the gradient theory
and the model of compressible liquid lattice where i is replaced by the segment density
of the t-th component. The simplified equations for interfacial tension and interphase
thickness (width) are:
( )
1/ 2
o a 1 / 2 *AB *AB
AB = 1 / 2 2m
112
Interface (Interphase) in Demixed Polymer Systems
o / 2a 1 / 2 )(v AB * v AB*)1 / 6
d int er = (m
Some values of the interfacial tension and of the interfacial width are given in Table 4.1.
113
Handbook of Polymer Blends and Composites
Copolymers are formed from at least two different monomers arranged in a specific
manner. They can be random, alternating, di-block, tri-block, multi-block, star-shaped,
graft, comb, etc. Their composition depends on the nature and the amount of
comonomers, polymerisation mechanism and conditions (temperature, pressure, etc.).
Each group exhibits a particular behaviour. Random graft copolymers [57] have the
ability to form micelles, which can be changed by adjusting the copolymer architecture
and hydrophilicity/hydrophibicity of backbone and grafts (polybutylene (PB)/PS). Critical
micelle concentration (CMc) as appears from spinodal is optimum for a certain number
of grafts. Selection of such systems, potentially applicable in drug delivery, is much
aided by molecular-thermodynamic calculation, following evolution of a system close
to spinodal.
Theories of block copolymers are complex involving computation of the domain size,
the interface thickness, the structure and the order-disorder transition. Helfand and
Wasserman [58-60], using narrow interface approximation, found the following
expression for interfacial thickness:
114
Interface (Interphase) in Demixed Polymer Systems
Several general theories regarding polymer interdiffusion and the techniques used to
study polymer-polymer interfacial mixing and to measure diffusion coefficients are
described in [61, 62].
Vilgis and Noolandi [9] derived a statistical thermodynamic theory based on the lattice
model and using mean-field approximation and reduced equation-of-state variables for
a binary blend with an arbitrary block copolymer or solvent (PS, PB and styrene-butadiene
block copolymer) and computed the interfacial tension, the width of the interphase, and
the concentration profiles of the components see Figure 4.1. The diffusion equations
for the density profiles are solved numerically. It has been established that the efficiency
of the block copolymer increases with the molecular weight of the blocks and this efficiency
is maximum near the CMc. Finite molecular weight and conformational entropy effects
were considered, but the excluded volume was not.
spherical
CMC = exp{1.72(Zc)1/3f4/9(1.74f-1/3 1)1/3-fZc}
where f is the fraction of A-polymer in the A-B copolymer and Zc is the degree of
polymerisation of the copolymer. The relationship was later extended to cylindrical and
lamellar micelle morphologies.
As it was shown in the Chapter 3, the phase diagrams of a block copolymer exhibit two
distinct differences compared to the phase diagram of two homopolymers. Because of
the chemical links between blocks, microdomains are present instead of macrophases
and the size of microdomain is controlled by the molecular mass of blocks and copolymer
composition, concentration and transition temperatures of individual polymers, so the
115
Handbook of Polymer Blends and Composites
phase diagram will be very complex. Most phase diagrams are characterised by UCST,
so the block copolymers form an isotropic melt, while below it they are phase separate,
the rigid blocks forming glass or semicrystalline solids. Using various approximations,
e.g., narrow interphase approximation, random phase approximation, etc., expressions
for domain size and shape and interfacial thickness have been given; theoretical treatments
predicting the dependence of these values on degree of polymerisation and magnitude of
repulsive interactions between blocks. The same factors also affect the interface in
copolymer/homopolymer blends. Blends containing graft copolymers exhibit unusual
morphologies. The most common of these were spherical structures (named onions)
consisting of alternating concentric layers of both components, such structures arising
from the incompatibility of the copolymer with the homopolymer.
The ternary systems formed from a copolymer and two homopolymers are of particular
interest. From a theoretical (also practical) point of view of compatibilisation, three
aspects must be known: copolymer configuration and conformation, composition and
concentration, all of these assuring the achievement of desired effect the morphology
corresponding to a certain complex of end-use properties. Addition of a block copolymer
either in a single homopolymer or a two-polymer mixture will change its conformation.
The copolymer chain can be located at the interface, dissolved in one of both phases and
it can form micelles in one or both polymeric phases. The morphological analyses showed
that di-block copolymers have higher interfacial activity than tri-block or grafted
copolymers, so it more readily interacts with the two homopolymer phases forming
appropriate entanglements that result in a decrease of the interfacial tension coefficient
and enhanced interfacial adhesion in the solid state. Analysing the concentration of a
block copolymer at the interface, Leibler [64, 65] showed that due to equilibrium
adsorption of copolymer at the interface, AB should decrease, therefore it is possible that
thermodynamically controlled stable droplets can exist, protected by an interfacial film
of a copolymer. The size distribution of the droplets is expected to depend on the rigidity
and spontaneous radius of curvature of the interfacial film that depends on the molecular
structure of copolymer. The various relationships for CMc has been also established [66,
67]. Semi-empirical equations between the interfacial coefficient and compatibiliser
concentration were derived that fit the experimental data well, such as:
At the surface or interface occurs a selective adsorption of the components of the mixture,
which leads to formation of adsorption layers with a different composition when compared
with that of the initial mixture. It will appear as an enrichment of surface layer by one
component or surface segregation. Surface segregation was shown for a polymer mixture
116
Interface (Interphase) in Demixed Polymer Systems
in contact with a solid surface [68], in air, at the interface between phases, and of special
importance is copolymer segregation at the interface between homopolymers.
Segregation increases with increasing incompatibility, the higher the interaction parameter,
the higher is the surface segregation. Surface segregation in interpenetrating networks
(IPN) arises during crosslinking and depend on reaction rates of components.
A curious consequence of the theoretical consideration is the fact that the interfacial
tension vanishes to some concentration of block copolymer. To explain this effect it has
been suggested that only some of block copolymer macromolecules find it energetically
favourable to settle at the interface, while the remainder are randomly dispersed in the
bulk phases.
where is density and C is copolymer concentration. The CMc [62, 63] is:
Leibler [64, 66] examined the emulsifying effect of the A-B copolymer in an immiscible
polymer A with polymer B blend. The block and graft copolymers which have chain
fragments of the same chemical nature as both homopolymers, are typically
compatibilisers. These fragments (block or grafts) are miscible with the corresponding
homopolymers. The factor which determines the choice of a compatibiliser: is its ability
to diminish the interfacial tension, because it acts as a surface active substance, and to
prevent the phase separation in the solid state, in other words it stabilises the morphology
against stress-destruction during formation. The compatibilising effect is obtained if the
copolymer is localised at the interface between two immiscible polymers, in other words
if it preferentially segregates at the interface. The penetration of blocks into phases makes
the interphase region much broader when compared with that of the binary mixture.
117
Handbook of Polymer Blends and Composites
CMc:
Tang and Huang [67] and Utracki and Shi [82] gave other semi-empirical expressions for
the relation between copolymer concentration and interfacial tension and showed that
the radius of the dispersed drop follows the same curve as the interfacial tension coefficient.
With an arbitrary X-Y block copolymer, Noolandi and Vilgis obtained [9, 77]:
118
Interface (Interphase) in Demixed Polymer Systems
or
The amount of the interfacial agent required to saturate the interface wcr, is related to the
molecular weight, the interface area, and the specific cross-sectional area of the copolymer
molecule (A) by the relationship proposed by Paul [83]:
wcr = 3 M/(RAN)
where is the volume fraction of the dispersed phase, M is the molecular weight, R is
the radius of the dispersed drop, N is the Avogadro number and A is the area occupied
by a copolymer molecule.
Vilgis and Noolandi [9] concluded on the basis of this theory that it is possible to design
a universal block compatibiliser of XY type operating on the principle of competitive
repulsion interactions between homopolymers and copolymer blocks. Several universal
compatibilisers are presently on the market. Experimentally it was shown that a di-block
copolymer has higher interfacial activity than a tri-block- or a graft- copolymers.
For the system PS/PB it was found that interfacial tension decreases with increasing
molecular weight of PS and decreases with increase in polydispersity. Block lengths just
119
Handbook of Polymer Blends and Composites
above the molecular weight of entanglements are the most practical for the
compatibilisation of polymer blends. The efficiency of a block copolymer is limited by
the formation of micelles in bulk phase and by the kinetic factors.
Noolandi found that for concentrations below the CMc, the interfacial tension linearly
decreases with copolymer concentration, and that the interfacial tension diminishes with
increasing molecular mass of the blocks. On the basis of this theory, Thomas and
Prudhomme [84] found a relationship between particle size and copolymer concentration.
Correlation of the theoretical results with the experimental data led to the conclusion
that the effect of incorporation of a compatibiliser in a mixture of two immiscible polymer
mixture causes the decrease both of interfacial tension and the dimensions of the phase-
separated domain. The domain dimensions depend not only on the interfacial tension
but on the time necessary for phase separation, on the mechanism of coarsening
(coalescence) domains, beginning of phase separation and other reasons.
Because of experimental difficulties due to the high viscosity of polymer melts, the first
reliable measurements of surface tension were reported in 1965 and in 1969 for interfacial
tension of polymer melts by Wu and Roe [85, 86].
Surface (interfacial) tension (i) can be calculated from the group theories by additivity,
the parachor method Pij, or using solubility parameters for dispersive, polar and hydrogen
interactions these values being tabulated [87, 88]:
n n
i = v
i =1
Pij /
i =1
4
ij
AB = A + B 2 (A B)1/2
Determination of interfacial tension from surface properties [10, 11, 89] is based on
Equations 4.1 and 4.2.
The methods for direct determination of interfacial tension coefficient can be classified
in two groups [90]:
Equilibrium methods such as pendant, sessile drop and spinning drop. These methods
are time consuming and are limited to low molecular mass resins.
120
Interface (Interphase) in Demixed Polymer Systems
Dynamic methods such as capillary breakup and deformed drop. The methods from
this group are especially recommended for polymer blends, the last one seems to
have a great potential.
In the sessile drop method the equilibrium shape of the stationary droplet of polymer
with higher density immersed in the melt of the second polymer is observed. Usually a
small cube of the denser material is first placed on the bottom of quartz container than
filled with grains of the less dense one. After melting (usually under a blanket of inert
gas) formation of an equilibrium droplet is recorded. The equilibration involves flow of
polymer melt under the influence of the interfacial forces. Since the viscosity of most of
the commercial resins at processing temperature are much higher than 1 kPa-s and the
interfacial tension coefficient is usually much lower than 0.4 Pa-cm, the equilibration
time can be of the order of hours or days. From the equilibrium droplet shape, the
interfacial tension coefficient can be calculated.
Pendent drop method is a relatively simple and reliable method [85, 86]. The profile of a
pendent drop at equilibrium is governed by the gravitation and the surface tension. The
surface (or interfacial) tension is given by:
= g dc2/H
where is the surface (or interfacial) tension, g is the gravitational constant, the
difference in the densities of the two phases (drop and surrounding), de is the equatorial
diameter of the drop, and H is a correction factor calculated by Adamson [91]. Due to
the bigger role of the gravitational forces than in sessile drop method, the equilibrium is
sooner reached. The method was computerised and the acquisition of the data allows
continuous analysis of the droplet shape.
(b-a)/2do
121
Handbook of Polymer Blends and Composites
where do is initial diameter of the fibre and b and a are maximum and minimum diameter
observed during the breakup process, respectively; q depends on interfacial tension,
viscosity and , t time.
Capillarity instability method the slope of the straight line is a measure of AB see
Figure 4.2b,
Deformed drop retraction method (DDRM). Figure 4.2c. A piece of polymer is inserted
between two 1 mm thick films of polymer 2. The sandwich is placed between
microscopic glass slides on a hot plate under an optical microscope. After the desired
temperature is reached, the central drop (d = 0.2-0.5 mm) is deformed by about 15%
by shearing or by elongation. Retraction of the deformed drop is recorded. After
retraction is complete, the drop can be deformed again as many times as desired. The
method makes possible to verify the relationship AB = BA. The time dependence (t) of
the deformability parameter (D) leads to determination of AB according to the
dependence established by microrheology or Taylors theory:
LB 80t AB ( + 1)
D= = D o exp = D o exp(t / )
L+B (2 + 3)(19 + 16)m R o
The drop follows an exponential decay. After a long time the diffusion stops and the
CMc plateau is reached.
Advanced techniques of modern spectroscopy have been used to elucidate the interface
role in multicomponent systems. An extensive effort has been made in developing methods
to verify the interfacial structures of multicomponent materials and improving the
analytical techniques for surface characterisation in order to produce better materials.
Currently used instrumental techniques for the analysis of interfacial structures are Fourier
transform-infrared spectroscopy (FT-IR), differential scanning calorimetry (DSC), dynamic
mechanical spectroscopy (DMS), nuclear magnetic resonance (NMR) [92], transmission
and scanning electron microscopy (TEM and SEM), X-ray photoelectron spectroscopy
(XPS) and gel permeation chromatography (GPC) [93]. Together with Tg, the T1 can be
considered as a measure of the degree of phase separation. Whereas a large difference in
proton T1 values shows phase separation to a large extent, a small difference indicates
less phase separation, because of spin diffusion in a more homogeneous mixture, in
addition, the possible formation of entanglements can also decrease the mobility of the
mixtures and prevent phase separation from occurring.
122
Interface (Interphase) in Demixed Polymer Systems
Diffuse reflectance infrared spectroscopy (DRIFT) was used to characterise directly the
structure of blends on molecular level. Agrawal and others [40] obtained results which
agreed well with theoretical results. The recent techniques for measuring depth profiles
include forward recoil spectroscopy (FRES), FT-IR and attenuated total reflectance (ATR),
small angle neutron scattering (SANS), etc. FT-IR and ATR are limited by the depth
resolution of about 1000-10 000 but provide useful information at large interdiffusion
depths. FRES and SANS and specular neutron reflection (SNR) do not measure depth
123
Handbook of Polymer Blends and Composites
profiles directly and assumptions must be made about the nature of the profile before it
can be constructed. SNR has been developed to examine short range interdiffusion in
polymers with a resolution of about 5-10 . The concentration profile may be also
determined by secondary ion mass spectroscopy (SIMS), neutron reflection (determines
motion of polymer chains at a penetration depth comparable with radius of gyration).
SIMS has ability to measure depth profiles directly (it is useful for tracer studies), its
depth resolution is about 100 for polymers, but not as good as the neutron reflectivity
(SNR) technique with a resolution of about 5 [94]. Raman microspectroscopy was
also used to determine the depth profile [95].
Only few measurements of the interphase thickness are available. Generally, values in
the range of 2-5 nm without copolymer have been reported Tables 4.2 and 4.3.
Generally the values of surface tension lie between 0.02-0.04 N/m at 140 C and the
values of interfacial tension vary from 0.01-0.015 N/m and both exhibit almost linear
variation with temperature. The observed temperature coefficients for polymer melts lie
between 5-10 x 10-4 N/m/degree for surface tension and are around 20 x 10-4 N/m/
degree for interfacial tension. Wu [10, 11, 96, 97] reported that molecular weight
dependence is significant only in the low molecular range.
Helfand and Tagami theory predicts the interface to be approximately 35 to 45 thick for
the system PS/PMMA that was confirmed by neutron reflectivity when a interfacial thickness
of 50 10 [98] or 42-66 [99] and an interfacial tension of 1-1.8 x 10-3 N/m.
124
Interface (Interphase) in Demixed Polymer Systems
These results showed a good agreement between experiment and theory. However, the
ellipsometry measurements give a larger thickness suggesting that the Helfand-Tagami
theory excludes highly diffuse interfaces.
On the basis of experimental data, some correlations were attempted between the
interfacial properties of multiphase polymer systems and sample characteristics, miscibility
of components and conditions of interphase formation and agreement with the predictions
of various theories. The interphase width increased with annealing time for PMMA/
SAN system from 20 nm to 60 nm after 12 hours at 303 K and remained almost constant
for the PMMA/PS system. Therefore this property depends on the compatibility of
components, time, temperature, interdiffusion coefficient and is proportional with the
125
Handbook of Polymer Blends and Composites
According to Helfand-Tagami theory the interfacial thickness is very thin for homopolymer
pairs polyacrylonitrile (PAN)/polymethylmethacrylate (PMMA), PS/PAN, PMMA/PS. The
incorporation of a di-block copolymer (S-co-MMA) in PS/PMMA system broadens the
interphase region up to certain limit. At high amounts the block copolymer forms its
own structure at the interface. Addition of copolymer slightly increased the interphase
thickness from 5 to 8.4 nm in the PS/PMMA system and from 10 to 12 nm for PS/PE
with block copolymer of hydrogenated polybutadiene with styrene. To increase the
dinterphase either tapered di-block copolymers or a mixture of block and random copolymers
are used. Tri-block copolymers are less efficient in lowering interfacial tension but they
may provide thicker interphase thickness and better adhesion between phases in the
solid state. Stabilisation of the morphology by reduction of coalescence can be achieved
by selective wetting by the third polymer. Addition of the third immiscible polymer C
may stabilise dispersion of an immiscible pair of polymer B in matrix A, if the condition
of complete wetting is achieved:
AB > AC + BC
AB < AC + BC
126
Interface (Interphase) in Demixed Polymer Systems
Though classic, the phase equilibrium of polymer solutions and blends is still an actual
subject of research. Academically, the most basic problem in its study is to find analytical
expressions for G which allows quantitative prediction of phase relationships of various
polymer solutions. Despite many efforts, the expressions derived so far for G of quasi
binary solutions remain semi-quantitative.
Non-equilibrium processes of surface segregation have not been yet studied. It is necessary
much effort to study the kinetically frozen state and to investigate the kinetics of phase
separation. As about 10% of commercial blends are compatibilised by addition rest are
reactively compatibilised, the effect of various kinds of compatibilisation methods should
theoretically be described in a quantitative manner.
It is hoped that, in the not too distant future, a sufficiently broad data base of
thermodynamic quantities covering the whole spectrum of mixtures from small molecules
to polymer blends will be available to verify theoretical statements.
There is no description of the effect of tri-block and linear multi-block copolymers on the
interfacial tension, CMc, interfacial thickness, etc. Limited experimental information is
available on their morphology and mechanical performance of blends. The core-shell
copolymers are mainly used as impact modifiers. No theoretical information is available
on their effect on the interphase properties. Many characteristic parameters of these
copolymers are important for the performance of the blends, such as core size, concentration
127
Handbook of Polymer Blends and Composites
The adhesion between polymeric domains in solid state can be demonstrated by the
methods developed for composites. The direct measurements of interfacial tension are
rare, and mostly were not done in the light of the theoretical development. At this very
moment it appears that the theoretical work is ahead of the experimental work. While
advances in the theories of interfacial phenomena, which attempt to identify important
factors involved and point the way for practical development have been made, there are
few quantitative data available which are suitable for drawing comparison between theory
and experiment. Theoreticians often look for additional reliable experimental data.
Much less attention has been given to the thermodynamic properties of complex materials
like gels, biomacromolecules, micelles, colloids, block copolymers and similar substances
that are often called soft materials Such materials abound in nature and technology.
Application of chemical thermodynamics to soft materials has been delayed because of
experimental difficulties and because, until recently, there were few theoretical models
available for describing assemblies of complicated molecules.
Thanks to recent advances in statistical mechanics and molecular physics, and thanks to
increasingly fast computers, it is now possible to develop a hard science for soft materials.
The creation and understanding of soft materials depends primarily on experimental
science. Now statistical mechanics is able to provide guidance toward interpreting
experimental results and toward reducing experimental effort. While thermodynamic
models are useful for suggesting what experimental work is most likely to lead to a
successful result, the hard science for soft materials is still in an early stage but there is a
good reason to expect it to grow dramatically in the near future thanks to the creative
efforts of dedicated researchers.
References
128
Interface (Interphase) in Demixed Polymer Systems
7. L.A. Utracki, Polymer Alloys and Blends: Thermodynamics and Blends, Hanser
Publishers, Munich, Germany, 1989.
11. S. Wu, Polymer Interfaces and Adhesion, Marcel Dekker Inc., New York, NY,
USA, 1979.
12. D.W. van Krevelen, Properties of Polymers: Their Estimation and Correlation
with Chemical Structure, Elsevier, Amsterdam, The Netherlands, 1976.
14. Y.S. Lipatov, Journal of Applied Polymer Science, 1978, 22, 1895.
16. P.G. de Gennes, Comptes Rendus de lAcademie des Sciences, 1989, 2, 300, 13.
20. E. Helfand and Y. Tagami, Journal of Chemical Physics, 1971, 57, 1812.
21. E. Helfand and Y. Tagami, Journal of Chemical Physics, 1971, 56, 3592.
129
Handbook of Polymer Blends and Composites
26. E. Helfand and A.M. Sapse, Journal of Chemical Physics, 1975, 62, 490.
27. E. Helfand and A.M. Sapse, Journal of Chemical Physics, 1975, 62, 1327.
28. E. Helfand and A.M. Sapse, Journal of Polymer Science, 1976, C-54, 289.
31. I.C. Sanchez and R.H. Lacombe, Journal of Macromolecular Science, 1980, B17,
565.
32. I.C. Sanchez and R.H. Lacombe, Journal of Chemical Physics, 1976, 80, 2352.
33. I.C. Sanchez and R.H. Lacombe, Macromolecules, 1978, 11, 1145.
34. C.I. Poser and I.C. Sanchez, Macromolecules, 1981, 14, 361.
39. J.L. Willet and R.P. Wool, Macromolecules, 1993, 26, 5336.
40. G. Agrawal, R.P. Wool, W.D. Dozier, G.P. Felcher, T.P. Russell and J.W. Mays,
Macromolecules, 1994, 27, 4407.
41. P.A. Mirau, S.A. Heffner and M. Schilling, Chemical Physics Letters, 1999, 313,
1-2, 139.
42. G.S. Grest, M.D. Lacasse and M. Murat, Springer Proceedings of Physics, 1998,
83, 23.
130
Interface (Interphase) in Demixed Polymer Systems
46. H.W. Kammer, Zeitschrift fur Physikalische Chemie - Leipzig, 1977, 258, 1149.
47. A. Ajji and L.A. Utracki, Progress in Rubber and Plastics Technology, 1997, 13,
3, 153.
48. A. Ajji and L.A. Utracki, Polymer Engineering and Science, 1996, 36, 1574.
49. H. Yamakawa, Modern Theory of Polymer Solutions, Harper & Row Publishers,
New York, NY, USA, 1971, 46.
51. G.D. Merfeld, A. Karim, B. Majumdar, S.K. Satija and D.R. Paul, Journal of
Polymer Science, Part B, Polymer Physics, 1998, 36, 3115.
52. J.F. Joanny and L. Leibler, Journal of Physics (Paris), 1978, 39, 951.
53. H.S. Lee and J.W. Ho, Polymer, 1998, 39, 2489.
55. I.C. Sanchez, Polymer Engineering and Science, 1984, 24, 79.
56. I.C. Sanchez, Polymer Engineering and Science, 1984, 24, 598.
57. J.M. Prausnitz, Pure and Applied Chemistry, 2000, 72, 1819.
61. E. Jabbari and N.A. Peppas, Journal of Macromolecular Science C, 1994, 34,
205.
62. H.H. Kausch and M. Tirrell, Annual Review of Materials Science, 1989, 19, 341.
131
Handbook of Polymer Blends and Composites
66. L. Leibler and L.A. Utracki and Z.H. Shi, Polymer Engineering and Science,
1992, 32, 1824.
68. Y.S. Lipatov, T.S. Khramova, T.T. Todosijchuk and E.G. Gudova, Journal of
Colloid and Interface Science, 1988, 123, 143.
78. K.R. Shull and E.J. Kramer, Macromolecules, 1990, 23, 4769.
79. K.R. Shull, E.J. Kramer, G. Hadziioannou and W. Tang, Macromolecules, 1990,
23, 4780.
81. A.C. Balasz, C. Singh, E. Zhulino, S.S. Chern, I. Lystskaya and G. Pickett,
Progress in Surfactant Science, 1997, 55, 181.
82. L.A. Utracki and Z.H. Shi, Polymer Engineering and Science, 1992, 32, 1824.
83. D.R. Paul in Polymer Blends, Volume 2, Eds., D.R. Paul and S. Newman,
Academic Press, New York, NY, USA, 1978, 35.
132
Interface (Interphase) in Demixed Polymer Systems
85. S. Wu, Journal of Colloid and Interface Science, 1969, 31, 153.
86. R.J. Roe, Journal of Colloid and Interface Science, 1969, 31, 228.
87. J.H. Hildebrand and R.L. Scott, The Solubility of Non-Electrolytes, Dover
Publishers, New York, NY, USA, 1964, Chapter 3.
88. M.M. Coleman, J.E. Graf and P.C. Painter, Interactions and the Miscibility of
Polymer Blends. Practical Guides for Predicting and Designing Miscible Polymer
Mixtures, Technomic, Lancaster, PA, USA, 1991, Chapter 3.
91. S. Adamson, Physical Chemistry of Surfaces, Wiley, New York, NY, USA, 1967.
94. S.J. Whitlow and R.P. Wool Macromolecules, 1991, 24, 5926.
98. S.H. Anastasiadis, T.P. Russell, S.K. Satija and C.F. Majkrzak, Journal of
Chemical Physics, 1990, 92, 5677.
99. D.W. Schubert and M. Stamm, Europhysics Letters, 1996, 35, 419.
133
Water Soluble Polymer Systems - Phase Behaviour and Complex Formation
5
Water Soluble Polymer Systems - Phase
Behaviour and Complex Formation
5.1 Introduction
Water soluble polymers have attracted an increasing interest during the last decades, due
to their numerous applications in products of everyday life (foodstuffs, cosmetics, drugs,
paints) and in various industrial processes (paper making, ceramic processing, textile
sizing, water treatment, drilling fluids, enhanced oil recovery). They respond to the desire
for replacing organic solvents by water for environmental reasons and are related to
biological systems.
They were recognised as a group in its own right in the 1960s after the publication in
1962 of a text edited by Davidson and Sittig concerned with the sources, properties and
applications of water soluble polymers, which was revised 6 years later [1]. A more
extensive handbook was edited in early 1980s by Davidson [2]. In the same period two
monographs have appeared on synthetic water soluble polymers, one by Molyneux, on
their properties and behaviour [3, 4] and another by Bekturov and Bakauova on their
solution properties [5]. A brief but concise review by McCormick, Bock and Schulz is
included in the Encyclopaedia of Polymer Science and Engineering [6].
What makes the study of water soluble polymers more exciting is the unique character
of water as a solvent [7, 8]. It is a hydrogen-bonded liquid and only polymers with polar
groups can interrupt its structure and be dissolved, while the presence of non-polar groups
results in the appearance of hydrophobic interactions between the solute molecules [9].
Such interactions, in co-operation with hydrogen bonding, dipole-dipole and Coulombic
interactions lead to the spontaneous self-association of certain amphiphilic molecules
and play a crucial role in protein structure [10].
5.2 Classification
Water soluble polymers are classified as neutral or charged polymers (polyelectrolytes)
depending on their chemical structure and to natural, semisynthetic (chemically modified
natural) and synthetic polymers according to their origin (see Table 5.1).
135
neutral and charged
natural
semisynthetic
Handbook of Polymer Blends and Composites synthetic
136
Water Soluble Polymer Systems - Phase Behaviour and Complex Formation
Another example of such a behaviour is PAAM, for which an ideal Flory () temperature
at 8 C by extrapolating methods has been determined [15]. On the contrary, polymers
with a hydrophobic character, present lower critical solution temperature (LCST)
behaviour, accompanied by an exothermic dissolution process [3] and precipitate by
increasing the temperature. A well known example of such a polymer, with inverse
solubility in water, is PNIPAM, exhibiting an LCST at 31 C [16], with a phase diagram
as shown in Figure 5.1b. Due to the hydrophobic hydration, dissolution of such polymers
in water is entropically disfavoured. On the contrary, their phase separation is entropically
driven, in accordance with a postulate that at the UCST, phase separation occurs mainly
for enthalpic reasons, while at the LCST, phase separation occurs mainly for entropic
reasons [17]. Such an LCST behaviour is exhibited by many of the non-ionic water
soluble polymers as shown in Table 5.2.
137
Handbook of Polymer Blends and Composites
Figure 5.1a Phase separation curve of PAA, of infinite molecular weight, in aqueous
0.4 M NaCl
Redrawn from R. Buscall and T. Corner, European Polymer Journal, 1982, 18, 967, with
permission from Elsevier Science
138
Water Soluble Polymer Systems - Phase Behaviour and Complex Formation
Finally, water soluble polymers can be characterised as hydrophilic if they phase separate
from aqueous solutions by decreasing temperature, as hydrophobic if they phase separate
by increasing temperature.
Segregative phase separation is most common with mixtures of two non-ionic polymers
or two similarly charged polyelectrolytes or mixtures consisting of a non-ionic polymer
and a polyelectrolyte. Mixtures of a polyelectrolyte and a non-polyelectrolyte are less
likely to phase separate (preservation of the electrical neutrality requires the charged
polymer to segregate with its associated counterions and, as the number of counterions
is much greater than the number of charged chains, the entropic disadvantage of
segregation is greater) than a mixture of two charged or two uncharged polymers [25-
28]. Phase separation can be induced in the first case by addition of extraneous salt [26,
28]. Segregative phase separation has been proposed for separations involving labile
biological materials [29].
139
Handbook of Polymer Blends and Composites
Associative phase separation, on the other hand, is the result of specific attractive
interactions such as electrostatic attraction between two oppositely charged
polyelectrolytes or hydrogen-bonding association between a proton donor and a proton
acceptor polymer. In such cases, we often speak about the formation of interpolymer
complexes, which may be soluble, compact precipitates or liquid coacervates (complex
rich phase).
The formation of complexes between water soluble polymeric species was firstly studied
in the 1970s [30-32]. Interpolymer complexes (IPC) are divided into two major classes,
the hydrogen bonding and the polyelectrolyte complexes. Hydrogen bonding IPC
formation takes place between proton-donors, such as PAA or PMAA (weak polyacids)
and proton-acceptors, such as PEG or PVP (Lewis polybases). The formation of
polyelectrolyte complexes (PEC) is due to attractive Coulombic interactions between a
polycation and a polyanion. IPC formation is considerably influenced by hydrophobic
interactions. As a result the PMAA/PEO hydrogen bonding complexes are stronger than
the PAA/PEO complexes. This difference in stability has been explained in terms of
hydrophobic interaction between the -methyl groups of PMAA and the ethylene
backbone of PEO, which might provide additional stability in the complex [33]. Moreover,
the composition of the complex of PAA with the polycation [-N+(CH3)2-(CH2)10-N+(CH3)2-
(CH2)10-], 2X- (10, 10 Ionene) is 1:1 while the composition of the complex of PMAA
with 10, 10-Ionene is 9:1 due to the hydrophobicity of the PMAA [34].
140
Water Soluble Polymer Systems - Phase Behaviour and Complex Formation
Interaction between a proton donor, such as PAA, and a proton acceptor, such as PEG,
polymer leads to the formation of a hydrogen-bonding interpolymer complex
schematically shown in Figure 5.2.
Viscometry and potentiometry are the most commonly used methods for the study of
hydrogen bonding interpolymer complexes.
141
Handbook of Polymer Blends and Composites
Potentiometry. By titrating a PAA dilute aqueous solution with PEO [36, 45] the
degree of complexation of the polyacid can be determined. As a result of the
complexation that occurs, the pH of the solution increases until the complete
complexation of the polyacid and attainment of the complex stoichiometry [46].
Potentiometry results have been used for the determination of the stability constants
of the complexes [47] and for the determination of the thermodynamic parameters
of the complexation [48, 33]. Another approach has also been proposed based on
the isoionic dilution method. In this method the pH of the polybase solution is adjusted
to the same value of that of the weak polyacid by means of HCl. Complexation leads
to higher pH values, while the determination of the complexation constant should
not depend on dilution effects [42, 43, 49, 50].
Turbidimetry. Turbidity appears with the formation of insoluble IPC [53, 54].
In a study on the complexation of PEO with PAA, PMAA and a styrene/maleic acid
copolymer [54] it was shown that, besides the major role of hydrogen bonding, the
influence of hydrophobic interaction was considerable, while a definite number of binding
sites was necessary for a stable IPC to be formed and cooperative interaction among
active sites plays an important role. In the IPC formation between PMAA and PVP, three
phase changes were observed by increasing polymer concentration: namely a homogeneous
solution, a precipitate and a gel. The interpolymer complex formation was examined in
142
Water Soluble Polymer Systems - Phase Behaviour and Complex Formation
several protic (water, methanol and ethanol) and aprotic (N,N-dimethyl foramide,
dimethyl sulfoxide and tetramethyl urea) solvents and the relationship of the complex
formation with the dielectric constant of the solvent was determined. The effect of
hydrophobic interactions and the role of the -methyl group of PMAA are also discussed
[60]. Moreover, the contribution of hydrophobic interaction in the complexation of PMAA
with PVP in aqueous solution has been shown by a calorimetric study [61].
The complexation of PAA with PEO or PVP in dilute aqueous solutions was monitored
from the segmental rotational mobility of spin-labelled PAA using the electronic
paramagnetic resonance technique. The segmental rotational dynamics revealed that PAA
forms strong complexes at low pH with PEO or PVP, whereas no complexes were observed
with highly neutralised PAA [62]. A light scattering investigation of the aggregation of
the IPC formed between PMAA and PVP showed that the fast complexation in aqueous
solution was followed by aggregation, sensitive to pH [63].
The complex formation between vinyl alcohol oligomers (OVA) and PMAA was studied
potentiometrically and viscometrically [64] and the influence of the molecular weight of
OVA on the characteristics of the complexes formed was elucidated. In another study
[65], the influence of the degree of saponification and of the molecular weight of poly(VAC-
co-VAL) on its complexation with PAA was investigated. It was shown that a lower
degree of saponification leads to the formation of stronger and more compact complexes.
The complex formation between PVP and poly(itaconic acid monomethylester) (PMMI)
or the copolymer of maleic anhydride and monomethylitaconate has been studied. The
molar composition of the first complex was found to be 1:1, while that of the second
was found to equal to 3:2. The complex particles were very compact and their intrinsic
viscosity in water was about 5-10 g/cm3 not depending upon the molecular weight of the
polymer components [66]. From diffusion light scattering measurements with the complex
formed between PVP and PMMI, it was found that the complex size distributions, two
distinct modes, increase when temperature increases. From these results it can be concluded
that the driving force for the interpolymer association is hydrogen bonding as much as
hydrophobic interactions [67].
Chatterjee and co-workers have presented many studies on hydrogen bonding interactions
and complex formation between polycarboxylic acids, or their statistical copolymers with
acrylamides (AM), and polybases, such as PVP, PEO and PEI in water [68-70]. They have
concluded that the complexation ability of PEI was greater than that of PVP [71]. A
comparative study of the interpolymer complex formation between PVAL-graft-PAAM
and PMAA or PEI showed that the complex formed between the graft copolymer and
PMAA was relatively more stable than the complex prepared by mixing stoichiometric
portions of the two homopolymers with PMAA. A reverse trend was found with PEI [71].
143
Handbook of Polymer Blends and Composites
A pair of polymers forming hydrogen bonding complexes, like PAA and poly(N-acryloyl-
glycinamide), was used to design interpenetrating network (IPN) hydrogels exhibiting positive
temperature-dependent swelling. It was found that hydrogen bonding between the polymers
improved the rate and the reversibility of swelling changes in response to temperature
fluctuations [74]. The temperature-dependence of the equilibrium swelling of IPN hydrogels
constructed with PAA and PDMAM or poly(DMAM-co-AM) was also investigated [75].
IPC between PDMAM and PAA are stable even at 70 C in aqueous solution, whereas those
containing poly(DMAM-co-AM) dissociate with increasing temperature. In these IPN, the
transition temperature shifted to higher values with increasing DMAM content. Reversible
and pulsative temperature-responsive solute release was achieved by using these IPN hydrogels.
The complex formation between PDMAM and PAA was further investigated by measuring
the turbidity of the mixture of PDMAM and PAA buffered solutions at various pH values.
The complexes were very stable even at 80 C in pure water (pH 3.2). In buffered solution at
pH 3.7 the polymer mixture showed an LCST near 60 C, although the solutions were
transparent till 80 C at pH 3.8 [76]. The influence of the molecular architecture was recently
investigated by studying the characteristic solution properties demonstrated by graft
copolymers, PAA-g-PDMAM, in aqueous media [77].
Karibyants and co-workers [78] have examined the effects of the molecular weight of PEO
and of temperature on the conformation of the complex formed in aqueous media with a
PMAA gel and they have found that a wide range of PEO concentration exists in which the
gel may occur in the swollen and in the collapsed state.
Kudaibergenov and co-workers [79] have studied the complex formation between
polyvinylether of ethyleneglycol (PVEEG), polyvinylether of diethyleneglycol (PVEDEG) and
copolymers of vinylether of ethylene glycol and vinyl butyl ether (poly(VEEG-co-VBE)),
with PAA and PMAA in aqueous solution. They showed that PVEEG and PVEDEG did not
form complexes with PAA. Nevertheless, introduction of the hydrophobic VBE into the
PVEEG chain enhanced the complexation with the polycarboxylic acids. The swelling/
deswelling behaviour of composite films, derived from IPC of PVEEG or of poly(VEEG-
144
Water Soluble Polymer Systems - Phase Behaviour and Complex Formation
co-VBE) with the acrylic acid-VBE copolymer (poly(AA-co-VBE)) was studied in water,
alcohol and water-alcohol mixtures. Moreover the formation of IPC between poly(AA-
co-VBE) and polyvinylether of monoethanolamine on a water-butanol interface was
studied potentiometrically [80].
Interpolymer complexation of PAA with PAAM and PNIPAM in dilute aqueous solution
was studied. The stoichiometry of the complexes formed was determined [82]. It was
shown that hydrogen bonding was the main factor stabilising the PAAM/PAA complex,
strengthened by decreasing the temperature, while the much stronger PNIPAM/PAA
complex, strengthened by increasing the temperature, was stabilised by hydrophobic
interaction [83].
Prevysh and co-workers [84] examined the effect of added salt on the stability of hydrogen
bonding IPC consisting of PAA and PEO or PVP. They found that addition of NaCl
resulted in interpolymer complex aggregation, on the contrary to what happens in the
case of PEC formed between oppositely charged polymers where addition of salt results
in the dissolution of the precipitate [85].
Due to the interpolymer complex formation between PMAA and PEG at low pH the
diffusional rates of solutes (drugs) in PMAA-graft-PEG hydrogels is much lower in acidic
than in neutral or basic media [86]. The pH dependent structural changes, attributed to
the formation/dissociation of hydrogen bonding IPC, have been studied as a function of
the copolymer composition and the PEG graft chain molecular weight. The largest degrees
of complexation were observed in gels containing nearly equimolar amounts of monomeric
units and the longest PEG grafts [87].
In many polymer blends hydrogen bonding interactions enhance miscibility in the solid
state, while interpolymer complexation is not probably detected in dilute solutions.
145
Handbook of Polymer Blends and Composites
PVAL and poly(VAC-co-VAL) are often used for the preparation of film composites and
membranes after mixing with other polymers. It seems that hydrogen bonding interactions
play an important role in these systems. The structure of film composites based on PVAL
and polyacids like CMC, PAA or PMAA, was studied using IR spectroscopy and dielectric
relaxation measurements. The electrophysical properties of the composites were shown
to be related to the formation of interpolymer hydrogen-bonding complexes [89]. The
specific interactions between poly(4-vinyl pyridine) (P4VP) and poly(VAC-co-VAL) were
studied by Fourier transform IR spectroscopy (FT-IR). Hydrogen bonding between the
two polymers led to significant spectral modification in hydroxyl, carbonyl and pyridine
ring spectral bands. Three equilibrium processes compete in the blends, hydroxyl-hydroxyl
(self-association), hydroxyl-carbonyl (self-association) and hydroxyl-pyridine (inter-
association). The first process was proved to be predominant [90]. Moreover, the
miscibility of P4VP and poly(2-vinylpyridine) (P2VP) with PVAL, poly(vinyl acetate)
(PVAC) and poly(VAC-co-VAL) was studied by using differential scanning calorimetry
(DSC) and FT-IR. The latter method was useful for the study of possible hydrogen bonding
interactions [91]. From the study of membranes of mixtures of poly(sodium styrene
sulfonate) (PNaSS) or PSSA with PVAL, cast from aqueous solutions, a partial miscibility
of the non-ionic polymer with the polyelectrolyte was concluded and it was attributed to
interactions between the -OH groups of PVAL and the ionic groups of PNaSS or PSSA
[92]. The miscibility of PVAL and PSSA in dilute aqueous solutions, studied by a
viscometric method, is favoured by increasing the molecular weight and the number of
residual acetate groups of PVAL [93].
The compatibility of PVAL/PVP blends has been studied by means of viscometric and
ultrasonic experiments [94]. The results obtained showed that the blends were miscible
over the entire composition range. The miscibility was attributed to hydrogen bonding
interaction between hydroxyl groups in the PVAL and the carbonyl group of PVP. PVAL/
PVP blends cast from aqueous solutions were further studied by DSC, FTIR spectroscopy
and X-ray photoelectron spectroscopy (XPS) [95]. A single glass transition temperature
(Tg) was observed for all the blends, suggesting that PVAL/PVP blends were miscible within
the whole composition range. As a result of the hydrogen bonding interaction between
PVAL and PVP the blends showed good miscibility in the bulk, even if, according to XPS
results, their surface was enriched with the lower surface energy component, PVAL. So
even if hydrogen bonding is the cause for miscibility in the bulk of PVAL/PVP blends, its
effect is not strong enough to prevent surface segregation of PVAL. From a DSC study of
blends of PVAL with different degrees of hydrolysis (88 and 99%) with PVP with different
molecular weights (10,000 and 360,000 g/mol), strong interaction between PVP and PVAL
was shown, while the density of interactions was larger in the case of the lower molecular
146
Water Soluble Polymer Systems - Phase Behaviour and Complex Formation
weight PVP and the PVAL with a lower degree of hydrolysis. The crystallinity of the blends
was decreased as the molecular weight of PVP was increased [96].
DSC measurements showed that PDMAM was immiscible with PVAC but was miscible
with PVAL and poly(VAC-co-VAL). IR spectroscopy studies revealed the existence of
specific interactions via hydrogen bonding between hydroxyl groups in vinyl alcohol
units and the carbonyl group in the tertiary amide, which appeared to be decisive for
miscibility [97].
Hydrogen bonding interaction has also been used to improve the interfacial bonding
between hydroxyapatite and polyactive 70/30, an ethylene glycol-butylene terephthalate
biodegradable block copolymer. Hydroxyapatite was first surface-modified by polyacrylic
acid or ethylene-co-maleic acid in aqueous solution. The surface-modified hydroxyapatite
was then used as filler in composites with polyactive 70/30. The strength and elastic
modulus of the composite were significantly improved. This improvement was attributed
to hydrogen bonding interactions between the polyacid chains and the polyethylene glycol
segments in the polymer matrix [99].
147
Handbook of Polymer Blends and Composites
A lot of work was done on the CS/PEG blends [109-111]. It was established that CS/PEG
aqueous solutions exhibit a pseudoplastic non-Newtonian behaviour described by the
Ostwald de Waele model, the rheological behaviour of CS/PEG mixture being determined
by CS [112].
Miscibility studies were also performed on CS with a deacetylation degree (DD) of 45-
55% in the mixture with PVAL [113] and it was found that the polymers show good
miscibility in the blends unlike the other chitosans of DD<40 or >60%. Nishio and co-
workers [114] performed miscibility studies of using PAAM, PDMAM, PVP, poly(N-
vinyl acetamide) (PNVA), poly(acryloyl morpholine) (PACMO) with CS. CS forms a
fairly miscible phase with PVP, PAAM and PNVA while CS/PDMAM or PACMO blends
exhibit immiscibility (because the Tg of the blends is not dependent on the blend
composition). In the CS/PVP blend, it was proved by nuclear magnetic resonance (NMR)
measurements that the upper limit of the heterogeneous domain size is < 2.5 nm, so the
blend is partially-miscible on the ~2.5 nm scale. For the CS/PAAM mixture the upper
limit of the heterogeneous phase is 1.9 nm [114]. They found that the number of hydrogen
atoms attached to the nitrogen in the amide group is not an essential factor controlling
the miscibility character of the polymer blends with chitosan. Miscibility [115] and
compatibility [116] investigations have been performed on the polysaccharides (dextran,
pulluan, HPC) and maleic acid copolymers with vinylacetate and styrene mixtures. It
was established that the blends with a higher than 85% content in polysaccharide present
an evenly fine particle dispersion morphology [117]. In the blends where one of the
partners presents LCST, i.e., HPC, hydrophobic interactions could improve the miscibility
of the components [118].
148
Water Soluble Polymer Systems - Phase Behaviour and Complex Formation
[120]. Veis [121] has made a study of polyelectrolyte interactions leading to phase
demixing, while compatibility, immiscibility and phase separation, complex coacervation
in low and high charge density complexes have been examined.
ii. Turbid colloidal systems with suspended complex particles (colloidal complexes);
iii. Homogeneous systems containing small complex aggregates (water soluble non
stoiciometric polyelectrolyte complexes - NPEC).
A 1:1 stoichiometry has been observed for the polysalt precipitates corresponding to a
random charge compensation between the ionic sites of the polyions [122, 123] and with
regard to supermolecular order of the polymer chains in a complex precipitate, the scrambled-
egg model, shown in Figure 5.4, has been considered to be closer to reality [122].
149
Handbook of Polymer Blends and Composites
There are two types of polycations, the pendant-type polycations which have charges in
the side groups and the integral-type polycations which have charges in the chain
backbone. Tsuchida and co-workers [127] observed that when a pendant-type polycation,
such as quaternised P4VP is used, an equimolar insoluble complex is formed with PNaSS.
On the other hand with an integral type polycation (-N+Cl-(CH3)2-(CH2)3-N+Cl-(CH3)2-
CH2-C6H4-CH2- = 3X) an equimolar insoluble complex is obtained only at [PNaSS]/
[3X] < 3, while at a polymer mixture composition [PNaSS]/[3X] > 3, a water soluble
complex should be formed with the composition [PNaSS]/[3X]=3.
When a weak polyelectrolyte is involved in a PEC formation, the dissociation state of the
weak polyelectrolyte directly affects the composition of the complex obtained. As a result
in the complexation of the polycation 3X with PAA, the composition r = [cation]/
[carboxylic acid] of the PEC formed, depends on the degree of neutralisation () of the
polycarboxylic acid. The value of r is 0.2 at = 0, increases with and becomes unity at
= 1 [34].
150
Water Soluble Polymer Systems - Phase Behaviour and Complex Formation
The SPEC are electroneutral, as the charges of the components are mutually neutralised,
and they precipitate from solution. They only can be swelled in water and could be used
as membranes [133].
The influence of the type of counterions and the concentration and type of added salt on
the stoichiometry has been examined in an early conductometric study of the reaction of
polystyrene sulfonates with the polyvinyl benzyl trimethylammonium [134]. Furthermore,
it has been shown that in addition to coulombic forces, the hydrophobic interaction
plays an important role in the complexation [135] as well as the chain length of the
polycation and the dissociation degree of the polyacid, when the polyanion is derived
from a weak polyacid, such as PAA, PMAA or poly(itaconic acid) [136]. A study on the
electrostatic interaction between poly(L-lysine) hydrobromide and poly(L-glutamic acid)
at various neutralisation degrees of the glutamic acid, showed that the binding process,
involving strong electrostatic interaction between the COO- and NH3+ groups, is strongly
cooperative [137].
Complexes formed between PMAA and P4VP or its methylvinylpyridinium methyl sulfate
copolymers have been studied by fluorescence quenching [144], while the fluorescence
151
Handbook of Polymer Blends and Composites
The thermodynamic parameters of the PEC formation reaction between CMC or partially
hydrolysed polyacrylamides of different charge densities with PDADMAC were
determined. CMC with low charge density reacted exothermally and the PEC formation
reaction was of the enthalpy dominant type, while for CMC with high charge density the
152
Water Soluble Polymer Systems - Phase Behaviour and Complex Formation
exothermic tendency was low and the PEC formation reaction was thus shown to be of
the entropy dominant type. The PEC formation reaction with the polyacrylamide systems
was slightly endothermic, while they gained large entropy during the PEC formation,
indicating that this reaction was of the entropy dominant type [154].
Water-insoluble SPEC redissolve upon salt addition. Trinh and Schnabel [155] studied
the redissolution of PEC when ionic strength, I, is increased. The critical ionic strength,
[I] crit , for redissolution was found to depend on the chemical nature of the
polyelectrolytes. Provided that the difference in [I]crit is sufficiently large, isolation of
charged polymers can be achieved by selective precipitation. The method was illustrated
by separating albumin at pH = 8 from polyphosphate via precipitation as PEC with
PNE4VPB. PEC of PNE4VPB with PNaSS and/or poly(sodium phosphate) (PNaP) were
prepared by precipitation upon mixing aqueous solutions of the corresponding
polyelectrolytes. Binary and ternary systems were studied and the chemical composition
of the PEC was determined as a function of the mixing ratio. The higher binding ability
of PNaSS as compared to that of PNaP was clearly demonstrated [156]. The stability
of PEC formed between the weak polyelectrolytes PNE4VPB and PNaMA has been
studied by time-resolved light scattering measurements performed in conjunction with
the stopped-flow method [157].
A complex formation study between weak polyanions and strong polycations with
cationic groups in the main chain showed an influence of the structure of the
complementary polymers and of their molecular weight. The greater the structural
differences, the higher the end point deviation from stoichiometry [158].
The response of PEC, formed between PNaSS or PNaMA and PDADMAC and its
copolymer with AM, to the addition of NaCl, was studied in relation to the anionic
group of the polyanion and the charge density of the polycation. The effects of
aggregation and macroscopic flocculation were discussed [161]. Schindler and
Nordmeier [160] studied the formation and aggregation of PEC formed between
PDADMAC and PSSA. The complexes were insoluble precipitates with a stoichiometry
of 1:1. The size of the aggregates increases with salt concentration, while a critical salt
concentration exists at which this increase becomes extremely abrupt.
PNaSS and PDADMAC have been deposited on spinning silicon wafers. The dependence
of polyelectrolyte multilayer thickness was evaluated. Film thickness is approximately
proportional to the number of layers and the salt concentration [161].
The complex formation between oppositely charged polyelectrolytes, such as PEI and
PDADMAC with maleic acid-propylene copolymers and maleic acid-styrene copolymers,
as well as their interaction with silicate powders were investigated [162].
153
Handbook of Polymer Blends and Composites
Complex formation between poly (sodium acrylate) PNaA or some copolymers of acrylic
acid with itaconic or maleic acid, used as anionic polymers, and cationic polyelectrolytes
with quaternary ammonium salt groups in the main chain, was studied [163].
Analytical ultracentrifugation was used for the determination of the complex composition
and characterisation of the particles of water insoluble polyelectrolyte complexes between
PNaSS and poly(DADMAC-co-AM) [165]. It was found that the complexes are
remarkably polydispersant and form non-stoichiometric particles including the major
component in considerable excess. These particles consist of a neutralised, relatively
compact core and a surrounding shell of the excess component, which stabilises the
particles by electrostatic repulsion. The stoichiometry depended strongly on the ionic
strength of the solution and the concentration of the components. Regimes of relative
monodispersity were found in water/salt media. PEC with the copolymer containing
50% DADMAC and AM form comparatively monodispersant species in aqueous salt
solution with sedimentation coefficients between 4 and 40 Sv. Deviations from the 1:1
stoichiometry increase with increasing amounts of AM in the polycation and with
increasing chain length of PNaSS.
Huglin and co-workers [132, 167] have studied the compexation of poly[sodium(2-
acrylamido-2-methyl propane sulfonate)] with poly(2-vinyl pyridinium chloride) and
poly(4-vinyl pyridinium chloride) (P4VPC) in dilute aqueous solution. Complex
154
Water Soluble Polymer Systems - Phase Behaviour and Complex Formation
composition departed from stoichiometry and depended on the order of mixing. Elemental
analysis of precipitates revealed that PEC were deficient in polyanionic units. The influence
of charge density on the stoichiometry of the polyelectrolyte complex reaction occurring
by mixing the cationic polymer P4VPC with poly(2-acrylamido-2-methyl propane
sulphonate-co-N,N-dimethylacrylamide) was also studied [168]. Insoluble SPEC were
formed by mixing 0.2 M aqueous solutions of PNaP with P4VPC. They were dissolved
by strong acids or bases and by weak acids [169].
Dautzenberg and co-workers [173] have shown that complex formation between PNaSS
and poly(DADMAC-co-AM) results in highly aggregated PEC particles, with an
aggregation number in the order of 103, while in highly dilute solutions it leads to quasi-
soluble particles on a colloidal level [122, 174]. Moreover the polyelectrolyte complex
formation in diluted aqueous solution was studied on the system PNaSS/PDADMAC
using static and dynamic light scattering [175] and it was shown that the polyelectrolyte
complex investigated exists as a highly polydisperse system of compact and nearly spherical
particles.
The effect of low molecular weight salt on the properties of PEC formed between
PDADMAC and copolymers of maleic acid with propylene or -methylstyrene has been
studied [176]. The effect of the ionic strength on the structure of the aggregates formed
between PDADMAC and the sodium salt of maleic acid/propene alternating copolymers,
155
Handbook of Polymer Blends and Composites
with a threefold excess of cationic groups, was studied [177]. The sedimentation coefficient
and hydrodynamic size of the aggregates increased with rising salt concentration, while
the electrophoretic mobility decreased gradually.
Stable and monodispersive polyion complex micelles were prepared in an aqueous medium
through electrostatic interaction between two oppositely charged block copolymers with
PEG segments: PEG-b-poly(L-lysine) and PEG-b-poly(,-aspartic acid) [178]. It was
shown that the polyion complex micelles were spherical particles in thermodynamic
equilibrium without any secondary aggregates.
The NPEC, containing an excess of one component, have a net charge of the same sign
as this excess component, are usually water soluble and can be treated as ordinary soluble
polyelectrolytes [179]. They can be obtained by a simple mixing of aqueous solutions of
oppositely charged polyelectrolytes taken in non-equivalent ratios. However, a number
of specific conditions must be met for the preparation of NPEC [127, 180]. It has been
shown that soluble products are formed only if the concentration of one polyelectrolyte,
P1, is higher than the concentration of the other, P2 [181]. P1, the polyelectrolyte included
in excess in the NPEC is referred as the host polyelectrolyte (HPE). P2, the polyelectrolyte
included in deficient amount is referred as the guest polyelectrolyte (GPE). Moreover, it
is also necessary for the solution to contain a certain amount of a low-molecular weight
electrolyte, for example 0.0020.1 M NaCl. In general, as the excess of the HPE is
larger, the concentration of the salt required for the formation of a water soluble NPEC
is smaller. Regarding the ratio = [GPE]/[HPE], it should be lower than a critical value,
c. If is higher than c, a soluble NPEC of the characteristic composition c and a
corresponding amount of an insoluble SPEC coexist in the reaction system. For most of
the systems investigated c varies between 0.2 and 0.5 depending on the polyelectrolytes
comprising the NPEC [134].
156
Water Soluble Polymer Systems - Phase Behaviour and Complex Formation
A study of the factors controlling the direction of the macro-substitution reaction of the
PNE4VPB polycation with the PNaMA and PNaP polyanions showed that these are the
polyanion chain length and the nature of the counterion (potassium, sodium or lithium) [185].
Zezin, Kabanov and co-workers have studied a large variety of water soluble NPEC,
such as complexes of PNaA with 5,6-ionene bromide [186], complexes of PMAA and
PAA with PNE4VPB [187, 188] and of PNaP or PAA with PDMAEM, ionene bromide
and poly(4-vinyl pyridinium bromide) (P4VPB) [189, 190]. The influence of the excess
ratio of hydrophobic or hydrophilic sequences of ionised groups not forming salt bonds
as well as the state of hydrophilic units on the degree of association of the complexes
PDMAEM/PNaP, PAA/ionene bromide, PAA/PDMAEM and PMAA/PNE4VPB) was
studied [191, 192]. The effect of the ratio of the degrees of polymerisation of the
components on the formation of NPEC between PAA and PDMAEM was investigated
[193]. A reaction mechanism based on the concept of a significant amount of defects
(loops) in the structure of the water soluble NPEC formed between PNaMA and P4VPB
has been proposed [194] and features of the phase transitions in aqueous salt solutions
of NPECs formed between PAA and PEI have been presented [195]. Moreover, exchange
reactions were studied in three-component mixtures of polyelectrolytes containing one
polyanion, PNaMA, and two polycations, PNE4VPB and 5,6-ionene bromide [196].
The kinetics and the mechanism of interpolyelectrolyte exchange was firstly studied for
the system of the oppositely charged polyelectrolytes PMAA and PNE4VPB in aqueous
salt solutions by a fluorescence quenching technique [197]. In another study the influence
of chain length of a competitive polyanion and of the nature of monovalent counterions
on the direction of the substitution reaction of polyelectrolyte complexes formed between
PNaMA or PNaP and PNE4VPB was investigated [198].
The influence of the nature of the counterions on the formation of NPEC of PAA (blocking
(guest) polyelectrolyte) with poly(N-ethyl-4-vinylpyridinium) salts (lyophilising (host)
polyelectrolyte) was investigated. It was shown that NPEC were formed only when the
degree of binding of the low molecular weight counterion was slight, otherwise SPEC
were formed. The stability of the NPEC depended on the degree of binding of the low
molecular weight counterions by the lyophilising polyelectrolyte [199].
157
Handbook of Polymer Blends and Composites
A study of the influence of salt on the behaviour in dilute aqueous solution of NPEC
showed that increase of the ionic strength resulted in a reduction of the complex particles
dimensions and of the second virial coefficients of the solutions. At a given salt
concentration, phase separation with a stoichiometric complex takes place [201]. The
effect of small ions, i.e., bromide, chloride, lithium, sodium, potassium and
tetramethylammonium, on the dissociation, ion binding and phase separation of the
NPEC formed between PMAA and PNE4VPB was studied by Kabanov and co-
workers [201].
A new family of mechanically reversible gels has been obtained by mixing an aqueous
solution of a NPEC of PNaMA and PNE4VPB containing a certain amount of covalent
links between the two oppositely charged polyelectrolytes, with an aqueous solution of
PKVS. Gelation is attributed to the partial replacement of the electrostatic contacts between
the polycation and the PMA anion in the original NPEC with those of polycation and
poly(vinyl sulfate) polyanion in the mixture [203]. A study of the behaviour of water
soluble NPEC containing lyophilising PMA anions and blocking PNE4VP cations in
sodium chloride solutions showed that as the degree of polymerisation of the polycation
decreased, the phase separation is prevented, whereas decreasing the charge of the
polycation, by decreasing the extent of alkylation, leads to phase separation, explained
by a strengthening of hydrophobic interactions in the NPEC particles [130].
A light scattering study of soluble PEC between PNaSS or PNaMA, used as polyanions,
and PDADMAC or a copolymer of DADMAC with AM containing 47 mol% DADMAC,
used as polycations, showed that very small amounts of sodium chloride lead to a drastic
decrease of the level of aggregation, while a higher ionic strength results in macroscopic
flocculation [206].
158
Water Soluble Polymer Systems - Phase Behaviour and Complex Formation
The formation of PEC between PMA or dextrans functionalised with anionic groups,
both polyanions being labelled with fluorescent pyrenyl groups, and PNE4VP as a
quencher in comparison with complex formation with highly sulphated or sulphonated
polyanion competitors, like heparin and polystyrene sulfonate, has been studied by means
of the fluorescence quenching technique. The data obtained showed that both the chemical
nature of the negatively charged groups and the charge density of the polyanion chains
control the equilibrium of competitive interpolyelectrolyte reactions [207].
A mathematical model describing water soluble NPEC has been recently presented [209].
It is a mixture of the Madelungs theory for ionic crystals and of the Mannings counterion
condensation theory. The predictions are compared with experimental results obtained
with non-stoichiometric complexes formed between ionene and PAA or PMAA. The
central parameters are the degree of complexation and the degree of counterion binding.
In a recent study Osada and co-workers [210] showed that the complexes formed between
the zwitterionic polymer, poly[3-dimethyl(methacryloyloxyethyl) ammonium propane
sulfonate] (PDMAPS) and the anionic polymer poly(2-acrylamido-2-methyl propane
sulfonic acid) (PAMPSA), or the cationic polymers poly-3-acryloylamino propyl trimethyl
ammonium chloride and the ionene bromides are soluble and exhibit an UCST behaviour,
dependent on the molar ratio. The PDMAPS/PAMPSA complex exhibits a dramatic
increase in viscosity due to a network structure through electrostatic interaction.
159
Handbook of Polymer Blends and Composites
The effect of the ionic strength and pH on the complex formation between BSA and
PNE4VPB [218] or copolymers of P4VP with 4-vinyl-N-cetylpiridinium bromide has
been studied [219]. The binding of haemoglobin with dextran sulfate and
diethylaminoethyl dextran has been studied. The results indicate that a complex is formed,
either soluble or insoluble depending on pH. Binding appears to be stronger for the
haemoglobin-dextran sulfate compounds [220].
A study of the complexation between potassium poly(vinyl alcohol) sulfate (KPVAS) and
papain, human serum albumin, lysozyme, ribonuclease, trypsin and pepsin showed that
in a salt free solution at pH = 2 electrically neutral protein-polyelectrolyte complexes
(aggregates) with a uniform size were formed from all the proteins, other than pepsin [221].
160
Water Soluble Polymer Systems - Phase Behaviour and Complex Formation
mixtures has been recently studied [224]. These reversible gels exhibited similarities with
chemically cross-linked macromolecules.
A complexation study of human serum albumin, hemoglobin and bovin trypsin with
PDADMAC and KPVS in salt-free conditions, showed that complexes with a 1:1 charge
stoichiometry are formed and that changes in the conformations of the protein molecules
caused by the complexation are not so large as to lose their biochemical capacity [225].
The interaction between three globular proteins of substantially different IEP (BSA, bovine
pancreas ribonuclease, and chicken egg lysozyme) and a number of synthetic cationic
and anionic polyelectrolytes of different charge densities was studied in dilute aqueous
solution. For each polyion-protein pair, there is a well-defined critical pH at which binding
starts by the formation of a soluble complex, while a further pH change produces phase
separation (complex coacervation) [129, 226-231]. In a study of the interaction of PAA
with egg white proteins (EWP) and lysozyme the following results were obtained: for
EWP precipitation, larger molecular weight PAA had a higher efficiency in protein removal,
while lysozyme removal was largely independent of PAA molecular weight, indicating
that the high positive charge density of this protein is sufficient to ensure strong
electrostatic binding. Particle size was also affected. Larger molecular weight PAA gave
larger precipitates. For turbidimetric titrations a larger molecular weight PAA gave higher
critical pH values [232]. Coacervation induced by pH in complexes of BSA and
PDADMAC has been studied. The state of macromolecular assembly of complexes formed
prior to and during the pH-induced coacervation could be characterised by specific pH
values at which recognisable transitions took place [233].
Patrickios and co-workers [234] studied the interaction of dilute solutions of synthetic
low-molecular weight block acrylic polyampholytes with proteins by turbidimetric
titration. As in the homopolyelectrolyte-protein systems, the onset of interaction is
manifested by a large increase in turbidity at a certain pH lying between the
polyampholytes self-aggregation pH and the proteins isoelectric point. Increasing salt
concentration suppresses the protein-polyampholyte and polyampholyte-polyampholyte
interactions, suggesting that the main driving force in these phenomena is electrostatic.
The interactions between soluble collagen from calfskin and PAA were studied. In the 2.5-
4.0 pH range PEC between the oppositely charged polymers are formed and precipitate.
Blending of collagen with PAA allows the possibility of producing bioartificial materials
with good hydrolytic stability and better physicochemical and mechanical properties [236].
161
Handbook of Polymer Blends and Composites
References
1. Water Soluble Resins, 2nd Edition, Eds., R.L. Davidson and M. Sittig, Van
Norstrand Reinhold, New York, NY, USA, 1968.
2. Handbook of Water Soluble Gums and Resins, Ed., R.L. Davidson, McGraw-
Hill, New York, NY, USA, 1980.
5. E.A. Bekturov and Z.K. Bakauova, Synthetic Water Soluble Polymers in Solution,
Huthig and Wepf Verlag, Basel, Switzerland, 1986.
10. Intermolecular and Surface Forces, Ed., J. Israelachvili, Academic Press, New
York, NY, USA, 1991.
162
Water Soluble Polymer Systems - Phase Behaviour and Complex Formation
12. R. Buscall and T. Corner, European Polymer Journal, 1982, 18, 967.
13. P.J. Flory, Principles of Polymer Chemistry, Cornell University Press, Ithaca, NY,
USA, 1953.
14. P-G. de Gennes, Scaling Concepts in Polymer Physics, Cornell University Press,
Ithaca, NY, USA, 1979.
15. A. Kanda, M. Duval, D. Sarazin and J. Francois, Polymer, 1985, 26, 406.
18. S. Saeki, N. Kuwahara, M. Nakata and M. Kaneko, Polymer, 1976, 17, 685.
19. R.A. Horne, J.P. Almeida, A.F. Day and N-T. Yu, Journal of Colloid and Interface
Science, 1971, 35, 77.
21. E.K. Just and T.G. Majewicz in Encyclopedia of Polymer Science and
Engineering, Eds. H.F. Mark, N.M. Bikales, C.G. Overberger and G. Menges,
John Wiley & Sons, New York, 1985, 3, 248.
23. E. Dickinson in Gums and Stabilisers for the Food Industry 4, Eds., G.O.
Phillips, D.J. Wedlock and P.A. Williams, IRL Press, Oxford, UK, 1987, 249.
24. L. Piculell and B. Lindman, Advances in Colloid and Interface Science, 1992, 41,
149.
25. L. Piculell, K. Bergfeldt and S. Nilsson, Biopolymer Mixtures, Eds., S.E. Harding,
S.E. Hill and J.R. Mitchell, Nottingham University Press, Nottingham, UK, 1995, 13.
163
Handbook of Polymer Blends and Composites
27. I. Iliopoulos, D. Frugier and R. Audebert, Polymer Preprints, 1989, 30, 371.
28. M.B. Perrau, I. Iliopoulos and R. Audebert, Polymer, 1989, 30, 2112.
29. P-A. Albertsson, Partition of Cell Particles and Macromolecules: Separation and
Purification of Biomolecules, Cell Organelles, Membranes and Cells in Aqueous
Polymer Two-Phase Systems and their Use in Biochemical Analysis and
Biotechnology, 3rd Edition, Wiley, New York, NY, USA, 1986.
30. V.A. Kabanov and I.M. Papisov, Polymer Science USSR, 1979, 21, 261.
31. E.A. Bekturov and L.A. Bimendina, Advances in Polymer Science, 1981, 41, 99.
38. V.Y. Baranovsky, S. Shenkov and G. Borisov, European Polymer Journal, 1993,
29, 8, 1137.
39. H. Ohno, K. Abe and E. Tsuchida, Die Makromolekulare Chemie, 1978, 179, 3,
755.
40. G. Staikos and K. Tsitsilianis, Journal of Applied Polymer Science, 1991, 42, 3, 867.
164
Water Soluble Polymer Systems - Phase Behaviour and Complex Formation
44. I. Iliopoulos, J.L. Halary and R. Audebert, Journal of Polymer Science: Polymer
Chemistry Edition, 1988, 26, 1, 275.
45. I. Iliopoulos and R. Audebert, European Polymer Journal, 1988, 24, 2, 171.
46. S. Shenkov and V.Y. Baranovsky, Journal of Polymer Science: Polymer Chemistry
Edition, 1994, 32, 1385.
47. Y. Osada and M. Sato, Journal of Polymer Science: Polymer Letters Edition,
1976, 14, 129.
48. Y. Osada, Journal of Polymer Science: Polymer Chemistry Edition, 1979, 17,
3485.
51. L.A. Bimendina, V.V. Roganov and E.A. Bekturov, Journal of Polymer Science:
Polymer Symposia Edition, 1974, 44, 65.
52. S.K. Chatterjee, A. Malhotra and D. Yadov, Journal of Polymer Science, Polymer
Chemistry Edition, 1984, 22, 12, 3697.
53. D.V. Subotic, J. Ferguson and B.C.H. Warren, European Polymer Journal, 1991,
27, 1, 61.
56. H.T. Oyama, W.T. Tang and C.W. Frank, Macromolecules, 1987, 20, 8, 1839.
58. K. Sivadasan, P. Somasundaran and N.J. Turro, Colloid and Polymer Science,
1991, 269, 2, 131.
59. M. Koussathana, P. Lianos and G. Staikos, Macromolecules, 1997, 30, 25, 7798.
165
Handbook of Polymer Blends and Composites
60. H. Ohno, K. Abe and E. Tsuchida, Die Makromolekulare Chemie, 1978, 179, 3,
755.
61. K. Abe, H. Ohno, A. Nii and E. Tsuchida, Die Makromolekulare Chemie, 1978,
179, 8, 2043.
63. A. Usaitis, S.L. Maunu and H. Tenhu, European Polymer Journal, 1997, 33, 2,
219.
65. G. Staikos and G. Bokias, Die Makromolekulare Chemie, 1991, 192, 2649.
66. L.A. Bimendina, E.A. Bekturov, G.S. Tleubaeva and V.A. Frolova, Journal of
Polymer Science: Polymer Symposia Edition, 1979, 66, 9.
67. A. Leiva, L. Gargallo and D. Radic, Polymer International, 1994, 34, 4, 393.
69. S.K. Chatterjee, A. Malhotra and D. Yadav, Journal of Polymer Science, Polymer
Chemistry Edition, 1984, 22, 12, 3697.
70. S.K. Chatterjee, D. Yadav, A.M. Khan and K.R. Sethi, Angewandte
Makromolekulare Chemie, 1989, 171, 191.
71. S.K. Chatterjee and N. Misra, Macromolecular Chemistry and Physics, 1996,
197, 12, 4193.
166
Water Soluble Polymer Systems - Phase Behaviour and Complex Formation
78. N.S. Karibyants, O.E. Filipova and S.G. Starodubtsev, Polymer Science Series B,
1995, 37, 7-8, 385.
79. S.E. Kudaibergenov, Z.S. Nurkeeva and G.A. Mun, Macromolecular Chemistry
and Physics, 1995, 196, 7, 2203.
80. S.E. Kudaibergenov, Z.S. Nurkeeva, G.A. Mun, V.V. Khutoryanskiy, and A.D.
Gazizov, Polymer Advances for Technologies, 2000, 11, 1, 15.
81. M. Koussathana, P. Lianos and G. Staikos, Macromolecules, 1997, 30, 25, 7798.
84. V.A. Prevysh, B.C. Wang and R.J. Spontak, Colloid and Polymer Science, 1996,
274, 6, 532.
85. C.K. Trinh and W. Schnabel, Macromolecular Chemistry and Physics, 1997, 198,
5, 1319.
86. A.M. Lowman and N.A. Peppas, Journal of Biomaterials Science: Polymer
Edition, 1999, 10, 9, 999.
87. A.M. Lowman and N.A. Peppas, Polymer, 2000, 41, 1, 73.
88. M. Jiang, M. Li, M. Xiang and H. Zhou, Advances in Polymer Science, 1999,
146, 122.
89. T.G. Lazareva, I.A. Ilyushchenco and I.F. Alimov, Polymer Science Series A,
1994, 36, 9, 1231.
90. L.C. Cesteros, J.R. Isasi and I. Katime, Macromolecules, 1993, 26, 26, 7256.
91. L.C. Cesteros, J.R. Isasi and I. Katime, Journal of Polymer Science: Polymer
Physics Edition, 1994, 32, 2, 223.
167
Handbook of Polymer Blends and Composites
92. M.O. David, and T.Q. Nguyen, European Polymer Journal, 1994, 30, 9, 1013.
94. R.V. Rao and P. Iatha, Journal of Materials Science Letters, 1999, 18, 457.
95. L. Li, C-M. Chan and L-T. Weng, Polymer, 1998, 39, 11, 2355.
96. S.N. Cassu and M.I. Felisberti, Polymer, 1997, 38, 15, 3907.
97. L.G. Parada, L.C. Cesteros, E. Meaurio and E. Katime, Polymer, 1998, 39, 5,
1019.
98. S-A. Chen and M-Y. Hua, Macromolecules, 1996, 29, 14, 4915.
99. Q. Liu, J.R. De Wijn, D. Bakker and C.A. Van Blitterswijk, Journal of Materials
Science: Materials in Medicine, 1996, 7, 9, 551.
100. W-M. Kulicke, D. Eidam, F. Kath, M. Kix and A.H. Kull, Starch, 1996, 48, 3,
105-114.
101. M. Alloncle, J. Lefebvre, G. Llamas and J.L. Doublier, Cereal Chemistry, 1989,
66, 90.
102. M. Yoshimura, T. Takaya and K. Nishinari, Carbohydrate Polymers, 1998, 35, 71.
103. M.T. Kalichevsky, P.D. Orford and S.G. Ring, Carbohydrate Polymers, 1986, 6, 145.
104. M.T. Kalichevsky and S.G. Ring, Carbohydrate Research, 1987, 162, 323.
105. P. Annable, M.G. Fitton, B. Harris, G.O. Phillips and P.A. Williams, Food
Hydrocolloids, 1994, 8,351.
106. B. Conde-Petit, A. Pfirter and F. Escher, Food Hydrocolloids 1997, 11, 393.
107. D. Yumin, X. Zuyong and R. Lu, Journal of Natural Science, 1997, 2, 2, 220.
108. S.S. Rashidova, N.L. Voropaeva, T.D. Kalantarova and A.A. Tager,
Vysokomolekulyarnye Soedineniya Seriya A and Seriya B, 1997, 39, 3, 556.
109. W.H. Jiang and S.J. Han, European Polymer Journal, 1999, 35, 11, 2079.
110. A. Nikolova, N. Manolova and I. Rashkov, Polymer Bulletin, 1998, 41, 1, 115.
168
Water Soluble Polymer Systems - Phase Behaviour and Complex Formation
111. J. Piekielna and M. Mucha, Inzynieria Chemiczna Procesowa, 1998, 19, 1, 145.
112. A. Nikolova, N. Manolova and I. Rashkov, Polymer Bulletin, 1998, 41, 1, 115.
115. G.G Bumbu, C.G Chitanu, A. Carpov, H. Darie and C. Vasile, Proceedings of the
4th International Conference on Hydrocolloids, 1998, Osaka, Japan, p.171.
116. A. Carpov, G.G. Bumbu, G.G. Chitanu, H. Darie and C. Vasile, Proceedings of
Fifth Mediterranean School on Science and Technology of Advanced Polymer
Based Materials, Capri Naples, Italy, 1997, 513.
117. G.G. Bumbu, C. Vasile, G.C. Chitanu and A. Carpov, Polymer Degradation and
Stability, 2001, 72, 1, 99.
118. G.G. Bumbu, G. Staikos, C. Vasile and G.C. Chitanu, unpublished data
119. A.S. Michaels and R.G. Miekka, Journal of Physical Chemistry, 1961, 65, 10,
1765.
120. M.J. Lysacht in Ionic Polymers, Ed., L. Holliday, Applied Science, London, UK,
1975, 281.
121. A. Veis in Developments in Ionic Polymers, Eds., A.D. Wilson and H.J. Prosser,
Applied Science Publishers Ltd., Barking, UK, 1983, 293.
123. A.S. Michaels, Industrial Engineering and Chemistry, 1965, 57, 10, 32.
124. A.S. Michaels, L. Mir and N.S. Schneider, Journal of Physical Chemistry, 1965,
69, 5, 1447.
125. J. Ktz, K-J. Linow, B. Philipp, L.P. Hu and O. Vogl, Polymer, 1986, 27, 10,
1574.
169
Handbook of Polymer Blends and Composites
128. H. Dautzenberg, Y. Gao and M. Hahn, Langmuir, 2000, 16, 23, 9070.
129. J. Xia, P.L. Dubin, Y. Morishima, T. Sato and B.B. Muhoberac, Biopolymers,
1994, 35, 411.
130. V.A. Izumrudov and S.K. Lim, Polymer Science Series A, 1998, 40, 3, 276.
131. Y. Itoh, K. Negishi, E. Iizuka, K. Abe and M. Kaneko, Polymer, 1992, 33, 14, 3016.
132. M.B. Huglin, L. Webster and I.D. Robb, Polymer, 1996, 37, 7, 1211.
133. V.A. Kabanov and A.B. Zezin, Pure and Applied Chemistry, 1984, 56, 3, 343.
134. A.S. Michaels, L. Mir and N.S. Schneider, Journal of Physical Chemistry, 1965,
69, 5, 1447.
135. E. Tsuchida and Y. Osada, Die Makromolekulare Chemie, 1974, 175, 2, 593.
136. E. Tsuchida, Y. Osada and K. Abe, Die Makromolekulare Chemie, 1974, 175, 2,
583.
139. K. Abe, H. Ohno and E. Tsuchida, Die Makromolekulare Chemie, 1977, 178, 8,
2285.
140. J.C. Salamone, S. Poulin, A.C. Watterson and A.P. Olson, Polymer, 1979, 20, 5,
611.
142. Y. Hirai and T. Nakajima, Journal of Applied Polymer Science, 1988, 35, 1325.
143. Y. Hirai and T. Nakajima, Journal of Macromolecular Science, 1989, 26, 10, 1465.
144. N.R. Pavlova, Y. E. Kirsh and V.A. Kabanov, Polymer Science USSR, 1979, 21, 9,
2276.
170
Water Soluble Polymer Systems - Phase Behaviour and Complex Formation
146. A.V. Kharenco, E.A. Neverova, R.I. Kalyuzhnaya, A.B. Zezin and V.A. Kabanov,
Polymer Science USSR, 1981, 23, 9, 2268.
147. E.N. Yermakova, P.V. Nuss, V.A. Kasaikin, A.B. Zezin and V.A. Kabanov,
Polymer Science USSR, 1983, 25, 7, 1605.
148. L.N. Yermakova, Y. G. Frolov, V.A. Kasaikin, A.B. Zezin and V.A. Kabanov,
Polymer Science USSR, 1981, 23, 10, 2529.
150. J. Ktz, J. Kunze, K.J. Linow and B. Philipp, Polymer Bulletin, 1986, 15, 3, 247.
155. C.K. Trinh and W. Schnabel, Angewandte Makromolekulare Chemie, 1993, 212,
167.
156. C.K. Trinh and W. Schnabel, Angewandte Makromolekulare Chemie, 1994, 221,
127.
157. C.K. Trinh and W. Schnabel, Macromolecular Chemistry and Physics, 1997, 198,
1319.
158. S. Dragan, M. Cristea, C. Luca and B.C. Simionescu, Journal of Polymer Science:
Polymer Chemistry Edition, 1996, 34, 3485.
171
Handbook of Polymer Blends and Composites
160. T. Schindler and E. Nordmeier, Polymer Journal, 1994, 26, 10, 1124.
161. S.T. Dubas and J.B. Schlenoff, Macromolecules, 1999, 32, 24, 8153.
163. S. Dragan, M. Cristea, C. Luca and B.C. Simionescu, Journal of Polymer Science:
Polymer Chemistry Edition, 1996, 34, 17, 3485.
167. L. Webster, M.B. Huglin and I.D. Robb, Polymer, 1997, 38, 6, 1373.
168. L. Webster and M.B. Huglin, European Polymer Journal, 1997, 33, 7, 1173.
169. N. Acar, M.B. Huglin and T. Tulun, Polymer, 1999, 40, 23, 6429.
172. E.A. Bekturov, S.E. Kudaibergenov, R.E. Khamzamulina, V.A. Florova, D.E.
Nurgalieva, R.C. Schulz and J. Zoeller, Die Makromolekulare Chemie, Rapid
Communications, 1992, 13, 4, 225.
173. H. Dautzenberg and G. Rother, Journal of Polymer Science: Physics Edition Part
B, 1988, 76, 353.
172
Water Soluble Polymer Systems - Phase Behaviour and Complex Formation
177. D.V. Pergushov, H-M. Buchammer and K. Lunkvitz, Colloid and Polymer
Science, 1999, 277, 2-3, 101.
179. E. Tsuchida and K. Abe, Developments in Ionic Polymers-2, Eds., A.D. Wilson
and H.J. Prosser, Elsevier Applied Science Publishers Ltd, Barking, UK, 1986,
191.
180. V.A. Kasaikin, O.A. Kharenco, A.V. Kharenco, A.B. Zezin and V.A. Kabanov,
Vysokomolekulyarnya Soedineiya, 1979, B21, 84.
181. O.A. Kharenco, A.V. Kharenco, R.I. Kalyuzhnaya, V.A. Izumrudov, V.A.
Kasaikin, A.B. Zezin and V.A. Kabanov, Vysokomolekulyarnya Soedineiya, 1979,
A21, 2719.
182. V.A. Kabanov, A.B. Zezin, V.A. Izumrudov, T.K. Bronich and K.N. Bakeev, Die
Makromolekulare Chemie, Supplement, 1985, 13, 137.
183. K.N. Bakeev, V.A. Izumrudov, S.I. Kuchanov, A.B. Zezin and V.A. Kabanov,
Macromolecules, 1992, 25, 4249.
184. Z.G. Gulyaeva, M.F. Zansokhova, I.V. Chernov, V.B. Rogacheva, A.B. Zezin and
V.A. Kabanov, Polymer Science, Series A, 1997, 39, 2, 213.
185. V.A. Izumrudov, T.K. Bronich, O.S. Saburova, A.B. Zezin and V.A. Kabanov, Die
Makromolekulare Chemie, Rapid Communications, 1988, 9, 1, 7.
186. Z.G. Gulyaeva, O.A. Poletaeva, A.A. Kalechev, V.A. Kasaikin and A.B. Zezin,
Polymer Science USSR, 1976, 18, 12, 3204.
187. V.A. Izumrudov and A.B. Zezin, Polymer Science USSR, 1976, 18, 11, 2840.
188. V.A. Izumrudov, V.A. Kasaikin, L.N. Yermakova and A.B. Zezin, Polymer
Science USSR, 1978, 20, 2, 452.
189. O.A. Kharenco, A.V. Kharenco, R.I. Kalyuzhnaya, V.A. Izumrudov, V.A.
Kasaikin, A.B. Zezin and V.A. Kabanov, Polymer Science USSR, 1979, 21, 12,
3002.
173
Handbook of Polymer Blends and Composites
190. O.A. Kharenco, A.V. Kharenco, V.A. Kasaikin, A.B. Zezin and V.A. Kabanov,
Polymer Science USSR, 1979, 21, 12, 3009.
191. O.A. Kharenco, V.A. Izumrudov, A.V. Kharenco, V.A. Kasaikin, A.B. Zezin and
V.A. Kabanov, Polymer Science USSR, 1980, 22, 1, 274.
192. V.A. Kabanov and A.B. Zezin, Makromolekulare Chemie Supplement, 1984, 185,
6, 259.
193. A.B. Zezin, V.A. Kasaikin, N.M. Kabanov, O.A. Kharenco and V.A. Kabanov,
Polymer Science USSR, 1984, 26, 7, 1702.
194. V.A. Izumrudov, A.B. Zezin and V.A. Kabanov, Polymer Science USSR, 1983, 25,
9, 2296.
195. V.B. Rogacheva, S.V. Ryzkhikov, A.B. Zezin and V.A. Kabanov, Polymer Science
USSR, 1984, 26, 8, 1872.
196. V.A. Izumrudov, A.P. Savitskii, A.B. Zezin and V.A. Kabanov, Polymer Science
USSR, 1984, 26, 8, 1930.
197. V.A. Izumrudov, A.P. Savitskii, K.N. Bakeev, A.B. Zezin and V.A. Kabanov, Die
Makromolekulare Chemie, Rapid Communications, 1984, 5, 709.
198. V.A. Izumrudov, T.K. Bronich, O.S. Saburova, A.B. Zezin and V.A. Kabanov, Die
Makromolekulare Chemie, Rapid Communications, 1988, 9, 7.
199. N.K. Nefedov, T.G. Yermakova, V.A. Kasaikin, A.B. Zezin and V.A. Lopyrev,
Polymer Science USSR, 1985, 27, 7, 1677.
200. V.A. Izumrudov, O.A. Kharenco, A.V. Kharenco, Z.G. Gulyaeva, V.A. Kasaikin,
A.B. Zezin and V.A. Kabanov, Polymer Science USSR, 1980, 22, 3, 767.
202. D.V. Pergushov, V.A. Izumrudov, A.B. Zezin and V.A. Kabanov, Polymer Science,
Series A, 1995, 37, 10, 1081.
203. V.A. Kabanov, A.B. Zezin, V.A. Izumrudov T. Bronich, N.M. Kabanov and O.V.
Listova, Die Makromolekulare Chemie, Macromolecular Symposia, 1990, 39, 155.
204. V.A. Kabanov, A.V. Kabanov and I.N. Astafyeva, Polymer Preprints, 1991, 32, 1,
592.
174
Water Soluble Polymer Systems - Phase Behaviour and Complex Formation
205. A.V. Kabanov, S.V. Vinogradove, Y.G. Suldaltseva and V.Y. Alakhov,
Bioconjugate Chemistry, 1995, 6, 639.
208. V.A. Kabanov, A.B. Zezin, V.B. Rogacheva, Z. G. Gulyaeva, M.F. Zansochova,
J.G.H. Joosten and J. Brackman, Macromolecules, 1999, 32, 6, 1904.
209. E. Nordmeier and P. Beyer, Journal of Polymer Science: Polymer Physics Edition,
1999, 37, 4, 335.
210. L. Chen, Y. Honma, T. Mizutani, D-J. Liaw, J.P. Gong and Y. Osada, Polymer,
2000, 41,1, 141.
211. J.Y. Gao and P.L. Dubin, Biopolymers, 1999, 49, 185.
213. J. Xia, P.L. Dubin and E. Kokufuta, Macromolecules, 1993, 26, 24, 6688.
214. P.L. Dubin, T.D. Ross, I. Sharma and B. Yegerlehner in Ordered Media in
Chemical Separations, Eds., W.L. Hinze and D.W. Armstrong, ACS Symposium
Series No.342, Washington, DC, USA, 1987, 162.
215. F. Petit, R. Audebert and I. Iliopoulos, Journal of Colloid and Polymer Science,
1995, 273, 777.
216. J. Xia and P.L. Dubin in Macromolecular Complexes in Chemistry and Biology,
Eds., P.L. Dubin, J. Bock, R.M. Davis, D.N. Schulz and C. Thies, Springer-
Verlang, Berlin, Germany, 1994, 247.
217. J.Y. Gao, P.L. Dubin and B.B. Muhoberac, Journal of Physical Chemistry, 1998,
102, 28, 5529.
218. V.A. Kabanov and M.I. Mustafaev, Polymer Science USSR, 1981, 23, 2, 280.
219. V.A. Kabanov, M.I. Mustafaev and V.V. Goncharov, Polymer Science USSR,
1981, 23, 2, 287.
175
Handbook of Polymer Blends and Composites
221. A. Tsuboi, T. Izumi, M. Hirata, J. Xia, P.L. Dubin and E. Kokufuta, Langmuir,
1996, 12, 26, 6295.
222. C. Tribet, R. Audebert and J-L. Popot, Langmuir, 1997, 13, 21, 5570.
224. R. Borrega, C. Tribet and R. Audebert, Macromolecules, 1999, 32, 23, 7798.
226. J.M. Park, B.B. Muhoberac, P.L. Dubin and J. Xia, Macromolecules, 1992, 25, 1,
290.
227. J. Xia, P.L. Dubin, Y. Kim, B.B. Muhoberac and V.J. Klimkowski, Journal of
Physical Chemistry, 1993, 97, 17, 4528.
229. L.S. Ahmed, J. Xia and P.L. Dubin, Journal of Macromolecular Science - Pure
and Applied Chemistry, 1994, A31, 1, 17.
230. J. Xia, P.L. Dubin, L.S. Ahmed and E. Kokufuta in Macro-Ion Characterisation,
From Dilute Solutions to Complex Fluids, Ed. K.S. Schmitz, ACS Symposium
Series No.548, Washington, DC, USA, 1994, 225.
231. Y. Li, K.W. Mattison, P.L. Dubin, H.A. Havel and S.L. Edwards, Biopolymers,
1996, 38, 527.
232. J-Y. Shieh and C.E. Glatz in Macromolecular Complexes in Chemistry and
Biology, Eds., P.L. Dubin and J. Bock, R. Davis, D.N. Schulz, C. Thies and D.
Schulz, Springer-Verlag, Berlin, 1994, 16, 273.
233. K. Kaibata, T. Okazaki, H.B. Bohidar and P.L. Dubin, Biomacromolecules, 2000,
1, 1, 100.
234. C.S. Patrickios, C.J. Jang, W.R. Herter and T.A. Hatton in Macro-Ion
Characterisation, From Dilute Solutions to Complex Fluids, Ed., K.S. Schmitz,
ACS Symposium Series No. 548, Washington, DC, USA, 1994, 257.
235. J.H.E. Hone, A.M. Howe and T. Cosgrove, Macromolecules, 2000, 33, 4,1199.
176
Water Soluble Polymer Systems - Phase Behaviour and Complex Formation
237. O.V. Kargina, O.V. Prazdnichnaya, Polymer Science, Series A, 1993, 35, 4, 561.
238. O.V. Kargina, O.V. Prazdnichnaya, I.D. Yurgens and E.Y. Badina, Polymer
Science, Series A, 1996, 38, 8, 937.
239. O.V. Kargina, E.Y. Badina, O.V. Prazdnichnaya and I.D. Yurgens, Polymer
Science, Series B, 1999, 41, 11-12, 331.
177
Water Soluble Polymer Systems - Applications of Interpolymer Complexes and Blends
6
Water Soluble Polymer Systems - Applications
of Interpolymer Complexes and Blends
6.1 Introduction
Water soluble polymers, being environmentally friendly substances, show good potential
for further development in industrial, agricultural and medical applications, either as
themselves or in blends. Blends based on water soluble polymers have been prepared to
improve certain of their properties used in various applications or to obtain new properties.
This chapter is not an exhaustive presentation of all water soluble polymer systems studied
in the literature. An outline of the specific processing and characteristics of such blends
in their main field of applications is presented.
179
Handbook of Polymer Blends and Composites
NPEC have been investigated as potential drug carrier systems for parenteral
administration. Complexes between the polycationic quaternised polyvinyl imidazole
and an excess of a higher molecular weight partially sulphonated dextran were designed
to inherit the biocompatible properties of dextran. Platelet aggregation studies showed
that toxic aggregatory effects normally induced by the polycations were eliminated in
vitro when they were a component of a soluble PEC. However this complexation was
not sufficient to prevent un-wanted interactions in vivo, leading to removal of the PEC
from the circulation [8]. In another study complexing of the DNA plasmid with poly(N-
ethyl-4-vinyl pyridinium) polycations results in the formation of a membrane active
polycomplex that is much more active in transformation of competent cells than pure
plasmid [9]. A PEC, comprised of poly(diallyldimethylammonium chloride) (PDADMAC)
and sodium cellulose sulfate, has been used for the microencapsulation of gonadotropin
releasing hormone [10].
The interactions of proteins with polyelectrolytes are important in natural biological systems
as well as in biotechnological applications. Protein-polyelectrolyte complex coacervation
provides an alternative to conventional protein separation and purification methods [11,
12]. The efficiency of separation was examined using the cationic polyelectrolyte PDADMAC
and the proteins bovine serum albumin, -lactoglobulin, -globulin and ribonuclease A
[13]. The existence of soluble protein-polyelectrolyte complexes makes polyelectrolytes
potentially important enzyme carriers, in which the enzyme activity could be controlled
[14]. Immobilisation of other active substances on PEC has been also studied. Complexation
of flavin-containing polycations to poly(methacrylic acid) (PMAA) influences favourably
the catalytic activity of the immobilised flavin units [15]. Drug delivery could be another
use of such systems [16, 17].
PEC have been used for the preparation of membranes [18, 19] of high water flux and
good selectivity. Combinations of anionic polyelectrolytes like PMAA or polystyrene
sulfonic acid (PSSA) with cationic ones like poly(vinyl benzyl trimethylammoniun chloride)
have been used. Moreover, the interfacial reaction between aqueous solutions of an ionic
and a cationic polyelectrolyte has been proposed for the microencapsulation of biologically
180
Water Soluble Polymer Systems - Applications of Interpolymer Complexes and Blends
active substances [20]. PEC formed between sodium cellulose sulfate and PDADMAC
have been used for encapsulation of artificial tissues [21]. A membrane consisted of
poly(maleic acid-co-styrene)polyethylene glycol (PEG) IPC has been used for the selective
permeation of carbon dioxide and oxygen [22]. Composite membranes containing a
separating layer of PEC consisting of polyacrylic acid (PAA) and a polycation were
developed for the separation of water-ethanol by pervaporation. Among the polycations,
ionenes with quaternary ammonium groups in the backbone chain were effective in giving
membranes of higher permselectivities [23].
PEC formed between a piperidinium cationic polymer and various polyanions, such as
polysodium acrylate (PNaA), polysodium styrene sulphonate, dextran sulfate and
carboxymethylcellulose (CMC), in aqueous solution have been used for binding of azo
dye anions (methyl orange and its homologues) [24].
An automatic colourimetric titration method has been described for the determination
of polyelectrolytes in aqueous solution, based on the formation of PEC between cationic
and anionic polymers. Examples are given for polyethylene imine (PEI), polyvinyl sulfate
and polyacrylate [25]. The number of basic groups of human carboxyhaemoglobin was
evaluated by colloid titration with potassium polyvinyl alcohol sulfate [26].
Interactions between branched PEI and PNaA with kaolin have been investigated for
structuring phenomena in pigment systems [31].
Complexation of polyaniline (PAn) with PSSA has been used to render PAn soluble in
water. It occurs as a result of the formation of a NPEC, with a structure resembling to
that of a block copolymer. The complexation was executed in N-methyl-2-pyrrolidone
181
Handbook of Polymer Blends and Composites
in the presence of lithium chloride. The precipitate, obtained with addition of acetone,
were soluble in water [32].
Stable dispersions of PEC nanoparticles can be used to remove dissolved organic molecules
from aqueous solution via hydrophobic or electrostatic interactions. The sorption
capability of such macromolecular assemblies increases with increasing molar mass and
hydrophobicity of the macromolecules used and new materials for removing organic
pollutants from wastewater can be designed [33].
Polysaccharides show large differences in solubility and in their solution or gel properties
due to a great variation in their chemical (primary) structure that determines the conformation
of the macromolecules adopted both in aqueous systems and in the solid state.
A particular feature of the polysaccharides is their ability to form gels in certain conditions
(concentration, temperature, ionic strength). The gelation phenomenon of polysaccharides
is of a major interest for their food and pharmacological applications.
A gel structure occurs when the permanent chain-chain interactions like hydrogen bonding,
dipole-dipole and ionic interactions as well as the interactions with the solvent are strong
enough to obtain an ordered polymer structure. A single polysaccharide chain could
participate in several ordered regions (junction zones), to form a three-dimensional
network, or gel structure.
182
Water Soluble Polymer Systems - Applications of Interpolymer Complexes and Blends
Reduction of the water activity. This causes an increase in interchain binding compared
to the chain-solvent association. Reduction of the water activity could be achieved
by addition of a low molecular weight, hydrophilic molecule, binding water in
competition with the polymer.
When one or both polymers of a mixture can gel, the phase separation phenomenon is of
a particular interest. If only one of the polymers can gel, the gelation will occur when the
concentration of this polymer in the continuous phase is above the minimum gelling
concentration [36], Figure 6.1. Segregative phase separation will prevent gelation of the
mixture if the non-gelling polymer forms the continuous phase and the concentration of
Figure 6.1 Phase diagram for a mixture of a gelling (P2) and a non-gelling polymer
(P1). Gelation occurs in the zones 2 and 6. Zone 3 would be the image of zone 2 for a
mixture of two gelling polymers. C2* is the minimum gelling concentration
Reprinted with permission from D.V. Zasypkin, E.E. Braudo and V.B. Tolstoguzov, Food
Hydrocolloids, 1997, 11, 159, Figure 1a. Copyright 1997, Elsevier Science
183
Handbook of Polymer Blends and Composites
the gelling agent in this phase is lower than the minimum gelling concentration, C2*. Gelation
of the overall mixture is therefore possible only in areas 6 and 2 (Figure 6.1) [37].
Charged polymeric cosolutes inhibit gelation of charged gelling agents and uncharged
cosolutes inhibit the gelation of uncharged gelling agents [36].
When both polymers form thermoreversible gels, clear mixtures could be obtained beyond
the gelation temperature in a large range of composition, but at low temperatures,
macroscopic phase separation is prevented by gelation, i.e., mixtures of lime hide gelatin
and iota-carrageenan [38].
The gels obtained from mixtures of two different biopolymers often have a biphasic
structure induced by segregative interactions (thermodynamic incompatibility) between
the constituent polymers [39]. The ultrastructure of the resulting co-gels can vary widely
depending on the time-temperature course of gelation in relation to the rate of the
segregation and network formation [40].
184
Water Soluble Polymer Systems - Applications of Interpolymer Complexes and Blends
group for every two disaccharide repeating units shows a greater ability to interact with
galactomanans than -carrageenan does. Substituents occurring on the outside surfaces
of agaran and -carrageenan decrease the binding of galactomannans.
Galactomannans and xanthan co-gel, too. Mixtures of xanthan and LBG form firm,
true, thermoreversible gels at low total polysaccharide concentration [47] by heating
and cooling their mixed aqueous solutions in 0.1 M NaCl [48]. The gels are thermally
stable, do not flow and do not recover from mechanical damage. These gels exhibit
sharp melting and gel setting over a narrow temperature range. Gel setting increases by
increasing the total polysaccharide concentration and shows little dependence on the
relative levels of concentrations of the two polysaccharides. The synergistic interaction
between xanthan chains and unsubstituted regions of galactomannan chains leads to
aggregated subunits that may stick together at rest or under shearing, forming shear
sensitive superaggregates and inducing thixotropic properties even at very low
concentrations.
Xanthan interacts with konjac mannan (KM) more strongly than with galactomannans
forming mixed gels [47]. They form gels at a total polysaccharide concentration of 0.02%,
the lowest gelling concentration observed for a carbohydrate system. A gel of 0.2%
xanthan with 0.25% KM melts at around 63 C while a gel of xanthan/galactomannan
melts at 41 C. By deacetylating xanthan, its gelation behaviour with LBG is improved,
while gel melting and setting temperatures of its mixture with KM are reduced by 20 C.
185
Handbook of Polymer Blends and Composites
Agarose and carrageenan, gel in the presence of a high level of arabic gum (14%) at the
highest level inclusion. The gelation of gellan is inhibited at low levels of arabic gum.
Higher molecular weight of gellan compared with carrageenan could be an explanation
for the inhibition of its gelation [36, 37].
Studies of the gelatinisation behaviour of the corn starch in the presence of xanthan,
guar, LBG and sodium alginate showed that the effect was more pronounced with ionic
polysaccharides [54]. The presence of another polysaccharide in the starch-based blends
has little effect on the gelatinisation temperature or the gelatinisation enthalpy of starch
[59-61]. Ferrero and co-workers [61] attributed the increases in the midpoint temperature
of the gelatinisation process, Tgel, and the final temperature of the gelation process, Tc,
to the reduced availability of water required for the disruption of the crystalline regions
within the granule. For the starch/KM system [60] and sweet potato starch in presence
of various cellulose derivatives [62] Tgel and T c increase by increasing the total
polysaccharide concentration.
Sudhakar and co-workers [63] studied the influence of the inorganic salts on the
gelatinisation or pasting behaviour of starch/xanthan systems and the pasting
characteristics of corn starch and waxy maize with guar gum and LBG. The increase in
the storage moduli of corn starch in the presence of guar, LBG or xanthan was attributed
to phase separation due to incompatibility of starch and added polysaccharides [56, 64].
186
Water Soluble Polymer Systems - Applications of Interpolymer Complexes and Blends
Alginate/starch gels have also been prepared. An effective diffusivity at the centre of the
gel of 4.6 x 10-10 m2/s was reported. Similar values of 7.8 x 10-10 m2/s and 3.3 x 10-10 m2/
s were reported for diffusion of glucose (25 C) into carrageenan (2%) [66] and sugar
(61%) in an agar gel/milk bilayer system [67]. Mixtures of alginate and high-methoxyl
pectin give firm gels at low pH and low solids (20% sucrose or less). Gelation is not
observed above a pH of 3.8. Alginate/pectin gels are thermoreversible; the melting point
increases as the pH of gel formation is reduced [68]. This kind of low pH gelling system
can be used for low calorie or diet jams and jellies, salad dressings and low pH fruit-
filled gels. A similar system using alginate/whey protein complex and fish protein can be
utilised to produce fish analogs from fish mince.
Proteins can co-gel as well with polysaccharides. For example, films of starch/gelatin
were prepared by mixing their aqueous solutions, at 80 C, at a total polymer
concentration of ctot = 5.0 %wt and then casting them from the gel state at 20 C. The
films have a crystalline structure and a greater thermodynamic stability [69]. The gelation
temperature and gel strength of the gelatin/starch gel were dominated by the gelatin
component, with no indication of network formation by the starch.
Because of the interactions between casein micelles and carrageenan (iota), carrageenan
is used in the food industry to make milk gels and to stabilise milk fat emulsions, ice
creams and chocolate milk. These kinds of gel have been intensely studied. Dalgleish and
Morris [70] established that , and -carrageenans may be absorbed on casein micelles
and their charge density influences their absorption. Snoeren [71] reported that -
carrageenan interacts with -casein, mostly at the periphery of the micelles, but not with
S1--caseins. Langendorff [72] demonstrated by electophoretic mobility determinations
that -carrageenan chains are absorbed progressively on casein micelles at 25 C while at
60 C only -carrageenan is absorbed. Coil-helix transition of -carrageenan is essential to
its electrostatic absorption on casein micelles, this transition resulting in an increased
charge density of the polymer chain [71, 72]. When -carrageenan is added to milk, two
modes of structure could occur: a network of -carrageenan chains reinforced by the
presence of casein micelles or another network, much weaker, involving -carrageenan
chains in helical conformation adsorbed on casein micelles.
One of the main directions of the applications of the water soluble polysaccharide-based
blends is in the food packaging industry as edible films (see Table 6.1). Food packaging
187
188
Soluble Casting of mixed solution Films prepared at low temperature Plasticiser: glycerol or [89]
gelatin/starch or of 5% aqueous gelatin and have higher tensile strength () and sugars.
hydroxypropyl 2% aqueous soluble potato percentage elongation at break () Conditioning the films at
starch starch solutions at 60 C or than those prepared at high relative humidity higher than
20 C. temperature. 60% increases their and
causes a drop in of about
20%.
Sodium Casting of mixed solution = 7.5-8.0 GPa [76]; Plasticiser: polyols. [88,
caseinate/starch followed by evaporation at Increasing the amount of sodium 99]
high temperature. caseinate in the blend, the WVTR
decreases from 7.3 to 5.4 x 10-11
g/m s Pa [92].
100 starch (corn, Biodegradable. Blend contains: plasticiser, [100]
wheat, potato)/ water (50-70 parts),
Food packaging
Edible films
(1-7) cellulose hygroscopic agent (2-6
(CMC, EC, MC, parts),
acetyl fibre)/(6- lubricant (0.5-2 parts),
18) water-soluble foaming agent (0.3-2 parts),
resin foaming aid (0.3-2 parts).
(polyethylene Materials may contain
glycol, PVAL, additives as: antioxidant,
alkyd resin). fungicide, brightener, and
coupling agent.
Water Soluble Polymer Systems - Applications of Interpolymer Complexes and Blends
Table 6.1 continued...
Blend Mixing process Characteristics Observations Ref.
(56-86) Laboratory scale: casting Wide range of good mechanical Plasticiser: glycerol. [58]
Pectin/(44-14) from 5-8 wt%, relatively properties [101-103]. Films are used for: water
starch aqueous solutions, dry at Tg ~ -50 C, flexible films at room soluble pouches for
room temperature. temperature. detergents and insecticides,
Industrial: extrusion on a Storage modulus = 103 MPa at flushable liners and bags and
twin screw extruder mass room temperature. medical delivery systems and
flow rate for blends: 27-102 Expected elongation at break devices.
g/min; water: solid feed ~ 10%-20%. Edible bags for soup and
rates = 0.43-2.85 cm3/g. Moisture content 30%-75%. noodle ingredients.
Six different extruder Excellent oxygen barrier properties
temperature profiles were [104].
used.
The screw speed: 350 rpm
for profiles I-III and 350
and 450 rpm for profile IV.
PVAL/CS Solution mixing under Up to 20% plasticiser in the blends Pressure (20-121 MPa) does [79]
vigorous stirring and does not show a phase separation. not affect the diffusion
heating; plasticiser is added of unplasticised blends is ~0.55 coefficients.
into the solution for 10-15 GPa for plasticised blends with 5% Chitosan does not affect the
minutes until it is dispersed. sucrose and 5% water, ~305 MPa. crystallinity of PVAL [105].
189
Handbook of Polymer Blends and Composites
material have to provide optimum properties (appropriate gas and water barrier properties
in order to prevent the destruction of food by microbial or insect attack) so that the
supplied product remains in a satisfactory condition for its anticipated shelf life. Consumer
demand for high-quality, long shelf life, ready-to-eat foods has initiated the development
of mildly preserved products that keep their natural and fresh appearance as long as
possible [73, 74]. Edible and biodegradable polymer films for food packaging offer
alternative packaging options, advantageous to the synthetic recalcitrant packaging
polymers, being environmentally non-polluting [75, 76], lowering in the same time the
moisture, aroma and lipid migration among food components [77] and improving the
textural properties and coherent structure [78].
The value of polysaccharides as edible films is well recognised [79, 80] but apart from
some special applications [81, 82], polysaccharide- and/or protein-based edible films
have not been used extensively in the food industry, yet [78]. The main drawbacks of
edible films are related to their poor moisture barrier performance (apart from wax),
colour, appearance and their mechanical and rheological properties [83-85]. The main
focus is to improve their water vapor transmission (WVT) barrier properties [79].
Although most edible films are not effective as WVT barriers, they are quite efficient as
gas barriers. Due to their potential application as films, the theoretical or empirical
calculation of gas permeation was the subject of several investigations [86, 87]. A
methodology to evaluate the gas permeability for amorphous synthetic polymers was
proposed by Salame [87].
Gas diffusivity of the films containing hydrogen-bonding groups increases with increasing
the water content [88]. Changes in the polarity of the hydrated matrix as a result of
water absorption may further complicate the solubility and diffusivity of gases and thereby
affect the permeability values. For example for polyol-plasticised starch/methylcellulose
blends an increase in gas diffusion rate of at least 3-4 orders of magnitude was noticed
compared to non-plasticised blends [89].
Natural polymers are characterised by extensive water clustering and plasticising of their
matrix properties, which is regarded as an important shortcoming for packaging
applications.
190
Water Soluble Polymer Systems - Applications of Interpolymer Complexes and Blends
Due to their processing difficulties, the edible films and coatings based on polysaccharide
(or protein) blends are still at an experimental level. The most extensively and successfully
polymers used in this field are starch and gelatin [79].
The presence of a single glass transition (Tg) for these blends could be due to both the
close proximity of the Tg of the individual polymeric components and their similar
behaviour in the presence of plasticisers [89].
CS/PVAL films plasticised and non-plasticised with sorbitol, sucrose or water [79] were
prepared and studied for food packaging applications. For these blends, a low sorption
of CO2 was found. The low levels of sorbed CO2 could be assigned to the high level of
hydrogen bonding in both plasticised and non-plasticised blends. Carbon dioxide
diffusivity was studied because of its strong effective bacteriostatic action toward
preventing microbial growth. The low CO2 transport for CS/PVAL blends could be
interesting for the food packaging companies, i.e., in the bakery industry.
191
Handbook of Polymer Blends and Composites
Proteins have good film forming properties and may be used in coating formulations for
fruit and vegetables as they are effective as gas barriers for O2 and CO2, but their WVTR
is high [74, 91]. For example, casein and casein derivatives with different molecular
weights (19,000-23,900) have been extensively used in the food industry (dairy, meat
and confectionery) as well as in medical and pharmaceutical applications [92, 93]. Gelatin
coatings with or without polyols, i.e., glycerol, carrying antioxidants were effective in
reducing rancidness when they were applied by spraying or dipping to cut up turkey
meat or to smoke-cured chicken [94].
For the charged polysaccharides, the compatibility with proteins decreases by increasing
the salt concentration [97, 98].
Cellulose derivative based blends are used for the formulation of delay or sustained drug
release. Cellulose ether-ester films [106] are flexible and transparent, exhibit moderate
strength, resistance to oil and fat migration. They act as moderate barriers to moisture
and oxygen [78, 107, 108].
Sakellariou and Rowe [109, 110] studied the compatibility of HPMC and PVAL without
or with plasticiser (glycerol) because these polymers are extensively used as film coatings
of solid pharmaceutical formulations. They have found that the polymers are totally
incompatible due to the significant discrepancies in the polar and hydrogen bonding
characteristics [111]. The plasticisers are often added to depress the Tg and the modulus
of the polymers by eliminating the internal stresses that are responsible for coating cracking
and bridging of the intagliations (engravings) [112].
192
Water Soluble Polymer Systems - Applications of Interpolymer Complexes and Blends
Film coatings for pharmaceutical purpose made of ethylcellulose (EC) with HPMC [113,
114] and of HPMC with sodium CMC [115] blends were prepared and studied. The
performance of the final product depends strongly on film thickness, rate of wetting,
solvent system, temperature profile and air flow conditions of the coating process. The
interactions and the morphology in the blend are of major importance for the drug
release profile [116].
In EC/HPMC film coatings, the drugs are released through pores created in the film
coating by the dissolution of the dispersed phase (HPMC). Sakellariou [117] reported on
the retention/leaching of water soluble (HPMC, HPC and PEG 6000) and pH-dependent-
soluble cellulose derivatives from the blends with 20% and 80%w/w ethylcellulose
content, used in delayed or sustained release film. A small amount of HPMC is retained
in the blends with high (80%w/w) EC content [116].
Influence of the polymer viscosity on the drug release from the hydrophilic erodable
matrix of HPMC and sodium CMC was investigated by Bonferoni and co-workers [115].
They found that when the erosion occurred there were obtained different release profiles
depending on polymer viscosity.
Hydrogels suitable for drug release were obtained from blends based on CMC with
agaragar. These blends form a strong structural network of the initial polysaccharides
and at temperatures lower than 36 C appears to be a non-thixotropic component [118].
Administration of the drug through buccal and sublingual sites [119] are advantageous
compared with other routes [120] due to the rapid onset of action, high level of the drug in
blood, avoidance of the first pass effect and the exposure of the drug to the gastrointestinal
tract. Therefore efforts are being focused on researching these kinds of tablets. In Japan are
now commercially available tablets named Aftach [121, 122] for the treatment of aphthous
ulcers. They have HPC and carboxyvinyl polymer as the major excipients and have
appropriate oral mucosal adhesion properties and controlled drug release features.
The CS complexes are well known through their application for the immobilisation of
enzymes, the micro-encapsulation of cells and for controlled drug release [123, 124]. For
example albumin was microencapsulated within a CS-alginate complex membrane in
the presence of calcium chloride [125]. Collagen (C)/CS composites, membranes and
hydrogels were studied for their potential applications in biomedical and pharmaceutical
field [126, 127].
193
Handbook of Polymer Blends and Composites
Many formulations for coatings with applications in colonic drug delivery, with a bimodal
drug release profile [128] were proposed based on pectin/CS blends [129, 130]. This
could be explained by the great degree of protection against premature drug release in
the gastrointestinal tract of pectin/CS compression-coated tablets when compared to
tablets using pectin alone.
For example Macleod and co-workers [131] prepared a coating from a mixture of pectin/
CS/HPMC/0.1 M HCl with glycerol as plasticiser. They found that a ratio of pectin/CS
(3:1) is favourable in terms of swelling and permeability due to the maximum interaction
between the NH3+ groups from chitosan and the COO- groups from pectin. The
permeability of the coating for a model drug is higher in the presence of pectinolytic
enzyme than in buffer alone [132]. A tablet coated with a film containing pectin/CS/
HPMC in a ratio 3:1:1 and subsequently coated with a methacrylic acid copolymer
would deliver the majority of the drug dose into the colon; the release of the drug will
start at a low rate in conditions that simulated the small intestine without pectinolytic
enzyme and accelerated in the conditions simulated the colon (that implies the presence
of pectinolytic enzyme) [131].
Fernndez-Hervs and Fell [129] have prepared a coating for tablets from pectin/CS by
compressing around the core mixture. They used normal concave punches at a core/coat
ratio of 1:5. The chosen weight ratio of pectin/CS was 10/1 because at this ratio the
maximum interaction between pectin and CS occurs. The coat thickness depends on the
type of the drug to be released, i.e., 1.59 mm for indomethacin and 1.48 mm for
paracetamol. They concluded that the coat is successful to protect and then release
relatively insoluble compounds. Soluble materials showed premature release in conditions
analogous to those in the upper gastrointestinal tract. These coatings are capable of
retarding the release of the tablet core materials until they reach the colon, an environment
rich in bacterial enzymes which degrade the coating and allow the release of drug [130].
Polymeric matrices for controlled drug release systems were prepared based on CS and
different polyacids (PAA and PSSA). They were obtained by radical polymerisation of
acrylic acid monomer [132] or sodium styrenesulfonate [133] in the presence of CS or
CS hydrocloride, respectively. CS/PAA systems have a crystalline structure, the
polycomplexes being quite ordered. CS/PSS systems have a scrambled egg structure.
CS/PAA have a negative influence on mitocondrial activity [132].
194
Water Soluble Polymer Systems - Applications of Interpolymer Complexes and Blends
from a solution of sodium alginate in distilled water, sprayed into 0.5% CaCl2 solution
with magnetic stirring and transferred to a chitosan hydrochloride solution (0.1-2 w/v%).
A porous structure was obtained by lyophilisation. The microspheres could be used for
immune activating growth factor interleukin-2 (Il-2) and bovine serum albumin-fluorescein
isothiocyanate release, in tumour immunotherapy [136] and they have been approved
for oral use like generally recognised as safe materials. The sustained release of proteins
from these microspheres is of longer duration than release from another kind of
microspheres based on alginate, i.e., alginate/polylysine or alginate/CaCl2 microspheres.
Miyazaki and co-workers [137] prepared from CS/alginate mixtures mucosal adhesive
tablets for delivery of diltiazem, a calcium channel blocker useful in cardiovascular diseases
such as angina and systemic hypertension. Tablets were made by compression of weighed
dry powder of CS (deacetylation degree = 80%), sodium alginate (mixing ratio CS/alginate
varied from 1:4 to 4:1) and the drug.
CS/protein blends have been investigated for drug encapsulation too. Remun-Lpez and
Bodmeier [138] studied the CS/gelatin type B coacervates for encapsulation of some drugs
like piroxicam, clofibrate, sulfamethoxazole. The coacervation was carried out by solution
mixing, at 40 C, for 4 hours, at a pH of 5.5 (optimum pH: 5.25-5.5) and a total polymer
concentration of ~3.25% w/w. Spherical and isolated coacervate CS:gelatin droplets were
obtained at a ratio of 1:20 till the total polymer concentration reaches 3.15% w/w. At
higher concentrations they become a gel. The amount of coacervate formed reaches the
optimum at a ratio CS:gelatin of 1:10, and is suppressed at a ratio of 1:50.
For the controlled release of propanolol hydrochloride, it was proposed that a C/CS
membrane be obtained by a solvent evaporation technique. For this purpose a 0.5% w/v
solution of chitosan and collagen in 0.5 M acetic acid were mixed in different ratios
(C/CS of 1:3, 1:1 or 3:1), poured on plate and dried at 4 C. The membranes obtained
were neutralised by immersion in 1% aqueous NaOH and dried at 4 C [139]. By varying
195
Handbook of Polymer Blends and Composites
the C/CS ratio in the membrane, its swelling potential and permeability to the drug
could be regulated. Partition of propanolol hydrochloride into the composite membranes
is much greater when compared to that of the drug into either collagen or chitosan
membranes and this is attributed to the lower polarity of composite membranes due to
charge neutralisation and thereby better interaction with the drug molecules. From this
membrane the drug is released following a zero-order kinetic law [139].
Properties of pure synthetic polymers and those of pure biological polymers alone are
often inadequate to produce materials with good chemical, mechanical, thermal and
biological performance characteristics. So researchers have tried to prepare blends of
synthetic polymers with biological macromolecules to obtain bioartificial polymeric
materials [140]. They are produced in different forms such as, films, sponges and hydrogels
and were evaluated as biomaterials for dialysis membranes [141], wound dressings [142]
and drug delivery systems [142, 143].
Cascone and co-workers [147-149] have prepared hydrogels based on PVAL and different
polysaccharides, such as CS, dextran, hyalouronic acid (HyA) and have tested them as
controlled release agents of human growth hormone (HGH) [142, 143] or as a drug
delivery system [149]. Hydrogels used for controlled release of HGH were obtained by
mixing a solution of the partners in different ratios: polysaccharide/PVAL (10-40)/(90-
60) followed by 8 cycles of freeze-thawing (from 20 C at 1 hour to room temperature
for 30 minutes); the first cycle of freezing at 20 C lasts 12 hours. The CS/PVAL hydrogel
is characterised by an increase in the Tg value up to a 10% content of biopolymer in the
blend, then remains constant. A decrease of the storage modulus value, E with the increase
in chitosan content was noted. Hydrogels have a macroporous structure with a low
degree of order. The degree of crystallinity of PVAL from the blends decreases, especially
by increasing the chitosan content of the blend.
The dextran/PVAL hydrogel is more compact and homogenous than the chitosan-containing
ones. The storage modulus value, E, and the crystallinity degree of PVAL increase with
196
Water Soluble Polymer Systems - Applications of Interpolymer Complexes and Blends
increasing the dextran content. In the HyA/PVAL hydrogel an increase in the elastic modulus
value and thermal stability was seen for a certain HyA/PVAL ratio [142, 143].
The controlled release of HGH depends on the content of the biological component. By
increasing the content of the biological component the HGH release is increased. In this
way the rate of release could be controlled [142, 143]. The amount and method of HGH
release depend, too on the polysaccharide type from the hydrogel. For the CS/PVAL
hydrogel with a CS content up to 30%, the total amount of the released HGH decreases,
and takes place in two steps [149]. However the total amount of HGH released is in a
physiological range (2.125100 ng/ml), enough to have an osteoblast proliferation [142].
In the dextran/PVAL blend up to 30% content in dextran, the total amount of the released
human growth hormone increases.
Blends of lysozyme/galactomannan (or xyloglucan) in a mixing ratio of 1/4 (or 1/8) are
dissolved in distilled water, lyophilised and then incubated at 60 C for 2 weeks at a
relative humidity of 78.9%. The protein-polysaccharide conjugates were separated by
gel filtration on columns of Sephadex G-50 and CM-Toyopearl 650 M [154]. They have
good emulsifying properties and are heat stable forms.
Alginate/protein blends can be used as emulsifying and foaming agents too [68].
197
Handbook of Polymer Blends and Composites
Thermoplastic starch (TPS) was blended with biodegradable polymers like polylactic
acid [155] or PVAL and plastifying agents (glycerol and its derivatives, sorbitol and its
derivatives, etc) in order to obtain bioplastics, i.e., biodegradable materials with high
water vapour permeability, a good oxygen barrier and not electrostatically chargeable.
Bioplastics can be processed by the existing plastic processing industrial production lines.
During the production of polymer blends a transesterification reaction takes place in situ
at different temperature and shearing conditions [156-159]. Bioplastics reduce the waste
amounts in an ecological manner by closing the natural material cycles [160]. Their
main applications are as compostable packaging (compostable in accordance with DIN
V 54900 [161]) or as short-term consumable articles and special products. Biodegradable
plastics known on their commercial name as Mater-bi AF10H, prepared from a blend of
starch (60%) and natural additives with (40%) modified PVAL and plasticisers are used
for packaging and agricultural mulch films [162, 163]. Mater-bi AF10H presents a low
biodegradability compared with other biodegradable plastics [164]. By melt-mixing and
extrusion of 70% starch/PVAL polymer alloy (Mater-bi) with 30% talc, a paper exhibiting
degradation after 8 months was obtained [165].
An eraser for rubbing off letters without damaging paper was obtained by mixing and
injection moulding of potato starch (100 parts), PVAL (20 parts), glycerin (50 parts) and
water (50 parts) [166].
Starch/PVAL blends are used as adhesive corrugated paperboard [167, 168]. Photo or
biodegradable films (with an optimal starch content of 50%) were produced by doctor
blade casting process [169].
Starch/alginate blends have applications in the most different fields. For example starch/
alginate can be used to produce crunchy products and cellular foods [170] while starch/
sodium alginate blend is a thickening agent in printing ink components for local bleaching
in grey linen fabric printing [171]. Cellular foods are manufactured by freeze-dehydration
of gels [172, 173].
198
Water Soluble Polymer Systems - Applications of Interpolymer Complexes and Blends
Chitosan is an attractive material for hydrophilic membranes [175, 176] due to the
presence of both amino and hydroxy groups in its structure. CS membranes crosslinked
with glutaraldehyde [177, 178] and sulfuric acid [179] were applied successfully in the
dehydration of alcohol-water mixtures. The membranes of PVAL/CS blends have good
performance for separation of ethanol-water mixtures [180]. Addition of
hydroxyethylcellulose (HEC) to the composition of the membranes affects their sorption
properties; the content of HEC that gives the highest sorption selectivity depends on the
concentration (or activity) and the type of mixture. CS/HEC membranes show a higher
sorption selectivity than the pure CS membranes, but a membrane with a higher HEC
content is not appropriate for separation of the higher water content mixture [181].
Huang and co-workers prepared membranes from CS and PAA for separation of ethanol-
water mixture, too [183]. The membranes were prepared by mixing, at room temperature
199
Handbook of Polymer Blends and Composites
of PAA (5%) and CS (0.5%) solutions made into 30 wt% acid acetic solution at a molar
ratio of mixing 1:1. The mixed solution was cast onto a polyvinyl chloride plate and
dried at room temperature for 24 hours. The thickness of this membrane is about 20-
40 m. By washing this membrane with water for 24 hours another type of membrane
(membrane thickness is 30 m) is obtained with a structure of a polyelectrolyte complex.
Heating the PEC membrane for 1 hour, at 180 C in an oven, produced another type
membrane, by the reaction between the amino groups of CS and the carboxyl groups of
PAA. The PEC membrane exhibits the higher water selectivity, the highest permeability
(1.4 x 10-3g/m h) and separation factor at feed solution of high ethanol content (over
2000 at a 95 wt% ethanol aqueous solution of 30 C) and exhibits a great potential for
the dehydration of organic solvents. The pervaporation performance can be improved
by using a high molecular weight polymer [184].
Protein separation and purification is generally realised by means of the so called aqueous
two-phase systems (ATPS). ATPS are easily applied as downstream procedures because
they do not require expensive specialised equipment, they can be scaled up and operated
200
Water Soluble Polymer Systems - Applications of Interpolymer Complexes and Blends
Most of the laboratory work was done with a system composed of fractionated dextran
and PEG [192]. Akashi and co-workers [193, 194] focused on dextran/
polyvinylalkylamide (thermosensitive polymer) as ATPS and they found that this system
has a higher separation ability for both polymers and biomolecules than that of the PEG/
dextran system, usually used. Because fractionated dextran has a high cost, in order to
obtain ATPS some other polysaccharides were tested: starch derivatives [195, 196],
maltodextrin [197], cellulose derivatives [198], PEI [199] agarose [200], guar gum [201],
and LBG [202]. By including a starch derivatives (hydroxypropyl starch crude or purified)
in polyethylene oxide, Almeida and co-workers [203] obtained an ATPS system used for
cutinase separation and purification.
Concluding Remarks
Water soluble polymer blends or formulations have been used in many applications especially
in the food, pharmaceutical and cosmetic industries and for medical uses, due to their
environmentally safe features and non-polluting technologies of the blends manufacture.
Although the number of water soluble polymer base systems is enormous, until now
only a relative low number of such blends (with applications in food industry as thickeners,
gels and pharmacy as controlled or sustained drug delivery matrix or coatings) are
commercialised. Nevertheless, the interest in such kinds of materials is continuously
increasing.
References
1. B. Philipp, J. Ktz, K-J. Linow and H. Dautzenberg, Polymer News, 1991, 16, 4,
106.
4. A.V. Kabanov and V.A. Kabanov, Polymer Science, 1994, 36, 2, 157.
201
Handbook of Polymer Blends and Composites
6. A.B. Zezin, V.A. Izumrudov and V.A. Kabanov, Die Makromolekulare Chemie -
Macromolecular Symposia, 1989, 26, 249.
8. C.J. Davison, K.E. Smith, L.E.F. Hutchinson, J.E. OMullane, L. Brookman and
K. Petrak, Journal of Bioactive and Biocompatible Polymers, 1990, 5, 3, 267.
11. P.L. Dubin, J. Gao and K. Mattison, Separation and Purification Methods, 1994,
23, 1.
12. J-Y. Shich and C.E. Glatz in Macromolecular Complexes in Chemistry and
Biology, Eds., P. Dubin, J. Bock, R. Davis, D.N. Schulz, C. Thies and D. Schulz,
Springer-Verlang, Berlin, Germany, 1994, 273.
13. Y. Wang, J.Y. Gao and P.L. Dubin, Biotechnology Progress, 1996, 12, 356.
14. A.L. Margolin, S.F. Sherstiuk, V.A. Izumrudov, A.B. Zezin and V.A. Kabanov,
European Journal of Biochemistry, 1985, 146, 625.
16. R.M. Ottenbrite and A.M. Kaplan, Annals of the New York Academy of Science,
1985, 446, 160.
17. I.C. Kwon, Y.H. Bae and S.W. Kim, Journal of Controlled Release, 1994, 30, 2,
155.
19. M.J. Lysacht in Ionic Polymers, Ed., L. Holliday, Applied Science Publishers,
London, UK, 1975, 281.
202
Water Soluble Polymer Systems - Applications of Interpolymer Complexes and Blends
22. J.H. Dong, C.K. Huang and Y.Y. Jiang, Polymer Bulletin, 1988, 20, 6, 521.
25. D. Horn, Progress in Colloid and Polymer Science, 1978, 65, 251.
27. N.A. Plate, E.D. Alieva and A.A. Kalachev, Polymer Science USSR, 1981, 23, 3, 720.
31. J. Ktz and S. Kosmella, Journal of Colloid and Interface Science, 1994, 168,
505.
32. Y-H. Liao, K. Levon, J. Laakso and J.E. Osterholm, Macromolecular Rapid
Communications, 1995, 16, 5, 393.
33. H-M. Buchhammer, G. Petzold and K. Lunkwitz, Colloid & Polymer Science,
2000, 278, 841.
34. C.L. McCormick, J. Bock and D.N. Schulz in Encyclopedia of Polymer Science and
Engineering, Eds., H.F. Mark, N. Bikales, C.G. Overberger, G. Menges and J.I.
Kroschwitz, John Wiley and Sons, New York, NY, USA, 1989, Volume 17, p.730.
35. Industrial Gums: Polysaccharides and their Derivatives, 3rd Edition, Eds., R.L.
Whistler and J.N BeMiller, Academic Press, Inc., San Diego, CA, USA, 1992, 21.
36. M. Puaud, S.E. Hill and J.R. Mitchell in Hydrocolloids, Volume 2, Ed., K.
Nishinari, Elsevier Science, Oxford, UK, 2000, 147.
203
Handbook of Polymer Blends and Composites
37. D.V. Zasypkin, E.E. Braudo and V.B. Tolstoguzov, Food Hydrocolloids, 1997, 11, 159.
39. V.V. Suchkov, V.Y. Grinberg and V.B. Tolstoguzov, Carbohydrate Polymers, 1,
1981, 39.
41. I.C.M. Dea, A.A. McKinnon and D.A. Rees, Journal of Molecular Biology, 1972,
68, 153.
42. V. Carroll, M.J. Miles and V.J. Morris in Gums and Stabilisers for the Food
Industry 2. Applications of Hydrocolloids, Eds., G.O. Phillips, D.J. Wedlock and
P.A. Williams, Pergamon, Oxford, UK, 1983, 501.
43. P.B. Fernandes, M.P. Goncalves and J.L. Doublier, Carbohydrate Polymers, 22,
193, 99-106.
44. P.B. Fernandes, M.P. Goncalves and J.L. Doublier, Food Hydrocolloids, 1994, 9, 1.
46. E.B. Bagley in Physical Properties of Foods-2, Eds., R. Jowitt, F. Escher, M. Kent,
B. McKenna and M. Roquers, Elsevier Applied Science, London, UK, 1987, 345.
47. I.C.M. Dea, E.R. Morris, D.A. Rees, E.J. Welsh, H.A. Barnes and J. Price,
Carbohydrate Research, 1977, 57, 249.
50. C. Tonon, G. Cuvelier and B. Launay in Physical Networks, Polymer and Gels, Eds.,
W. Burchard and S.B. Ross-Murphy, Elsevier Applied Science, London, 1990, 335.
51. I.H. Smith, K.C. Symes, C.J. Lawson and E.R. Morris, International Journal of
Biological Macromolecules, 1981, 3, 129.
204
Water Soluble Polymer Systems - Applications of Interpolymer Complexes and Blends
53. L. Lundin and A-M. Hermansson, Carbohydrate Polymers, 1995, 26, 129.
54. P.A. Williams and F.B. Ahmad in Hydrocolloids, Volume 2, Ed., K. Nishinari,
Elsevier Science, Oxford, UK, 2000, 165.
55. C.B. Closs, B. Conde-Petit, I.D. Roberts, V.B. Tolstoguzov and F. Escher,
Carbohydrate Polymers, 1999, 39, 67.
56. M. Alloncle, J. Lefebvre, G. Llamas and J.L. Doublier, Cereal Chemistry, 1989,
66, 90.
57. Z.H. Mohammed, M.W.N. Hember, R.K. Richardson and E.R. Morris,
Carbohydrate Polymers, 1998, 36, 27.
58. M.L. Fishman, D.R. Coffin, R.P. Konstance and C.I. Onwulata, Carbohydrate
Polymers, 2000, 41, 317.
59. P. Annable, M.G. Fitton, B. Harris, G. O. Phillips and P.A. Williams, Food
Hydrocolloids, 1994, 8, 351.
61. C. Ferrero, M.N. Martino and N.E. Zaritzky, Journal of Thermal Analysis, 1996,
47, 1247.
62. K. Kohyama and K. Nishinari, Journal of Food Science, 1992, 57, 128.
63. V. Sudhakar, R.S. Singhal and P.R. Kulkarni, Food Chemistry, 1995, 53, 405.
67. F. Warin, V. Gekas, A. Voirin and P. Dejmek, Journal of Food Science, 1997, 62, 454.
68. K. Clare in Industrial Gums. Polysaccharides and their Derivatives, 3rd Edition,
Eds., R.L. Whistler and J.N BeMiller, Academic Press, Inc., San Diego, CA, USA,
1993, 97.
205
Handbook of Polymer Blends and Composites
70. D.G. Dalgleish and E.R. Morris, Food Hydrocolloids, 1988, 2, 311.
74. E.A. Baldwin, M.O. Nisperos-Carriedo and R.A. Baker, Critical Reviews in Food
Science and Nutrition, 1995, 35, 509.
77. Edible Coatings and Films to Improve Food Quality, Eds., M. Krochta, E.A.
Baldwin and M.O. Nisperos-Carriedo, Technomic, Lancaster, PA, USA, 1994.
78. J.J. Kester and O. Fennema, Food Technology, 1986, 40, 47.
81. J.W. Park, R.F. Testin, H.J. Park, P.J. Vergano and C.L. Weller, Journal of Food
Science, 1994, 59, 916.
82. H.J. Park, C.L. Weller, P.J. Vergano and R.F. Testin, Journal of Food Science,
1993, 58, 1361.
206
Water Soluble Polymer Systems - Applications of Interpolymer Complexes and Blends
84. J.M. Krochta, A.E. Pavlath and N. Goodman in Engineering and Food, Volume
2, Preservation Process and Related Techniques, Eds., W.E. Spiess and H.
Schubert, Elsevier Science Publishers, Oxford, UK, 1990, 329.
85. J.M. Krochta in Advances in Food Engineering, Eds., R.P. Singh and M.A.
Wirakartakusumah, CRC Press, Boca Raton, FL, USA, 1992, 517.
86. D.W. van Krevelen, Properties of Polymers: their Correlation with Chemical
Structure; their Numerical Estimation and Prediction from Additive Group
Contributors, 3rd Edition, Elsevier, Amsterdam, The Netherlands, 1990, 189.
89. I.S. Arvanitoyannis and C.G. Biliaderis, Carbohydrate Polymers, 1999, 38, 47.
91. S. Guilbert in Food Packaging and Preservation, Theory and Practice, Ed., M.
Mathlouthi, Elsevier Applied Science, London, UK, 1986, 371.
93. C.R. Southward in Developments in Dairy Chemistry - 4, Ed., P.F. Fox, Applied
Science Publishers, London, UK, 1989, 173.
94. M.N. Moorjani, K.C.M. Raja, P. Puttarajapa, N.S. Khabade, V.S. Mahendrakar
and M. Mahadevaswamy, Indian Journal of Poultry Science, 1978, 13, 1, 52.
95. D. Lourdin, G. della Valle and P. Colonna, Carbohydrate Polymers, 1995, 27, 261.
97. D.A. Ledward in Protein Functionality in Food Systems, Eds., N.S. Hettiarachchy
and G.R. Ziegler, IFT Basic Symposium Series, Marcel Dekker, New York, NY,
USA, 1994, 225.
207
Handbook of Polymer Blends and Composites
98. D.A. Ledward, Trends in Food Science and Technology, 1993, 4, 402.
99. I.S. Arvanitoyannis and C. Biliaderis, Food Chemistry, 1997, 62, 3, 333
101. D.R. Coffin and M.L. Fishman, Journal of Agriculture and Food Chemistry,
1993, 41, 1192.
102. M.L. Fishman and D.R. Coffin, inventors; no assignee; US Patent 5,451,673,
1995.
103. M.L. Fishman, D.R. Coffin, J.J. Unruh and T. Ly, Journal of Macromolecular
Science: Pure and Applied Chemistry, 1996, A33, 639.
104. D.R. Coffin and M.L. Fishman, Journal of Applied Polymer Science, 54, 1994,
1311.
105. M. Miya, R. Iwamoto and S. Mima, Journal of Polymer Science: Polymer Physics
Edition, 1984, 22, 1149.
107. R.D. Hagenmaier and P.E. Shaw, Journal of Agriculture and Food Chemistry,
1990, 38, 1799.
109. P. Sakellariou and R.C. Rowe, Macromolecular Reports, 1994, A31, 6&7, 1201.
110. P. Sakellariou, A. Hassan and R.C. Rowe, Colloid and Polymer Science, 1994,
272, 1, 48.
111. P. Sakellariou, A. Hassan and R.C. Rowe, Polymer, 1993, 34, 6, 1240.
113. N.B. Shah and B.B. Sheth, Journal of Pharmaceutical Science, 61, 1972, 412.
208
Water Soluble Polymer Systems - Applications of Interpolymer Complexes and Blends
115. M.C. Bonferoni, C. Caramella, M.E. Sangalli, U. Conte, R.M. Hernandez and
J.L. Pedraz, Journal of Controlled Release, 1992, 18, 3, 205.
117. P. Sakellariou, R.C. Rowe and E.F.T. White, Journal of Controlled Release, 1988,
7, 147.
119. D. Harris and J.R. Robinson, Journal of Pharmaceutical Science, 1992, 81, 1.
121. T. Nagai, Y. Machida, Y. Suzuki and H. Ikura, inventors; Trjin Limited, assignee;
US 4226848, 1980.
122. T. Nagai, Y. Machida, Y. Suzuki and H. Ikura, inventors; Trjin Limited, assignee;
US 4250163, 1981.
125. Polk, B. Amstel, K. De Yao, T. Peng and M.F.A. Goosen, International Journal of
Pharmaceutics, 1994, 83, 178.
128. G.S. Macleod, J.T. Fell and J.H. Collett, International Journal of Pharmaceutics,
1999, 188, 11.
130. G.S. Macleod, J.T. Fell, J.H. Collett, H.L. Sharma and A.M. Smith, International
Journal of Pharmaceutics, 1999, 187, 251.
209
Handbook of Polymer Blends and Composites
131. G.S. Macleod, J.T. Fell and J.H. Collett, Journal of Controlled Release, 1999, 58,
303.
132. G.D. Guerra, P. Cerrai, M. Tricoli, S. Maltinti and R. Sbarbati del Guerra,
Journal of Materials Science: Materials in Medicine, 1998, 9, 73.
136. L-S. Liu, S-Q. Liu, S. Y. Ng, M. Froix, T. Ohno and J. Heller, Journal of
Controlled Release, 1997, 43, 65.
140. P. Giusti, L. Lazzeri and M.G. Cascone in The Polymeric Materials Encyclopedia,
Ed., J.C. Salamone, CRC Press, Boca Raton, FL, USA, 1996.
142. M.G. Cascone, B. Sim and S. Downes, Biomaterials, 1995, 16, 569.
143. M.G. Cascone, L. Di Silvio, B. Sim and S. Downes, Journal of Materials Science:
Materials in Medicine, 1994, 5, 770.
144. R.W. Baker, M.E. Tuttle and R. Helwing, Pharmaceutical Technology, 1984, 8, 26.
210
Water Soluble Polymer Systems - Applications of Interpolymer Complexes and Blends
148. R. Sbarbati del Guerra, M.G. Cascone, N. Barbani and L. Lazzeri, Journal of
Materials Science: Materials in Medicine, 1994, 5, 613.
149. M.G. Cascone, Journal of Materials Science: Materials in Medicine, 1999, 10, 301.
154. Y-W. Shu, S. Sahara, S. Nakamura and A. Kato, Journal of Agriculture and Food
Chemistry, 1996, 44, 2544.
155. P.J. Wuk, L.D. Jin, Y.E. Sang, I.S. Soon, K.S. Hyun and K.Y. Ha, Korea Polymer
Journal, 1999, 7, 2, 93.
162. T. Iwanami and T. Uemura, Japanese Journal of Polymer Science and Technology,
1993, 50, 767.
211
Handbook of Polymer Blends and Composites
165. M. Sakamoto, inventor; Japan Kokai Tokkyo Koho, assignee; JP 175288, 1998.
166. K. Suzuki and S. Mitsuhiro, inventors; Japan Kokai Tokkyo Koho, assignee JP
132525, 2000.
168. Y.P. Nekrasov, O.N. Belyatskaya, L.I. Bulatnikova and L.V. Bykova, Plastiche
Massey, 1997, 4, 29.
170. D.K. Rassis, I.S. Saguy and A. Nussinovitch in Hydrocolloids Volume 2, Ed., K.
Nishinari, Elsevier Science, Oxford, UK, 2000, 203.
171. E.L. Vladimirtseva, L.V. Sharnina, I.B. Blinicheva and B.N. Melnikov, Izvestiya
Vysshikh Uchebangkh Zavedenii Khimiya Ikhimicheskaya Tekhnologiya, 1998,
4, 50-53.
172. D.K. Rassis, I.S. Saguy and A. Nussinovitch, Journal of Agriculture and Food
Chemistry, 1998, 46, 2981.
174. H-J. Steiger, G. Arnold and C. Gass, inventors; IBN GmbH Dresden, assignee;
DE 19,751,234, 1999.
176. R.Y.M. Huang and X. Feng, Journal of Membrane Science, 1996, 116, 67.
212
Water Soluble Polymer Systems - Applications of Interpolymer Complexes and Blends
179. Y.M. Lee, S.Y. Nan and D.J. Woo, Journal of Membrane Science, 1997, 133, 103.
180. L.G. Wu, C.L. Zhu and M. Lu, Journal of Membrane Science, 1994, 90, 199-205.
182. F. Jun, H. Jicai, C. Liankai and G. Qunhui, Mo Kezue Yu Jishu, 1997, 17, 5, 60.
183. J-J. Shieh and R.Y.M. Huang, Journal of Membrane Science, 1997, 127, 184.
185. Z. Xi, S. Yanqiao and C. Guanwen, Gongneng Gaofenzi Xuebao, 1998, 11, 3,
358.
186. W.H. Jiang and S.J. Han, Journal of Polymer Science: Polymer Physics Edition,
1998, 8, 1275.
187. X.P. Wang, F.Y. Zhang and Z.Q. Shen, Chinese Membrane Science and
Technology, 1996, 16, 62.
188. S. Cao, Y. Shi and G. Chen, Journal of Applied Polymer Science, 1999, 74, 1452.
189. S. Suto and N. Ui, Journal of Applied Polymer Science, 1996, 61, 13, 2273.
190. M-R. Kula, K.H. Kroner and H. Hustedt, Advances in Biomedical Engineering,
1982, 24, 74.
191. P.A. Albertsson, Partition of Cell Particles and Macromolecules, 3rd Edition,
Wiley, New York, NY, USA, 1986.
213
Handbook of Polymer Blends and Composites
195. A. Venncio, J.A. Teixeira and M. Mota, Biotechnology Progress, 1993, 9, 635.
197. D.C. Szlag and K.A. Giuliano, Biotechnology Techniques, 1988, 2, 277.
198. D.R. Skuse, R. Norris-Jones, M. Yalpani and D.E. Brooks, Enzyme and
Microbial Technology, 1992, 14, 785.
200. A.S. Medin and J-C. Janson, Carbohydrate Polymers, 1993, 22, 127.
203. M.C. Almeida, A. Venncio, J.A. Teixeira and M.R. Aires-Barros, Journal of
Chromatography B, 1998, 711, 151.
214
Reactive Blending
7
Reactive Blending
Cornelia Vasile
7.1 Introduction
Many reactive processes involving multicomponent systems are known, such as:
polymerisation, modification during polymerisation of one or both components, alloying
polyolefins (PO) during polymerisation, blending, interpenetrating polymer network (IPN)
preparation, filling of polymers with conductive particles, controlled rheology, by which
completely new (innovative) products have been obtained and, in some cases starting
from existing polymers [1-3].
An unfavourable interaction causes a large interfacial tension in the melt, poor interfacial
adhesion in the solid state, makes it difficult to finely disperse the components of a blend
and, particles coalescence leads to phase separation and as a consequence to inferior
mechanical properties of the blends relative to its components [4]. For effective
compatibilisation, architecture and molecular weight of block or graft copolymers previously
synthesised must be carefully controlled. Such tailor-made compatibilisers are rarely available
commercially and are quite expensive and therefore have not been extensively used. One
difficulty is adequately dispersing a block copolymer near the interface between two phases
due to its high viscosity and sometimes the added block copolymer can localise in a
homopolymer phase in micelle form rather than at the interface [5, 6].
215
Handbook of Polymer Blends and Composites
The use of a polymer having hydrogen bonding, ionic bonding and acid-base complex
functional groups to improve the miscibility of two immiscible polymer blends has also
been studied [7, 8]. These kinds of compatible blends show phase separation when the
temperature is higher than the dissociation temperature.
Reactive blending can be used to change the structure of polymers and thus increase
their attraction for each other. An attractive technique is to synthesise the compatibilisers
by a chemical reaction directly in the melt blending process. Chemists refer to this as in
situ synthesis, while engineers generally use terms such as reactive processing or reactive
extrusion [9]. The functional units are easily introduced by a copolymerisation reaction
step or by grafting reaction during the extrusion process. Compared with conventional
monomeric reactions, analogous polymer reactions are hindered by slow diffusion, low
concentration of reactive groups, steric hindrance and the short time in the extruder.
However the reaction time in the extruder is shorter than in other type of reactors. For
example only 4 minutes were required for the reaction between polyethylene terephthalate
(PET) and polyamide (PA) in extruder compared with 36 hours in an unstirred reactor.
The reaction time can be controlled and minimised by using the catalysts.
216
Reactive Blending
Reactive processing is now the most active aspect of compatibilisation research. Reactive
processing is also a method for producing commercially useful blends when components
are incompatible by making a controlled stable morphology of a significantly reduced
domain size relatively to the incompatible blends primarily by suppressing domain
coalescence in the melt and for improving interfacial strength, increasing interphase
thickness and strengthen the interface in solid state so preventing mechanical failure due
to the weak defects between phases. The process may be defined as a reaction between
two or more polymers to form a copolymer so, integrating polymer chemistry with polymer
processing. Many commercial blends exist because of reactive processing. The reactive
processing during compounding or forming stage assures a controlled chemical reaction
between components. The technology opens up new possibilities for old polymers but
also allows preparation of blends which could not be economically made before. Super-
tough PA/PO, polyphenylene ether (PPE)/PA, polycarbonate (PC)/PA, polybutylene
terephthalate (PBT)/PET, polyoxymethylene (POM)/PC are only a few examples. From
the economic as well as the performance point of view, the technique is more interesting
than the addition of a compatibiliser.
Intensive and controlled mixing to achieve the desired dispersed morphology of one
polymer in another;
217
Handbook of Polymer Blends and Composites
High reactivity to react across the phase boundary and within the residence time in
the extruder;
Optimal operational conditions must be established for each process and each type of
blend. Several ways to perform reactive blending and average values of parameters are
described here.
Functionalisation can be performed before the blending process by solution grafting, melt
grafting, solid state functionalisation, copolymerisation, end-capping, physical procedures,
surface functionalisation, mechanochemical [9-14] and others. A Brabender mixer operating
under a nitrogen atmosphere at a rotation speed of ~50 rpm and
T = 200 220 C may be used. Maleation of PO can be carried out by melt mixing initiated
by peroxides (as dicumyl peroxide; DCP) in Brabender mixing chamber at 180 C. After
the polymer is melted, a mixture of monomer and initiator is added [13]. The grafting of
PO with maleic anhydride (MA) was done using a co-rotating intermeshing twin-screw
extruder (ZSK 30; Werner & Pfleiderer) with screw configuration adapted for grafting. A
masterbatch of PO granules and 0.5 wt% di-tert-butyl peroxide (Trigonox B; Peroxide
Chemie) and 0.2 wt% MA (Merck) was fed to extruder. A melt temperature profile from
180 to 210 C for PO and 215-260 C for other polymers, an output of 6 kg h-1 and a
screw speed of revolutions 150 min 1 were used; or feeding the first the polymer then the
MA and finally the peroxide; residence time was ~5 minutes. For neatness, these could be
carried out at different times but this would add to the complexity of the process and to the
final cost. Thus some processors feed the first ingredients to the back of the screw and then
inject other ingredients along the screw, to separate and run all reactions in their proper
sequence. The simplest is to feed all the ingredients to the back of the extruder.
The extruder, considered as another form of a chemical reactor provides extensive and
intensive mixing, good temperature control and controlled residence time distribution
218
Reactive Blending
(residence time ranges from ~100 to 200 seconds) affecting the molecular weight
distribution of the final product. The mixing they offer and their superior heat and mass
transfer capability [15, 16]. The process can takes place in three, two or one step, e.g.,
functionalisation of one of the other component or preparing a copolymer, blending and
extrusion or feeding together all polymers and functionalising agents. Because of the
residence time limitations in the extruder, high conversion rates are required. These can
be achieved either by using a relatively high concentration of the reactive groups, e.g.,
high concentration of chain ends, i.e., low molecular weight of the components, highly
reactive functional groups, or efficient catalysts.
Although both single and twin-screw extruder configurations are used in reactive
processes, twin screw ones are increasingly favoured over the single screw ones [15].
Figure 7.1 Simple, double and triple lobe intermeshing twin screw
219
Handbook of Polymer Blends and Composites
fact, this design can allow high speeds and thus strong shearing forces and high outputs.
With the co-rotating screw the material is transferred from one screw to the other and
undergoes a constant mixing.
The screw configuration has different conveying elements for feeding, melting reaction
and discharge, thorough mixing of the reactants was accomplished through a set of
kneading blocks just prior to the reaction zone. The reaction zone length estimated from
the previous batch studies is more accurately determined experimentally and could be
controlled by the barrel temperature and a certain temperature profile. Favourable
degassing conditions are easily created. The modular design and assembly arrangement
of the screw and barrel sections of the machine, the screw profile, along with the use of
special feeding and venting ports provide adequate flexibility for specific reactive processes.
Two screw configurations A and B are depicted in Figure 7.2 for co-rotating twin-screw
extruder (Werner & Pfleiderer ZSK 30) with the following characteristics, screw diameter
D = 30 mm, length to diameter ratio (L/D) =42 [18-21].
The operational conditions are: average feed rate 5.5 kg/h; screw speed 75-150 rpm,
mean residence time 150 seconds. The residence time has to be correlated with the lifetime
of the peroxide used (see Table 7.1) [18-21].
220
Peroxides
lifetime of
grafting temperature
Reactive Blending
Before feeding the materials into the extruder, the monomer(s) and peroxide were absorbed
into the small porous polymer pellets for 30 minutes at room temperature. Reactive
blending may carried out in the same twin-screw extruder as is used for grafting (A) but
screw configuration are different (B). The line may be equipped with a vacuum pump to
remove products as residual monomer (devolatilisation zone) that could alter the final
performance of the product. Before blending all materials must carefully dried.
The ordering processes are the basis of morphological architecture for multiphase
polymeric materials based on reacting polymer blends. Reactive polymers each bearing a
large number of reactive groups are important for new compatibilised blends with the
graft copolymers generated in situ. The availability of more and more functional polymers,
whether they are made by copolymerisation or chemical modification provides new
opportunities to make new alloys by coupling two reactive polymers. The complexity of
the graft copolymers grows as the grafting reaction continues and gelation occurs when
the extent of grafting reaction goes beyond a certain limit. The amount that can be
formed is limited by the gelation, but it is a need for enough graft copolymers to be
produced for a homogeneous microphase size. It is therefore necessary to stop the grafting
reaction at some point if gelation is not desirable. The material properties such as
mechanical properties, transport properties, optical properties, etc., depend very much
221
Handbook of Polymer Blends and Composites
on phase size and phase homogeneity. The amount and the structure of the copolymers
play a major role in that regard. Certain degrees of demixing between the chain segment
of graft copolymers and the solubilised homopolymers within the microdomains can
appear, even with homogeneous solubilisation. At low grafting conversion, i.e., < 10 wt%
graft copolymers in the products, the graft copolymers act as a compatibiliser. The reactive
polymers are often highly polydispersed so it is also important to know the effect of
polydispersity [22-24].
The molecular architecture of compatibiliser may significantly affect the final morphology
by changing the particle size of dispersed phase (see Figure 7.3) as even compatibiliser
exhibits various morphological aspects [25].
It is well known that block and graft copolymers are very efficient as compatibilisers for
a high number of polymer blends of decreasing particle size. However, in the case of the
polyolefin modification with acrylic acid (AA) is has been established [25] that in random
Figure 7.3 The number-average diameter of the dispersed phase versus the amount of
PS-GMA for four different blend systems indicated in the legend
Redrawn with permission from S. Kim, J.K. Kim and C.E. Park, Polymer, 1997, 38, 8,
1809, Figures 1 and 3. Copyright 1997, Elsevier
222
Reactive Blending
copolymers the AA units are well distributed across the entire chain and thus the grafting
reaction between polyethylene (PE)-r-AA and PS-GMA occurs uniformly giving rise to a
significant decrease in domain size. Contrary, because the AA units in PP-g-AA exist at
very limited positions, they are not well distributed, the reaction between PP-g-AA and
PS-GMA is highly constrained at certain places and thus the graft copolymer has a lower
ability to considerably reduce the domain size.
Critical surface concentration (CSC) is the amount of compatibiliser needed for absorption
of one monolayer of compatibiliser molecules on the dispersed polymer surface. The
calculations are based on the effective area of adsorption for compatibiliser molecule
calculated from the hydrodynamic radius and the total area to be stabilised in conditions:
CSC >1 means multiplayer adsorption and CSC <1 means surface is not completely
covered or stabilised. CSC can be used to establish the compatibiliser efficiency [30].
The interfacial area (A) stabilised per molecule for a graft copolymer located at the
interface can be estimated by [31]:
s
A = 3Mn /(RWcN) (7.1)
Where is volume fraction of the dispersed phase, Mn is the average molecular weight
of the compatibiliser, R is the particle radius, Wc is the mass of compatibiliser per unit
volume of blend and N is Avogardos number.
Only low concentrations of block or graft copolymers are required to saturate the interface
and produce optimum compatibilisation. As little as 0.5-2 wt% of such copolymer is sufficient
for optimisation of blends performance but more frequently 10-20 wt% are used.
The effect of the compatibiliser can also be established by a rheological Cole-Cole plot
[30] By this method, the loss modulus is plotted as a function of the elastic modulus or
223
Handbook of Polymer Blends and Composites
more commonly, the imaginary viscosity is plotted as a function of the real part. A
continuous or homogeneous system is represented as one uniform half-circle while for
immiscible blends the continuity is broken.
The impact strength is the most sensitive to blend compatibilisation. The addition of the
compatibiliser changes the morphology of the blends considerably from a coarse
morphology with average domain size as large as tens of microns (incompatibility) to
much finer, with the average particle size of about 1 micron [32]. When finely dispersed,
the rubbery particles act as effective stress concentrators initiating both crazing and
shear yielding from many sites of the matrix, favouring the dissipation of large amounts
of impact energy and slowing down the growth of cracks, hence increasing significantly
the toughness and impact strength. In the case of the polypropylene (PP)/PBT blend
reactive compatibilised with PP-g-oxazoline both impact strength and elongation at break
significantly increased with graft copolymer content, see Figure 7.4.
Sometimes phase separation is associated with chemical reactions [9, 33-35]. Miscible
polymer blends are destabilised by crosslinking reactions and undergo phase separation.
These phenomena are often observed during the preparation of IPN where the morphology
resulting from phase separation plays a crucial role in modifying the physical properties
of these polymer materials.
Figure 7.4 Notched Charpy impact strength and elongation at break versus graft
copolymer content [18-21]
Reproduced with permission from T. Vainio, G-H. Hua, M. Lambla and J. Seppala,
Journal of Applied Polymer Science, 1997, 63, 7, 883, Figure 5. Copyright 1997, Wiley
224
Reactive Blending
When a polymer does not have enough natural reactivity for reactive processing, the
polymer may be modified ahead of time to add the functional groups that will be needed
(functionalisation). This may be done during polymerisation by adding a comonomer to
supply the functional groups or it may be done by specific post-polymerisation reactions
for example olefin-acrylic copolymer, MA or acrylic onto a PO backbone [35-38].
Carboxyl groups are introduced by grafting or copolymerisation with MA, fumaric acid
and related compounds, acrylic acid (AA) and methacrylic acid, acrylic esters, etc.
Locations A to F refer to successive collection ports along the screw. Melting takes place
between locations A and B. The grafting profiles are similar, showing high conversion
rates from melting up to L/D=15, where the peroxide decomposition is completed. The
highest value of grafting was obtained for PE (1.7 wt%) and the lowest for PP (0.64 wt%).
For PE and EPR a crosslinking was observed after melting and for PP degradation.
Owing to reversibility of the reaction with the equilibrium shifted toward MA, the extent
of grafting is small. MA unit can be attached either in form of single succinic anhydride
ring (in PP and ethylene-propylene (EP) copolymers) as well as short oligomers grafts, at
the chain end (as in PPE).
Modifications with maleic acid anhydride and maleic esters substantially change the
polymers reactivity and chemical and mechanical properties.
225
Handbook of Polymer Blends and Composites
Figure 7.5 MA content along the extruder for modification of PE, EPR and PP [50] and
oxazoline (Oxa) content along the screw for grafting of PP, [ROOR] 0.3 phr [18-21]
Redrawn with permission from T. Vainio, G-H. Hua, M. Lambla and J. Seppala, Journal
of Applied Polymer Science, 1997, 63, 7, 883, Figure 4. Copyright 1997, Wiley
Due to their hydrophobic nature, poor dispersability with inorganic fillers and poor
miscibility in blends even with themselves, PO have restricted use in several new engineering
technologies. One way to overcome these drawbacks is the introduction of a small amount
of polar groups onto polymer backbone [51]. By controlling the extent of reaction, properties
can be changed in a targeted way to improve biocompatibility, fire retardancy, adhesion or
their ability to blend with other polymers. Grafted copolymers of PO are widely used as
compatibilisers in blends of PO with PA and polyesters (PES) [52].
Initiator concentration was found to be the predominant parameter controlling the flow
properties of the polymer. A commercial product from BP Chemicals, Polyblend is
maleated selectively at the chain ends. Ionic crosslinked rubber-like polymers were
obtained from maleic anhydride grafted onto a polyolefin (MA-g-PO) with some alkali
metal compounds. Gaylord [40] obtained alternating styrene-maleic anhydride copolymer
(S/MA) chain onto PO backbone. The succinic anhydride formed is highly reactive being
a starting point for synthesis of new polymers [53-55].
226
Reactive Blending
An alternative route to solution and bulk (melt grafting) methods has been proposed by Lee
[56] and Borsig [11, 12, 57-59] who used a solid phase polymerisation (modification) technique.
This technique offers the advantage of low temperatures and low operating pressures.
7.5.1.2 Photofunctionalisation
A PP-g-AA copolymer is a commercial one with trade name Polybond 1003 used as
compatibilising agent. The acrylate group can be either in the main backbone or can be
introduced by grafting. Many random copolymers and terpolymers of MA are tested as
compatibilising agents as: S/MA, S/MMA/MA, S/AN/MA, MA/VA [61] for polyvinyl
chloride (PVC)/collagen blends [62-63] and polysaccharide-based blends [64, 65] and
also graft copolymers such as SEBS-g-MA, styrene-butadiene-styrene three block
copolymer (SBS)-MA, ABS-g-glycidyl methacrylate (GMA), styrene-butadiene copolymer
(SB)-MMA-g-GMA used for PA/PES toughening. They react rapidly with amine groups
of polyamides, with hydroxyl groups of polyesters, polysaccharides, epoxy or oxazoline
groups, etc., of other polymers, [61-65].
Scheme 7.2
227
Handbook of Polymer Blends and Composites
bonds in styrene-butadiene copolymers does not allow the direct use of DEM in
combination with free radical initiators, because it rapidly crosslinks [66]. Such
functionalisation takes place preferentially at the aliphatic carbons of the PO blocks. To
favour functionalisation and crosslinking in certain position versus degradation, some
different approaches have been used. Among these, the reaction of SBR with thioglycolic
acid and its aliphatic ester occurs mainly by addition at the vinyl double bonds of the
1-2 butadiene units with high selectivity. Thiols are very suitable because they combine a
very high transfer capacity of hydrogen to formed radicals with the capacity of the free
radicals formed on the sulfur atom to add a double bond [66].
Epoxy groups are mainly introduced by grafting of GMA (Tnb = 189 C) onto PO. The
grafting of GMA onto ABS or PPE leads to good compatibilising agents [67, 68]. The
GMA functionalised polymers exhibit two different, independent kinds of functionality:
that of free radical of the methacrylate group and of the epoxy group of glycidyl. Epoxy
group reacts very easily with carboxyl, amine or other hydrogen active groups, end-groups
of PA, PES, etc., giving good compatibilising agents for PS/PO, PBT/PS, PA blends, etc.
Epoxy groups can be also introduced by reaction with epichlorohydrin onto natural
polymer backbone (lignin or starch) to improve their compatibility with synthetic polymers
(PO) [68, 69].
PP, PE, EPR or SEBS react with ricinol oxazoline maleinate on a co-rotating twin screw
extruder at 170-260 C at a screw speed of 50-65 rpm in the presence of peroxides. Both
batch and continuous mixing can be used, grafting time 5 minutes, under a nitrogen
atmosphere. The monomer and peroxide were absorbed into the powdered matrix material
at room temperature for at least 15 minutes. Degradation of PP limits the degree of its
grafting with oxazoline [18-21].
228
Reactive Blending
Scheme 7.3
By virtue of its tetra-functionality, bismaleimide (BMI) was [74] used as a reactive modifier
for polymer blends and as a crosslinking additive for elastomers [75]. The effect of BMI
on the morphology and properties of PP/EPDM blends is particularly noteworthy [76].
The BMI is activated to form a charge transfer complex with an accelerator (poly (2,2,4-
trimethyl-1,2-dihydroquinoline)). This system was found to increase the interfacial
adhesion between the dispersed EPDM phase and the PP continuous phase enhanced
nucleation of PP matrix. Functionalised PO with maleimides (1%) showed significantly
improved hydrophilicity [47, 77].
Melt state chemical reactions carried out in the extruder have been studied and
commercialised at Exxon Chem. Co., for over 30 years, under the name EXXELOR.
The characteristic bands for the monomer are: 1671 cm-1 for C=N stretching and oxazoline
ring bending; 1737 cm-1 for ester group, 1200-1100 cm-1 epoxy, 1090 cm-1 -C-O-C
asymmetric stretching is related to either chemical reactions or very strong physical
229
Handbook of Polymer Blends and Composites
interactions such as hydrogen bonding. The band, 720 cm-1CH2-; 700 cm-1 is characteristic
for grafted PS. Reference 2732 cm-1 for E/P copolymer, 1601/1463 block of styrene. The
grafted MA content was determined by quantitative IR spectroscopy using the intensity of
the absorption band at 1792 cm-1. It commonly varies between 0.5 wt% and 4.7 wt%.
Carboxyl groups can be also determined by a simple titration by KOH in n-butanol [13,
14, 56-59].
The chemical processes by which inter-chain copolymers can be formed [80-84] are:
reaction between end group of a polymer with pendant functionality of the other
copolymer resulting in a graft copolymer (PA/PO and their copolymers),
reaction between pendant groups or main chain of the two polymers, as a result
either graft copolymer or crosslinked network are formed (E-co-S/PP, natural rubber/
PE, PVOH/PE, PPE/PP or E or S copolymers), and
The radicals of the two polymers recombine with each other producing block or graft
copolymers. Free radical reactions can be produced by thermal decomposition of initiators
(peroxides, azo compounds), redox sources, photochemical decomposition, high-energy
radiation (-rays), thermal and/or mechanical shear. For saturated PO, reactivity is limited
primarily to the tertiary hydrogen atoms at branch point. These can be abstracted by
peroxides or radiation, producing free radicals, which can then graft to each other or to
unsaturated monomers or polymers in the vicinity (LLDPE-g-PS, HDPE/LLDPE/PS block
copolymers, LDPE-crosslinks or graft-PP, LDPE or PP/EPDM, ethylene-methyl acrylate
copolymer/EPDM or graft LDPE/PS, LDPE/HDPE/PP, HDPE/LDPE/EPDM, PO-g-MA
or other polar monomers) [43-49, 79].
230
imide group formation
ring opening
urea formation
urethane formation
Reactive Blending
The reaction of the two chain ends leads to diblock copolymers and seems to be the most
desirable one but reaction is not always limited to just this. The block and graft copolymers
formed during processing have molecular weights close to the sum of the two reacting chains.
More typically a polymer has multiple reactive groups along its chains so graft copolymers
result when it reacts with another polymer having a single reactive chain end. The number
of grafts can be controlled by the number of functional groups. If both chains have
multiple reaction sites then crosslinking occurs by extensive mixing.
Most combinations of polymers derived from a condensation reaction (PA and PES) which
exhibit nucleophilic, (i.e., electron donor), end groups such as NH2 (or NH), COOH (or
derivatives) and OH. They can react with electrophilic groups such as cyclic anhydride,
epoxy, isocyanate, carbodiimide, etc., placed along the chain by copolymerisation, end
capping or grafting. The possible reactions leading to the block or graft copolymers during
processing or reactive blending classified by the reacting groups are [26-30, 36, 84-85]:
231
Handbook of Polymer Blends and Composites
Three pairs of reactive groups are presently used in industry: anhydride/amine (primary
and secondary); carboxylic acid/epoxy; oxazoline/anhydride or carboxylic acid [86, 87].
The reaction of cyclic anhydride with polyamides has been the basis for numerous
polyamide-based blends (PA 6/PA 12, PP/PA, PE/PA, PA 6/PC, PA 6/SEBS, PA 5/styrene-
acrylonitrile (SAN), PA/ liquid crystalline polymer (LCP)) [3, 9, 15, 26, 27, 83].
Some workers consider that the MA content in the system is much more important than
that of the compatibilising agent [88, 89]
Acid + Amine Reactions. PA/PO are immiscible. PO offer melt strength, flexibility,
lubricity, impact strength, electrical resistance, low dielectric constant and loss, water
resistance and low price. PA offer melt fluidity, rigidity, strength, high temperature
performance and solvent resistance. MA-functionalised PO have been widely used as
232
Reactive Blending
compatibilising precursors during the reactive blending of various PO and PA. Carboxylic
groups are grafted onto PO and then reacted with PA end groups leading to PA-g-PO
copolymer. At high PO/PA ratio the PA generally contributes to impermeability to gasoline
while at high PA/PO ratios, PO contributes impact strength and water resistance.
Examples: PP, HDPE, LDPE, LLDPE, EPR, EPDM, metallocene PO, ethylene-vinyl alcohol
(EVOH); carboxylating agents PO-g-MA or with acrylic acid, methacrylic acid, succinic
acid, PO-ionomers.
233
Handbook of Polymer Blends and Composites
e) Oxazoline reacts with carboxylic groups, amine, phenol, and mercaptan to form an
esteramide so it is an effective compatibiliser. The oxazoline-modified PO have been
used for compatibilisation of PO with PA or PET in a one or two-step process. The
blends exhibited high impact strength and elongation at break.
Scheme 7.8
234
Reactive Blending
Transesterification has been used for years to manufacture polyesters or to modify the
properties of miscible polyester blends comprise of at least two components as: PES, PC,
polyethylene glycol (PEG), aromatic polyamides (PAr), LCP and polymers with side-
chain ester groups (PMMA, EVA, acrylic impact modifiers, etc). Transesterfication may
involve chain cleavage followed by recombination of the end groups to change molecular
weight distribution (MWD) and randomise the monomeric sequences in the polymers. It
235
Handbook of Polymer Blends and Composites
is catalysed by metal compounds, e.g., organo-tin, SnO2, Ti(OBu)4 and hindered tri-
phenyl phosphite, p-toluene sulfonic acid [80]. Compatibilisation of PET with PBT
depends on the extent of transesterification. In a twin-screw extruder the compatibilisation
of PC with PAr occurs to a greater extend than in a single extruder or solution casting. In
another procedure PET has been mixed in Brabender plasticorder with sodium salt of n-
butyl-2-sulfobenzoate resulting new types of chain ends [90].
Redistribution between two PA is a trans reaction between amine and amide groups.
Trans-amidation products of PA 6 and polyamide 66 (PA 66) are commercially available.
They show improved processability, enhanced mechanical and barrier properties. Blends
of aromatic PA are usually immiscible. Compatibilisation by trans-reaction may reduce
crystallinity. The reaction is controlled by temperature and intensity of mixing.
Commercial examples are PA 6 and PA 46 and PA 6 with poly (m-xylene adipamide).
Redistribution between PA and PES must be strictly controlled by process variables such as
screw speed, T, P, residence time, and catalysts. The reaction can be catalysed by p-toluene
sulfonic acid, dimethylol propanoic acid, etc. The reaction product should be impact modified.
The reactions during reactive blending of PBT and PA 66 [91] Scheme 7.11 - are:
236
Reactive Blending
7.6 Morphology
The final morphology of a blend obtained by reactive processing results from two
competitive phenomena: reduction in particle size and coalescence.
237
Handbook of Polymer Blends and Composites
number and their density increase. In fact, some of factors that favour drop breakup
such as high shear rates and reduced dispersed phase viscosity, lead also to increased
coalescence of polymer blends. Reactive compatibilisation lowers interfacial tension,
producing a steric stabilisation by reducing the size of dispersed phase particles. Due to
these competitive phenomena the relationship (Equation 7.1) is not always valid.
For example, the dispersed-phase particle size dramatically decreases as the amount of
MA grafted onto PP increases in spite of the fact that the viscosity ratio is 0.1 [97-99].
The decrease is very important at a MA content below 1-2 wt% or compatibilising agent
below 2-5 wt% in various polymer blends as it appears from examination of Figure 7.6
(as Nylon 6/PP-g-MA, PPE/PA-g-MA, SEBS/SEBS-g-MA (2%)/PA 6). Further increase of
MA concentration in copolymer or of compatibiliser in blends resulted in only minimal
additional decrease in particle size [26, 27].
For ABS-containing blends (ABS/PA, ABS/PES, ABS/PC, ABS/PBT), that are multiphase
blends, a functionalised polymer that is miscible with one of the phases is recommended
[26, 27, 100-106].
Figure 7.6 Effect of the amount of MA content in PP-g-MA on the particle size of the
dispersed phase in PA matrix and of SEBS-g-MA content on particle size of the dispersed
hase in Nylon 6 matrix and dispersed phase weight average particle size for mixture of
PA 6 with SEBS and SEBS-g-MA (2% MA). Redrawn from [26, 27]
238
Reactive Blending
The block and graft copolymers act by both lowering interfacial tension (or reducing
dispersed-phase particles) and making coalescence of particles that makes steric
stabilisation more difficult. In the latter, it is considered that coalescence preventing
lowering of interfacial tension is more important than the former. In the presence of a
reactive compatibiliser, the particle does not grow with concentration or in time the
morphology being stabilised by chemical bonds formed [26, 27, 104, 105] the coalescence
process being retarded.
The functionality or the number of reactive groups per chain is an important factor in
generation the morphology in reactive systems. Sometimes, reactive compatibilisation is
based on the reactive chain ends resulting from condensation. For example, Nylon 66 is
difunctional having two amine groups as end chains, while Nylon 6 is monofunctional.
Some kinds of Nylon-6-based materials could have a greater amine functionality [107].
Comparing the blends of Nylon 6 and Nylon 66 with SEBS-g-MA or ABS [29], Paul and
co-workers [22, 26, 108, 111] found that the rubber particles were extremely small and
spherical in Nylon 6 containing blends, while in the Nylon 6,6 containing blends, the
rubber domain were much larger and more complex in shape. The explanation could be
that the monofunctional chains simply graft onto the maleated rubber particles with one
attachment per chain, while the difunctional chains can form bridges or loops by
crosslinking so the polyamide can be entrapped within rubber particles resulting in
complex morphologies. To produce the fine dispersion of maleated rubber in difunctional
PA needed for efficient toughening it is necessary to use intense mixing capabilities of a
fully intermeshing, co-rotating twin-screw extruder, while very fine dispersions with
monofunctional PA can be achieved in single-screw extruder [111]. On the contrary, it has
been shown [112] that reactive compatibilisers with epoxide groups are not very effective
in Nylon 6/ABS blends, because they react with both the amine and acid end group. In this
case, the Nylon 6 is difunctional and it is difficult to obtain a good dispersion with ABS.
Therefore, there are severe limits in respect with functionality for PA and PES, too.
In the PA 6/SEBS blends, the particle size decreases with increase of molecular weight of
PA 6, increase of MA content of SEBS-g-MA and increase of intensity of mixing, the
dependence is continuous both for maleated and unmaleated SEBS [113-115] without a
minimum as predicted by Wus relationship [96].
239
Handbook of Polymer Blends and Composites
If the compatibiliser (C) is fully miscible with the rubber phase (B), one may expect it to
reside primarily within the rubber phase, with its functional groups forming chemical
bonds to thermoplastic chains (A) at the interface. This situation is expected to provide
the most efficient reduction in rubbery domain size through a decreased interfacial tension
and increased steric stabilisation.
If the functionalised compatibiliser resides at the rubber interface because it has some
thermodyamic affinity for this phase, but not enough to be miscible with it, than the
reduction of domain size can be expected, but it is not very effective. The third case is
when compatibiliser is even less attached to the rubber phase and it remains in the
thermoplastic matrix away from the interface, but it can increase the matrix viscosity
which, in turn, will lead to change in the morphology of the blend. That is why, the
reactive compatibilisation depends both on the content of reactive groups and also on
the order of the mixing steps.
The reactive polymer is a precursor for reactive compatibiliser formation and this at its
turn determines (by its structure and distribution between phases) the morphology and
the end-use of the compatibilised blend. In the case of the ABS/PA 66 system, in presence
of styrene-acrylonitrile (SAN)-co-GMA, the epoxy groups of GMA react with amine
end-groups of PA 66, so a grafted SAN-co-GMA-g-PA 66 copolymer is formed. Two
ways of mixing have been tested [116]: (1) Premixing of the SAN-co-GMA with PA 66
to form grafted copolymer followed by mixing with ABS. If the SAN-co-GMA contains
only up to 2% GMA, the grafted copolymer will be easily formed and it will contain
many free segments of SAN and SAN-co-GMA, which will migrate in ABS phase in the
mixing step. The grafted copolymer chains across traversing the interface efficiently
improve the interface adhesion and interface will have a large width. Alternatively, if the
GMA content is ~ 10%, a high content of grafted copolymer will be formed with high
molecular weight and low mobility, so it will reach with difficulty the interphase to
penetrate the ABS phase. The length of the free fragments is smaller and the interface
weaker than in the first case. If this mixing way is preferred, a copolymer with low
degree of grafting should be chosen.
(2) The second way consists in premixing of SAN-co-GMA with ABS and then mixing
with PA 66. The ABS is miscible with SAN-co-GMA at low GMA content and the
distribution will be uniform. The copolymer molecules in the vicinity of the interface are
able to directly contact and react with amino-end groups of PA 66. The grafting reaction
takes place at interface so the number of the grafting points is low and interface is weak.
The copolymer with high GMA content will be much efficient in this case.
Morphology of the blends evolves as a function of time in batch mixers and along the
length of extruders [117, 118]. It was found that the morphology evolved in a complex
240
Reactive Blending
Studies showed that interphase thickness increased with time to a plateau, whose level
depends on temperature and concentration of reactive sites.
Immiscible 2
Block copolymers 4 to 6
Polymer/copolymer 30
Reactive Compatibilisation 30 to 60
7.7.1 Rubber-Toughened PA
In 1979, Epstein from DuPont [200] and Houghton [201] patented the blends of
polyamides with functionalised elastomers. This opened a way for commercially successful
supertough Nylon products that means products with Izod impact strength higher than
800 J/m, value that is obtained for a minimum particle size of about 1 m [26, 27, 113-
115] and for a certain interparticle distance [108-110, 202].
241
reactive compatibilised Polymer
blends
Handbook of Polymer Blends and Composites
Table 7.3 Reactive Compatibilised Polymer Blends [9, 10, 15, 78, 79]
Blend Reactive processing procedure Some characteristics and Ref.
conditions and compatibiliser enhanced properties*
used or synthesised in situ
PP/EVA PP-g-MA Reduced permeability to
gases and liquids
PO/Elastomers Improved impact
resistance
PA 6 or PA 66/PP Poly(E-co-acrylic ester-MA) Recyclable, high impact [124]
or EPR or EPDM, and SEBS-MA. PP-g-MA-PA strength, particle size
or nitrile rubber 6, oxazoline-functionalised ~ 1 m
NBR
PO/PA or PES PO-g-GMA or MA Rigid
PO/EPDM Melt grafting 200-210 C, Three times increased [43-45]
DCP or BPO, roll-mill/press impact strength
method. PO functionalised
oligomers with MA diallyl
maleate, AA, ethylenglycol
dimethacrylate (EGDM),
bismaleimide (BMI)
PA 6/PO Melt grafting 200-210 C, Increased thermal [46]
DCP or BPO, roll-mill/press properties
method. Functionalised PO
and oligomers with MA
diallyl maleate, AA, EGDM,
BMI
iPP/LLDPE or Maleated pyrolysis products Increased degree of [47-49]
HDPE, Mixed of PE/PP blends crystallinity
plastics
80 PA 6 (62 000; Extrusion functionalisation: Improved toughness, [125,
NH2 content 32 35 rpm; temperature profile impact strength; fracture 126]
meq/kg/20 VLDPE 110/130/145/150 C behaviour; particle size ~
(885.6 kg/m3, 1.77 Blending: 110 rpm; twin 5 m; = 120-140% in
g/10 min; C3/C4 = screw, temperature profile: respect with 5% for
9:4, PA 6/ 170/210/235/230/220 C unfunctionalised
maleated PE octene 0.8 pph MA, 8 pph DEM
elastomer VLDPE-g-MA or DEM 0.43
%mol DEM;
242
reactive compatibilised Polymer
blends
Reactive Blending
243
reactive compatibilised Polymer
blends
Handbook of Polymer Blends and Composites
244
reactive compatibilised Polymer
blends
Reactive Blending
245
reactive compatibilised Polymer
blends
246
reactive compatibilised Polymer
blends
Reactive Blending
247
reactive compatibilised Polymer
blends
Handbook of Polymer Blends and Composites
248
reactive compatibilised Polymer
blends
Reactive Blending
249
reactive compatibilised Polymer
blends
Handbook of Polymer Blends and Composites
250
Reactive Blending
The size range for supertough PA widens as the molecular weight of the matrix increases
[113-115]. The worsening of properties observed in the region of lower particle size
were explained either by the changes in the orientation of polyamide crystallites [203] or
by their incapacity to cavitate [113-115, 203].
Figure 7.7 Extent of reaction in blends of Nylon x,y with SEBS-g-MA and estimated
interfacial thickness [3]
Reproduced with permission from D.R. Paul, Macromolecular Symposia, 1994, 78, 83,
Figure 5. Copyright 1994, Wiley-VCH
251
Handbook of Polymer Blends and Composites
To obtain commercially useful blends, all of the characteristics have to be carefully designed
such as rubber particle size, graft ratio, rubber content, viscosity of components, and
especially to optimise the composition and functionality of the compatibilisers, end group
type (monofunctional versus difunctional).
Toughened PA with maleated EPDM or EPR elastomers and PP-g-MA were developed
first, than the second generation of PA blends such as PA/PPE, PA/ABS, and PA/PC were
introduced. As a common feature, these blends synergistically combine the mechanical,
thermal and chemical resistance typical of PA with superior toughness, higher dimensional
stability at high temperature and lower moisture absorption provided by the second
polymer [204]. When grafting degree increases crosslinking of elastomers appears (EPR-
g-Ma (1.3%) crosslinked with hydroxyl-terminated PB). Elastomeric network could be
formed with thermoreversible linkages. These networks are receiving more interest, as it
is possible to process them several times by simple reheating.
The internal carbonate groups of PC may react with the amine terminal groups of PA 6
during the melt mixing process at 240 C. The formation of block PA 6-b-PC copolymers
suggested that these copolymers can act as compatibilisers, inducing compatibility of the
heterogeneous blends of PA 6/PC [83].
7.7.2 PC/ABS
PC offers, transparency, dimensional stability, flame retardance, high HDT and high
impact strength. ABS is a successful rubber-toughened thermoplastic, low cost, notch
insensitive/low thermal stability, poor thermal and chemical resistance. Their combination
gives materials with good properties.
Reactive core-shell impact modifiers are based for example, on crosslinked butadiene or
n-butyl acrylate rubber as core, while the shell consists of grafted chains that chemically
interact with the matrix. They are developed mainly for toughening polyamide-based
252
Reactive Blending
systems, which require strong interactions. Usually, the reactive grafts are chemically
incorporated during the final stage of emulsion polymerisation. The reacting groups can
be chemically bonded either in the shell or are part of an added polymer such as S/MA
copolymer, which is miscible with the shell chains. Due to the independent control of
particle size, supertough polyamide compositions can be obtained [151, 202-209].
Another example is: PBAc-rubbery core, PMMA-glassy shell compatibilised with epoxy
resin DGEBA [208]. The compatibilisation of digycidyl ether of bisphenol A (DGEBA) is
achieved by the reaction of glycidyl groups with amine group of Nylon 6 and hydrogen
bonds may be also generated between the hydroxyl groups and the carbonyl groups on
PMMA. That leads to a dramatic increase of the impact strength and a finer dispersing
of the core-shell particles in the Nylon 6 matrix. ABS is a core-shell particle impact
modifier [100-103].
When a core-shell impact modifier is used, the problem of compatibility between the
matrix and the polymer shell arises. In several cases, the shell is compatible with the
matrix and the core-shell particle well dispersed. Otherwise, the particles exist as lumps
instead of individual particles, which resulted in poor mechanical properties.
The Monsanto Company patented a low cost method as reactive extrusion process for
complete recycling of post consumer carpets, including Nylon 66 face fibres, PP backing
and other components in presence of 3-10% PP-g-MA and SEBS as impact modifiers
(reactive fillers), to produce pellets for use in injection moulding counter-rotating non-
intermeshing extruder [210]. No fibre separation or latex removal steps are required.
Post-consumer carpets (PCC) account for ~ 1 wt% of municiple solid waste in the US.
They contain Nylon 6, PES or PP. PP-g-MA (1-1.5%), eliminated the PP/Nylon 66
253
Handbook of Polymer Blends and Composites
One major problem that limits the direct blending of PO with engineering polymers is
the nonpolar structure of PO. However, reactor-made olefin based co- and terpolymers
with functional groups are commercially available and many functionalities can be
introduced into the structure of POs through post-modification [9, 51]. The presence of
branched olefin blocks favours a high degree of functionalisation (FD) values due to the
easier formation of macromolecular free radicals by hydrogen abstraction from tertiary
C-atoms (LLDPE, very low density PE (VLDPE), EPR, PP). In presence of double bonds
(as in EPDM), FD increases due to a combination of ene reaction between the double
bonds present on the polymer and the monomer; in the same time the crosslinking also
takes place. The PO which degrade by chain scission, like as PP or SEBS showed the
lowest FD values due to low permanence time of free radicals on the backbone [211].
Popular PO/PA blends are usually reactively compatibilised in three steps: (1) modification
of PO by grafting acid or anhydride moieties; (2) mixing the PO-adduct with PA; (3)
addition of main PO and PA components. Two-step blending by addition of the first-
stage acidic copolymer, e.g., PO-MA, or SEBS-MA, to a PA/PP mixture has been carried
out as well. The method can be extended to simpler system, e.g. impact modification of
PA, PC or PES, or to more complex systems e.g. PO/PA/PES.
To improve the impact resistance, lower water absorption and reduce material costs of
PA, various elastomers containing acrylate (methyl, butyl), acrylic acid or MA,
diethylmaleate, have been blended in the PA melts [9, 212-214].
254
Reactive Blending
7.7.6 PPE/PA
PPE is an amorphous resin with Tg = 210 C with good rigidity, high melting point.
PPE offers high rigidity, high Tg, good heat resistance, dimensional stability, outstanding
electrical properties required for automotive and electronics but it exhibits poor
processability, susceptibility to oxidative, UV and visible light degradation, low impact
strength, etc, so it can not be used as a neat resin. These last properties are offered by
PA. PA absorb water and have poor dimensional stability; reactive compatibilisation
of PPE has been at the forefront of its technology in the blends such as: PPE/PA/PPE-
MA, PPE/PBT/PC in 1970 and with PA in 1979. In 1985, Torray introduced high
performance blends for automotive applications PPE/PBT/PC/E-co-GMA. Hydroxyl-
terminated PPE was end capped with trimellitic anhydride acid chloride. The resulting
copolymer was used as compatibiliser for PPE/PA/SEBS blend. Mouldings showed high
impact and tensile strength [26, 27].
255
Handbook of Polymer Blends and Composites
Poly (hydroxybutyrate) (PHB) and its copolymers are environmentally degradable and
their mechanical properties are comparable with iPP and PET. Their toughening must be
achieved keeping their biodegradability (Biopol material) [194, 195]. The suitable
components for such purpose are poly (butyl acrylate) and polycaprolactone, both being
biodegradable polymers. First was polymerised in presence of PHB and the second was
melt blending in presence of peroxide. In both cases grafted species were formed, the end
products being completely biodegradable materials.
Compatibilisation by using ionomers has been used for impact modification of PA with
multifunctional ionomeric acrylic elastomers. The incorporation of ionic groups in a
polymer backbone induces drastic modifications in physico-chemical properties of the
original materials; molecular flexibility and network structure may be strongly modified,
providing a wide range of properties.
The ionic crosslinks are temporary and exist only at low temperatures. Poly (p-phenylene
terephthalamide) K-amide salt with P-4 vinylpyridine (VP) or P-2-VP by formation of
clusters and PA 6 and PA 66 with sulfonated styrenics containing 10 mol% of SO3H are
examples. The ethylene copolymers with acrylic and methacrylic acid partially neutralised
with Zn+2 or Na+ can be used to control the compatibility degree of non-miscible polymers
such as PO with PA, PET and PC. Incorporation of ionic groups (PS with 5 mol%
- SO3H groups; ethyl acrylate with 5 mol% VP; sulfonated polyisoprene + PS 5 mol%
VP); ion-dipole (P(S-co-lithium methacrylate) with 9.5 mol% of ionic groups
+ polyalkylene oxides). Feng and co-workers [218-220] found for the miscible lightly
(4.4 9.5 mol % sulfonate groups) sulfonated PS ionomers and N-methylated polyamide
256
Reactive Blending
Some areas of the phase separation in such systems are outside the scope of this chapter,
so only several aspects are presented.
Thermosets and thermoplastics can be successfully combined to bring out the best
properties of each system [227]. At high enough loading levels of reactive thermoplastics
in a thermoset matrix, the systems exhibit toughness and solvent insensivity properties
with retaining the processability of the neat thermoset.
Thermosetting polymers like phenolics, epoxies, unsaturated polyesters, etc., are frequently
used in formulations containing a low or high-molecular mass-rubber, a thermoplastic
polymer, etc., in an amount of 2-50 wt% compared to the thermoset. This extra
component, called the modifier, may initially be miscible or may phase separate during
cure (reaction induced phase separation). Toughnening of the polymer network means
increased values of fracture energy but decrease in the elastic modulus and yield stress.
This situation is typical of rubber-modified epoxies used as coatings, structural adhesives
or matrices of composites materials. When engineering thermoplastics like polyetherimide
or polyether sulfone are used the decrease of mechanical properties is avoided [228]. The
brittleness of thermosets (epoxies and polyesters) at room temperature can be overcome
by adding reactive low molecular mass rubbers, by reactive blending procedures, to the
brittle matrix [51, 229] especially to enhance the toughness of tri- and tetra-functional
epoxies and of unsaturated polyester resins.
257
Handbook of Polymer Blends and Composites
To obtain the highest modifying effect of an additive strong adhesive, interaction between
included particles and matrix is often a necessity. Obviously, the interfacial interaction is
the strongest when chemical bonds form between these phases. Therefore the modifiers
must contain functional groups that can react with one of the components of the oligomer
system being cured. Modifying additives are usually miscible with the initial oligomer,
however, such a single-phase system undergoes phase separation during the cure. As a
result, two phases are formed: a matrix phase, which is enriched with the cross-linked
polymer and a disperse phase rich in modifying additive. It is quite obvious that chemical
bonds between the phases of a heterophase polymer must appear before the phase
decomposition takes place in the course of curing. The rate of the reaction involving
functional groups on the surface of particles that have separated from solution during
the cure decreases sharply, as compared to the reaction rate of these groups in solution.
Controlling kinetics of phase separation during the formation of the IPN (or semi-IPN)
provide the method of generation of desired morphology and properties. One of the
most important characteristics of the morphological structure of a heterophase system is
the interfacial interaction. The miscibility with monomers may be varied significantly by
changing the nature of end-groups. Increasing polarity of rubber through incorporation
of polar comonomers in butadiene/acrylonitrile copolymers increased miscibility with
DGEBA-based epoxy resin [228] in the order amino, epoxy, carboxyl and non-functional
end-capped copolymers. Carboxylic acid terminated PS or PB increased energy per unit
area required to fracture at epoxy-polymer interface. Core-shell particles bearing
functional groups such as GMA can help to disperse thermoplastic phase in the
thermosetting matrix.
The reaction becomes diffusion-controlled and the concentration of the reacting groups
becomes lower. Therefore, the rate of reaction tends to zero. The second condition that
must be satisfied is that the reactivity of functional groups of modifying additive with
one of the components of the oligomer system must significantly exceed that in reactions
leading to the formation of a network.
258
Reactive Blending
process. A mixture of PEI and amine-terminated PEI gave the best result in the epoxy resin.
High performance material must meet the following criteria: high Tg, high modulus, good
adhesion to a variety of substances, processing easy, resistance to solvents.
Grafting of light stabilisers as hindered amine light stabilisers and antioxidants [233,
234], good adhesion filler/matrix (example EPR-g-DEM with silica are some examples
of various applications of reactive blending) [234].
The problems related to morphology stabilisation against damage during processing and
adhesion between the phases in the solid state still need basic research. In order to a
better understanding, control and modelling both of functionalisation and reactive
blending as an interdisciplinary research (chemistry, physics and engineering) should be
developed. To improve theoretical insights kinetics of process, mass and heat transfer for
many types of systems has to be known. Search for new functional, very reactive groups
is in progress. New solutions for scrap reclamation could be found.
References
1. Handbook of Polyolefins, 2nd Edition, Ed., C. Vasile, Marcel Dekker Inc., New
York, NY, USA, 2000, Chapter 1, p.17; Chapter 23, p.633; Chapter 26, p.723.
3. R.D. Paul, Die Makromolekulare Chemie - Macromolecular Symposia, 1994, 78, 83.
259
Handbook of Polymer Blends and Composites
8. E.J. Moskala, S.E. Howe, P.C. Painter and M.M. Coleman, Macromolecules,
1984, 17, 9, 1671.
10. Polymer Blends and Alloys, Eds., M.J. Folkes and P.S. Hope, Blackie Academic,
London, UK, 1993, 491.
14. V.P. Volkov, A.N. Zelenetskii, M.D. Sizova, N.Y. Artemeva, N.A. Egorova and
V.P. Nikolovskaya, 1st International Conference on Polymer Modification,
Degradation and Stabilisation, Palermo, Italy, 2000, (MoDeSt), Symposium 7,
Paper No. P7Th2100.
17. D.B. Todd in Reactive Extrusion: Principles and Practice, Ed., M. Xanthos, Carl
Hanser Publishers, Munich, Germany, 1992, Chapter 5, 203.
18. T. Vainio, G-H. Hu, M. Lambla and J.V. Seppala, Journal of Applied Polymer
Science, 1996, 61, 5, 843.
260
Reactive Blending
19. T. Vainio, G-H. Hu, M. Lambla and J.V. Seppala, Journal of Applied Polymer
Science, 1997, 63, 7, 883.
20. P. Hietaoja, M. Heino, T. Vainio and J.V. Seppl, Polymer Bulletin, 1996, 37, 3,
353.
21. H. Cartier and G-H. Hu, Journal of Materials Science, 2000, 35, 8, 1985.
22. A. J. Oshinski, H. Keskkula and D. R. Paul, Polymer, 1992, 33, 2, 268 and 284.
23. G.H. Hu, H. Cartier and C. Plummer, Macromolecules, 1999, 32, 14, 1713;
25. S. Kim, J.K. Kim and C.E. Park, Polymer, 1997, 38, 8, 1809.
26. R. Majumdar and D.R. Paul in Polymer Blends, Volume I, Formulation, Eds.,
D.R. Paul and C.B. Bucknall, John Wiley & Sons, New York, NY, USA, 2000,
Chapter 17, 540-673 and references herein.
27. E. Passaglia, S. Ghetti, F. Picchioni and G. Ruggeri, Polymer, 2000, 41, 12, 4389.
29. K. Dijkstra, J. ter Laok and R.J. Gaymans, Polymer, 1994, 35, 1399.
30. K.L. Borve, H.K. Kotlar and C.G. Gustafson, Journal of Applied Polymer
Science, 2000, 75, 3, 355.
31. G. Chen and J. Liu, Journal of Applied Polymer Science, 2000, 76, 6, 799.
32. K.L. Borve, H.K. Kotlar and G.C. Gustafson, Journal of Applied Polymer
Science, 2000, 75, 3, 361.
34. S.B. Brown and C.M. Orlando in Encyclopedia of Polymer Science and
Engineering, Volume 14, John Wiley & Sons, New York, NY, USA, 1988, 169.
35. J.L. Lee in Comprehensive Polymer Science, Eds., G.G. Allen and J.C. Bevington,
Pergamon Press, Oxford, UK, 1989, Volume 7, 379.
36. N.C. Liu and W.E. Baker, Advances in Polymer Technology, 1992, 11, 4, 249.
261
Handbook of Polymer Blends and Composites
37. M. Xanthos and S.S. Dagli, Polymer Engineering and Science, 1991, 31, 3, 929.
38. L. Mascia and M. Xanthos, Advances in Polymer Technology, 1992, 11, 4, 237.
39. N.G. Gaylord and M.K. Mishra, Journal of Polymer Science, Polymer Letters
Edition, 1986, 21, 23.
40. N.G. Gaylord and R. Mehta, Journal of Polymer Science: Polymer Chemistry,
1989, 26, 4, 1189.
41. N.G. Gaylord, M. Mehta and R. Mehta, Journal of Applied Polymer Science,
1987, 53, 7, 2549.
42. W. Heinen, C.H. Rosenmoller, C.B. Wenzel, H.J.M. Degroot, J. Lugtenburg and
M. van Duin, Macromolecules, 1996, 29, 4, 1151.
46. M. Dascalu, H. Darie, C,M. Tabarna, R. Darie and C. Vasile, Bulletin of the
Polytechnical Institute of Iasi, 1999, XLV(IL), 1-2, 281.
48. C. Vasile, R.D. Deanin, M. Mihaies, C. Roy, A. Chaala and W. Ma. International
Journal of Polymeric Materials, 1996, 37, 3-4, 173.
50. A.V. Machado, J.A. Covas and M. van Duin, 1st International Conference on
Polymer Modification, Degradation and Stabilisation, Palermo, Italy, 2000,
(MoDeSt), Symposium 7, Paper No. 07M1645.
262
Reactive Blending
52. S. Datta and D.J. Lohse, Polymeric Compatibilisers: Uses and Benefits in
Polymer Blends, Carl Hanser, Munich, Germany, 1996.
53. N.G. Gaylord, Journal of Polymer Science, Part C 1970, 31, 247;
54. N.G. Gaylord, H. Andripova and B.K. Patnak, Journal of Polymer Science,
Polymer Letters, 1971, 9 387.
55. N.G. Gaylord, H. Andripova and B.K. Patnak, Journal of Polymer Science,
Polymer Letters, 1972, 10, 2, 95.
58. E. Borsig and L. Hrckova, Journal of Macromolecular Science, 1994, A31, 10,
1447.
60. D. Curto, A. Valenza and F.P. La Mantia, Journal of Applied Polymer Science,
1990, 39, 4, 865.
61. B.C. Trivedi and B.M. Culbertson, Maleic Anhydride, Plenum Press, New York,
NY, USA, 1982, 296.
62. M. Lungu, G.G. Bumbu, G.C. Chitanu and C. Vasile inventors, Romanian
Academy, assignee; Romanian Patent A/00557/2000.
63. M. Lungu, G.G. Bumbu, M. Precup, I. Moldovanu, G.C. Chitanu, H. Darie and
C. Vasile, 1st International Conference on Polymer Modification, Degradation
and Stabilisation, Palermo, Italy, 2000, (MoDeSt), Symposium 7, Paper No.
02Tu1145.
64. G.G. Bumbu, C. Vasile, G.C. Chitanu and A. Carpov, Polymer Degradation and
Stability, 2001, 72, 99.
263
Handbook of Polymer Blends and Composites
65. A. Carpov, G.G. Bumbu, H. Darie and C. Vasile, 5th Mediterranean School on
Science and Technology of Advanced Polymer-Based Materials, Eds., E.
Martuscelli, P. Musto and G. Ragosta, 1998, Naples, Italy, 513.
68. C. Vasile, Milton Downey, Betty Wong, M. M. Macoveanu, M. Pascu, J-H. Choi,
C. Sung and W. Baker, Cell Chemistry and Technology, 1998, 32, 61.
71. U. Anttila, C. Vocke and J. Seppala, Journal of Applied Polymer Science, 1999,
72, 7, 877.
72. U. Anttila, C. Vocke and J. Seppala, Journal of Applied Polymer Science, 1999,
72, 43.
74. T. Inoue and T. Suzuki, Journal of Applied Polymer, 1999, 56, 9, 1113.
75. C.M. Liauw, V. Khunova, G.C. Lee and R.N. Rothon, Macromolecular Materials
and Engineering, 2000, 279, 6, 34.
78. R.D. Deanin, M.A. Manion, C.H. Chuang and K.N. Tejeswi in Handbook of
Polyolefins, Second Edition, Ed., C. Vasile, Marcel Dekker Inc., New York, NY,
USA, 2000, Chapter 22, 613.
264
Reactive Blending
79. R.D. Deanin and M.A. Manion in Handbook of Polyolefins, Second Edition, Ed.,
C. Vasile, Marcel Dekker Inc., New York, NY, USA, 2000, Chapter 23, 633.
80. A. Ajji and L.A. Utracki in Progress in Rubber and Plastics Technology, 1997,
13, 3, 153.
81. A. Ajji and L.A. Utracki, Polymer Engineering and Science, 1996, 36, 12, 1574.
82. S.B. Brown in Reactive Extrusion: Principles and Practice, Ed., M. Xanthos,
Hanser Publishers, Munich, Germany, Chapter 4, 1992, 75.
83. L.A. Utracki, Polymer Alloys and Blends. Thermodynamics and Rheology,
Hanser Publishers, Munich, Germany, 1989.
84. D.R. Paul in Thermoplastic Elastomers, Eds., G. Holden, N.R. Legge, R.P. Quirk
and H.E. Schroeder, Hanser Publishers, Munich, Germany, 1996, Chapter 15.
85. J.H. Anastasiadis, I. Gancarz and J.T. Koberstein, Macromolecules, 1989, 22, 3,
1449.
92. S.C. Tjong, R.K.Y. Li and X.L. Xie, Journal of Applied Polymer Science, 2000,
77, 9, 1964.
93. K-C. Chiou and F-C. Chang, Journal of Polymer Science: Polymer Physics, 2000,
38, 1, 23.
265
Handbook of Polymer Blends and Composites
95. N.A. Plate, A.D. Litmanovich, V.V. Yashin, I.V. Ermakov, Y.V. Kudryavtsev and
E.N. Govorun, Polymer Science Series A, 1997, 39, 1, 3.
97. A. Gonzales-Montiel, H. Keskkula and D.R. Paul, Polymer, 1995, 36, 24, 4587.
98. A. Gonzales-Montiel, H. Keskkula and D.R. Paul, Polymer, 1995, 36, 24, 4605.
99. A. Gonzales-Montiel, H. Keskkula and D.R. Paul, Polymer, 1995, 36, 24, 4621.
100. B. Majumdar, H. Keskkula, and D.R. Paul, Polymer, 1994, 35, 15, 3164.
101. B. Majumdar, H. Keskkula, D.R. Paul and N.G. Harvey, Polymer, 1994, 35, 20,
4263.
102. B. Majumdar, H. Keskkula, and D.R. Paul, Polymer, 1994, 35, 20, 5453.
103. B. Majumdar, H. Keskkula, and D.R. Paul, Polymer, 1994, 35, 25, 5468.
104. W. Hale, H. Keskkula, and D.R. Paul Polymer, 1999, 40, 21, 365, 3665.
105. W. Hale, L.A. Pessan, H. Keskkula, and D.R. Paul Polymer, 1999, 40, 15, 4237.
106. G. Wildes, H. Keskkula and R.D. Paul, Polymer, 1999, 40, 20, 5609.
107. A.J. Oshinski, H. Keskkula, and D.R. Paul, Journal of Applied Polymer Science,
1996, 61, 4, 623.
108. H. Keskkula and R.D. Paul in Nylon Plastics Handbook, Ed., M.I. Kohan,
Hanser Publishers, Munich, Germany, 1995.
110. C-C. Huang and F-C. Chang, Polymer, 1997, 38, 17, 4287.
111. B. Majumder, B.H. Keskkula and D.R. Paul, Journal of Applied Polymer Science,
1994, 54, 3, 339.
112. R.A. Kudva, H. Keskkula and D.R. Paul, Polymer, 1998, 39, 12, 2447.
266
Reactive Blending
113. O.J. Oshinski, H. Keskkula and D.R. Paul, Polymer, 1996, 37, 22, 4891.
114. O.J. Oshinski, H. Keskkula and D.R. Paul, Polymer, 1996, 37, 22, 4909.
115. O.J. Oshinski, H. Keskkula and D.R. Paul, Polymer, 1996, 37, 22, 4919.
116. F.C. Chang in Polymer Blends and Alloys, Eds., M.J. Folkes and P.S. Hope,
Blackie, London, UK, 1993, 491-521.
117. C. Maier, M. Lambla and K. Ilham, Proceedings of Antec 95, Boston, MA, USA,
Volume 2, 2015.
119. B. Majumdar, D.R. Paul and A.J. Oshinski, Polymer, 1997, 38, 8, 1787.
126. Z.Z. Yu, Y.C. Ke, Y.C. Ou and G.H. Hu, Journal of Applied Polymer Science,
2000, 76, 1285.
129. J.T. Yeh, C.C. Chao and C.H. Chen, Journal of Applied Polymer Science, 2000,
76, 1997.
130. D.J. van der Wal and L.P.B.M. Janssen, Proceedings of ANTEC 94, San
Francisco, CA, USA, 1994, 46.
267
Handbook of Polymer Blends and Composites
131. H. Li, T. Chiba, N. Higashida, Y. Yang and T. Inoue, Polymer, 1997, 38, 15, 3921.
132. X.M. Zhang and J.H. Yin, Polymer Engineering and Science, 1997, 31, 1, 197.
133. P. Hing and T.M. Ko, Polymer Engineering and Science, 1997, 37, 1326.
140. J.M. Felix and P. Gatenholm, Journal of Applied Polymer Science, 1991, 42, 609.
141. B. Majumar, H. Keskkula and D.R. Paul, Journal of Polymer Science, Part B:
Polymer Physics, 1994, 32, 2127.
142. B. Majumar, H. Keskkula and D.R. Paul, Journal of Polymer Science, Part B:
Polymer Physics, 1994, 857.
268
Reactive Blending
150. D.A. Costa and C.M.F. Oliveira, Journal of Applied Polymer Science, 1998, 69,
857.
151. X. Wang and H. Li, Journal of Applied Polymer Science, 2000, 77, 24.
153. C.G. Cho, T.H. Park and Y.S. Kim, Polymer, 1997, 38, 4687.
154. S.C. Tjong and Y.Z. Meng, European Polymer Journal, 2000, 36, 123.
155. G.L. Mantovani, L.C. Canto, E.H. Junior and L.A. Pessan, Macromolecular
Symposia, submitted.
156. M.K. Akkapeddi, B.V. Buskirk and G.J. Dege, Proceedings of ANTEC 94, San
Francisco, CA, USA, 1994, Volume 2, 1509.
157. C.R. Chiang and F.C. Chiang, Journal of Applied Polymer Science, 1996, 61,
2411.
158. C.R. Chiang and F.C. Chiang, Polymer, 1997, 38, 3807.
160. E. Yilor, I. Yilgor and T. Sumazcelik, Polymer Preprints, 1998, 39, 1, 552.
161. W.T.W. Tseng and J.S. Lee, Journal of Applied Polymer Science, 2000, 76, 1280.
162. C.Y. Han and W.I. Gately, inventors; General Electric, assignee; US Patent
4,689,372, 1987.
163. F.F. Khouri, R.J. Halley and J.R. Yates, inventors; General Electric, assignee; US
Patent, 5,212,255, 1993.
269
Handbook of Polymer Blends and Composites
164. F.F. Khouri, R.J. Halley and J.R. Yates, inventors; General Electric, assignee; US
Patent 5,247,006, 1993.
165. M.K. Akkapeddi and B. Van Buskirk, ACS Polymer Preprints, 1992, 67, 317.
166. A. Sugio, M. Okabe and A. Amagai, inventors; Mitsubishi Gas Kagaku KK,
assignee; EP 148,774, 1985,
167. S.B. Brown and R.C. Lowry, inventors; General Electric, assignee; US Patent
5,153,267, 1992.
168. S.B. Brown, C.F.R. Hwang, F.F. Khouri, S.T. Rice, J.J. Scobbo and J.B. Yates,
inventors; General Electric, assignee; US Patent 5,223,351, 1996.
169. C.Y. Han, S.B. Brown, E.W. Walles, T. Takekoshi and A.J. Caruso, inventors;
General Electric, assignee; US Patent, 5,122,578, 1992.
170. H. Izutsu, K. Yamamura and E. Tanigawa, inventors; Dainippon Ink & Chemical
Inc., assignee; JP 6166783, 1994.
171. S.B. Brown, K.H. Dai, C.F.R. Hwang, S.T. Rice, J.J. Scobbo and J.B. Yates,
inventors; General Electric, assignee; US Patent, 5,504,165, 1996.
172. S.B. Brown, K.H. Dai, C.F.R. Hwang, S.T. Rice, J.J. Scobbo and J.B. Yates,
inventors; General Electric, assignee; US Patent, 5,612,401, 1997.
174. M. Okamoto and T. Inoue, Polymer Engineering and Science, 1993, 3, 176.
175. S.C. Tjong and Y.Z. Meng, Polymer, 1997, 38, 4609.
176. S.C. Tjong and Y.Z. Meng, Polymer International, 1997, 42, 209.
177. S.C. Tjong and Y.Z. Meng, Polymer, 1998, 39, 99.
180. H.J. ODonnell and D.G. Baird, Polymer, 1995, 36, 3113.
270
Reactive Blending
183. W.E. Baker and M. Salem, Polymer Engineering and Science, 1988, 28, 1427.
184. W.E. Baker and M. Salem, Journal of Applied Polymer Science, 1990, 39, 655.
185. N.C. Liu, H.Q. Xie and W.E. Baker, Polymer, 1993, 34, 4680.
186. N.C. Liu, H.Q. Xie and W.E. Baker, Polymer, 1994, 35, 988.
188. K. Oksman and C. Clemans, Journal of Applied Polymer Science, 1998, 67, 1503.
190. E. Passaglia, S. Ghetti, F. Piechioni and G. Ruggeri, Polymer, 2000, 41, 4389.
192. M. Chiriac, B.S. Munteanu, G.G. Bumbu, M. Burlacel, A. Ioanid and C. Vasile,
Macromolecular Materials Engineering, 2000, 283, 26.
197. R.W. Venderbosch, H.E.H. Meijer and P.J. Lemstra, Polymer, 1994, 35, 4349.
198. K-J. Eichhorn, I. Hopfe, P. Ptschke and P. Schmidt, Journal of Applied Polymer
Science, 2000, 75, 1194.
199. J.Z. Liang and R.K.Y. Li, Journal of Applied Polymer Science, 2000, 97, 409.
271
Handbook of Polymer Blends and Composites
203. O.K. Morantoglu, A.S. Argon, R.E. Cohen and M. Weinberg, Polymer, 1995, 36,
25, 4771.
205. R.J.M. Barggreve and R.J. Saymans, Polymer, 1989, 30, 63.
206. R.J.M. Barggreve and R.J. Saymans, Polymer, 1989, 30, 71.
207. R.J.M. Barggreve and R.J. Saymans, Polymer, 1989, 30, 78.
208. M. Lu, H. Keskkula and D.R. Paul, Polymer, 1993, 34, 9, 1874.
209. M. Lu, H. Keskkula and D.R. Paul, Polymer Engineering and Science, 1994, 34,
1, 33.
210. C.G. Hagberg and J.L. Dickerson, Proceedings of ANTEC 96, Indianapolis, IN,
USA, 1996, Volume 1, 288.
212. M. van Duin and R.J.M. Borggreve in Reactive Modifiers for Polymers, Ed., S.
Al-Malaika, Kluwer Academic Publishers, Dordrecht, The Netherlands, 1997,
Chapter 3, 133.
214. J.Z. Liang and R.K.Y. Li, Journal of Applied Polymer Science, 2000, 77, 408.
216. R.J. Clark, Proceedings of ANTEC 94, San Francisco, CA, USA, 1994, Volume
3, 2572.
272
Reactive Blending
218. Y. Feng, A. Schmidt and R.A. Weiss, Macromolecules, 1996, 29, 3909.
219. Y. Feng, R.A. Weiss and C.C. Han, Macromolecules, 1996, 29, 3925.
220. Y. Feng, R.A. Weiss, A Karim, C.C. Han, J.F. Ankner, H. Kaiser and D.G. Peiffer,
Macromolecules, 1996, 29, 3918.
221. M.C. Pascu, G. Popa, M. Grigoras and C. Vasile, Synthetic Polymer Journal,
1998, 2, 158.
222. U. Epple and H.A. Schneider, Thermochimica Acta, 1987, 112, 123.
223. U. Epple and H.A. Schneider, Thermochimica Acta, 1990, 160, 103.
224. M. Bolsinger and H.A. Schneider, Macromolecular Chemistry and Physics, 1994,
195, 2683.
225. M. Bolsinger and H.A. Schneider, Journal of Thermal Analysis, 1998, 51, 643.
226. M. Bolsinger and H.A. Schneider, Journal of Thermal Analysis, 1998, 52, 115.
227. C.D. Athanasiou and R.J. Farris, Proceedings of ANTEC 95, Boston, MA, USA,
1995, Volume II, 2766.
228. R.J.J. Williams, B.A. Rozenberg and J.P. Pascault, Advances in Polymer Science,
1997, 128, 97.
230. B.A. Komarov, E.A. Dzhavadyn, V.I. Irzhak and B.A. Rozenberg, Polymer Science
Series A, 1997, 39, 2, 153.
231. B.A. Komarov, E.A. Dzhavadyn, V.I. Irzhak and B.A. Rozenberg, Polymer Science
Series A, 39, 2, 1997, 237.
232. R. Subramanian, Y.H. Huang, S. Zhu, A.N. Hrymak and R.H. Pelton, Journal of
Applied Polymer Science, 2000, 77, 1154.
273
Handbook of Polymer Blends and Composites
235. E. Passaglia, J.B. Beirao de Veiga, M. Aglietto and F. Ciardelli, 1st International
Conference on Polymer Modification, Degradation and Stabilisation, Palermo,
Italy, 2000, (MoDeSt), Symposium 7, Paper No. P7 Th1140.
274
Advanced Polymers: Interpenetrating Networks
8
Advanced Polymers: Interpenetrating
Networks
The term interpenetrating network (IPN) was introduced for the first time by Millar in
1960s in a scientific study about polystyrene networks [1]. Since that time, the field of
IPN has expanded dramatically. Reviews of the field include numerous articles and books
by Sperling [2, 3] and Kim [3], Klempner and co-workers [4, 5], Lipatov and Sergeeva
[6] and Bischoff and Cray [7]. However, it is difficult to identify quickly, many IPN-
related works because many articles, patents and commercial products on IPN do not
explicitly contain the terms interpenetrating polymer networks or IPN.
One can ask: why IPN? Is it not enough to know how many classes of polymers, blends,
alloys and composites exist? One can successfully search for simple polymeric materials
with all the properties that one needs for ones applications. Furthermore, IPN complexity
does not mean that the polymer will be more expensive or sensitive to deterioration.
Examples of IPN are seen everyday without realising it. Like Jonas Aylsworth, Thomas
Edisons chief chemist, who in 1914 mixed rubber and sulfur with a crosslinked phenol-
formaldehyde resin (which was brittle and easily broken), in the hope of obtaining a
material with more toughness for phonograph records. It was a great success, his lucky
star rising in that day from the constellation of interpenetrating polymer networks, but
no one named this constellation until the 1960s and even today many look at it, without
knowing his name.
However, the extraordinary properties of what Millar first called IPN could be of great
interest to someone who was experiencing difficulty in using traditional polymeric
materials for certain applications.
The extreme broadening of the range of polymeric material applications needs new,
simple, flexible and efficient solutions. Surprisingly or not, these solutions are quite
frequently found by using complex and rigid IPN structures.
275
Handbook of Polymer Blends and Composites
Some hybridation features could be extrapolated from the live organisms to these
polymeric systems. So, an IPN resulting from the combination of two different polymers
could exhibit not only a mixture of the individual, but also a synergism of their properties.
Furthermore, an IPN may look like one of the constitutive polymers but act as the other,
or it may be quite different from both. It is also important to know that controlling the
IPN synthesis one can easily control the properties of the end product. Moreover, IPN
are intercompatible, self-priming, self-repairing and very stable over time. As consequence,
one IPN can replace with high efficiency, many competitive and expensive materials for
a particular job.
8.2 Introduction
Reciprocally permanent mutual entanglements that occur when at least one network is
synthesised and/or crosslinked in the intimate presence of the other will prevent the total
phase separation and stabilise the morphology by keeping the networks together. A
schematised spatially distribution of networks chains in an IPN is depicted in Figure 8.1.
Because there is no chemical bonding between the networks, each network may retain its
individual properties independently of its individual proportion in the blend. As result
an improvement can be attained in properties such as mechanical strength, impact
resistance, toughness etc, with respect to the corresponding homopolymer networks by
use of IPN.
Interpenetrating polymer networks can be distinguished from the other multiphase systems
through their bicontinuous structure ideally formed by two crosslinked polymers that
are in intimate but not chemical contact, yielding materials with properties ranging from
elastomeric to high impact plastics depending on the composition and the degree of
crosslinking. So, IPN are not simple, macroscopically homogeneous, physical mixtures
of different polymer species, like polymer blends. They are also different from graft
copolymers and polymer complexes that involve either chemical bonds and/or low degree
of crosslinking. From this point of view only (the preparative steps and most of the
structure and properties presented being quite different), IPN can be generically named
polymer alloys through which polymer blends can be made chemically compatible to
achieve the desired phase morphology. Furthermore, IPN swell and creep and flow are
suppressed, features which are not presented by other macromolecular assemblies. Despite
the general occurrence of some degree of phase separation because of the combined
276
Advanced Polymers: Interpenetrating Networks
277
Handbook of Polymer Blends and Composites
The IPN network structure resulting through either chemical or physical crosslinks among
the polymer molecules consists of a complex architecture of crystalline, glassy or ionic
phase domains that contain multiple segments of polymer chains. While most IPN do
not interpenetrate at the molecular level, when properly engineered they may form finely
divided co-continuous phase domains in the nanometer size range, behaviour often
responsible for their unique properties. Independent of the chosen method of synthesis,
the miscibility decreases, and a phase separated IPN will be obtained during the
polymerisation of the monomers. Since crosslinking reduces phase separation and, in
particular, reduces domain size, it often results in finely dispersed phase domains ranging
from about 10 to 100 nm. The kinetics and extent of crosslinking and phase separation
have great influence over the phase morphology [2, 3].
The co-continuous nature of IPN can lead to materials having a greater toughness than
each of the constituent polymers. When microheterogeneous phase domains are in the
range from ~10 to ~20 nm, the whole material is essentially a interphasic one, so the
glass transition temperatures (Tg) tend to be very broad, intermediate between those of
the two component polymers, giving materials that can absorb energy or are useful as
damping materials over a broad temperature and frequency range.
A variety of methods and techniques are used for studying the main IPN features such as
miscibility, morphology, synthesis and curing processes, physical, thermo-mechanical,
swelling and electrical properties.
The most common procedures for studying miscibility in IPN consist of macroscopic
determination of the glass transition temperature and compatibility behaviour by
differential scanning calorimetry (DSC) and dynamic mechanical analysis (DMA)
techniques.
When IPN constitutive polymers are completely miscible, a single glass transition
intermediate between the Tgs of the homopolymers appears, which indicates an infinite
number of phases of a different composition. Partially miscible polymers may show one
broad glass transition or two sharp transitions, shifted inward compared with those of
the homopolymers, each phase containing a certain amount of the other component. For
immiscible polymers, two glass transition temperatures can be observed, separate and
similar with those of the homopolymers. Thus, the presence of one or two glass transition
temperatures, their sharpness and positions gives an indication of characteristic
278
Advanced Polymers: Interpenetrating Networks
DSC techniques can also be used, together with Fourier transform infra red spectroscopy
(FT-IR), for controlling the curing processes. So, the heats of reaction, together with the
variations in Tg value and shape during IPN formation gives information about curing
evolution and thermal characteristics of both the final material, and the starting and
intermediate components. The DSC data can be correlated with the IR evolution of the
functional groups characteristic bands, allowing the determination of the reaction rates
and transformation efficiency of the starting materials.
DSC has been also used in studying the thermal stability of the systems, while the heat of
photocure can be measured by the differential photocalorimetry (DPC). DMA has been
applied to determine the viscoelastic properties of cured IPN. Thermal-modulated
differential scanning calorimetry (M-TDSC) could be used for the analysis of IPN
interphases and quantitative calculation of interphase mass fraction, interfacial thickness
composition and distribution, for the estimation of vitrification during isothermal curing.
M-TDSC results could be correlated with transmission electron microscopy (TEM) and
dynamic mechanical thermal analysis (DMTA) data.
DMTA analysis of thermal relaxation, associated with DSC and DMA data, gives
information on elasticity modulus changes that, compared with those obtained from
modulus/composition theories [9, 10], allow relative heterogeneity and mechanical
damping determination.
Damping is the most sensitive indicator of all kinds of molecular motions displayed by a
polymer in a solid state and can be correlated to many transitions, relaxation processes,
structural heterogeneities, and to the morphology of multiphase and crosslinked systems [11].
The integration of the area under the loss modulus curve has been suggested as a
quantitative measurement of the damping capacity (tan ) of the material by many authors
[12-14], supported by the fact that the loss modulus is directly proportional to the heat
279
Handbook of Polymer Blends and Composites
dissipated per cycle of dynamic deformation. The experimental Tg data could be correlated
using the equation developed by Brekner [15]:
(T
g ) = (1 + K )w
Tg1
(K1 + K 2 )w22c + K 2 w23c (8.1)
(Tg2 T )
g1
1 2c
K GT w2
w2 c = (8.2)
w1 + K GT w2
1 1 G
= + 1 f (8 .3)
G Gm G m G + (G f G)
2( 4 5v)
where = , v is the Poisson ratio of the system;
15(1 v)
G shear modulus;
Gf, Gm shear modulus for the minor component polymer chains viewed as
filler and matrix network, respectively;
280
Advanced Polymers: Interpenetrating Networks
The Budiansky model can also be expressed for a rigid/soft type composition of the IPN
through the following equation:
v1 v2
+ =1 (8.4)
G1 G2
1 + 1 +
G 1 G 1
The change in chemical potential of individual IPN, 2, may be estimated with the
equation of Gibbs-Duhem:
1 2
w1 + w2 =0 (8.5)
w1 w2
w1, w2 the mass fraction of the solvent and polymer in a swollen system,
respectively.
The values of the polymer-filler interactions in filled IPN, gP-F, can be calculated based
on the analysis of the solvent sorption by an unfilled polymer, the filler and filled polymer
according to the equation [18-20]:
where: g1 free energies of the filled polymer interaction with a large amount of
solvent;
g11 free energy of the filler with a large amount of solvent;
281
Handbook of Polymer Blends and Composites
g111 free energies of the unfilled polymer interaction with a large amount of
solvent;
n mass fraction of the polymer
m mass fraction of filler in filled polymers
Temperature dependence of the peak frequency of the dielectric loss for the - and
-relaxation mechanisms of the IPN can be determined by using the Vogel-Tammann-
Fulcher equation [21, 22]:
= 0[ ( 0 ) ]
B TT
(8.7)
From this equation can be calculated the apparent activation energy, E [22, 23]:
E ln BT2
= = (8.8)
(T T0 )
k 1 2
The characteristic relaxation parameters for -relaxation in IPN can be extracted from a
single Havriliak-Negami function [22, 24]:
( ) =
(8.9)
1 + (i)1
The contribution of each relaxation can be separated by fitting the experimental curves
to a model consisting of the sum of two Havriliak-Negami equations, one for each
overlapped relaxation:
282
Advanced Polymers: Interpenetrating Networks
( ) = a
(8.10)
b
1 + i
0
The degree of segregation in IPN can be calculated from the maximum relaxation
parameters [25]:
h1l1 + h 2 l2 + h m l m
SG = (h1 + h 2 ) (h10 + h 02 ) (8.11)
L
where: h1, h2 mechanical loss for each component at different degree of segregation;
h01, h02 maximum values of mechanical loss for the pure components;
hm maximum values of mechanical loss for the relaxation transition;
L the interval between the glass transition temperatures of the pure components;
l1, l2, lm the shifts in temperature scale.
Theoretical conversion to gelling for each IPN phase could be determined using the
Flory statistical theory [8, 26]:
1
1 2
gel = (8.12)
(fm1 1)(fm2 1)
The segmental mobility in IPN can be estimated by the calculation of parameter with
the Fox and Frisch equations [27, 28]:
Tg Tg( est.)
= (8.13)
Tg( est.) +1
283
Handbook of Polymer Blends and Composites
F
= x 1 c (8.15)
m
F
Small angle neutron scattering (SANS), small angle x-ray scattering (SAXS), atomic force
microscopy (AFM), TEM, and nuclear magnetic resonance (NMR), are also used to
characterise the morphology, particularly the extent of mixing related to the domain size
and to the extent of interfacial mixing, of IPN made by a wide variety of synthesis
procedures.
NMR can be successfully used to probe the dimensions of phase-separated regions and
also the intimately mixed fraction of the IPN materials. Meanwhile, the electron
microscopy techniques provide enough details about the structural features and distances
of the order of ~100 nm to allow a comparison with the NMR results given by NMR
spin diffusion techniques. Mulder and co-workers [29] have used rapid 1H NMR spin
diffusion to bridge distances up to 100 nm coupled with high resolution 13C NMR that
affords the selection and detection of 13C label polarisation. 13C NMR spectroscopy is
furthermore required to discriminate between rigid IPN material components, 1H NMR
resolution being generally not sufficient due to the strong dipolar couplings and limited
chemical shift dispersion.
When crosslinking affects only one polymer component, the resulting material is a semi-
IPN, while a full-IPN is made when all species are in network form. If all constitutive
284
Advanced Polymers: Interpenetrating Networks
polymers are treated at the same time, one in the presence of the other(s), a simultaneous
IPN is obtained; otherwise the IPN is called sequential. These procedures of synthesis are
comparatively presented in Figure 8.2.
Figure 8.2 The two major pathways for synthesis of an IPN system
285
Handbook of Polymer Blends and Composites
The IPN constitutive polymer species can be crosslinked during or after the mixing, in a
post-curing process.
These major categories can be furthermore categorised into specific groups like
thermoplastic, hydrogel, latex, grafted, hybrid, filled, conductive, elastomeric, gradient
and homo IPN, based on their characteristic features. In many cases the borders between
different types of IPN are also interpenetrated, making an exhaustive classification
difficult. The most commonly polymer networks used in IPN formation are those based
on polyurethanes, polysiloxanes, polyesters and epoxy resins, polyacrylates, carbohydrate
polymers and rubbers.
When polymer networks are simultaneously formed and subsequently crosslinked one in
the presence of the other, in different and non-interfering reactions, a simultaneous IPN
(sim-IPN) is obtained. Compatibility is greater in a mixture of monomers and prepolymers
than in a mixture of polymers, so sim-IPN generally exhibit a high degree of intermixing
and require crosslinking of two mostly compatible polymers because the degree of
thermodynamic incompatibility of simultaneously growing chains greater affects the extent
of phase separation.
In the preparation of sim-IPN the degree of phase segregation and thus the final
morphology attained (ranging from very fine phase dispersion to more or less
individualised domains) are principally influenced by the relative rate of the two
competitive kinetic processes of phase separation and network formation. Both processes
are controlled by the reaction conditions such as composition and miscibility of
components, amount of catalyst, temperature, viscosity and thermodynamic compatibility
of contained polymers, all of which determine boundary layer composition and the
properties of the resulting materials. A selective adsorption of some components by the
others takes place simultaneously, a phenomenon which is responsible for incomplete
phase separation at boundary layers and no uniform polymer distribution in the thickness.
While equilibrium thermodynamics describes the overall tendency for a system to remain
miscible or to realise a phase separation, it is dynamics that determines the speed of a
possible phase separation. Spinodal decomposition together with nucleation and growth
are the two types of phase separation dynamics generally involved. The average size of
the molecules increases with increasing time and therefore the diffusion coefficients
decrease rapidly, so the dynamics of the phase separation become very slow and may
facilitate an irreversible freezing of the spinodal decomposition in an stationary state. As
a consequence, the formation of homogeneous sim-IPN can be interpreted as a freezing
of spinodal decomposition by irreversible chemical reactions [30-33].
286
Advanced Polymers: Interpenetrating Networks
Three important events take place in the process of sim-IPN synthesis, gelation of the
first polymer, gelation of the second and phase separation of former from the latter
polymer. By controlling their order of occurrence one can manage the resulting sim-IPN
morphology and properties. Based on flow-gelation characteristics, turbidity and FT-IR
analysis of the extent of polymerisation of each component, metastable phase diagrams
can be constructed which describe the condition under which gelation of one polymer
may be caused to precede or follow the other, and under which conditions phase separation
precedes or follows one or both gelations [34, 35].
The time of network formation, associated with the physical interlock of components, is
the time when both polymers reach the gel point and thus the phase domain size and
surface segregation cannot increase much further beyond this point [36, 37]. By reaching
simultaneously the gel point of the two essentially independent polymerisation reactions,
simultaneous network formation provides the best interchain contact for a given
composition. Some examples of sim-IPN are presented in Table 8.1.
287
Handbook of Polymer Blends and Composites
compositions and properties of resulting seq-IPN. The first formed network is stretched
due to the swelling step and tends to form the matrix, whereas in sim-IPN both networks
are more or less relaxed and the polymer that gels first forms the more continuous phase.
288
Advanced Polymers: Interpenetrating Networks
8.4.3 Semi-IPN
These polymer networks are distinguished from full interpenetrating polymer networks
because the constituent linear or branched polymers may be separated from the constituent
polymer networks without breaking chemical bonds, only by swelling in an appropriate
289
Handbook of Polymer Blends and Composites
solvent, so that they could be considered as intermediates between IPN and blends. Semi-
IPN exhibit minor swelling capacity compared to the corresponding full-IPN.
290
Advanced Polymers: Interpenetrating Networks
obtained with relatively low molecular weight NKGM, which plays an important role in
plasticising, cure accelerating and strengthening of the semi-IPN product. The PHB/net-
PEG semi-IPN show improved strength and toughness, associated with a crystallinity
reduced below 24.1% in comparison with 67.7% for a linear PEG macromer.
291
Handbook of Polymer Blends and Composites
8.4.4 Full-IPN
Full IPN represent the most complete case of an IPN according to its definition, in that
all polymers are crosslinked in network form. The two networks are ideally juxtaposed,
which generates a lot of entanglements and interaction between them [58] (Figure 8.4).
A higher degree of crosslinking restricts the mobility of polymer chains and suppresses
the phase segregation and the size of the final morphology. Linear chains in polymer
networks are diminished in size with increasing crosslink density until they collapse into
aggregates, even at extremely low linear chain concentrations, affecting the homogeneity
of polymer networks. In some cases a synergistic improvement of properties like toughness,
elongation at break, tensile and impact strength, mechanical damping, etc., can be
observed.
292
Advanced Polymers: Interpenetrating Networks
When both polymers form structurally regulated IPN with the same composition over
all, the full-IPN so formed is named homogenous IPN (homo-IPN). This could happen
when it is very difficult to modify the position of one polymer chain as a consequence of
a high crosslink density. So the second polymer chains will grow and interpenetrate the
existing network, causing a forced compatibilisation of both polymers to take place
[59]. On contrary, if the polymers are immiscible and the crosslink density of the first or
more rapid formed polymer network is small, the growing chains of the other will push
apart the already existing chains and a phase-separated IPN will be obtained.
A good opportunity for studying the intimate factors that affects the behaviour of full-
IPN materials is done by the synthesis of different architectures based on the crosslinking
of two polymers belonging to the same class of macromolecules. Duenas and co-workers
[52] did work with PBA/poly butyl methacrylate (PBMA) IPN made by a sequential
method, using dielectric and dynamo-mechanical techniques for the determination of
miscibility and molecular mobility.
Sometimes, the distinction between semi- and full-IPN can only be made by the ratio of
mixed polymers. Mao and co-workers [60] related such behaviour for calcium chloride
crosslinked high acyl gellan/low acyl gellan gels IPN. Each acyl gellan polymer type that
is in a higher amount, i.e., in a ratio of 75/25, forms a continuous gel matrix, so that the
other one is dispersed discretely. However, at a high acyl gellan/low acyl gellan ratio of
50/50, both polymers form separate, continuous networks, resulting in full-IPN structures.
For this reason, a further separation of IPN by the polymerisation method, composition
and type of constitutive polymer species, gives a better distinction between IPN.
Such classification is shown in Table 8.4 using examples of binary IPN systems:
293
elastomeric hydrogel grafted hybrid conductive filled homogeneous
gradient heterogeneous latex
Some of the most representative subtypes of IPN materials will be discussed in the
following sections.
When networks are formed by physical crosslinks rather than chemical crosslinks between
constituent polymers, an IPN is obtained, which flow at elevated temperature in a similar
way to thermoplastic elastomers, while behaving like conventional thermoset IPN at their
application temperature [7]. Usually, at least one component is a block copolymer and the
other one a semi-crystalline or glassy polymer. Typical physical crosslinks include glassy
blocks, ionic groups or crystalline segments, thermoplastic IPN being hybrids between IPN
and polymer blends. Depending on the continuity and proportion of phases, this kind of IPN
can exhibit a wide range of properties, from reinforced rubber to high impact plastics.
294
thermoplastic
usually either polyacrylates [34, 35, 66-70], polyester [8, 71, 72], or polystyrene [46]
(Table 8.5, Table 8.6).
295
Handbook of Polymer Blends and Composites
When both IPN components have thermoplastic behaviour, the resulting thermoplastic
material is usually called elastomeric IPN. An example is 2-hydroxyethyl methacrylate-
terminated polyurethanes/polyetherurethanes (HPU/PEU) IPN [73]. Enhanced mechanical
strength was also reported for semi-IPN composed from elastomers only, (i.e.,
thermoplastic PVC/urethane) [28, 74].
Grafted IPN are usually made by simultaneous synthesis and exhibit complex structure
and morphologies; therefore it is difficult to control the length and the efficiency of the
grafts. Degree of grafting and microstructure of the resulting graft IPN are affected by a
series of interconnected factors like reactivity and amount of grafting agent and host
polymer, network formation behaviour, grafting temperature, time and reaction rates.
In most cases, grafting reactions occur to a low extent during IPN synthesis as a secondary
process. Grafting may be neglected only if the concentration of crosslinks significantly
exceeds the concentration of graft sites and if the morphology and physical properties
are unaffected.
Induced grafts may act as compatibilisers between phases. When the grafting reaction
between the phases is promoted, phase separation is much weaker, and entirely suppressed
in some cases. Thus, slight grafting could improve interfacial bonding and interpenetration
between networks and thereby their physical and mechanical properties.
Sung [77, 78] and Han [79-81] have synthesised a series of graft-IPN with epoxy resins
and polyurethane derivatives. These phase separated systems have better compatibility,
toughness, high impact strength and tensile strength than the original polymers.
Polyether and polyester type PU graft agents were used for synthesis of diglycidyl ether
of bisphenol A/urethane modified bismaleimide (DGEBA/UBMI) [81], see Table 8.7.
IPN systems with polyether type PU graft agent presents a heterogeneous phase with
UBMI particles dispersed in the epoxy matrix, whereas those with polyester type PU
graft agent may be homogeneous. Diglycidyl ethers of bisphenol A/polydimethylsiloxane/
296
grafted epoxy/urethane
polypropylene glycol (DGEBA/PDMS/PPG) grafted IPN have broad and high Tg regions,
so could be applied as effective absorbing materials. The degree of compatibility is strongly
affected by the polysiloxane/polyurethane ratio. Grafted chain length could act as a
regulator of phase separation and mechanical properties of graft IPN. Longer graft chains
could serve as a plasticiser that increase the flexibility of the polymer network but also
diminish the crosslinking density. The interpenetration between phases for epoxy/urethane
acrylate resin (UAR) IPN was appreciably improved due to the excellent miscibility
between the poly(oxypropylene) (PPO) grafts and PPO segments existing in the graft
epoxy and the UAR network, respectively.
Using grafting techniques, luminophoric groups can be covalently linked to one IPN
component, allowing dynamic and static photophysical and spectroscopic measurements
of the degrees of domain interpenetration and permeability [76].
When one of the IPN components is a hydrophilic polymer the resulting system may
form a hydrogel when soaked in water. The most suitable systems for obtaining latex
and hydrogel IPN with advantageous properties was until now those containing at least
one acrylate-type component (Table 8.8).
297
Hydrogel formation full
Table 8.8 Examples of hydrogel formation for the case of full-IPN systems
Sim-IPN Network I Network II Refs.
type
Monomer Crosslinker Initiator/ Monomer Crosslinker Initiator/-
catalyst catalyst
2A 2C t
P t = ln 1 (8.16)
V C0
Pl
Dm = (8.17)
Kd
298
Advanced Polymers: Interpenetrating Networks
Cm
Kd = (8.18)
Cs
Furthermore, it was shown that the semi-IPN hydrogels have improved mechanical
properties compared to the homopolymer hydrogels, but without affecting the gel
collapsing temperature [84]. Muniz and Geuskens [38, 39] have synthesised PAAm/
PNIPAM semi-IPN hydrogels with good qualitative mechanical properties, even in the
swollen state, and only low shrinking behaviour at temperatures above the lower critical
solution temperature (32 C) of the linear PNIPAM. This low shrinking behaviour is
determined by the physical restrictions of the PAAm network, which prevent the PNIPAM
chains from collapsing. The equilibrium swelling ratios markedly decrease by increasing
the amount of the less hydrophilic PNIPAM chains (by 18% for 5 wt% PNIPAM in the
PAAm gel), as result of higher polymer volume fraction. This could be due to the
entanglements of the PNIPAM chains that may act inside the semi-IPN hydrogels.
The elastic modulus, E, and the apparent crosslinking density, e, for IPN hydrogels
could be determined with the equations:
=
f
S0
(
= E 2 ) (8.19)
(
= RT p ,0 / p )
23
(
p e 2 ) (8.20)
299
Handbook of Polymer Blends and Composites
The IPN hydrogels of PNIPAM with hydrophobic polymers can be used for surface
modification of rigid hydrophobic polymers through coating [85]. Temperature/pH
sensitive comb-type graft and core-shell type IPN hydrogels composed from PNIPAM
and calcium-alginate were also prepared [40, 86].
Lee and Kim [49, 50, 87-89] synthesised electrical responsive PVA/PAA IPN hydrogels
that exhibit different swelling patterns and permeabilities to solutes as function of pH,
temperature, ionic strength of the external solution and ionic groups content.
The release of drugs incorporated into these hydrogels was strongly influenced by the
degree of swelling and show pulsatile patterns as a response to pH and temperature. When
a PVA/PAA IPN is swelled in a sodium chloride electrolyte solution and placed between a
pair of electrodes, it exhibits bending behaviour upon applied electric field. The electric
responses depend on applied voltage, charge density of ionic groups within the IPN, ionic
strength of the electrolyte solution and greatly affect the drug release behaviour. The amount
of loaded drug, increases with both IPN and drug ionisable groups.
To study the release of drugs as function of the electrokinetic processes, Lee and Kim
[49, 50, 87-89] have determined the equilibrium conditions for PVA/PAA hydrogels
through the osmotic pressure , which is the sum of the osmotic pressure due to the
rubber elasticity, 1, the solubility of the solvent in the polymer chain, 2, and the difference
in ionic concentration between the gel and surrounding medium, 3:
1 3 v
=
[ ]
ln(1 v) + v + v2 RT
+
v RTv e
2
+ ( C C )RT
i j (8.21)
V1 V0
300
Advanced Polymers: Interpenetrating Networks
At equilibrium, the osmotic pressure of the gel is equal to that of the surrounding aqueous
solution:
= 0 = 1+ 2 + 3
The swelling behaviour of the hydrogel IPN could be influenced by the IPN composition,
the degree of crosslinking and the swelling medium characteristics. For poly(acrylamide-
co-acrylic acid)/poly(vinyl alcohol) hydrogel IPN, incorporation of acrylic acid raises the
swelling ratio several times while crosslinking of PVA suppresses it, therefore they could
be used as modulating factors. Furthermore, the IPN being ionic, swelling is affected by
the pH and ionic strength of the medium [91].
Various chitosan-based hydrogel type IPN were also synthesised [41, 45, 92]. These IPN
generally exhibit good mechanical properties and an equilibrium water content increasing
with the degree of crosslinking of the networks from under 60% to over 95%. For a
PEG/chitosan hydrogel full-IPN two Tg values were observed, with a value between those
of the individual components, indicating the presence of phase separation. The tensile
strength and elongation at break in the swollen state was in the range 0.06-0.18 MPa
and 18%-48%, respectively.
301
Handbook of Polymer Blends and Composites
Silica and silicon compounds are the main inorganic components used for making hybrid
IPN. Silicon-containing organic/inorganic architectures manifest the desirable properties
of both an organic polymer (light-weight, flexible, mouldable) and an inorganic network
(high thermal stability and strength), but also offer the promise of producing low-cost
high performance materials.
IPN hybrids of poly(N,N-dimethylacrylamide) gel and polystyrene gel with silica gel
were prepared by polymerisation of N,N-dimethylacrylamide and N,N-
methylenebisacrylamide, respectively, by the polymerisation of styrene monomer in the
presence of divinylbenzene in methanolic solutions of tetramethoxysilane (TMOS) [96].
The loss of organic elements could be prevented in both cases by the incorporation of
crosslinking points in the organic network. The resulting homogeneous glassy materials
present high surface areas, large pore volumes and a sharp distribution of pore size
below 2 nm, indicating a molecular-level integration of the organic gel and silica gel in
the IPN polymer hybrids.
The most homogeneous hybrid morphology with the smallest silica domains of size
10-20 nm appears in the sequential IPN, where the development of the silica structures
is restricted by the rigid reaction medium of the preformed epoxide network.
302
Advanced Polymers: Interpenetrating Networks
Reversible, transparent and homogenous hybrid IPN of organic gel and silica gel were
synthesised by Imai and co-workers [99, 100] from TMOS and poly(2-methyl-2-oxazoline)
having coumarine or thymine moieties as reversibly photocrosslinkable side groups.
Furthermore, when maleimide and furan groups were introduced in the side chain of poly(2-
methyl-2-oxazoline), the resulting IPN was proved to be thermally reversible [101].
IPN structures can evolve through chelate complexation of linear polymer chains. Such
materials were obtained by the treatment of alginate (Alg)/PVA films cast from mixed
aqueous polymer solutions with calcium tetraborate solutions [102]. In this case IPN
formation takes place by simultaneous occurrence of a chelate complexation of alginates
with Ca2+ cations and a borate ion-aided crosslinking between PVA chains [102].
Both condensation and hydrosilylation reactions have been successfully applied to the
synthesis of silicon-based IPN consisting of a stable Si-O and/or a Si-C linkage, i.e., a
ladder silsesquioxane oligomer and a polycarbosilane were used as starting materials for
IPN formation through hydrosilylation polymerisation of bifunctional Si-H and Si-vinyl
monomers [105].
In latex IPN both networks are included in a single latex particle, usually by emulsion
polymerisation. In the sequential method, if the monomers corresponding to the second
polymer react near the surface of the first polymer, a latex IPN with shell/core morphology
will be obtained. The core could be a hard polymer and the shell a soft polymer, or vice
versa. If the second monomer is not compatible with the first polymer a compatibiliser has
to be added. In some cases both polymer networks can be preformed and subsequently
mixed and crosslinked. A more homogenous morphology can be obtained if the second
monomer can diffuse fast in the first polymer latex, and by simultaneous network formation.
However, the domain size and hence the properties of latex IPN, such those of PU ionomers
with acrylate and styrene polymers, can be easily controlled by managing the molecular
weight between crosslinks [51]. Some PS-based IPN systems are presented in Table 8.9.
303
Handbook of Polymer Blends and Composites
The IPN core/shell polymers were found to be very good dampers due to their higher
miscibility. The highest level of damping may be achieved by inverting core/shell particles
with dual-phase continuity compared to normal core/shell particles, as was related for
acrylate/SAN latex IPN with cores from PBA or poly(ethylhexyl methacrylate) (PEHMA)-
based copolymers and glassy SAN shells [106]. The better miscibility of PBA-based
copolymers was associated with their higher grafting efficiency.
Introduction of inert fillers into the reaction system at the stage of crosslinked polymer
formation essentially affects the reaction kinetics, inhibits the process of microphase
separation, and modifies the viscoelastic properties of the system.
The tensile strength, flexural strength, hardness, tensile modulus, dynamic storage modulus
and flexural modulus of PU/PVE filled IPN increase with kaolin content, reaching a
maximum for 20%-25% filler [110]. An increase of the filler content lowered the impact
strength and shifted the Tg to a higher temperature.
304
Advanced Polymers: Interpenetrating Networks
Based on studies on PU/PEA sim-IPN, Sergeeva and co-workers [20] concluded that
introducing fillers into the IPN at their formation stage influences both the chemical
crosslinking and the microphase separation. Thus, fillers will change the chemical reaction
rates, the structure of networks being formed and the microphase separation rate, giving
non-equilibrium structures with various segregation degrees by adsorption on one or
several components. If there is an affinity between the filler and the IPN components,
more thermodynamically stable could be realised [19].
Table 8.10 Filled IPN. Comparative examples for the case of PU systems
Filled IPN IPN type IPN procedure Fillers Refs.
PU/PEA Full Simultaneous Aerosil-300, Al2O3 and PEA particles [19, 20]
3%, Carbon fibres 2%-10%
PU/PBMA Semi Simultaneous Talc 20%-40%, TEGMA 10%-40% [25]
8.5 Applications
IPN have improved properties compared to their constituent polymers. The wide range
of these kinds of materials, the possibility of controlled synthesis and design, the long
time resistance and stability, make them the most suitable solutions for various
applications.
Classical ways for IPN use include sheet moulding (SMC) and reinforced injection
moulding (RIM) compounds, and coatings and adhesives, damping, sound insulation
and acoustic absorbing materials, ion exchange resins, permselective/permeable
membranes, dental fillings, toughening of rubbery and plastic materials, and as impact
modifiers for thermoset materials.
305
Handbook of Polymer Blends and Composites
However, the emerging need for performance materials in all technological fields push
the polymers and their derived materials far away from their traditional uses. In this
way, IPN architectures could be found in applications like nonlinear optical materials,
electro-conductive and medical devices, intelligent polymers with improved
biocompatibility for biomedical and pharmaceutical applications like molecular
separation, enzyme activity controlling systems, controlled drug delivery, tissue culture
substrates. Representative examples are given in Table 8.11.
Major advances were made in recent years in applying IPN type materials to biological
applications, opening a wide and promising way for the future.
PAA-co-PEG/polyamide (PA) IPN thin films covalently grafted to metal oxides, silicon,
quartz or glass prevents non-specific protein adsorption and specifically promotes cell
attachment [117]. Furthermore modifications of these IPN by grafting peptides that mimic
the cell binding, facilitate osteoblast adhesion, proliferation and formation of a mineralised
tissue. Peptide-IPN could be also used in biomimetic surface engineering as a coating to
improve the initial wound healing and stability of metallic and polymeric implants [118].
IPN materials having both high elasticity and high tensile strength, especially those
containing PU or polysiloxane, are widely used in the manufacture of medical devices
such as catheters, tubing, films, balloons and the like.
The most wanted IPN for biological applications are currently the hydrogel type semi-IPN,
because of their specific properties, i.e., changing behaviours as response to variations in pH
and temperature or to electric stimuli, swelling and shrinking capabilities, controlled
permeability and better mechanical properties than each individual component hydrogel.
Major applications are in the field of size exclusion and biomimetic membranes, encapsulation
of bioactive species like drugs and enzymes, biosensors and controlled drug delivery.
Thermoplastic IPN composed from castor oil-based polyurethane and methyl acrylate
(MA) were coated onto the surface of regenerated cellulose film to obtain biodegradable
water-resistant films with properties dependent on the MA content [113].
306
applications
307
Handbook of Polymer Blends and Composites
Surprising, only a few IPN commercial products are explicitly named as IPN, therefore
one can find true IPN-type materials under the guise of alloys, composites and even
blends. One important consequence of this fact is a greater difficulty to try to make a
commercial analysis of the market. From this point of view, IPN remain one of the few
major groups of chemical products with significant discrepancies between industrial,
commercial and scientific nomenclatures.
However, the large number of patents focused on them, shows the great global impact of
IPN products associated with their broad, high technology applications. Some
representative examples of patents in the field of IPN materials are reviewed in the
Tables 8.12 8.14.
308
Full IPN applications
309
Full IPN applications
310
Full IPN applications
Advanced Polymers: Interpenetrating Networks
311
Full IPN applications
312
Semi-IPN applications
313
Semi-IPN applications
314
IPN applications
315
IPN applications
316
Advanced Polymers: Interpenetrating Networks
8.6 Conclusions
Interpenetrated polymer networks are advanced materials with complex, rigid architecture,
resulting from the irreversible interlocking of the growing chains of each individual
polymer network, without any chemical bonding between them.
Reciprocally permanent mutual entanglements prevent total phase separation and stabilise
the morphology holding together these networks. So, IPN control the phase separation
in a multi-component polymer mixture by the formation of individual polymer networks,
allowing multi-phase materials to be made, which have useful mechanical, permeation,
optical, and biological properties. However, the suppression of phase-separation is not
absolutely complete in practice.
This particular use of networks in IPN synthesis allows the making of materials with
bicontinuous structure by forced compatibilisation of incompatible polymers and from
polymers insoluble in solvents, giving a clear distinction between IPN and blends.
Obviously, IPN materials are also different from alloys and composites that involve only
mixing of at least partially miscible polymers by physical means.
The IPN morphology and properties depend and could be modulated by a large group of
factors, such as: method of synthesis, polymer type, amount and ratio, kinetics of
polymerisation and network formation, crosslinker, catalyst and initiator type and amount,
additives, cure and post-cure conditions, compatibility of the monomers and polymer
systems used, crosslinking density in each of polymer networks and crosslinking degree
reached in IPN and inter-network grafting extent.
IPN analysis can be done with various techniques, such as: DSC, DMA, DMTA, M-
TDSC, FT-IR, NMR, DPC, SEM, TEM, SANS, SAXS and AFM.
Two major methods are used for making IPN: simultaneous and sequential. If only one
polymer is crosslinked the resulting material is a semi-IPN, while a full-IPN is made
when all species are in network form.
Function of criteria like phase segregation extent, type of polymer network, number and
types of crosslinks that occur, and the final product properties, both semi- and full-IPN
major types could be further split into thermoplastic, hydrogel, latex, grafted, hybrid,
filled, conductive, elastomeric, gradient and homo-IPN subtypes.
All polymers have advantageous and disadvantageous properties. For example, acrylics
have good UV resistance and colour clarity, but they make poor bonds to wet surfaces,
are brittle, flammable and shrink. Epoxy resins have good chemical resistance and good
adhesion on both dry and wet surfaces, but are brittle and poor resistant against UV.
317
Handbook of Polymer Blends and Composites
Polyurethanes are flexible and tough, but allow moisture penetration, can blister and
bubble. Silicones have good flexibility and heat resistance, but they make poor bonds
with surfaces and other polymers.
By making IPN systems from such starting polymers, one has the opportunity of obtaining
materials with a range of properties that overcome the disabilities of individual polymer
components and perhaps generating a synergistic effect on one or more of the properties.
Along with the improvements of some components properties, minimising of unwanted
ones and synergistic findings, the IPN-type structure itself gives to this kind of material
the gift of long life, stability and resistance to physical and chemical agents.
References
3. IPNs Around the World, Science and Engineering, Eds., S.C. Kim and L.H.
Sperling, Wiley, New York, NY, USA, 1997.
7. R. Bischoff and S.E. Cray, Progress in Polymer Science, 1999, 24, 185.
8. X. Ramis, A. Cadenato, J.M. Morancho and J.M. Salla, Polymer, 2001, 42, 9469.
9. S.C. Kim, D. Klempner, K.C. Frisch and H.L. Frisch, Macromolecules, 1977, 10,
1187.
10. D.J. Hourston and F.U. Schfer, Polymer, 1996, 37, 3521.
11. L.E. Nielsen and R.F. Landel, Mechanical Properties of Polymers and
Composites, 2nd Edition, Marcel Dekker, Inc, New York, NY, USA, 1994.
318
Advanced Polymers: Interpenetrating Networks
12. D.J. Hourston and F.U. Schfer, Journal of Applied Polymer Science, 1996, 62, 2025.
13. R. Hu, V.L. Dimonie, M.S. El-Aasser, R.A. Pearson, A. Hiltner, S.G. Mylonakis
and L.H. Sperling, Journal of Polymer Science, Part B, Polymer Physics, 1997,
35, 1501.
14. Y.C. Chern, S.M. Tseng and K.H. Hsieh, Journal of Applied Polymer Science,
1999, 74, 335.
15. M.J. Brekner, H.A. Schneider and H.J. Cantow, Polymer, 1988, 29, 2085.
18. C. Clark-Monks and B. Ellis, Journal of Polymer Science, Polymer Physics, 1973,
11, 2089.
19. Y.S. Lipatov, L.V. Karabanova and L.M. Sergeeva, Polymer International 1994,
34, 7.
20. L.M. Sergeeva, S.I. Skiba and L.V. Karabanova, Polymer International, 1996,
39, 317.
21. P. Hedvig, Dielectric Spectroscopy of Polymers, Adam Hilger, Bristol, UK, 1977.
23. J.H. Ambrus, C.T. Moynihan and P.B. Macedo, Journal of Physical Chemistry,
1992, 176, 3287.
24. S. Havriliak and S.J. Negami, Journal of Polymer Science, Polymer Symposium,
1966, 14, 89.
25. Y.S. Lipatov, T.T. Alekseeva, V.F. Rosovitsky and N.V. Babkina, Polymer
International, 1995, 37, 97.
26. P.J. Flory, Principles of Polymer Chemistry, Cornell University Press, Ithaca, NY,
USA, 1953.
27. H.L. Frisch, K.C. Frisch and D. Klempner, Polymer Engineering and Science,
1974, 14, 648.
319
Handbook of Polymer Blends and Composites
28. C.U. Pittman Jr., X. Xu, L. Wang and H. Toghiani, Polymer, 2000, 41, 5405.
29. F.M. Mulder, W. Heinen, M. van Duin, J. Lugtenburg and H.J.M. de Groot,
Macromolecules, 2000, 33, 5544.
33. Y.C. Chou and L.J. Lee, Polymer Engineering and Science, 1994, 34, 1239.
34. V. Mishra, F.E. Du Prez, E. Gosen, E.J. Goethals and L.H. Sperling, Journal of
Applied Polymer Science, 1995, 58, 331.
35. L.H. Sperling and V. Mishra, Macromolecular Symposia, 1997, 118, 363.
37. Y.S. Lipatov and G.M. Semenovich, Polymer, 1999, 40, 6485.
38. E.C. Muniz and G. Geuskens, Journal of Membrane Science, 2000, 172, 287.
40. T.G. Park and H.K. Choi, Macromolecular Rapid Communication, 1998, 19, 167.
41. S.J. Lee, S.S. Kim and Y.M. Lee, Carbohydrate Polymers, 2000, 41, 197.
43. L.Y. Chiang, L.Y. Wang, C.S. Kuo, J.G. Lin and C.Y. Huang, Synthetic Metals,
1997, 84, 721.
45. J.W. Lee, S.Y. Kim, S.S. Kim, Y.M. Lee, K.H. Lee and S.J. Kim, Journal of
Applied Polymer Science, 1999, 73, 113.
320
Advanced Polymers: Interpenetrating Networks
47. H.S. Min and S.C. Kim, Polymer Bulletin, 1999, 42, 221.
49. Y.M. Lee, S.H. Kim and C.S. Cho, Journal of Applied Polymer Science, 1996,
62, 301.
50. S.Y. Kim, H.S. Shin, Y.M. Lee and C.N. Jeong, Journal of Applied Polymer
Science, 1999, 73, 1675.
52. J.M.M. Duenas, D.T. Escuriola, G. Gallego Ferrer, M.M. Pradas, J.L.G. Ribelles,
P. Pissis and A. Kyritsis, Macromolecules, 2001, 34, 5525.
54. S.S. Hou and P.L. Kuo, Polymer, 2001, 42, 9505.
57. W.M. de Azevedo, J.M. de Souza and J.V. de Melo, Synthetic Metals, 1999,
100, 241.
59. D.G. Fradkin, J.N. Foster, L.H. Sperling and D.A. Thomas, Polymer Engineering
and Science, 1996, 26, 730.
60. R. Mao, J. Tang and B.G. Swanson, Carbohydrate Polymers, 2000, 41, 331.
61. M. Song, D.J. Hourston, M. Reading, H.M. Pollock and A. Hammiche, Journal
of Thermal Analysis and Calorimetry, 1999, 56, 991.
321
Handbook of Polymer Blends and Composites
65. I. Alig, W. Jenninger, and J.E.K. Schawe, Thermochimica Acta, 1999, 330, 167.
66. V. Mishra and L.H. Sperling, Journal of Polymer Science, 1996. 34, 883.
67. X.Q. Yu, G. Gao, J.Y. Wang, F. Li and X.Y. Tang, Polymer International, 1999,
48, 805.
68. Y. Rharbi, A. Yekta, M.A. Winnik, R.J. DeVoe and D. Barrera, Macromolecules,
1999, 32, 3241.
69. S.H. Wang, S. Zawadzki, and L. Akcelrud, Materials Research, 2001, 4, 27.
71. L.H. Fan, C.P. Hu and S.K. Ying, Polymer Engineering and Science, 1997, 37, 2.
72. G.Y. Wang, Y.L. Wang and C.P. Hu, European Polymer Journal, 2000, 36, 735.
73. T.T. Hsieh, K.H. Hsieh, G.P. Simon and C. Tiu, Polymer, 1999, 40, 3153.
75. S. Dadbin, R. P. Burford and R. P. Chapin, Polymer Gels and Networks, 1995, 3,
179.
77. P.H. Sung and W.G. Wu, European Polymer Journal, 1994, 30, 905.
78. P.H. Sung and C.Y. Lin, European Polymer Journal, 1997, 33, 231.
79. J.L. Han and Y.C. Chem, K,Y. Li and K.H. Hsieh, Journal of Applied Polymer
Science, 1998, 70, 529.
80. J.L. Han and K.Y. Li, Polymer Journal, 1999, 31, 401.
81. K.H. Hsieh, J.L. Han, C.T. Yu and S.C. Fu, Polymer, 2001, 42, 2491.
322
Advanced Polymers: Interpenetrating Networks
82. F.J. Hua and C.P. Hu, European Polymer Journal, 1999, 35, 103.
83. F.J. Hua and C.P. Hu, European Polymer Journal, 2000, 36, 27.
84. B.C. Shin, M.S. Jhon, H.B. Lee, and S.K. Yuk, European Polymer Journal, 1998,
34, 1675.
86. H.K. Ju, S.Y. Kim and Y.M. Lee, Polymer, 2001, 42, 6851.
87. H.S. Shin, S.Y. Kim and Y.M. Lee, Journal of Applied Polymer Science, 1997, 65,
685.
88. H.S. Shin, S.Y. Kim, Y.M. Lee, K.H. Lee, S.J. Kim and C. Rogers, Journal of
Applied Polymer Science, 1998, 69, 479.
89. S.Y. Kim and Y.M. Lee, Journal of Applied Polymer Science, 1999, 74, 1752.
90. G. Gallego Ferrer, M.M. Pradas, J.L.G. Ribelles and P. Pissis, Proceedings of the
Workshop Dielectrica 98, Monte de Caparica, Portugal, 1998.
91. D. Dhara, C.K. Nisha and P.R. Chatterji, Journal of Macromolecular Science -
Pure and Applied Chemistry, 1999, A36, 197.
92. S.M. Cho, S.Y. Kim, Y.M. Lee, Y.K. Sung and C.S. Cho, Journal of Applied
Polymer Science, 1999, 73, 2151.
94. F.Yang, Y. Ou and Z. Yu, Journal of Applied Polymer Science, 1998, 69, 355.
95. W. Yin, J. Li, J. Wu and T. Gu, Journal of Applied Polymer Science, 1997, 64,
903.
96. R. Tamaki, T. Horiguchi and Y. Chujo, Bulletin of the Chemical Society of Japan,
1998, 71, 2749.
97. Y. Imai, N. Yoshida, K. Naka and Y. Chujo, Polymer Journal, 1999, 31, 258.
98. L. Matejka, K. Dusek, J. Plestil, J. Kriz and F. Lednicky, Polymer, 1999, 40, 171.
99. Y. Imai, K. Naka and Y. Chujo, Polymer Journal, 1998, 30, 990.
323
Handbook of Polymer Blends and Composites
100. Y. Imai, T. Ogoshi, K. Naka and Y. Chujo, Polymer Bulletin, 2000, 45, 9.
101. Y. Imai, H. Itoh, K. Naka and Y. Chujo, Macromolecules, 2000, 33, 4343.
103. L.H. Gan, Y.Y. Gan and W.S. Yin, Polymer International, 1999, 48, 1160.
104. L.H. Gan, Y.Y. Gan and W.S. Yin, Polymer, 1999, 40, 4035.
105. M. Tsumura and T. Iwahara, Journal of Applied Polymer Science, 2000, 78, 724.
106. M.S. El-Aasser, R. Hu, V.L. Dimonie and L.H. Sperling, Colloids and Surfaces A:
Physicochemical and Engineering Aspects, 1999, 153, 241.
107. C.D. Vo and R.P. Burford, Journal of Applied Polymer Science, 1999, 74, 622.
108. G.H. Ma, M. Nagai and S. Omi, Journal of Colloid and Interface Science, 1999,
214, 264.
110. C.H. Chen, W.J. Chen, M.H. Chen and Y.M. Li, Journal of Applied Polymer
Science, 1999, 71, 1977.
111. L.Z. Zhang, Z.G. Cai, Q.D. Ying and Z.X. Liang, Chinese Journal of Polymer
Science, 1999, 17, 435.
112. S.Y. Kim and Y.M. Lee, Journal of Applied Polymer Science, 1999, 74, 1752.
113. J.H. Yu, Y.M. Du and F.Y. Cheng, Journal of Macromolecular Science - Pure and
Applied Chemistry, 1999, 36, 1259.
114. L.H. Fan, C.P. Hu and S.K. Ying, Polymer Engineering and Science, 1997, 37, 338.
115. S.Y. Kim, J.C. Ha and Y.M. Lee, Journal of Controlled Release, 2000, 65, 345.
116. K.C. Gupta and M.N.V.R. Kumar, Journal of Applied Polymer Science, 2000, 76,
672.
117. J.P. Bearinger, D.G. Castner and K.E. Healy, Journal of Biomaterials Science,
Polymer Edition, 1998, 9, 629.
324
Advanced Polymers: Interpenetrating Networks
118. K.E. Healy, Current Opinion in Solid State & Materials Science, 1999, 4, 381.
120. J.A.J. Schutyser and A.J.W. Buser, inventors; Akzo Nobel NV, assignee; World
Patent 9,506,075, 1995.
121. J.A.J. Schutyser, A.J.W. Buser, and A. Steenbergen, inventors; Akzo Nobel NV,
assignee; US Patent 5,728,468, 1998.
122. L. Tateosian and F.D. Roemer, inventors; Dentsply Research & Devevelopment,
assignee; US Patent 4,698,373, 1987.
124. Kane J.F. and N.R. Mowrer, inventors; Ameron International Corporation,
assignee; US Patent 5,736,619, 1998.
125. S-D. Tong, L-C. Hsu and C.B. Hu, inventors; Edwards Lifesciences Corporation,
assignee; World Patent 0,108,718, 2001.
126. S-C. Kim, H-W. Roh, M-J. Song and Y-C. Shin, inventors; Korea Institute Science
Technology, assignee; US Patent 6,156,344, 2000.
127. D.R. Ingenito, H.F. Rugge, D.S. Soane and W.L. Sturm, inventors; Rasor
Associates, Inc, assignee; US Patent 5,986,001 1999.
128. C-T. Lee, M-Y. Lee and C-J. Ryu, inventors; Korea Chemical Company Ltd,
assignee; Korean Patent 9,104,647, 1991.
130. N.E. Blank, R.C. Hartwig and C. Vu, inventors; W.R. Grace and Company,
assignee; US Patent 5,091,455, 1992.
131. L.H. Sperling, C.J. Murphy and V. Mishra, inventors; Schneider USA, Inc.,
assignee; US Patent 5,783,633, 1998.
132. K.C. Frisch, P. Geng and H.X. Xiao, inventors; Hi Tex Inc, assignee; US Patent
5,747,392, 1998.
325
Handbook of Polymer Blends and Composites
133. S.D. Rizk, N.B. Shah and J.W. Powers, inventors; Essex Specialty Production,
assignee; US Patent 4,842,938, 1989.
134. S.D. Rizk and N.B. Shah, inventors; Essex Specialty Production, assignee; US
Patent 4,766,183, 1988.
135. J.J. Zupancic, R.J. Swedo and S. Petty-Weeks, inventors; UOP, Inc., assignee; US
Patent 4,708,981, 1987.
136. T. Anazawa and T. Takada, inventors; Dainippon Ink & Chemical, Inc. and
Kawamura Institute of Chemistry Research, assignees; Japanese Patent
11,209,648, 1999.
137. J.A.J. Schutyser, H.J. Slots, A.J.W. Buser and P.H. Zuuring, inventors; Akzo NV,
assignee; European Patent 0,417,837, 1991.
138. J.A.J. Schutyser, H.J. Slots, A.J.W. Buser and P.H. Zuuring, inventors; Akzo NV,
assignee; European Patent 0,413,386, 1991.
140. U.A. Sorathia, W.L. Yeager and T.L. Dapp, inventors; US Navy, assignee; US
Patent 5,237,018, 1993.
142. S.V. Davis, J. E. Mathers, B.H. Mills, B. Tan and J-H. Chen, inventors; Eastman
Kodak Company, assignee; US Patent 6,067,438, 2000.
145. Y. Duan, Y. Wei, Y. Zhu, S.E. Stammler, B.L. Marty, G. Haider, R.R. Davies and
M. Maksymkiw, inventors; H.B. Fuller Licensing & Financing, Inc., assignee; US
Patent 6,017,998, 2000.
146. D.R. Ingenito, H.F. Rugge, D.S. Soane and W.L. Sturm, inventors; Rasor
Associates, Inc., assignee; US Patent 5,789,483, 1998.
147. H.R. Allcock, K. Visscher and Y.B. Kim, inventors; The Penn State Research
Foundation, assignee; US Patent 5,747,60, 1998.
326
Advanced Polymers: Interpenetrating Networks
148. G.A. Hill, K.C. Frisch, V. Sendijarevic and S-W. Wong, inventors; Johnson &
Johnson Vision Products, Inc., assignee; US Patent 5,674,942, 1997.
149. G.A. Hill, K.C. Frisch, V. Sendijarevic and S-W. Wong, inventors; Johnson &
Johnson Vision Products, Inc., assignee; US Patent 5,656,210, 1997.
150. S.K.Tripathy, R-J. Jeng, J. Kumar, S. Marturunkakul and J.I. Chen, inventors;
University of Massachusetts Lowell, assignee; US Patent 5,532,320, 1996.
152. T. Sato, inventor; Nisshin Spinning, assignee; European Patent 1,057,869, 2000.
153. J.A. Dillon and M.E. Dillon, inventors; Tetratec Corporation, assignee; US Patent
4,945,125, 1990.
155. Y-Y. Hsu, D.L. Wise, J.D. Gresser and D.J. Trantolo, inventors; Cambridge
Scientific, Inc., assignee; US Patent 6,071,982, 2000.
156. R.E. Harelstad, L.A. Pavelka, D.A., Barrera and W.D Joseph, inventors; 3M
Innovative Properties Company, assignee; US Patent 6,001,936, 1999.
157. M.E. Dillon, inventor; Bio Medical Sciences, Inc., assignee; US Patent 5,980,923,
1999.
158. M.A. Perez, W.D. Coggio, M.C. Palazzotto and D.S. Parker, inventors; Minnesota
Mining & Manufacturing, assignee; World Patent 9,808,906, 1998.
160. B.C. Arkles and R.A. Smith, inventors; no assignee; US Patent 4,970,263, 1990.
161. E.S. Hsiue, D. Ziatyk, G.R Stone and B.T. Debona, inventors; Allied
Corporation, assignee; US Patent 4,623,577, 1986.
162. L.H. Sperling, C.J. Murphy and V. Mishra, inventors; Schneider USA, Inc.,
assignee; US Patent 5,786,426, 1998.
163. R.S. Langer, J.H. Elisseeff, K. Anseth and D. Sims, inventors; Massachusetts
Institute of Technology, University Technology Corporation and The General
Hospital Hospital Corporation, assignees; US Patent 6,224,893, 2001.
327
Handbook of Polymer Blends and Composites
164. Y. Wang, R. van Boxtel and S.Q. Zhou, inventors; Pharmacia & Upjohn AB,
assignee; US Patent 6,011,082, 2000.
165. V.R. Shastri, R.S. Langer and P.J. Tarcha, inventors; Massachusetts Institute of
Technology, assignee; US Patent 5,837,752, 1998.
166. M.A. Perez, D.A. Ylitalo, T.M. Clausen, R J. DeVoe, K.E. Kinzer and M.D.
Swan, inventors; Minnesota Mining & Manufacturing Company, assignee; US
Patent 5,709,948, 1998.
167. Y.H. Bae, T. Okano and S.W. Kim, inventors; University of Utah, assignee; US
Patent 4,931,287, 1990.
168. N. Shirai, T. Takada and T. Anazawa, inventors; Dainippon Ink & Chemical, Inc.
and Kawamura Institute of Chemical Research, assignees; Japanese Patent
11,293,147, 1999.
170. T. Nguyen, H. Hu, E. Rossberg, H. Tran and C.R. Briggs, inventors; Sunsoft
Corporation, assignee; World Patent 0,002,937, 2000.
171. M.E. Dillon, inventor; Biomed Sciences, Inc., assignee; World Patent 0,149,228,
2001.
173. No inventor; The Dow Chemical Company, assignee; Spanish Patent 8,701,208,
1987.
174. A.O. Egli and T.L. St Clair, inventors; NASA, assignee; US Patent 4,695,610,
1987.
175. J.M. Crosby and M.K. Hutchins, inventors; LNP Corporation, assignee; US
Patent 4,695,602, 1987.
177. W.H. Mandeville, III, S.R. Holmes-Farley, T.X. Neenan and G.M. Whitesides,
inventors; GelTex Pharmaceuticals, Inc. assignee; US Patent 5,925,379, 1999.
328
Advanced Polymers: Interpenetrating Networks
178. R.H. Pater and M.G. Hansen, inventors; The USA as represented by the
Administrator of the NASA, assignee; US Patent 5,648,432, 1997.
181. R.H. Pater, inventor; The USA as represented by the Administrator of the NASA,
assignee; US Patent 5,492,979, 1996.
329
Heterofibres
9
Heterofibres
The technology for the manufacture of man-made fibres has developed very expensive
and new generic types, but increasingly sophisticated techniques can combine different
polymers or variants of the same polymer into a single fibre to take advantage of the
special characteristics of each polymer.
HF, MUCF or MUF are physical combinations or mixtures of two or more chemically/
physically distinct polymers combined by coextrusion (cofilate, conjugate or twin fibres)
or before extrusion (polyblends or composite fibres). The components are not always fibre-
forming, they can be thermoplastic alone. Bicomponent fibres (BF) are composed of two
or more regions {side-by-side (S/S), core-sheath (C/S) and matrix-fibril (M/F) configurations},
placed into any number of possible cross-sectional shapes that are commonly used in
monocomponent fibres (round, trilobal, quadrilobal, cross-sections) [1-3].
Definitions of the American Society for Testing Materials (ASTM) are as follows:
BF can be made from two variants of the same generic fibre [two types of: polyester
(PES), polyethyleneterephthalate (PET), polyamide (PA), copolyester (coPES),
polyacrylonitrile (PAN), polyethylene (PE), polypropylene (PP)] or from two generically
different fibres, e.g., PET/PA, PA/Spandex, PE/PP; PP/PES, etc.
331
Handbook of Polymer Blends and Composites
Biconstituent fibres are defined as fibres consisting of a continuous matrix of one polymer
in which a different fibre-forming polymer is dispersed as a second, distinct, discontinuous
phase. This type of fibre is also referred to in the USA as a matrix fibre in which the
fibre-forming components are dispersed one within the other before they are extruded.
The use of the term biconstituent fibre for this type of fibre is generally avoided, remaining
in use only for the general sphere of HF or MUCF fibres.
Under the Textile Fibre Products Information Act of USA, these products made from HF
that are composed of different generic fibres must have listed on their labels the generic
fibres presented after predominance by weight and must state the mass percentage of such
components. For M/F configuration the term matrix-fibril or matrix fibre may be used.
BF made from the same generic fibre type with different characteristics (shrinkage, viscosity
levels, melting points) are not covered by this regulation, and the products made from
them would be labelled as 100 percent of the generic fibre (PA, PAN, PES, PP, etc).
After the Act and Regulations (1997) of USA the legal terminology designated for the
fibre manufactured from two or more distinct polymer is biconstituent or MUCF fibres,
although the most frequently used terms are bicomponent, cofilate or cospun (in Europe
and the USA) and conjugate fibres in Asia.
The classification of hetero- or MUCF fibres is complex and presented in Figure 9.1.
S/S structure, where each polymer is divided along the length into two ore more
distinct regions of the cross-section (Figure 9.2).
C/S, cover-core (C/C), or mantle-core (M/C) structure, where one of the components
(core) is fully surrounded by another component (sheath, mantle or cover) (Figure 9.3).
One type of C/S fibre is self-bonding fibres consisting of a low melting temperature sheath
surrounding a higher melting temperature core, e.g., PE/PP, PE/PET, coPET/PET, PA/PET.
A second type of C/S fibre is filled fibres consisting of a core product from recycled or
conductive material that is covered by a sheath that possesses desired aesthetics or other
properties.
A third type of C/S is additivated fibres consisting of a sheath containing expensive additives
useful on the surface of the filaments for tailoring properties.
M/F BF containing many fine fibrils of one polymer dispersed randomly in size and
location but with axial alignment in a matrix of another component (Figure 9.4).
332
Heterofibres
Figure 9.2 S/S fibre cross-sections. a-h = round areas; i = quadrilobal area; j, k =
trilobal areas; l = cross area; m = banana-shaped area
333
Handbook of Polymer Blends and Composites
Figure 9.3 C/S fibre cross-sections. a-g = round areas; h, i = bicircular areas; j =
eliptico-circular area; k = elliptical area; l = trilobal area
Polyblend or composite fibres derived from polymer blends or alloys contain homogeneous
or heterogeneous mixtures of structurally different homo- or copolymers, to obtain fibres
for specific uses by tailoring one or more properties with minimum sacrifice of another
properties.
Homogeneous polymer blends behave as a single polymer having single glass transition
and melting temperatures.
334
Heterofibres
Heterogeneous polymer blends are separated into different spatial regions. The polymers
are immiscible but they can be compatible or incompatible. For the latter compatibilisation
treatment is necessary to obtain good mechanical properties of the blend.
9.2.1 Background
Fibre processing is the science of producing and controlling the anisotropy of molecular
microstructure and geometry that determines the anisotropy of properties and
performances.
The manufacture of chemical homo- and hetero-fibres consists of four main steps [2-12]:
1. Transition of solid polymers into liquid state (either melts or solutions with isotropic
or liquid crystal properties);
2. Transition of liquid state into solid fibres (spinning, i.e., extrusion-stretching - fibre
consolidation).
For melts, the term is melt spinning, for the concentrated solution, if the solvent has a
low boiling point the term is dry spinning (the solvent flashes off after extrusion with
335
Handbook of Polymer Blends and Composites
air). If the solvent has a high boiling point and it must be removed in an extraction with
a non-solvent the term is wet spinning. For the dilute solution of polymer, e.g., high
molecular weight polyethylene (HMWPE), the process is known as gel spinning,
processing this solution into an unoriented precursor fibre that will be carefully
superdrawn to draw ratios above 50% to perform fibres with the highest specific tensile
properties of any known material.
3. Drawing reduces fibre entropy by stretching the spun fibres to increase their molecular
orientation and to improve the mechanical properties (especially tenacity and
modulus);
4. Heat treating or annealing lowers the fibre internal energy and relaxes residual local
stresses stabilising the fibres microstructure (molecular orientation, crystallinity,
crystalline morphology, interchain forces) and therefore its properties.
Besides these main steps, the fibre processing history may contain additional processes
i.e., cutting for staple fibres, texturing for bulk continuous filament (BCF) yarns or cording
for tyre yarns.
The scheme of the general spinning process is drawn in Figure 9.6 and the steps of fibres
processing combined with the parameters for different spinning methods are indicated in
Table 9.1 [2-6].
The spinning of BF adds complexity and creates unique problems that are not normally
associated with monocomponent fibre spinning.
The extensive literature and patents on the technology of BF have been reviewed in
several articles and books [2-24].
Taking very different polymers and combining them into a single fibre, displaying the
best properties of each material is enticing. There are some limitations however on the
combinations of materials used in a bicomponent system [25-33]. Large differences in
viscosity complicate the production of the desired cross-area, and in extreme cases make
the system unspinnable. If the final goal is not a splittable fibre, the two polymers must
be sufficiently compatible at the interface to have adequate interfacial adhesion. In the
blending of the polymers, to achieve systems with adequate mechanical properties,
complete miscibility is not necessary, but improved characteristics appear, as the two
polymers become more compatible. As internal adhesion and compatibility increase, the
average domain size of the dispersed phase will decrease and the ability to transfer stresses
from one domain to another without separation will increase [34-39].
As both polymers traverse the capillary, they will spontaneously rearrange themselves
until the configuration which produces the least amount of resistance to flow is obtained,
336
Heterofibres
thus the material of lower viscosity will move to regions of greater shear, near the walls,
causing an encapsulation effect.
Improved uniformity of S/S cross-section for each component is achieved with separate
entry of the two polymers, using separate extruders and metering pumps, followed
by final introduction of the two streams together in the same capillary.
337
Fibres
process
Handbook of Polymer Blends and Composites of
More uniform C/S cross-sections are obtained when the core component is introduced
centrally into the spinning orifices. The polymer with the lower softening temperature
tends to be tacky at take-up and therefore it will be chosen as fibre core; this component
must be more than the 50% as compared with a S/S fibre for the same coil size and
flattening of the stress-strain curve.
Conventional drawing or spin-draw procedures are used to produce BF. After the
drawing step, the yarn passes through a tensionless heating zone, for developing the
spiral crimp.
338
Heterofibres
The first bicomponent application was introduced in 1960 by DuPont (S/S elastic fibre
called Cantrese made from two types of PA polymers with different retraction) [9].
Very complex and expensive spinning packs used in the manufacturing process were found
to be technically and economically unsatisfactory. After 1990, using thin flat plates
(developed by many firms, i.e., Hills Inc. or Enca Inc.), with grooves and holes to conduct
the polymer, the process became very flexible and quite price effective. Production of BF
involves two polymer sources, a system to conduct each polymer to the spinning packs
and a distribution system to meter the both polymers to the die. Polymer sources can be
two continuous polymerisation processes, two extruders or one of each. The main polymer,
e.g., PET, PA, PP, etc., is normally supplied from a continuous polymerisation process
either directly or as a side stream of a production unit. The minor polymer, e.g., polyvinyl
alcohol (PAV), polystyrene (PS), PE is furnished from chip through an extruder. The heart
of the spinning machine is the bicomponent blocks and spinning packs. After the investment
for bicomponent facilities, the production cost is comparable with that of a
monocomponent line of same capacity. The facility can spin conventional monocomponents
or reclaimed polymer chips when bicomponent production is not required.
The spinning pack is the most important part of the fibre spinning process. It includes
metering pump, filters, extruders with distribution plate, dies.
Extruders for BF are basically the same used for producing thin films and common
fibres. Generally, a single screw extruder is normally satisfactory but a recent trend is to
use twin-screw extruders, that will allow use of incompletely dried or additivated polymers.
Dies
Since the early days of the man-made fibre industry, it has been recognised that the die is one
of the most important components in ensuring successful production, being regarded as the
heart of the spinning operation. Fibre uniformity depends on capillary geometry [40].
The number of holes can range from one to tens of thousands; for staple fibres, die hole
counts will be as high is operationally possible and for multifilament yarns this number
will be a small integral number times the number of filaments in the yarns.
The design of the die will depend upon the product to be obtained, the polymer being
used, the difference being how the polymers are distributed to the die [41-43]. Round,
elliptical and rectangular spinning packs are used for producing BF.
The total die hole length depends on the pressure being exerted by the metering pump.
Therefore, the die plate has to be designed to withstand the pressure being exerted. For
339
Handbook of Polymer Blends and Composites
large diameter or wide dies, the die plates are much thicker. Since the length of the capillary
is fixed, the extra thickness must also contain a longer lead hole down to the capillary.
Separate melt streams enter the top of the pack, where the screens serve to filter and to
help uniformly distribute the flows across the width of the pack. From here the streams
are split and routed to match the number of die orifices. Finally, the two polymers are
brought together just prior to exiting the die.
Considerable ingenuity has been devoted to the design of distribution systems, giving good
control of the proportions of the two polymers between and along filaments [44-52].
Many spinning packs are quite versatile, allowing a wide range of fibres to be made, e.g.,
interchangeable plates allow various C/S and S/S configurations to be spun, or alternate die
faces can also be used to produce any desired overall fibre cross-sections (round, multilobal,
rectangular, etc.), or a uniform mixture of fibres of different cross-sections [53].
BF having a common core polymer and different sheath polymers can be extruded from
alternate die orifices in the same die plate. Multiple distribution plates have surface
grooves or depressions to introduce polymers from different sources to only selected die
orifices while maintaining the polymer flows separated from each other.
The shape of the die orifices can be chosen to correspond the cross-sections desired for
the extruded fibres. While circular die orifices are commonly utilised, other non-round
cross-sections may be provided for special applications. Multi-lobal fibres, i.e., tri- or
quadrilobal, X- or Y-shaped, or other area are possible.
Die manufacture sees fashion and technology trends, as do other parts of the man-made
fibre industry and its suppliers. Especially, oriented fashion is the demand for bicomponent
packs. On a technological level, the demand for die coating is increasing, to prevent
deposits on the die, for durability, cleanability, compatibility with cleaning processes
[54-58].
The thick distributing plate and the accurate machining are both expensive and preclude
any realistic possibility of rendering the plates disposable as an option to periodic cleaning.
340
Heterofibres
250 m, with accurate flow distribution paths defined therein at maximally high density.
The distribution plan paths may be photochemical etched. This etching process is much
less expensive than the drilling, milling, reaming or other machining/cutting processes
utilised for thick distribution plates. The precision of etching assures fibre uniformity in
shape and denier.
The resulting low cost etched plate can, therefore, be discarded and economically replaced
during the periodic cleaning of the spinning pack.
Many inventors have dedicated their creative work to the spinning installation for HF
(Table 9.2).
An early method of producing a S/S BF was devised by Sisson and Moorhead [15] in
1953 who channelled two polymer streams separated by a knife edge or a septum into a
die orifice. Today this basic principle is still used, introducing polymer streams into die
through narrow channels but without knife edge [92].
Higher processing speeds, reduced equipment investment, more bulk, improved handle,
aesthetics, and elimination of down stream processes are among the advantages of this
technology. spinning intallations for
341
Handbook of Polymer Blends and Composites
The components must show very good adhesion to each other. The two components
may be connected mechanically or compatibilised chemically.
Two components (solution or melt) are fed to the die orifices and BF as a combined
streams are formed near the orifices during extrusion or at the surface of the die.
Two components are firstly formed into a multilayered structure and slowly fed into
the orifices that intersect with the polymer interface layers.
Two components are formed into a layered structure and fed into the orifices that
dont intersect with polymer layers so that a wide range of combination ratios of two
polymers can be obtained.
Two components are slit-extruded into a layered film, which is then cut into strips,
drawn, cut into staple and fibrillated and then crimped by heat relaxation.
Adhesion of the polymers is not always essential, a highly contoured interface between
core and sheath can mechanically interlock both components [93, 94].
Two streams of polymers are separately led to the die orifices and then extruded
together in C/S form, with a very different geometry of spinning holes shape
(concentric, excentric, tipped, eliptical, trilobal or other types).
For a concentric core, the flow conditions of the polymer are strictly controlled to maintain
the central position of the spinning orifice outlet.
For an excentric core one can modify: the position of the inner polymer channel; the
supply rates and ratios of the two polymers; the spun concentric filament passage over a
hot edge. The fibres obtained have an inherent latent crimp property, which can be
developed after proper treatments, such as:
342
Heterofibres
Adhesion between the matrix and fibril is not essential because higher interfaces of polymer
are presented [95-102].
The mixture of the two polymers in the appropriate concentrations (under 20%) is spun
and one polymer is suspended in droplet form (globules) and then elongated during
subsequent drawing within the second melt, so it appears as fibril within the final fibre.
Fibres with small amounts of added polymer (10% or less) are also produced, the polymer
is present to improve dyeing, antistatic or other physical properties [103-108].
Another method for producing M/F BF injects streams of core polymer with many tubes,
into a sheath polymer stream.
During the last decade, man-made fibres with deniers finer than silk have became available
from the major fibre producers. These low deniers called microfibres are defined as
fibres having deniers below one. Although PES dominates this market, a significant number
of products contain PA and PP and a few contain acrylic or rayon fibres.
Recently, the techniques used to spin BF have been refined and extended to allow the
production of submicron or nanometer size fibres, applied to filament, staple, spunbond,
and meltblown processes [109, 110].
There are many ways to obtain submicron diameter fibres: direct spinning,
microfibrillation by bicomponent techniques [(for mechanically splittable fibres (pie,
stripe, citrus or hollow types) and dissolvable ones (islands-in-the-sea or tipped types)];
spunbond and meltblown processes.
One of the oldest methods used to make micron-sized fibres is to spin a splittable
segmented pie BF. Segmented pie cross-section is an example where the wedge-shaped
343
Handbook of Polymer Blends and Composites
segments closely touch in the middle of each fibre. This high perfection is required to
control fibre properties such as splittability and dyeability. Realisation of this perfection
requires precision in the polymer distribution system, melt viscosity control, heat transfer
uniformity and uniform residence time from the initial melt to the die face. The number
of pie-segments can range between 8 and 36 or more but for commercialised fibres the
common limit is 16, in staple or filament form, using PES and PA. Because of cost, the
amount of PA used is only 15 to 20% and the amount of PES used is 80 to 85%. After
splitting, this results in wedge-shaped filaments of varying cross-sizes ranging from 0.5 to
2 m across to 6 m in length. Some producers are also using newly developed polymers
to make these fibres, e.g., water-soluble or easily-dyeable modified PES.
For this technology it is possible to spin a hollow segmented pie fibre [111, 112], easily
splittable into small micron size rectangular filaments using an air jet.
The application of this technology to the spunbond process involves the high speed
spinning of 16 segmented pie fibres into a web, lightly bonded and then hydroentangled
to obtain a fabric composed of micron size filaments.
For melt blowing process it appears to be an easy step to apply the segmented pie
technology. Since melt blown filament diameters are already of 1 to 5 m, only
four segments would be used to attain this dimensions after spinning.
The citrus type BF can be made by polymer distribution and/or die modification of pack
assemblies. By introducing polymer streams alternately through eight radial channels
toward the die hole, the resultant fibre is an eight segment citrus type fibre.
If the die orifice has the configuration of some slots on a circle, the fibre is a hollow
citrus type.
Tipped fibres, derived from C/S structure, can be utilised in a very sophisticated technology
developed by Hills Inc., for special products with aesthetics, bonding or other properties.
The minor component is placed in small quantities on the tips of trilobal or cross-section
fibres. For example, the core polymer is a melt spinnable polymer, e.g., polyurethane
(PU), and the tip polymer is another thermoplastic one, e.g., PP. The ratio of the two
polymers is about 70:30. After spinning the fibres are twisted and then wet-heat treated
(dyeing). The PP polymer on the tips of the fibres breaks apart into MF of approximately
0.2 denier each and spiral around the PU core that shrank during the dyeing. The resulting
yarn has 100% elastic stretch and looks similar to a conventional core/spun yarn except
the processing costs are substantially reduced and the fibres ringing the core are
microdenier (~0.2 dpf).
344
Heterofibres
Using this same technique in combination with blocking plates can be used to make
intricate patterns of any shape (letters for identification for product privacy or for brand
identification) of one coloured island polymer inside of the other can be obtained.
The etched distributor plate facilitates extrusion of about 0.1 denier MF staple. Each die
orifice issues a master fibre made up of multiple S/S loosely bonded subfibres, which can
be separated into multiple, low denier uniform MF made from multicomponent polymers
(with mechanical working) or from the same polymer type (with dissolving and recovering
of one polymer from the solvent after the fibre extrusion) [90].
Nowadays there are many ways of providing an improved melt/solution spinning for
extrusion plural-component fibres. The density of the die orifices in a spinning pack can
be maximised, e.g., 1768 master fibres, each having a drawn denier of 6.4 and finally 64
micro-fibres of 0.1 denier issued from 64 segments in a checkboarded pattern.
In this type of the bicomponent manufacture, the fibre segments are separated, by
dissolving one of the components. Islands-in-the-sea configuration has similar uses to
the C/S type or may be used as an intermediate in the production of other specialised
types of fibre, i.e., MF or porous fibres where one component is dissolved away leaving
the other. Most of the cross-sections similar to those obtained for mechanically spittable
fibres can be realised by alternating the soluble and insoluble segments.
The solvent can be introduced at any processing step after spinning, but most conveniently
during fabric finishing, eliminating the problems in the handling of the fine fibrils. Since
the solute must be recovered for economical and ecological reasons, it is desirable to
minimise the soluble portion. Multiple filaments of 2 to 5 dpf (12 to 20 m) are spun
into an expandable matrix or a soluble sea that is dissolved away generally after the
knitting or weaving to leave submicron island filaments on the material surface. Twenty-
four and 32 islands-in-the-sea fibres are used for various products, i.e., ultra-suede and
artificial leather. Extrusion of multiple streams into each filament can be done in a number
of ways (by multiple injection tubes, funnel combining chambers or static mixers that
subdivide the incompatible polymer multiple times in the matrix stream). PET, PA or PP
and PS are used as the islands, and water soluble PES and plasticised or saponified PAV
are the dissolvable portions. Typical polymer ratios are 20:80 sea-islands. When the
island polymer is greater than 65% of the total filament mass, the island filaments become
square in shape due to the packing density.
345
Handbook of Polymer Blends and Composites
A dissolvable sea must be relatively cheap and recoverable, easy to dissolve and to handle
(non-toxic, non-corrosive, etc.), and stable and melt spinnable at a temperature compatible
with the island polymers (PET, PA, PP) [113].
This technology has not yet been applied industrially to spunbonds or melt blowns. This
extension will need further development.
Within the past few years, developments have been greatly expanded for the islands-in-
the-sea technology. It was started with spinning a hundred islands and a new technique
appeared later in pack design, that allowed the island count to approach 1000, and the
technology goes further. The interesting fact is that this can be done in fibres with a total
denier as small as 2 dpf (12 m). After the drawing and the sea dissolving, the island
fibre diameters range from 100 to 800 nm. The spinning and processing of the fibres up
and until the dissolving processing are quite similar to melt standard spinning.
Beside the classical methods of obtaining BF, in the last years there has been increased interest
in polymer blends because the synthesis of new homopolymers is very expensive [114].
Research on the bi- and multicomponent fibres obtained from polymer blends are oriented
to revaluation of existing polymers, for example the natural ones, that are in abundance
and have continuously regenerating sources.
In addition, these polymers have strong chemical bonds that will improve the interactions
between the blend components.
Examples of natural cellulosic materials used for polymer blends are wood [115, 116],
bast fibres as flax [117, 118] kenaf, jute [119] and sisal [120-122] and pure cellulose [123].
The main advantages of natural fibres are: acceptable specific strength properties, low
cost, low density, high toughness, good thermal properties, reduced toolwear, low dermal
and respiratory irritation, ease of separation, enhanced energy recovery, and
biodegradability.
The main limitations in the broad use of these materials have been the poor compatibility
between the natural and synthetic polymers and their difficulties of processing.
As it is well known, when two different polymers are mixed, the result is a multiphase
system, where the improved adhesion results from the reactions and enhanced polar
interactions at phase boundaries.
346
Heterofibres
The complexities involved with the compatibilisation of these materials and the different
techniques used to understand the interfacial interaction are reviewed by Maldas and
Kokta [124] (see also Chapter 7 on Reactive Blending).
There are two principal ways to process the nonmiscible polymer blends, namely:
The first one involves chemical modification, e.g., of cellulose, allowing film or fibre
production followed by a regeneration step [120] (Figure 9.7).
The second one consists of dissolving both polymers in two appropriate solvents,
then producing fibres or films, and finally precipitating the solution in a nonsolvent
of cellulose (Figure 9.8).
The multiphase blends seem to be more promising for obtaining some synergistic
improvement for mechanical behaviour (for example cellulose/PA 66 blends). Cellulose
has a relatively high tensile modulus, but a very small tenacity (brittle behaviour). On
the contrary, PA 66 has a good tenacity due to its large strain at break.
Among the various ways to obtain polyphase polymer blends, the phase separation
controlled by the crystallisation of one parent polymer is original and new [125, 126].
This step results in well-dispersed PA domains in a continuous cellulose matrix and
because the rest of the process (extrusion, casting, etc.) is made at a temperature much
347
Handbook of Polymer Blends and Composites
lower (~ 65 C) than the melting temperature (for PA 66 about 250 C) the spatial
distribution of those domains remains unchanged. However, to have good mechanical
properties, it is important that interfaces have strong adhesion [37, 96, 127, 128]. This
can be achieved if a certain miscibility exists, leading to a more or less narrow composition
gradient between domains. A constant amount of PA 66 (~ 5% of blends) does not
crystallise and is mixed with amorphous cellulose and the depression of the PA 66 melting
point indicates that strong interactions between cellulose and PA 66 exists [114].
Studies on PA [131] and PP [132, 133] indicate that blending of a small percentage of
high MW material does not affect the melt viscosity of the parent component to any large
extent but properties of drawn filaments can be significantly improved. These
improvements are related to a distorted spherulitic structure at the spinning stage, increased
amorphous orientation, long chain molecules acting as tie chains and a possible reduction
in crystal size.
348
Heterofibres
Optimum breaking stress (up to 0.74 GPa) and initial modulus (up to 7.34 GPa) of the
drawn samples are achieved at about 6% high molecular weight PP blend composition
when the draw temperatures change between 55 C and 127 C.
The increased values are a result of long molecular chains of the high molecular weight
PP fraction acting as interconnecting links between crystalline regions [120].
On the other hand, inferior strength and modulus at a relatively high percentage of the
second component are related to phase segregation of long chain molecules of high
molecular weight PP fractions.
During heat setting, the taut ties of macromolecules rearrange themselves in a state of
lowest potential energy. This process results in a detectable change in amorphous
orientation and crystallinity and consequently mechanical properties of the filaments.
Heat setting of drawn filaments shows little change up to 140 C and a rapid drop in
mechanical properties for 150-160 C [144].
Long siloxane chains of silanised bast fibres form a ductile interface along with good
fibre-matrix interfacial adhesion.
The problems associated with simple mechanical blending were overcome by reactively
compatibilising the blend components.
Chemical modification of the polar or hydrophilic fibres makes them more hydrophobic,
improving their compatibility with non-polar thermoplastic polymer, e.g., PP.
The same way to compatibilisation can be chosen and for polymer matrix and many
research works have been studied these possibilities [120, 145-149].
A combination of fibre and matrix modifications favours the various reactions and
enhanced polar interactions at phase boundaries which improve the interfacial adhesion
significantly, as opposed to simple mixing of the two components, since new covalent
bonds between the fibre surface and matrix are created in the former case.
A promising new trend is the development of polymer blends containing industrial wastes
of low density polyethylene (LDPE), PS and LDPE-PS mixtures or other combinations
[150, 151]. These recycled polymers show that virgin ones undergo chemical changes
under the action of external factors, i.e., ultra violet (UV) radiation, oxygen or water, so
that reactive groups (vinyl, ketones, ether, ester) form in them.
The reactive groups formed in the polymers during their processing and service life lead
to chemical and specific interactions at the polymer interfaces. It is natural to assume
349
Handbook of Polymer Blends and Composites
that these polymers are capable of specific bonds (for example hydrogen bridges) and
chemical interaction with other polar components, i.e., cellulose or proteic fibres) [115,
120, 152-156].
Figure 9.9 Mixed fibres. a = different colours; b = different colours and deniers;
c = different colours and cross-sections; d = BF and homofilaments mixture
350
Heterofibres
This entire new range of man-made fibres has many properties varying from very weak
to very strong, highly brittle to highly extensive, highly hydrophobic to hydrophilic,
highly resilient to no resiliency at all, highly absorbent to highly repellent [176-179].
BF are created by spinning two different polymers together in different structures for the
quality advantages one can add to the other.
C/S configuration may be used in the symmetrical or asymmetrical forms to combine the
properties of the two components, perhaps taking advantage of particular physical
properties, i.e., the strength, conductivity or reduced costs of the inner component and
the aesthetics, lustre, dyeability, stability, bondability of the outer component.
Typically, PES, PP and PA are used as the core material. This core is surrounded by a
sheath of more expensive polymers, e.g., coPES, polyolefin (PO) and coPA with special
characteristics. The sheath can control the thermal, chemical, electrical, optical and other
characteristics, i.e., fragrance, of the combined fibre.
If, for example, a manufacturer wants to introduce an antimicrobial agent into the fibre,
the antimicrobial cannot be added to the PES during the extrusion phase because of the
extreme heat used to produce the fibre. However, in the bicomponent process, the
microbial agent is put into a PO separate stream, then blended with the PES for a brief
period at the end of the process.
For C/S structure, as the radius of the interface increases so does the area of contact
between the phases, and finally the surface area for adhesion, i.e., for bonding. There are
two potential disadvantages [34].
The economic benefit of producing a BF is lost with decreasing sheath thickness, when
the sheath polymer, e.g., PP, is much less expensive than the core polymer, e.g., PA or
PES, from a raw materials standpoints.
Secondly, as the sheath become thinner, scuffing and other wear processes, e.g., for carpets,
which remove the fibre material will expose the core more quickly to support and absorb
the energy during the shearing and crushing action of the wear test. The main purpose
351
Handbook of Polymer Blends and Composites
for producing a BF with a staining resistant PP sheath and a wearing resistant PA core
was to obtain a carpet with improved stain and wear resistance. But in time the wear
that leads to failure of the interface allows for spaces into which the staining liquid may
seep into the core of the fibre.
BF wet-spun from soybean protein (core) and polyvinylalcohol (cover or sheath) or vice
versa, have excellent wet strength, shrinkage and very desirable hydrophilic properties [183].
S/S and asymmetrical C/S configurations, recalling the bilateral structure of wool, are
mainly used to generate a helical self-crimping and consequently extrabulk in the fibre
through the differential shrinkage in the two sides of the filament as a result of the
difference in physical properties of the two components (comonomer content, melting
and glass transition temperatures, MW, etc). To achieve a desirable coil size and frequency,
one component should have at least 30% greater shrinkage than the other component
[184-186]. Bulk and crimping are then readily developed at a later heating that allow the
filaments to relax [12, 187-193].
As spun these fibres show no crimping but upon drawing a very tight crimp is achieved,
because the sPP retracts more than iPP. This is likely to be due to the lower crystalline
content after spinning for the syndiotactic component, which has a greater rubbery,
amorphous character above the glass temperature domain. The effect of drawing stress
or only of slight heating without tension will be a high shrinkage of sPP portion of fibre
and consequently a high degree of crimp.
352
Heterofibres
This fibre does passes certain limitations in comparison with a wholly iPP fibre. The
melting point of the sPP is lower than that of the iPP, lowering the maximum service
temperature of these fibres. The lower crystallinity of sPP will weaken the fibre tenacity.
The possible failure at the interface of the two polymers may be an added complication.
But the excellent crimp stability is a powerful reason for using such fibres when they are
required for special end-uses.
The BF made from (PE/PP) have good cordability, environmentally friendly bondability
and consistent quality reflected in softness, wicking speed, strength and liquid management
properties for baby diapers, sanitary napkins and filters.
An artificial smart fibre system based on bicomponent design of pine cones is proposed.
One fibre-forming polymer has a higher coefficient of hygroscopic expansion and a lower
tensile stiffness than the other. By co-extrusion it is possible to produce a BF system that
responds to changes in relative humidity (RH) by bonding. Differential speed of extrusion
of the two polymers comprising the system gives a spiral-shaped BF that will amplify the
movement with changing RH. Such a fibre system could be incorporated into fabrics,
that would be able to change their breathability in response to changes in the local
micro-environment.
Acrylic yarns from water reversible crimp (WRC) BF have superior bulk and resilience
allowing elongation, but at the same time recovering the original shape and size even
after machine washing and tumble drying.
Matrix-fibril (or fibril-matrix) BF has fine fibrils of one component embedded randomly
in a matrix of the other. The individual fibrils are of varying but very restricted length
and do not extend along the full length of the fibre. As with the C/S BF, this configuration
is used to combine the properties of the two components [194-205].
With the advent of man-made fibres in the twentieth century there are rapid developments
in their various end-uses [206-221], e.g., agro- and geotextiles, medical and automotive,
architectural and constructional textiles, reinforcements for composites, other than apparel
[222, 223] and furnishing products.
353
Handbook of Polymer Blends and Composites
Although the market for BF is very rich and the inventors are very creative, there are
only a few commercial names found in the patent or firms references. Production and
applications expanded, so that today, the BF have a legitimate existence. There are over
30 firms in the world that manufacture BF as staple or filaments, such as from USA
(Akzo Nobel, DuPont, Fibre Innovation Technologies, Hills Inc., Hercules Inc.,
Intercontinental Polymers Inc., Solutia, etc.), from Japan (Chisso, Daiwabo, Kuraray,
Teijin, Toray, ToyoBo, Unitika, KoSa, Mendels Ltd., etc.), from China (Zhouquing
Microfibre Co., Chengdu Henyi Microfibre Co., etc.), from Taiwan (Nan Ya, Shingkong,
Tuntex, Lea Lea, Far Eastern Textiles, etc.), from Denmark (Danaklon, M&J Fibretech
a/s), from Germany (BASF, Bayer, Hoechst), from Ireland (Wellman International), from
Italy (Bausano SpA, Attrezzature FPA Srl, Montefibre), etc.
Table 9.3 summarises some commercial BF and Tables 9.4, 9.5 and 9.6 present the
successful work of the worldwide inventors for creating new HF.
From these tables one can see the principal directions for end-uses of BF. There is a vast
patent literature without chemical structure indications of HF but with many applications
ideas, such as: absorbent articles [298-300], coloured BF [70, 301], medical and sterilisable
barrier articles [302-304], multilobal BF [256, 305], nonwovens [306-316], spunbonded
fabrics [317, 318], tobacco filter [319] and MUF [320-331].
Today, the largest use for C/S technology is to produce crimping and bonding fibres for
nonwoven industry (carpet and upholstery fabrics). The fibre sheath has a lower melting
point than the core and at heat treating the mantle melts, creating bridges with another
adjacent fibres.
The concentricity/eccentricity of the core plays an important role in final fibre use. For
the strength concern concentric fibres are used, but for the bulkiness requirement eccentric
types are used.
354
Bicomponent fibre structure
Heterofibres
355
Bicomponent fibre applications of
The main characteristics of the sheath (absorptive power, dyeability, soil resistance, heat
insulating properties, bondability, etc.), improve the overall fibre properties which can
lead to multiple applications.
A wide variety of mostly low priced/low quality fibres such as coir, cotton and now PP or
bicomponents are also used in furniture construction in woven or nonwoven forms in
non-visible areas, i.e., linings, cushion and mattress tickings and platform cloths. In
addition, narrow woven webbings, often with PU coated PA yarns, are used as strapping
for seats, etc.
356
C/S structure patents
Heterofibres
357
C/S structure patents
Handbook of Polymer Blends and Composites
358
C/S structure patents
Heterofibres
359
C/S structure patents
360
S/S structure patents
Heterofibres
361
Island-in-the-sea fibre structure patents
improved manufacture techniques, e.g., the use of BF reduces or eliminates the resins
for the bonding step;
the environmental requirement, using recycled fibres, e.g., from plastic bottles, in the
fibrefill;
the inclusion of anti-allergenic additives to reduce further the risk of asthma and
other allergic reactions;
the flammability requirement with regard to PU foam, and in particular the risk of
toxic fumes, both during house fires and in disposal by incineration.
362
Heterofibres
Bicomponent fibres are most often used as an adhesive delivery system, especially in
nonwovens for industrial filtration and personal care products (baby diapers or adult
superabsorbents), for the automotive and the apparel industries, and in surgical supplies,
i.e., gowns, drapes and packs.
The first system (PP/PE) can be thermally bonded at the melt temperature of the PE
sheath but below that of the PP core, without fibre shrinkage or PP degradation. The
high degree of fibre to fibre bonding eliminates the need for a core support and any
possibility of media migration.
The second system (PES/CoPES) has wide solvent and chemical compatibility; then it is
suitable for many high temperature applications. The BF can be utilised for the construction
of an electrochemical system to concentrate and remove ions from water because reverse
osmosis, electrodialysis, multi-stage distillation and other water purification processes are
costly, energy intensive and require substantial maintenance due to chemical activities.
The C/S configuration of BF builds a stable, robust fibrous structure for the immobilisation
of functional particles (such as activated carbon, zeolites, ion exchange agents) which
are secured, yet fully accessible. Maximum accessibility depends on the fluid flow through
the material, when the molecular contamination removes from streams.
Another solution with reduced cost is a homogenous mix of BF having alternating sheaths
of fluoropolymers and PA over a relatively inexpensive common core material such as PP.
The mixtures of fibres of different cross-sections, e.g., PA, PET, PO, are useful as high
efficiency filters in various environments, coalescent filters, reservoirs for marking and
writing instruments, wicks and other elements designed to hold and transfer liquids for
medical and other applications, heat and moisture exchangers (HME) and other diverse
fibrous matrices [53].
363
Handbook of Polymer Blends and Composites
For coalescent-type filters such as those used to separate water from aviation fuel, a
silicon coated fibreglass is used or a mixture of fluoropolymer fibre and a bonding fibre,
i.e., PP, PE and PET. PET has a higher surface energy than the PO, which allows it to
absorb more liquids.
The island-in-the-sea technology is now at the point where the fibres cross-section round
or square, uniform or nonuniform, of any size down to submicron can be produced
commercially at reasonable costs and certainly filtration is an excellent prospect for
the future.
The barrier nonwovens for medical applications would be a low cost material that is
breathable, sterilisable, flexible and extremely resistant to blood and viral penetration.
A 30/70 C/S PE/PP can be used to give to the materials improved softness as well as good
barrier properties and low coverage index.
A splittable fibre such as a segmented pie can be provided as a staple fibre, carded and
then hydroentangled, to give a soft, highly pliable fabric with excellent barrier properties.
In the near future this process will be combined with spunbond fabrics to improve the
cost even further for disposable, lightweight fabrics with excellent barrier properties.
The C/S (PE/PP or PES) BF created many new product development opportunities and
represented the start of thermal bonding of cellulose wood pulp webs, known as thermally
bonded airlaid (TBAL) utilised in the absorbent hygiene products.
A hybrid bonding system known as multi-bonded airlaid (MBAL) has been developed
from mixtures of synthetic fibres and cellulose fluff pulp that offers high bulk, low density,
fast penetration, high absorbency, good tensile strength and a very low lint or dusting level.
The ultrasonically bonded PES/PP blend offers the breathability to provide comfort.
S/S or C/S configurations with a combination of PES, PP, PE, coPES are popular choices
for the airlaid bonding process because they provide the lower melting point polymers
required creating a bond.
364
Heterofibres
Several properties are used to obtain self-crimping fibres, namely: different shrinkage
elastomeric or swelling characteristics, different moduli of the components, the overall
and individual cross-sections of the fibre and of each component.
Thermobondable fibres have polymers with different melting points on the two sides,
e.g., PE, PP/PET.
Splittable fibres can split apart in certain processing steps (dissolution, flexion, etc.), yielding
very fine filaments of a sharp-edged cross-section. Conjugate rayon, acrylic, PA fibres may
be fibrillated provided the adhesion between the two polymers is not too high. For viscose
rayon, the other component should be an alkali-soluble polymer, i.e., hydroxyethyl-,
carboxymethyl- or carboxyethyl cellulose, PAV or polyacrylic acid (PAA) [12].
S/S bicomponents on retraction create a spiral or helical crimp. Any two polymers that
differ in chemical composition or response to processing conditions can be used. The
highest retractive forces of current commercial products are generated by PET/
polybutyleneterephthalate (PBT) compositions. These are being sold as low cost Spandex
substitutes for use in bathing suits. Varying the proportion of each polymer can
control crimp.
The fine fibres produced by this method are used in upholstery, synthetic leather, speciality
wipes, ultra-high filtration media, artificial arteries and many other specialised applications
[335-337].
In 1991, US producers predicted that in 5 years MF would account for almost 50% of
the American PES apparel filament market. In 1993, MF achieved only 2-4% of the total
USA, PES market, 6-7% in Europe and 10-11% in Japan [338].
365
Handbook of Polymer Blends and Composites
Although the market is not expected to reach anywhere near the early estimation, the
myriad of new products and end-uses introduced in the last few years indicates a great
future and an accelerated growth for MF products.
New production techniques especially using splittable fibres have led to decreased cost
of manufacture of finer filaments.
The most commercially successful use of C/S technology is the islands-in-the-sea variety
used for suedes and leather-like applications. The lower bending rigidity, high surface
area and greater covering factor enable the production of fabrics which are softer, quick
drying, lighter and cooler handle, with thermal insulation properties. In the outer layer
of rainwear the yarns can be packed closer to achieve wind and water resistance while
still maintaining the ability to breathe. However, these same properties also produce
disadvantages such as decreased dye capacity, higher polling, roll wrapping and nep
formation during carding of staple fibres.
The high fashion market appreciates these products because they imitate silk at a fraction
of the price. For example, the hosiery industry has been selling luxury PA sheers, opaques,
knee-highs and tights [339].
Around the turn of the century, the world advance technology and equipment for
differential and functioning chemical fibres adopt the integrated processes for the
elimination of inhomogeneities and defects. The preferences will be given to shorter
manufacture lines, increased computerisation and automation, smaller capacity and
volume, high production effectiveness and flexibility, wide-range of raw materials, low
cost, much diversity of products and high added value.
366
Heterofibres
The present production of BF worldwide is only 25 million tons in the man-made fibre
market, but in the next ten years there will be a significant growth especially in the high
volume bondable fibre markets.
As the man-made fibre industry was more commodity-oriented, and with big-volume
producers becoming even larger, speciality fibres have been neglected and it is necessary
to create a niche for the future.
The trend in the fibre industry in many of the industrially developed countries is that
production of conventional fibres remains at a static level, whereas many companies are
switching over to much higher value added products like speciality fibres for high
performance, constantly developed for non-conventional uses.
In the future the heat-textured yarn market will concentrate on the development of
commercial BF.
The fibre industry is undergoing an intense reshaping process. Some traditional chemical
companies, for example, are spinning off their fibre enterprises, while other companies are
acquiring new partners. New manufactures also continue to enter the market, their challenges
include new methods of extension and spinning by focusing on customer needs.
There is a competition among the different synthetic fibres, and when a very high volume
of petroleum-based raw materials is used, a great disadvantage appears in competing
with natural fibres.
367
Handbook of Polymer Blends and Composites
There is a multitude of information concerning HF, not included in this chapter. More
information can be found on the Internet. The most important web sites containing up-
to-date details are listed after the references.
References
1. Textile Terms and Definitions, 10th Edition, Eds., J.E. McIntyre, P.N. Daniels,
Textile Institute, Manchester, UK, 1995.
4. W. Bessey and M. Jaffe in Solid Phase Processing of Polymers, Eds., I.M. Ward, P.D.
Coates and M.M. Dumoulin, Hanser Publishers, Munich, Germany, 2000, 85, 13.
5. I.M. Ward in Solid Phase Processing of Polymers, Eds., I.M. Ward, P.D. Coates
and M.M. Dumoulin, Hanser Publishers, Munich, Germany, 2000, 120, 79.
10. G.R. Cuthberson, R.J. Kelly, J.R. Logan, D.S. Porter and P.D. Brass in Man-made
Fibres, Science and Technology, Ed., H.F. Mark, Interscience Publishers, New
York, NY, USA, 1968, Volume 3, 401, 78.
11. E.M. Hicks, E.A. Tippetts, J.V. Hewett and R.H. Brand in Man-made Fibres,
Science and Technology, Ed., H.F. Mark, Interscience Publishers, New York, NY,
USA, 1968, Volume 1, 375, 54.
12. H.F. Mark and N.G. Gaylord, Encyclopedia of Polymer Science and Technology,
Plastics, Resins, Rubbers, Fibres, Volumes 1-14, Wiley Inc., New York, NY, USA,
1968.
368
Heterofibres
13. R.A. Buckley and R. Phillips, Chemical Engineering Progress, 1969, 65, 10, 41.
15. W.A. Sisson and F.F. Moorhead, Textile Research Journal, 1953, 23, 1, 152.
16. P.A. Koch, Zeitschrift fur die Gasamte Textilindustrie, 1970, 72, 2, 253.
17. B.T. Hayes, Chemical Engineering Progress, 1969, 65, 10, 50.
18. K. Dietrich and S. Tolsdorf, Faserforschung und Textiltechnik, 1970, 21, 3, 295.
19. W. Berger and J. Mellentin, Faserforschung und Textiltechnik, 1970, 21, 3, 288.
20. C.W. Park and W.S. Lee, Journal of Applied Mechanics, 1995, 62, 2, 511.
21. C.W. Park and W.S. Lee, International Polymer Processing, 1994, 9, 3, 359.
22. C.W. Park, Polymer Engineering and Science, 1991, 31, 2, 197.
23. C.W. Park and K.S. Yoon, Polymer Engineering and Science, 1992, 32, 11, 1771.
25. S. Backer and S.K. Batra, Textile Research Journal, 1971, 41, 12, 1008.
26. A.E. Everage, Transactions of the Society of Rheology, 1975, 19, 4, 509.
27. J.L. White and B.L. Lee, Transactions of the Society of Rheology, 1975, 19, 3,
457.
29. A.E. Everage, Transactions of the Society of Rheology, 1973, 17, 4, 629.
30. D.L. MacLean, Transactions of the Society of Rheology, 1975, 17, 3, 385.
31. F.N. Cogswell, Polymer Melt Rheology, Wiley Inc., New York, NY, USA, 1981.
32. S.L. Rosen, Fundamental Principles of Polymeric Materials, 2nd Edition, Wiley
Inc., New York, NY, USA, 1993.
33. B.B. Bird, R.C. Armstong and O. Hasseger, Dynamics of Polymeric Liquids,
Volume 1, Wiley Inc., New York, USA, 1987.
369
Handbook of Polymer Blends and Composites
35. J. Varga, A.F. Breining and G.W. Ehrenstein, Muanyag es Gumi, 1997, 34, 1, 11.
36. T. Komatsu, Journal of Polymer Science: Polymer Physics, 1997, 35, 1, 107.
37. Y. Hong, J.J. Cooper-White, M.E. Mackay, C.J. Hawker, E. Malmstrom and N.
Rehnburg, Journal of Rheology, 1999, 43, 3, 781.
38. S.E. Barbosa and J.M. Kenny, Journal of Reinforced Plastic Composites, 1999,
18, 5, 413.
43. C.J. Wust Jr., inventor; Hercules Inc., assignee; EP662,533B1, 1999.
46. D.E. Wright and W.C. Flynn, inventors; BASF Corp., assignee; EP747,516A3,
1997.
47. D.E. Wright and W.C. Flynn, inventors; BASF Corp., assignee; EP747,516A2,
1996.
48. D.E. Wright and W.C. Flynn, inventors; BASF Corp., assignee; EP747,516B1,
2000.
50. C.J. Wust Jr., inventor; Hercules Inc., assignee; EP662,533A1, 1995.
51. B.L. Davies, inventor; Hoechst Celanese Corp., assignee; US5,256,050, 1993.
370
Heterofibres
52. D.E. Carter, A.L. McPeters and H.L. Skeen, inventors; Monsanto Co., assignee;
US5,017,116, 1991.
53. R.M. Berger, inventor; Filtrona International Ltd., assignee; US6,103,181, 2000.
55. J. Radhakrishnan, T. Kikutani and N. Okui, Seni Gakkaishi, 1996, 52, 11, 618.
56. T. Kikutani, S. Arikawa, N. Okui and A. Takaku, Seni Gakkaishi, 1995, 51, 7, 408.
57. G. Grover, S. Zhu, M.C. Jenkins and A.B. Presley, International Journal of
Nonwovens Research, 1993, 5, 1, 22.
67. J.P. Craig Jr., inventor; Monsanto Co., assignee; US4,284,598, 1981.
69. D.E. Carter, A.L. McPeters and H.L. Skeen, inventors; Monsanto Co., assignee;
EP376,911B1, 1993.
371
Handbook of Polymer Blends and Composites
70. D.E. Carter, A.L. McPeters and H.L. Skeen, inventors; Monsanto Co., assignee;
EP376,911A3, 1991.
71. D.E. Carter, A.L. McPeters and H.L. Skeen, inventors; Monsanto Co., assignee;
EP376,911A2, 1990.
72. J.P. Craig, Jr., inventor; Monsanto Co., assignee; EP034,937B1, 1985.
73. J.P. Craig, Jr., inventor; Monsanto Co., assignee; EP034,937A3, 1983.
74. J.P. Craig, Jr., inventor; Monsanto Co., assignee; EP034,937A2, 1981.
77. B.L. Davies, inventor; Hoechst Celanese Corp., assignee; US5,505,889, 1996.
78. H.R.E. Frankfort, B.H. Knox and G.A. Pai, inventors; E.I. Du Pont de Nemours
Co., assignee; US5,505,894, 1996.
84. R.L. Tabor, G.M. Lancaster, J.O. Bieser and M.F. Finlayson, inventors; Dow
Chemical Co., assignee; EP496,734B1, 1999.
85. B. Marcher, E. Nielsen and P.H. Hansen, inventors; Danaklon A/S, assignee;
WO9,301,334A1, 1993.
88. R.A. Houpt and D.C. Souers, inventors; Owens-Corning Fibreglas Technology
Inc., assignee; US5,595,766, 1997.
372
Heterofibres
89. D.E. Wright and W.C. Flynn, inventors; BASF Corp., assignee; US5,595,699,
1997.
92. M.R. Gownder, inventor; Fina Technology Inc., assignee; US6,074,590, 2000.
94. D.R. Kent, M.B. Hoyt, O.M. Ilg and M.K. Queen, inventors; BASF Corp.,
assignee; US5,885,705, 1999.
95. Y.H. So, A. Sen, P. Kin and V.S. Jeor, Journal of Polymer Science: Polymer
Chemistry, 1995, 33, 17, 2893.
98. H. Miller and F. Harris, International Fibre Journal, 1998, 13, 3, 38.
100. C.F. Helms Jr., D.R. Kent, M.B. Hoyt, J.R. Bristow and P.E. Wilson, inventors,
BASF Corp., assignee; US6,017,479, 2000.
101. D.R. Kent, M.B. Hoyt and C.F. Helms Jr., inventors, BASF Corporation, assignee;
US6,010,654, 2000.
102. S. Iannace, G. Nocilla and L. Nicolais, Journal of Applied Polymer Science, 1999,
73, 4, 583.
104. Y. Longueville, Composites Plastiques Reinforces Fibres Verre Textile, 1994, 34,
5, 7.
105. G. Li, J. Yin, B. Li, G. Zhuang, Y. Yang and L. Nicolais, Polymer Engineering
and Science, 1995, 35, 8, 658.
373
Handbook of Polymer Blends and Composites
106. S.S. Morye, P.J. Hine, R.A. Duckett, D. Carr and I.M. Ward, Composites, Part A:
Applied Science, 1999, 30, 5, 649.
107. P.E. Lopes and J.A. de Sousa, Polimeros: Ciencia e Tecnologia, 1999, 9, 1, 85.
108. W. Youshan, X. Tao, L. Ying and W. Yanlin, Luntai Gongye, 1999, 19, 3, 137.
109. C.F. Helms, Jr., O.M. Illg, D.R. Kent, M.B. Hoyt and J.A. Hodan, inventors,
BASF Corporation, assignee; US6,153,138, 2000.
110. D.L. Krueger and J.F. Dyrud, inventors; 3M Innovative Properties Co., assignee;
US6,057,256, 2000.
111. D.R. Kent, M.B. Hoyt and C.F. Helms, Jr., inventors; BASF Corporation,
assignee; US5,904,982, 1999.
112. D.R. Kent, M.B. Hoyt and C.F. Helms, Jr., inventors; BASF Corporation,
assignee; US6,017,478, 2000.
115. B.P. Mamunya, V.D. Mishak, G.M. Semenovich and E.V. Lebedev, Polymer
Science, 1994, 36, 8, 1125.
119. A.K. Rana, B.C. Mitra and A.N. Banerjee, Journal of Applied Polymer Science,
1999, 71, 4, 531.
121. P.V. Joseph, K. Joseph and S. Thomas, Composites Science and Technology, 1999,
59, 11, 1625.
374
Heterofibres
123. L.Y. Mwaikambo and E.T.N. Bisanda, Polymer Testing, 1999, 18, 3, 181.
124. D. Maldas and B.V. Kokta, Composite Interfaces, 1993, 10, 1, 87.
126. M.U. Otaigbe, C.J. Quinn and G.H. Beall, Polymer Composites, 1998, 19, 1, 18.
128. Y. Seo, S.M. Hong, S.S. Hwang, T.S. Park, K.U. Kim, S. Lee and J. Lee, Polymer,
1995, 36, 3, 515.
129. I. Hamerton, J.N. Hay, B.J. Howlin, J.R. Jones and L. Shui-Yu, Journal of
Materials Chemistry, 1998, 8, 6, 1333.
130. I. Hamerton, J.N. Hay, B.J. Howlin, J.R. Jones, L. Shui-Yu, G.A. Webb and M.G.
Bader, Journal of Materials Chemistry, 1997, 7, 1, 74.
132. B.L. Deopura and S. Radam, Journal of Applied Polymer Science, 1986, 31, 10,
2145.
137. J. Bidaux, G.D. Smith, N. Bernet and J.E. Manson, Polymer, 1996, 37, 11, 1129.
138. S. Al-Malaika, Reactive Modifiers for Polymers, Chapman and Hall, London,
UK, 1997.
375
Handbook of Polymer Blends and Composites
140. D.R. Paul, A.J. Oshinski and H. Keskkula, Polymer, 1992, 33, 2, 284.
141. M.M. Miller, J.M.G. Cowie, J.G. Tait, D.L. Brydon and R.R. Mather, Polymer,
1995, 36, 16, 3107.
142. S.C. Tjong, Y. Xu and Y.Z. Meng, Journal of Applied Polymer Science, 1999, 72,
13, 1647.
144. S.J. Mahajan, K. Bhaumik and B.L. Deopura, Journal of Applied Polymer
Science, 1991, 43, 1, 49.
145. A.R. Samadi, D.F. Chaulfield and R.M. Rowell, Plastics Engineering, 1994, 25,
4, 27.
147. S.C. Wong and Y.W. Mai, Polymer Engineering and Science, 1999, 39, 2, 356.
150. A.M. Kotliar, Polymer Plastics Technology and Engineering, 1999, 38, 3, 513.
152. T. Tanaka, T. Tanigami and K. Yamaura, Polymer International, 1998, 45, 2, 175.
153. A.R. Horrocks, S.C. Anand and D. Sanderson, Polymer, 1996, 37, 15, 3197.
154. B.P. Mamunya, V.D. Mishak, E.V. Lebedev and V.F. Annenkov, Plasticheskie
Massy, 1989, 8, 39.
155. B.P. Mamunya, V.D. Mishak, V.F. Shumskii and E.V. Lebedev,
Vysokomolekulyarnye Soedineniya, 1991, 32B, 11, 839.
376
Heterofibres
156. B.P. Mamunya, V.D. Mishak, G.M. Semenovich and E.V. Lebedev, Plasticheskie
Massy, 1990, 9, 49.
162. D. Gregor-Svetec, Journal of Applied Polymer Science, 2000, 75, 10, 1211.
163. S. H. Zeronian, M.K. Inglesby, N. Pan, D. Lin, G. Sun, B. Soni, K.W. Alger and
J.D. Gibbon, Journal of Applied Polymer Science, 1999, 71, 7, 1163.
164. H.H. Cho, K.H. Kim, Y.A. Kang, H. Ito and T. Kikutani, Journal of Applied
Polymer Science, 2000, 77, 10, 2267.
165. H.H. Cho, K.H. Kim, Y.A. Kang, H. Ito and T. Kikutani, Journal of Applied
Polymer Science, 2000, 77, 10, 2254.
166. H. Limei, D. Zhang and Q. Wang, Journal of Applied Polymer Science, 2000, 77,
13, 3010.
167. S. Natarajan and S. Michielsen, Textile Research Journal, 1999, 69, 11, 903.
171. J. Radhakrishnan, T. Kikutani and N. Okui, Textile Reserach Journal, 1997, 67,
9, 684.
377
Handbook of Polymer Blends and Composites
172. H.S. Kim, H. Ito, T. Kikutani and N. Okui, Journal of the Textile Institute, 1999,
90, 1, 4, 508.
173. H.S. Kim, H. Ito, T .Kikutani and N. Okui, Journal of the Textile Institute, 1997,
88, 3, 3, 37.
175. M.T. Webster, inventor; Burlington Industries Inc., assignee; US4,137,615, 1979.
176. P.V. Papero, E. Kubu and L. Roldan, Textile Research Journal, 1967, 37, 10,
823.
177. Y.B. Choi and S.Y. Kim, Journal of Applied Polymer Science, 1999, 74, 8, 2083.
178. H.S. Kim, H.H. Cho, H. Ito, T. Kikutani and N. Okui, Seikei-Kakou, 1997, 9, 6,
449.
180. E. Morales and J.L. Acosta, Angewandte Makromolecular Chemistry, 1996, 238,
31.
181. C. Zhang, X.S. Yi, S. Asai and M. Sumita, Composites International, 1999, 6, 4,
287.
182. G. Wu, S. Asai and M. Sumita, Macromolecules, 1999, 32, 10, 3534.
187. C. Pieper, A. Nogaj, H. Lohwasser, K. Hurm and H. Bong, inventors; Bayer AG,
assignee; EP032,215B1, 1983.
188. C. Pieper, A. Nogaj, H. Lohwasser, K. Hurm and H. Bong, inventors; Bayer AG,
assignee; EP032,215A1, 1981.
189. E.H. Olson, inventor; E.I. DuPont de Nemours Co., assignee; US3,779,853, 1973.
378
Heterofibres
190. J.T. Summers, inventor; E.I. DuPont de Nemours Co., assignee; EP044,221B1,
1984.
191. J.T. Summers, inventor; E.I. DuPont de Nemours Co., assignee; EP044,221A1,
1982.
192. J.R. Neely, D.F. Clark, T.J. Stokes, C.M. Freese and R.W. Griffin, inventors;
Kimberly-Clark Worldwide Inc., assignee; WO028,123A1, 2000.
193. X. Ning and S.E. Marmon, inventors; Kimberly-Clark Worldwide Inc., assignee;
US5,876,840, 1999.
194. Y.S. Kim and S.C. Kim, Macromolecular Symposia, 1997, 118, 371.
195. N.V. Ionov, V. Shershnev and I.D. Agabibullaev, Kauchuk i Rezina, 1999, 3, 4.
197. K.K. Byung, S.O. Young and M.L.P. Young, Journal of Macromolecular Science,
1999, 38B, 4, 367.
198. S-H. Hwang, Y-S. Kim, H-C. Cha and J-C. Jung, Polymer, 1999, 40, 21, 5957.
199. L. Daniel, P.J. Hogg and P.T. Curtis, Composites, Part B, 1999, 30, 3, 257.
200. J.P.F. Inberg, P.H. Hunse and R.J. Gaymans, Polymer Engineering and Science,
1999, 39, 2, 340.
201. G. Freddi, M. Tsukuda and S. Betetta, Journal of Applied Polymer Science, 1999,
71, 10, 1563.
202. N. Saha, D. Basu and A.N. Banerjee, Journal of Applied Polymer Science, 1999,
71, 4, 541.
203. H. Veenstra, J. Van Dam and A. Posthuma de Boer, Polymer, 1999, 40, 5, 1119.
204. A. Kalamkarov and H.Q. Liu, Composites International Journal, 1998, 29B, 5,
643.
205. Y.C. Chu and S.I. Rokhlin, Metallurgical and Materials Transactions, 1996, 27A,
1, 165.
379
Handbook of Polymer Blends and Composites
206. L.I. Irish, Proceedings of NeedlePunch 2000, Greenville, South Carolina, USA,
2000, Paper No.15
208. D.R. Kent, M.B. Hoyt and C.F. Helms Jr., inventors, BASF Corp., assignee;
US6,162,382, 2000.
212. G. Grover, S. Zhu, A.B. Jenkins, S.O. Hyun and N.R. Bertoniere, Proceedings of
the International Meeting of the Fiber Society, Raleigh, North Carolina, USA,
1993.
213. E. Gebauer and D. Gajewski, Proceedings of the 154th Technical Meeting &
Rubber Mini Expo 98, Nashville, Tennessee, USA, 1998, Paper No.29.
215. M.C. Kenney, S.K. Barlow and S.L. Eikleberry, Tappi Journal, 1997, 80, 3, 77.
216. K.E. Duckett, L.C. Wadsworth and V. Sharma, Tappi Journal, 1995, 78, 3, 169.
217. K.E. Duckett, V. Sharma and L.C. Wadsworth, Tappi Journal, 1995, 78, 3, 195.
220. M.I. Zeidman, M.W. Suh and S.K. Batra, Textile Research Journal, 1990, 60, 1, 1.
380
Heterofibres
224. H.G. Elias, Macromolecules, Volume 1, Wiley Inc., London, UK, 1977.
225. J.H. van Leeuwen and K.A. Weigand, inventors; Akzo Inc., assignee;
US4,473,617, 1984.
231. V.S. Ellis and K.W. Mieszkis, inventors; ICI Ltd., assignee; US4,388,370, 1983.
232. R.A. Breznak, J.A. Foldhazy, H.L. LaNieve and W.A. Piesczek, inventors;
Hoechst Celanese Corporation, assignee; WO9,814,647A1, 1998.
233. E. Rodriguez, J.W. Lindsay and W.E. Streetman, inventors; Sterling Chemical
International Inc., assignee; WO9,855,672A1, 1998.
234. D.R. Kent, M.B. Hoyt, O.M. Ilg and M.K. Queen, inventors; BASF Corporation,
assignee; US6,004,674 , 1999.
235. D.R. Kent, MB. Hoyt, O.M. Ilg and M.K. Queen, inventors; BASF Corporation,
assignee; US6,039,903, 2000.
237. W.B. Segraves and K.L. Mulholland, inventors; E.I. DuPont de Nemours Co.,
assignee; US4,069,363, 1978.
238. Y. Sakamoto, S. Miyahashi and H. Sekino, inventors; Nippon Ester Co. Ltd.,
assignee; JP8,323,070A2, 1996.
381
Handbook of Polymer Blends and Composites
239. B. Marcher, E. Nielsen and P.H. Hansen, inventors; Danaklon A/S, assignee;
US5,540,992, 1996.
240. E.J. Powers, S.F. Nielsen, J.E. Smith and T.S. Thornburg, inventors; Hoechst
Celanese Corporation, assignee; US5,580,459, 1996.
243. R.E. Golz, inventor; Murdock Webbing Co. Inc., assignee; US5,419,951, 1995.
244. R.D. Harris Jr., J.P. Parker and C.L. King, inventors; BASF Corporation, assignee;
US4,732,809, 1988.
245. R.D. Harris Jr., J.P. Parker and C.L. King, inventors; Akzona Inc., assignee;
US4,552,603, 1985.
246. R.E. Golz, inventor; Murdock Webbing Co. Inc., assignee; US5,677,056, 1997.
247. B.I. Goodwin and M. Hottner, inventors; W.L. Gore & Associates GmbH,
assignee; WO9,916,620A1, 1999.
248. W. Kaegi and W. Stibal, inventors; EMS-Inventa AG, assignee; US5,716,568, 1998.
249. R.L. Tabor, Z. Jezic, G.M. Lancaster, G.P. Young, J.O. Bieser and M.F. Finlayson,
inventors; Dow Chemical Co., assignee; US5,372,885, 1994.
254. J.S. Denton, D.B. Eagles, J.G. OConnor and R.B. Davis, inventors; Albany
International Corporation, assignee; EP991,799A1, 2000.
382
Heterofibres
255. J.S. Denton, D.B. Eagles, J.G. OConnor and R.B. Davis, inventors; Albany
International Corporation, assignee; US5,888,915, 1999.
256. J.S. Denton, D.B. Eagles, J. OConnor and R.B. Davis, inventors; Albany
International Corporation, assignee; WO9,812,370A1, 1998.
258. A.J. Hastie, inventor; HNA Holdings Inc., assignee; US5,948,529, 1999.
259. S.F. Nielsen and B.L. Davies, inventors; Hoechst Celanese Corporation, assignee;
US5,167,765, 1992.
260. P.H. Hansen and A.M. Larsen, inventors; Fibrevisions A/S, assignee;
WO9,845,519A1, 1998.
261. R.D. Pike, K.L. Brown and P.W. Shipp Jr., inventors; Kimberly-Clark Worldwide
Inc., assignee; US5,855,784, 1999.
262. B. Marcher, E. Nielsen and P.H. Hansen, inventors; Danaklon A/S, assignee;
EP522,995A2, 1993.
264. M.B. Hoyt, D.R. Kent, C.F. Helms and O.M. Ilg, inventors; BASF Corporation,
assignee; US5,888,651, 1999.
383
Handbook of Polymer Blends and Composites
273. F.F. Kochesky and B.F. Paschall, inventors; American Felt and Filter Co., assignee;
US5,284,704, 1994.
275. R.M. Berger, inventor; Filtrona International Ltd., assignee; US6,026,819, 2000.
276. R.K. Gupta, J.E. Mallory and K. Takeuchi, inventors; Hercules Inc., assignee;
US5,888,438, 1999.
278. A. Talley, A.E. Wilkie and K.H. Buchanan, inventors; BASF Corporation,
assignee; US6,158,204, 2000.
280. T. Omae, M. Sakurai and K. Asao, inventors; Sumitomo Chemical Co. Ltd.,
assignee; JP1,314,729A2, 1989.
281. R.M. Berger, inventor; Filtrona International Ltd., assignee; EP881,889A1, 1998.
282. J.T. Summers, inventor; E.I. DuPont de Nemours Co., assignee; US4,457,974, 1984.
283. J.R. Collier and B.J. Collier, inventors; Louisiana State University and
Agricultural and Mechanical College, assignee; US5,387,383, 1995.
284. M.M. Zwick and W.J. van Loo, inventors; American Cyanamid Co., assignee;
US4,278,634, 1981.
286. D.L. Krueger and J.F. Dyrud, inventors; Minnesota Mining and Manufacturing
Co., assignee; US4,729,371, 1988.
384
Heterofibres
287. T. Fletemier, R. Ette and R. Long, inventors; Findlay Industries Inc., assignee;
US6,156,682, 2000.
288. D.L. Krueger and D.E. Meyer, inventors; Minnesota Mining and Manufacturing
Co., assignee; US4,795,668, 1989.
290. D.L. Krueger and D.E. Meyer, inventors; Minnesota Mining and Manufacturing
Co., assignee; US4,547,420, 1985.
291. M. Yamaguchi and H. Ohkouchi, inventors; Teijin Ltd., assignee; WO9, 702,
377A1, 1997.
293. L.H. Sawyer, G.W. Sawyer and M.A. White, inventors; Dow Chemical Co.,
assignee; US5,126,199, 1992.
294. Z. Jezic and G.P. Young, inventors; Dow Chemical Co., assignee; US5,133,917,
1992.
296. K.S. Baron, M.F. Custer and A.G. Hegedus, inventors; Hexcel Corporation,
assignee; US5,147,721, 1992.
297. M.T. Okamoto, inventor; Toray Industries Inc., assignee; US4,381,335, 1983.
299. D.M. Jackson, W.S. Pomplun, P.S. Mumick and P.W. Estey, inventors; Kimberly-
Clark Worldwide Inc., assignee; US5,916,678, 1999.
300. D.M. Jackson and W.S. Pomplun, inventors; Kimberly-Clark Worldwide Inc.,
assignee; EP836,656A1, 1998.
301. M.B. Hoyt, D.R. Kent, C.F. Helms and O.M. Ilg, inventors; BASF Corporation,
assignee; EP899,364A2, 1999.
302. M. Steckel and L.F. Price, inventors; Ethicon Inc., assignee; EP523,743A1, 1993.
385
Handbook of Polymer Blends and Composites
303. S.L. Gessner and J.D. Gillespie, inventors; Fibreweb North America Inc.,
assignee; US5,503,907, 1996.
304. S.L. Gessner and J.D. Gillespie, inventors; Fibreweb North America Inc., S.L.
Gessner and J.D. Gillespie, assignee; WO9,503,172A1, 1995.
305. C.F. Helms Jr., O.M. Ilg, D.R. Kent, M.B. Hoyt and J.A. Hodan, inventors; BASF
Corporation, assignee; US5,948,528, 1999.
306. T.F. Gilmore, S.K. Batra, N.B. Timble and Y. Feng, International Nonwovens
Journal, 1994, 6, 4, 28.
307. S.L. Gessner, inventor; Fibreweb North America Inc., assignee; US5,593,768,
1997.
308. S.L. Gessner, inventor; Fibreweb North America Inc., assignee; US5,294,482,
1994.
309. D.M. Jackson, W.S. Pomplun, P.S. Mumick and P.W. Estey, inventors; Kimberly-
Clark Worldwide Inc., assignee; US5,916,678, 1999.
310. R.D. Pike, K.L. Brown, S.W. Gwaltney, T.A. Herschberger and S.D. Siegel,
inventors; Kimberly-Clark Corporation, assignee; US5,382,400, 1995.
311. J.M. Duellman, B.D. Haynes, M.B. Lake, J.L. McManus and K.E. Smith,
inventors; Kimberly-Clark Worldwide Inc., assignee; WO037,723A2, 2000.
312. R.K. Gupta and J.R. Williams, inventors; Hercules Inc., assignee; US5,554,441,
1996.
313. R.K. Gupta and J.R. Williams, inventors; Hercules Inc., assignee; US5,582,667,
1996.
316. D.C. Strack, L.A. Connor, S.W. Gwaltney, A.L. McCormack, S.E. Shawver and
J.S. Shultz, inventors; Kimberly-Clark Corporation, assignee; US5,336,552, 1994.
386
Heterofibres
319. R.M. Berger, inventor; American Filtrona Corp., assignee; EP737,038A1, 1994.
320. F-J. Tsai and B.T. Etzel, inventors; Kimberly-Clark Worldwide Inc.,assignee;
US5,952,088, 1999.
322. F-J. Tsai and B.T. Etzel, inventors; Kimberly-Clark Worldwide Inc., assignee;
US5,698,322, 1997.
324. P.G. Martin, G.L. Olson and D.G. Welygan, inventors; 3M Innovative Properties
Co., assignee; US5,972,463, 1999.
325. P.G. Martin, G.L. Olson and D.G. Welygan, inventors; Minnesota Mining and
Manufacturing Inc., assignee; US5,811,186, 1998.
326. P.G. Martin, G.L. Olson and D.G. Welygan, inventors; Minnesota Mining and
Manufacturing Inc., assignee; US5,733,825, 1998.
328. B.T. Etzel and F-J. Tsai, inventors; Kimberly-Clark Worldwide Inc., assignee;
WO9,829,585A2, 1998.
329. F-J. Tsai and B.T. Etzel, inventors; Kimberly-Clark Worldwide Inc., assignee;
WO9,824,951A1, 1998.
330. B.T. Etzel and F-J. Tsai, inventors; Kimberly-Clark Worldwide Inc., assignee;
EP951,593A1, 1999.
331. F-J. Tsai and B.T. Etzel, inventors; Kimberly-Clark Worldwide Inc., assignee;
EP941,377A1, 1999.
332. S. Hyang, H.K. Dong, T. Kitano and I.M. Suk, Kobunshi Ronbun, 1998, 55, 7,
378.
387
Handbook of Polymer Blends and Composites
334. H. Sun, K.B. Rhee, T. Kitano and S.I. Mah, Journal of Applied Polymer Science,
1999, 73, 11, 2135.
338. H.B. Goldstein, Textile Chemist and Colorist, 1993, 25, 2, 16.
Useful Websites
For Patents
Delphion Research; www.delphion.com
United States Patent and Trademark Office; www.uspto.gov
World IP Directory; www.patentcafe.com/patents_etc_cafe/patent_search.html
For Periodicals
World Textile Publications; www.worldtextile.com
International Fiber Journal; www.ifj.com
Wiley Publications Online; www.interscience.wiley.com
ChemWeb; www.chemweb.com
ChemWeb Databases; www.chemweb.com/databases
General Links
American Fiber Manufacturers Association; www.afma.org
Fiber Information Resources; www.afma.org/links.html
Institute of Textile Technology; www.itt.edu/WB links.asp
388
Glass Transition in Polymer Blends
10
Glass Transition in Polymer Blends
Hans A. Schneider
10.1 Introduction
Taking into account the brittleness of polymer glasses, the processing of amorphous
polymers requires the exact knowledge of the associated glass transition temperature
(Tg), to assure the flawless manufacturing of polymeric materials. Although the glass
transition shows many characteristics of a thermodynamic second order transition,
(e.g., continuity of volume, enthalpy, and other first order derivatives of the Gibbs free
energy, but discontinuity of the related second order derivatives, i.e., expansion coefficient,
heat capacity etc.), experimental findings, (i.e., operational and above all heating and
cooling rate dependences), point to a kinetically controlled freeze-in process. Accordingly
any polymer glass is thermodynamically metastable showing characteristic time-dependent
behaviour, well known as ageing of polymers, accompanied by more or less important
changes mainly of mechanical properties. Conditional on this dual aspect of the glass
transition in literature two main models have been proposed for the theoretical description
of the glass transition phenomenon: the kinetic free volume [1-4] and the thermodynamic
conformational entropy [5] approaches.
The free volume model assumes that polymer mobility is controlled by the specific free
volume, defined as vf = (voverall - voccupied), which is frozen in at Tg. The kinetic dependence
of the glass transition is responsible for the experimentally observed relaxation and
ageing of polymer glasses.
The thermodynamic conformational entropy model accepts on the other hand, that
conformational changes are controlled by flexible bonds (by definition those simple
bonds which by rotation induce conformational rearrangements within the polymer coil).
The glass is considered a thermodynamically stable fourth state of matter possessing a
second-order phase transition situated at T2 ~ Tg - 50 K and characterised by zero-
conformational entropy. Due to kinetics, however, experimentally the second-order
transition temperature will be inaccessible in finite time.
It has been shown that the glass temperature of polymers increases with internal stiffness,
(i.e., double bonds in the main chain and steric hindrances to rotations around simple
389
Handbook of Polymer Blends and Composites
Two empirical rules correlate Tg with polymer characteristics. According to the Simha-
Boyer rules [5], the products between Tg and the increments at Tg of the expansion
coefficients, = (l - gl)Tg are constant, where subscript l is for liquid and gl is for glass
and of the heat capacities cp = (cp,l - cp,gl)Tg, respectively.
The Schneider - DiMarzio rule [6] stipulates the proportionality between Tg and the
ratio of the mass, , versus the number of flexible bonds, , of monomeric unit:
Tg ~ / (10.2)
Concerning the glass temperature of polymer blends it is supposed that miscible polymers
(in the following compatible polymer blends) are monophasic, showing a single,
composition dependent, glass temperature, whereas immiscible polymer systems are,
depending on the number of components, bi- or multi-phasic, showing the more or less
shifted glass temperatures of the blend components and sometimes additionally
characteristic intermediate Tg depending on the thickness of interphase. But as a
consequence of the different sensitivity of the various measurement techniques towards
both the interval between glass transition temperatures and the physical dimensions of
phases, the Tg-criterion is, unfortunately, ambiguous [7].
Taking into account that compatible polymer blends show a single Tg, the following the
discussion will concentrate on the composition dependence of the glass temperature of
compatible polymer blends.
Instead of the use of an universal additivity rule based on weight fractions of components,
Tg,add=w1Tg,1+w2Tg,2, in the literature are recommended several models to predict the
composition dependence of the glass temperature of compatible polymer blends based
on additivity assumptions of certain properties of the blend components. Most of these
models were originally developed to predict the glass temperature of random copolymers
and then subsequently extended without further modifications to compatible polymer
390
Glass Transition in Polymer Blends
blends. Thus, for instance, Gordon and Taylor [8] supposed the additivity and continuity
of volumes at glass temperature, meaning in fact really the additivity of free volumes.
The result of this assumption is the well-known Gordon-Taylor equation:
with wi and Tg,i the weight fractions and the glass temperatures, respectively, of the
blend components, the index i = 1 referring to the blend component with the lower Tg,
whereas the index i = 2 denotes the component with the higher Tg,2. K is a model-specific
parameter. For free volume additivity KG-T=(12/21). i are the densities (as a
result of substitution in the final expression (10.3) of volume fractions by weight fractions)
and i the increments of the expansion coefficients at Tg. The same Gordon-Taylor
equations results accordingly to DiMarzio [9] for additivity of flexible bonds of the
blend components, but with another meaning of the model specific parameter K:
KDM = [(1/1)/(2/2)]
391
Handbook of Polymer Blends and Composites
392
Glass Transition in Polymer Blends
glass temperatures of real compatible polymer blends show more or less pronounced
deviations from predicted Tg additivity. Experimentally, both positive and negative
deviations of the glass temperature are observed and Figure 10.1 shows some typical
experimental data for the following blend pairs:
As can be seen from the experimental Tg-data shown in Figure 10.1, compatible polymer
blends exhibit either positive or negative deviations of the Tg from values predicted by
the additivity rules. The additive Tg behaviour is rather the exception.
393
Handbook of Polymer Blends and Composites
Taking into account the thermodynamic requirement of strong interactions between the
blend components the Gordon-Taylor equation (10.3) was extended by Kwei [18] with
an additional square term, qw1w2, comprising the interaction dependent parameter, q.
Second power equations for the composition dependence of the blend Tg were also
obtained by DiMarzio [9] assuming beside flexible bond additivity, effects of the different
specific volumes of the blend components and by Kanig [19], considering the contributions
to the respective Gibbs energy of specific interaction energies necessary for generating
one mole of holes in the polymer melt.
Assuming that interactions between the blend components lead to binary hetero-contact
formation which may induce conformational rearrangements in the neighbourhood of
the contacts, Brekner and others [20, 21] suggested that the glass temperature of
compatible polymer blends is controlled above all by the probability of hetero-contact
formation which determines both the free volume distribution and the related
conformational mobility in the polymer blend. Conformational rearrangements in the
immediate neighbourhood of the contacts are accompanied by conformational entropy
changes, which contribute additionally to the Gibbs energy of mixing. The idea of possible
conformational entropy contributions to polymer miscibility is also suggested by the
experimentally observed temperature dependence of the Flory-Huggins interaction
parameter suggesting its free energy character, i.e., comprising beside the initially assumed
enthalpic part, an entropic part too [22].
The reflections of Brekner and co-workers [20, 21] upon the composition dependence of
the blend Tg results in a third power virial like extension of the concentration
equation 10.3 based on additivity. The second power concentration term k1w1cw2c,
characterises the energetic interactions, whereas the third power concentration terms are
caused by conformational redistribution proved by the energetic interaction:
Tg,blend = w1cTg,1 + w2cTg,2 + k1w1cw2c + k2w1c2w2c + k3w1cw2c2 (10.8)
Gordon-Taylor equ. interaction conformational redistribution
additivity energetic effects conformational entropy effects
394
Glass Transition in Polymer Blends
w1c and w2c are the corrected weight fractions to account for the different expansivity of
the blend components, i.e., w2c = Kw2/(w1+Kw2) and w1c = (1-w2c) with K ~ (2/1) =
Tg,1/Tg2. The initial parameters k1 = [(E12-1-E11-1)+(E12-2-E22-2)] characterises the energetic
effects of the substitution of homo-contacts by hetero-contacts and depends on the
difference between the interaction energy of hetero-contacts and cohesive forces of homo-
contacts. The virial parameters k2 and k3 comprise the energetic effects of the
conformational mixing entropy due to conformational changes induced by hetero-contact
formation in compatible polymer blends, i.e., k2 = [(e12-1-e11-1)-(e12-2-e11-2)] respective
k3= [(e12-2-e22-2)-(e12-1-e22-1)].
Using corrected weight fractions, wic, the Gordon-Taylor equation (10.3) may be
formulated as follows:
Using the simplified Gordon-Taylor expression (10.9), the non-additivity effects of the
glass temperature of compatible polymer blends may be expressed as:
Substituting subsequently w1c = (1-w2c) and defining (k1+k2) = K1 and (k2-k3) = K2 the
non-additivity Tg-effects of compatible polymer blends may be expressed by the following
third power virial-like regression:
which may be used to fit experimental Tg versus composition data of compatible polymer
blends. K1 and K2 are the fitting parameters of the 3rd order regression (10.10).
K1 comprises the effects of the enthalpic hetero-interaction expressed by k1, but due to
the assymmetry of the 3rd order regression (10.10), it includes additionally via k2 shares
of the conformational mixing entropy. The difference between the two experimental
estimated fitting parameters, (K1-K2)=(k1+k3), comprises beside the same effects of the
enthalpic hetero-interaction via k3 the complementary shares of the conformational
entropy contributions. Unfortunately, the two effects are not separable. For further details
concerning the physical meaning of these parameters see [20, 23, 24].
395
Handbook of Polymer Blends and Composites
The kinetic character of the glass transition is supported by the experimentally observed
dependence of Tg on thermal and processing history, ageing and annealing, as well on
heating and cooling rate, respectively. Due to the kinetic effect, any difference between
heating and cooling rates leads to hysteresis phenomena, causing additionally as the case
may by endotherms respective exotherms in the corresponding differential scanning
calorimetry (DSC) traces. For specifying the glass transition temperature it is recommended
to indicate beside the temperature of half-vitrification the used heating/cooling rates. The
width of the glass transition range depends on the accompanying relaxation phenomena
and some authors claim consequently the specification in addition of the corresponding
temperatures of beginning, Tg,i, and end, Tg,f, of the glass transition range. It has been
observed that in compatible polymer blends showing negative deviations from additivity
of the glass temperature the enlargement of the Tg range, (Tg,f - Tg,i), may be important
[25]. This is illustrated in Figure 10.2 by the DSC traces for PVME/PS blends in comparison
with a polydonor/polyacceptor acrylate, i.e., poly([N-2-hydroxyethyl)carbazole]-) (PDNBA)/
i.e., poly([-hydroxyethyl)-3,5-dinitrobenzoate]acrylate) (PHECA), blend.
Figure 10.2 A. Shape of DSC-curves of miscible polymer blends in the glass transition
range; B. composition dependence of the width of the glass transition range
396
Glass Transition in Polymer Blends
The heating rate dependence of the glass temperature is exhibited in Figure 10.3 for
poly(vinylchloride)/poly(methylmethacrylate) (PVC/PMMA) blends, using data of Karlou
and others [26]. Taking into account that not only the glass temperatures are shifted to
higher temperatures with increasing heating rate but also the deviations from additivity
are affected, it is recommendable to use for comparison Tg data measured at identical
heating rate, assuring simultaneously the same experimental definition for Tg and identical
history of the samples. The latter may be realised by heating first the polymer glass into
the equilibrium melt state followed by the subsequent cooling at controlled cooling rate.
The Tg will than be read from the second heating run. This complex heating/cooling rate
and history dependences explains the large scatter of glass temperature data reported in
literature for the same polymeric system and complicates comparative studies.
To support the above affirmation, based on literature data are presented in Figure 10.4
the glass temperatures of the high molecular compatible polymer blends PVME/PS
(Figure 10.4a) and PS/poly(-methylstyrene) (PaMS), (Figure 10.4b), respectively.
Characteristics of the blend components are listed in Table 10.1.
Figure 10.3 Heating rate dependence of the Tg versus composition curves of PVC/PMMA
blends. A. Tg versus composition curves; B. Tg versus composition curves
397
Handbook of Polymer Blends and Composites
Figure 10.4 Glass temperatures of miscible polymer blends (from literature data).
A. PVME PS blends; B. PS/P(-MS) blends
Some regularities may be emphasised in spite of the large scatter of the Tg data. Thus, for
instance the PVME/PS blend Tg data support a remarkable influence of the heating rate
(compare, for instance, the Tg data of the blends L97.5/73.5 and SB73/75). On the other
hand, the exhibited Tg data of the PS/PaMS blends suggest a possible influence of the
mode of preparation of the samples used for DSC measurements. This is supported by
the data of Schneider and Dilger [27] presented in Figures 10.5a and 10.5b for the blend:
PS (Mw = 12,000)/PaMS (Mw = 25,000). The glass temperatures of the blend were
determined by both thermal dilatometry and by DSC. The polymer blend samples were
prepared by precipitation from common solvent followed by freeze drying. For DSC
these freeze dried samples were used at first without further treatment. For dilatometry,
films of the freeze dried samples were prepared by press-moulding using a pressure of
5 MPa. These press moulded blend samples were investigated in addition in a second
DSC run. The data presented in Figure 10.5 confirm the overwhelming influence of the
pressure used during sample preparation. It may thus be assumed that pressure acting
398
Glass Transition in Polymer Blends
during sample preparation squeezes out shares of frozen, non-equilibrium free volume,
enhancing thereby the probability of hetero-contact formation. The effect will be the
larger the higher the pressure and the less intense the energetic hetero-interaction. The
resulting conclusion is that the glass temperature of polymeric systems may be influenced
seriously by the processing conditions. The higher the processing pressure the more
399
Handbook of Polymer Blends and Composites
Figure 10.5 Pressure influence on the glass temperature of PS/P(-MS) blends shown
by DSC Tg - Data extrapolated for zero heating rate. a) Tg versus composition curves;
b) Tg versus composition curves
important the squeezing out of free volume, i.e., the higher is the experimentally observed
glass temperature.
Taking into account the kinetic governed relaxation phenomena observed during passing
the glass transition range, Rehage [36] recommended to characterise frozen polymer
glasses by an additional internal ordering parameter. It will be, however, hard to quantify
this ordering parameter, taking into account the thermodynamic metastability of
polymer glasses.
400
Glass Transition in Polymer Blends
Figure 10.6 Pressure influence on pV-T curves of the 75 w/w PVC / 25 w/w PMMA
blend. A. shape of the pV-T curves through the glass transition range; B. composition
and pressure dependence of the width of the Relaxation Zone during the glass
transition
It is interesting to notice that the observed enlargement of the relaxation zone is not
only composition dependent but in the same time much more accentuated in the PVC
rich blends at higher pressures. The asymmetric enlargement of the relaxation zone is
reflected in the composition dependence of the respective glass temperature of PVC/PMMA
blends as illustrated in Figure 10.7. S-shaped Tg versus composition curves seem to be
the result of this asymmetric pressure-composition relaxation effect. Based on the similarity
of the relaxation zone versus composition and related Tg versus composition curves
(compare Figures 10.6b and 10.7b) it may be assumed that conformational rearrangements
are allowed in the non-equilibrium frozen polymer glass during passing the glass
transition range. This is confirmed by the respective fitting parameters of the 3rd order
Tg versus composition regression (10.10) as illustrated in Figure 10.8. The result is that
the pressure influence is the higher on the K2 parameter reflecting conformational entropy
effects. Asymmetry effects, though less accentuated, are also shown by the heating rate
dependent Tg versus composition curves of the PVC/PMMA blends (see Figure 10.3).
401
Handbook of Polymer Blends and Composites
Figure 10.7 Pressure influence on the shape of isobaric pV-T curves of PVC/PMMA
blends. A. Tg versus composition curves; B. Tg versus composition curves
It has been shown that the ultimate condition of polymer miscibility is the existence of
predominant hetero-interactions to overcome the intramolecular cohesive forces acting
within polymers. Hetero-interactions between polymers supposes, however, the
accessibility of the interacting groups. It may thus be assumed that the stronger the
energetic interaction and the more symmetric the arrangement of the interacting groups,
the higher the probability of hetero-contact formation, accompanied by a corresponding
reduction of the free volume within the polymer blend and increase above additivity of
the glass temperature. A model of two limiting possibilities concerning the probability of
interaction in miscible polymer blends is presented in Scheme 10.1.
402
Glass Transition in Polymer Blends
Scheme 10.1
Assuming the strong hetero-interactions acting between the phenylsulfone groups of PES
and the phenylketone groups of the poly(phenylether-phenylketone)s, PEK and PEEK,
respectively, are decisive for polymer miscibility, it is evident that the symmetry of
arrangement of the interacting groups is the higher in the poly(phenylether-phenylsulfone)/
PEEK pair.
403
Handbook of Polymer Blends and Composites
Figure 10.8 Heating rate and pressure dependence of the K-parameters of the 3rd
Order Tg versus composition equation applied to PVC/PMMA blends
polymer backbones has been confirmed by both mechanical-dynamic studies [41] and
by DSC [13, 42]. Thus, for instance, it has been shown by mechanical-dynamic analysis
that the attachment of about 15 mol% of donor-acceptor interacting groups assures the
compatibilisation of donor-poly(dimethylsiloxane) with acceptor cellulose [43].
The interaction between donor and acceptor groups attached to polymer backbones
depends, however, on the availability of the groups for hetero-contact formation and it
404
Glass Transition in Polymer Blends
has been shown that the probability of CT-I depends on the spacer length between polymer
backbone and interacting groups. Figures 10.10a and 10.10b, show, respectively, the Tg
and Tg versus composition dependences of the polyacceptor/polydonor blends: poly([-
(hydroxyalkyl)-3,5-dinitrobenzoate]-methacrylate) (PDNBM-n) and poly([N-2-(hydroxy-
ethyl)-carbazolyl]-methacrylate (PHECM-2) [44]. From the data presented in Figure 10.10
it appears that the availability of the acceptor interacting groups for hetero-contact
formation increases with the spacer length between interacting group and polymer
backbone. The accompanying increase of the mobility of the acceptor side-group increases
the accessibility for interaction with the donor group of the second blend component.
The increased probability of hetero-contact formation is accompanied by a reduction of
the free volume in the blend in spite of the increased length of the spacer between
interacting group and polymer backbone. For the different polydonor/polyacceptor blends,
however, has to be find out experimentally in each case the most favourable spacer lengths
to assure the highest probability of hetero-contact formation between the interacting
groups.
405
Handbook of Polymer Blends and Composites
Kwei and others reported the miscibility enhancement of polymers via hydrogen bonding
[29, 45, 46, 47]. It has been shown, that the probability of hydrogen bonding promoting
polymer miscibility depends on both the common solvent used for blend preparation
and the annealing time of the solution casted blend films [48, 49]. According to Pehlert
and others [50] steric shielding effects play an essential role for the group accessibility in
favour of hetero-contact formation in hydrogen-bonded polymer blends.
Zhou, Jiang and others [51, 52] have shown that intermolecular complexation due to
specific interactions depends on the concentration of interacting groups available for
hydrogen bonding. In Figures 10.11a and 10.11b the respective Tg and Tg data concerning
the miscibility of the proton donating poly[styrene-co-(p-(2-hydroxypropane-2-yl)styrene]
(PS(t-OH)) with the proton acceptor poly(n-butylmethacrylate-co-4-vinylpiridine) (BVPy)
[49, 53, 54] are illustrated. From the data shown in Figure 10.11 the stronger probability
of hydrogen bonding is reflected in increased positive deviation from additivity of the
blend Tg with increasing content of proton acceptor groups of BVPy. It may thus be
406
Glass Transition in Polymer Blends
assumed that compatible polymer blends can be tailor-made by appropriate choice and
concentration control of the strongly interacting groups, for example electron-donor/-
acceptor of proton-donor/-acceptor groups, respectively.
Taking into account the high mobility of PVME, the probability of hetero-contact
formation will depend above all on the accessibility of the interacting phenyl-group of
407
Handbook of Polymer Blends and Composites
Figure 10.12 Dependence on molecular weight of the shape of Tg-curves and the
deviation from additivity of the glass temperature of polymers blends bearing weak
interacting groups. A. Tg versus composition curves; B. Tg versus composition curves
The influence of the Mw of the blend components is also supported by the Tg-data of the
blends of tetramethyl bis-phenol-A-polycarbonate (PC) with polystyrene [57, 58] and
oligo poly(-methylstyrene) [57], respectively. The small structural changes, produced
by the substitution of H in polystyrene with CH3 in poly(-methylstyrene) reduces
substantially the compatibility with PC, the high molecular poly(-methylstyrene) being
incompatible with PS. Nevertheless, the Tg of the oligo(-methylstyrene)/PC blends shows
because of the reduced Mw of PMS rather positive deviations from additivity, suggesting
an enhanced accessibility of the interacting phenyl-groups, whereas the Tg of the high
molecular PS/PC blends exhibit negative deviations from additivity (see Figure 10.13).
408
Glass Transition in Polymer Blends
Figure 10.13 Influence of structure changes within the monomeric unit on the
miscibility of polymers exemplified by blends of PC with PS and P(a-MS). A. Tg versus
composition curves; B. Tg versus composition curves
Fernandes and others [57] have also shown that PC is miscible with copolymers of styrene
with acrylonitrile, having an acetonitrile (AN) content smaller than 13%, with maleic
anhydride content of 18% and ally alcohol content less than 19.1%. Generally, polymer
miscibility can be improved by using copolymers, the interacting comonomer content
being, however, limited for each copolymer/polymer or copolymer/copolymer pair to a
well defined composition range. Such polymer blends are characterised by miscibility
windows which are delimited in phase diagrams by binodals. Phase diagrams of PMMA/
P(S-co-AN) blends, for instance, are exhibited in a paper of Suess and others [59].
Barlow, Paul and others have published a series papers concerning the Tg-behaviour of
blends of polyesters with tetramethyl bisphenol-A polycarbonate [60], polyhydroxyether
of bisphenol-A [61], poly(epichlorohydrine) [62] and PVC [63, 64]. To illustrated the
influence of the structure of the repeating unit on Tg versus composition behaviour of
polymer blends in Figure 10.14 are exhibited the respective Tg (Figure 10.14a) and Tg
curves (Figure 10.14b) of blends of PVC with the polyesters poly(1,4-butylene sebacate)
409
Handbook of Polymer Blends and Composites
Figure 10.14 Influence of the chemical structure of the polyesteric blend component on
the shape and deviation from additivity of the glass temperature of PVC blends. A. Tg
versus composition curves; B. Tg versus composition curves
410
Polymer blends characteristics Solubility parameters
411
Handbook of Polymer Blends and Composites
Kwei and others [18], on the other hand, have demonstrated in a series of papers, that
the polymer compatibility depends not only on the nature of the repeating unit, but also
considerably on the tacticity of the polymer used and for illustration are shown in
Figure 10.15 the Tg and Tg versus composition curves of the blends of Novolac B,
prepared from a mixture of 2-tert-butyl phenol (15%), m-cresol (17%) and o-cresol
(68%) with PMMA of different tacticity. It is interesting to notice that iso- and syndio-
PMMA show a quite different miscibility behaviour, expressed by distinct Tg versus
composition dependences. The Tg data suggest the higher accessibility for hetero-
interaction with Novolac B of the iso-tactic disposed PMMA-units. It is interesting to
notice that the situation is reversed in blends of stereo-regular PMMAs with poly(styrene-
co-p-(hexaflouro-2-hydroxy-2-propyl)styrene) [68]. It seems thus that in hydrogen bonded
polymer blends the nature of the second blend component influences in addition the
intensity of the hetero-interaction with stereo-regular polymers. This assumption is
supported by a paper of Jong and others [69] concerning the miscibility of stereo-regular
Figure 10.15 Influence of the tacticity of the blending component on the miscibility of
polymer blends. A. Tg versus composition curves; B. Tg versus composition curves
412
Glass Transition in Polymer Blends
An survey of the glass temperatures of polyolefin blends has been included in the
Handbook of Polyolefins [72]. Taking into account the overwhelming number of
published papers on the glass transition of miscible polymer systems the reader is directed
to the comprehensive review paper on miscible polymer blends, including copolymer
systems, published by Krause in the Polymer Handbook [73].
Although in the above section were discussed only a small number of typical examples
concerning the composition dependent Tg behaviour of miscible polymer blends some
general rules may be derived from the exhibited Tg and above all Tg curves. The main
conclusion which results from the reproduced Tg data point to the fact that the glass
temperature of miscible polymer blends depends not only on the thermodynamically
imposed condition that the hetero-interaction energy has to overcome the intramolecular
cohesive forces, but also to a high degree on the conformational mixing entropy produced
during hetero-contact formation. The stronger the hetero-interaction energy, i.e., hydrogen
bonding, electron donor-electron acceptor, as well as ion-ion and ion-dipole interaction
[74], the higher the probability of denser packing of the polymer blend induced by hetero-
contact formation which is accompanied by a corresponding increase of the glass
temperature above values predicted by additivity rules. On the other hand the probability
of hetero-contact formation, and the related packing density, depend not only on the
intensity of the hetero-interaction energy but also on the mobility of the interacting
polymers (which favours the decoiling of polymers and thus the accessibility of the
interacting groups), on symmetry factors concerning the spatial and energetic arrangement
of the interacting groups, on sterical shielding effects and additionally in copolymers on
the content of the interacting group bearing component. The smaller the interaction
energy, the more reduced the symmetry and the higher the sterical shielding effects, the
higher will be the probability of induced conformational rearrangements in the
neighbourhood of the binary hetero-contacts contributing to a proportionate increase of
the conformational mixing entropy. The result will be an increase of the free volume of
413
Handbook of Polymer Blends and Composites
the blends, expressed by a decrease of the glass temperature below values predicted by
additivity rules. It may thus be supposed that Tg-additivity in polymer blends is the result
of compensation of the energetic interaction effects and induced conformational entropy
contributions to polymer miscibility.
Taking into account that in a first approximation the mixing enthalpy of polymers is
connected to the reciprocal of the solubility of the blend components it has been suggested
that the K1-parameter of the 3rd order concentration [Equation (10.8)] which depends
to a high degree on the hetero-interaction energy, may be correlated with the mutual
solubility, i.e. with the difference between the solubility parameters of the blend
components [23, 24, 75]. The values of the K1-parameters of the 3rd order Tg versus
composition equation of the polymer blends discussed in this section are listed in
Table 10.2 together with the differences of the solubility parameters of the monomeric
units, m = (m,2-m,1) related to the differences of the blend component solubility
parameters, (2-1), the index 2 referring to the component with the higher Tg,2.
Accordingly to the group contribution method [76] the solubility parameters may be
estimated via the molecular attraction constants, F, of Small [77]:
i = ijFj / Mi (10.11)
where i is the polymer density and Mi the polymer molecular weight. The molecular
attraction constant is defined as:
j = (Ei, Vi,j)1/2
Ei,j being the energy of vapourisation contribution of the groups j belonging to the
monomeric unit, i, and Vij the respective molar volumes. The molecular attractions are
tabulated in [77].
Neglecting in a first approximation the influence of the mostly very similar densities of
polymers, Equation (10.11) was used to evaluate a solubility parameter of the monomeric
unit of the blend components, =Fj/i. Although the K1 parameter of Equation (10.8)
comprises not only energetic interaction contributions but also shares of conformational
entropy effects, the correlation with = (,2-,1), i.e., the difference between the
solubility parameters of the monomeric units of the blend components, is unexpectedly
good as results form the data presented in Figure 10.16.
It may thus be assumed that positive values of the difference of the solubility parameters
of the monomeric units will sooner by accompanied by positive deviations of the glass
temperature from Tg values predicted by additivity and vice versa. The scatter of the
data is actually larger as it is shown by the respective data of 49 polymer blends analysed
414
Glass Transition in Polymer Blends
Figure 10.16 Correlation between the K1-parameter of the 3rd order Tg versus
composition curves of miscible polymer blends with the difference of solubility
parameters of the monomeric units of the blend components
recently [74]. Nevertheless, the tendency depicted in Figure 10.16 remains valid, but it
has to be pointed out that the above attempt to predict glass temperatures of polymer
blends fails in describing molecular weight as well as pressure and tacticity influences
and is not applicable for blends comprising copolymers.
415
Handbook of Polymer Blends and Composites
References
3. R. Simha and R.F. Boyer, Journal of Chemical Physics, 1962, 37, 1003.
5. J.H. Gibbs and E.A. DiMarzio, Journal of Chemical Physics, 1958, 28, 373, 807.
6. H.A. Schneider and E.A. DiMarzio, Polymer, 1992, 33, 16, 3453.
8. M. Gordon and J.S. Taylor, Journal of Applied Chemistry, USSR, 1952, 2, 493.
10. P.R. Couchman and F.E. Karasz, Macromolecules, 1978, 11, 1, 117.
12. T.G. Fox, Bulletin of the American Physical Society, 1965, 1, 123.
14. M.A. de Araujo, R. Stadler and H-J. Cantow, Polymer, 1988, 29, 12, 2235.
15. H.A. Schneider and B. Leikauf, Thermochimica Acta, 1987, 114, 165.
16. J.P. Flory, Principles of Polymer Chemistry, Cornell University Press, Ithaca, NY,
USA, 1953.
18. T.K. Kwei, Journal of Polymer Science: Polymer Letters, 1984, 22, 6, 307.
20. M-J. Brekner, H.A. Schneider and H-J. Cantow, Polymer, 1988, 29, 1, 78.
416
Glass Transition in Polymer Blends
21. M-J. Brekner, H.A. Schneider and H-J. Cantow, Die Makromoleculare Chemie,
1988, 189, 9,2085.
22. D.R. Paul in Polymer Blends and Mixtures, Ed., D.J. Walsh, J.S. Higgins and A.
Maconnachie, Martinus Nijhoff Publishers, Dordrecht, The Netherlands, 1985,
Chapter 1, 1.
23. H.A. Schneider, Journal of Research of the National Institute of Standards and
Technology, 1997, 102, 229.
26. K. Karlou and H.A. Schneider, Journal of Thermal Analysis and Calorimetry,
2000, 59, 1-2, 59.
27. H.A. Schneider and P. Dilger, Polymer Bulletin, 1989, 21, 3, 265.
29. T.K. Kwei, T. Nishi and R.F. Roberts, Macromolecules, 1974, 7, 5, 667.
30. J-L. Halary, F.B.C. Larbi, P. Oudin and L. Monnerie, Die Makromolekulare
Chemie, 1988, 189, 9, 2117.
31. H.A. Schneider and M-J. Brekner, Polymer Bulletin, 1985, 14, 173.
32. S-F. Lau, J. Pathak and B. Wunderlich, Macromolecules, 1982, 15, 5, 1278.
33. S. Saeki, J.M.G. Cowie and I.J. McEwen, Polymer, 1983, 24, 1, 60.
34. J-L. Lin and R-J. Roe, Polymer, 1988, 29, 7, 1227.
35. P. Zoller and H.H. Hoehn, Journal of Polymer Science, Polymer Physics, 1982,
20, 8, 1385.
38. H.A. Schneider, F. Glatz and R. Mlhaupt, New Polymeric Materials, 1995, 4, 4, 289.
417
Handbook of Polymer Blends and Composites
39. V. Percec, A. Natanson and C.I. Simionescu, Polymer Bulletin, 1982, 5, 217,
40. V. Percec, A. Natanson and C.I. Simionescu, Polymer Bulletin, 1982, 5, 225.
41. H.A. Schneider, H-J. Cantow and V. Percec, Polymer Bulletin, 1982, 6, 11-12, 617.
42. U. Epple and H.A. Schneider, Thermochimica Acta, 1990, 160, 103.
43. H.A. Schneider, U. Epple, B. Leikauf and H. N. Neto, New Polymeric Materials,
1992, 3, 2, 115.
44. M. Bolsinger and H.A. Schneider, Makromolekulare Chemie, 1994, 195, 7, 2683.
45. S.P. Ting, E.M. Pearce and T.K. Kwei, Journal of Polymer Science: Polymer
Letters, 1980, 18, 3, 201.
46. S.P. Ting, B.J. Bulkin, E.M. Pearce and T.K. Kwei, Journal of Polymer Scienc:
Polymer Chemistry, 1981, 19, 6, 1451.
47. E.M. Pearce, T.K. Kwei and B. Min, Journal of Macromolecular Science A, 1984,
21, 8-9, 1181.
48. T.P. Yang, E.M. Pearce, T.K. Kwei and N.L. Yang, Macromolecules, 1989, 22, 4,
1813.
49. L.F. Wang, E.M. Pearce and T.K. Kwei, Journal of Polymer Science: Polymer
Physics, 1991, 29, 5, 619.
50. G.J. Pehlert, P.C. Painter and M.M. Coleman, Macromolecules, 1998, 31, 13,
8423.
52. M. Jiang, M. Li, M. Xiang and H. Zhou, Advances in Polymer Science, 1999,
146, 121.
55. J.D. Ferry, Viscoelastic Properties of Polymers, 3rd Edition, J. Wiley & Sons Inc.,
New York, NY, USA, 1980, 374.
418
Glass Transition in Polymer Blends
56. W.W. Graessley in Physical Properties of Polymers, Eds., J.E. Mark, A. Eisenberg,
W.W. Graessley, L. Mandelkern, E.T. Samulski, J.L. Koenig and G.D. Wignall,
ACS, Washington, DC, USA, 1984, Chapter 3, 97.
57. A.C. Fernandes, J.W. Barlow and D.R. Paul, Polymer, 1986, 27, 1778.
59. M. Suess, J. Kressler and H.W. Kammer, Polymer, 1987, 28, 957.
60. A.C. Fernandes, J.W. Barlow and D.R. Paul, Polymer, 1986, 27, 11, 1799.
61. J.E. Harris, S.H. Goh, D.R. Paul and J.W. Barlow, Journal of Applied Polymer
Science, 1982, 27, 3, 839.
62. A.C. Fernandes, J.W. Barlow and D.R. Paul, Journal of Applied Polymer Science,
1984, 29, 6, 1971.
63. E.M. Woo, J.W. Barlow and D.R. Paul, Polymer, 1985, 26, 5, 763.
64. J.J. Ziska, J.W. Barlow and D.R. Paul, Polymer, 1981, 22, 7, 918.
65. B. Wunderlich, Thermal Analysis, Academic Press Inc., Boston, MA, USA, 1990.
68. D. Luo, E.M. Pearce and T.K. Kwei, Macromolecules, 1993, 26, 23, 6220.
69. L. Jong, E.M. Pearce and T.K. Kwei, Polymer, 1993, 34, 1, 48.
70. H-J. Cantow and O. Schulz, Polymer Bulletin, 1986, 15, 5, 449.
71. H-J. Cantow and O. Schulz, Polymer Bulletin, 1986, 15, 6, 539.
72. H.A. Schneider in Handbook of Polyolefins, 2nd Edition, Ed., Cornelia Vasile,
Marcel Dekker, New York, NY, USA, 2000, Chapter 6, 207.
73. S. Krause in Polymer Handbook, 3rd Edition, Eds., J. Brandrup and E.H.
Immergut, John Wiley & Sons, New York, NY, USA, 1989, Chapter 6, 371.
75. H.A. Schneider, Journal of Thermal Analysis and Calorimetry, 1999, 56, 3, 983.
419
Handbook of Polymer Blends and Composites
76. D.W. van Krevelen and P.J. Hoftyzer, Properties of Polymers: Correlations with
Chemical Structure, Elsevier, New York, NY, USA, 1972, Chapters 6 and 7.
77. E.A. Grulke in Polymer Handbook, 3rd Edition, Eds., J.E. Brandrup and E.H.
Immergut, John Wiley & Sons, New York, NY, USA, 1989, Chapter 7, 519.
420
Crystallisation, Morphology and Melting in Polymer Blends
11
Crystallisation, Morphology and Melting
in Polymer Blends
11.1 Introduction
The driving force for the development and use of polymer blends is generally cost reduction
and the synergistic property profiles achievable. When is it convenient to combine a
crystalline polymer with an amorphous polymer? The most used crystalline polymers,
such as polyolefins, polyamides and polyacetals, generally present excellent chemical
resistance and good heat distortion and mechanical properties and moreover exhibit low
viscosity above their melting point. The most used amorphous polymers in blends; such
as acrylonitrile-butadiene-styrene (ABS), polycarbonate (PC), polysulfone, and polyarylate
generally have good dimensional stability, warpage resistance and impact strength. So
when combining a crystalline polymer with an amorphous polymer, the resulting blend
provides good dimensional stability, ease of processing, chemical resistance as well as
mechanical properties that can be tailored for a specific application.
To show the advantages provided by the use of polymer blends a few examples are
reported.
Bioriented films of isotactic polypropylene (iPP) blended with natural and synthetic resins
such as polyterpenes and hydrogenated oligocylopentadiene were recently introduced in
the packing industry. The blend films possess a reduced permeability to oxygen and
aromas and a higher thermal stability and can be processed at lower temperature compared
to plain iPP films [12-14].
421
Handbook of Polymer Blends and Composites
greatly improved over that of acrylates. The addition of PVDF results in materials that
are more resistant to environmental stress cracking or crazing. They offer combinations
of transparency, toughness, weatherability and are being used in the locomotive area for
decorative stripping and labelling, where clarity, chemical resistance and adhesion to
acrylic paint are required.
The blends containing a crystallisable polymers are currently attracting much research.
For these blends two kinds of phase separation can be observed: liquid-liquid separation
and liquid-solid demixing due to crystallisation of the crystallisable polymer. The
experimental phase diagram may therefore present overlaps between the two demixing
curves. These two curves offer great flexibility in controlling the morphology and hence
the properties of the material.
In this chapter the up-to-date results on the most important factors that control the
isothermal and non-isothermal crystallisation process, the morphology, and the melting
behaviour of binary crystallisable blends constituted by a crystallisable polymer and an
amorphous one are discussed.
422
Crystallisation, Morphology and Melting in Polymer Blends
polymer blends, the investigation of the miscibility is focused only on the amorphous
phase of the material. Below the melting temperature, a separate crystalline phase can
form and segregate and the miscibility degree concerns only the remaining amorphous
phase. Only above the melting point of the crystallisable component a complete
homogeneous mixtures can be present.
The state of miscibility is governed by the Gibbs free energy of mixing. The thermodynamic
criteria for spontaneous miscibility is that the Gibbs free energy of mixing is negative
[19, 20]:
Gm = Hm - TSm < 0
2Gm
and 2 >0
2
The determination of the miscibility range is very important and can be studied using
experimental and theoretical approaches. The experimental methods have been extensively
reviewed by MacKnight and co-workers [21] and Olabisi [19]. The most common
experimental method used for establishing miscibility in polymer-polymer blends is the
determination of the glass transition temperature (Tg). A miscible polymer blend will
exhibit a single Tg, composition dependent. For a phase-separated blend, two or more
glass transitions may be present. The use of Tg in the determination of polymer/polymer
miscibility is based on the fact that a single Tg indicates that the size of the domains is
below 15 nm. A review on the relation between Tg and the structure in blends is given by
Cowie [22]. There are several equations relating the glass transition temperature to
composition Tg of the blends, e.g., those derived by Couchman [23], Utracki-Jukes [24],
Gordon-Taylor [25] revised by Wood [26], Kwei [27] and Fox [28]. These equations,
with different approximations, predict a monotone dependence of Tg upon composition.
The most used expression of Tg-composition dependence of miscible blends is the Fox
equation [28]. For crystallisable polymer blends, the use of any expression relating Tg to
composition, must be done taking into account the possibility of crystallisation. In the
event of crystallisation, in fact, the composition of the amorphous phase is different
from the nominal one. The real amorphous composition can be calculated knowing the
amount of the crystallised material. A list of crystallisable polymer blends miscible in the
melt according to the presence of one Tg is reported in Table 11.1.
423
Polymer blends crystallisable
-polymethylmethacrylate 0.34 2
-Polyhydroxybutyrate 203 53
-Polymethylmethacrylate 113
Polyepichlorohydrin 0.068 57
Polyethylene oxide1 203 0.096 108
Polyethyleneoxide Polyvinylacetate 65 0.18 193
Polyvinylchlorure .3 194
0.091 194
Polyethylmethacrylate 43
Polyester-co-polyurethane 194
Polybenzyl methacrylate 60 0.1 195
Isotactic polystyrene -Polymethylmethacrylate 70 0.15 40
-Polystyrene 0.003 196
424
Polymer blends crystallisable
giving rise to a very complex phase diagram, (see Figure 11.1). In blends with a miscibility
gap, the morphology can be also influenced by the kinetics of crystallisation and the
demixing processes. In fact, depending on the type of phase diagram and on the relative
position of the crystallisation curve and the miscibility gap a large number of processes
can proceed. In systems with an upper critical solution temperature (UCST), demixing-
induced crystallisation as well as crystallisation induced demixing can occur. In systems
exhibiting a lower critical solution temperature (LCST), the crystallisation can be inhibited
by homogenisation or the homogenisation can be inhibited by the crystallisation process.
Therefore new structures and morphologies can be generated by exploiting the competition
between crystallisation and phase separation.
425
Handbook of Polymer Blends and Composites
Figure 11.1 a) Phase diagram of the PVF2/polyethyl acrylate (PEA) blends: spinodal;
binodal melt crystal equilibrium temperatures [167]. b) Phase diagram of PHB/
aPMMA blends: one phase; two phases () experimental glass transition
temperatures of quenched blends [16]
Figure 1b, Reproduced with permission from A. Siciliano, A. Seves, T. De Marco, S.
Cimmino, E. Martuscelli and C. Silvestre, Macromolecules, 1995, 28, 24, 8065, Figure 6.
Copyright 1995, American Chemical Society
In PHB/aPMMA blends, for example, the crystallisation and the separation process are
in competition with each other. Starting from the melt, where the system is homogeneous,
a liquid/liquid separation occurs before the crystallisation. From the glass state the
crystallisation preserves the homogeneity of the amorphous phases [17, 18].
Beside the method using the Tg determination to prove the miscibility of polymer pairs,
there are other several ways of obtaining an estimate of the interaction between two
polymers. They include measurements of the heat of mixing and viscosity, inverse gas
chromatography, solvent vapour adsorption, light, X-ray and neutron scattering and
NMR [31, 32].
426
Crystallisation, Morphology and Melting in Polymer Blends
Details on the procedure to obtain 12 by using the equilibrium melting point depression
are reported in details in Section 11.6.
427
Handbook of Polymer Blends and Composites
The crystallisation rate of pure crystalline polymer is often analysed by using the Hoffman,
Laurtitzen and Miller (HLM) kinetic theory of crystallisation [35, 36]. According to this
theory, the spherulite radial growth rate is written as:
U Kg
G = Go exp exp
( )
R Tc T ( )
Tc f T
where T is the degree of supercooling (Tmo - Tc), U* is the energy required for the
transport of the crystallisable macromolecules in the melt; R the gas constant, T is the
temperature below which all viscous flow ceases, and f a function that accounts for the
temperature dependence of the heat of fusion (f = 2Tc/(Tmo+Tc). Kg is the nucleation
constant and is defined as:
Kg = nibe Tm / kHf
Where ni is 4, for crystallisation conforming to regime III and I and 2 for crystallisations
conforming to regime II, and e are the interfacial free energies for unit area parallel
and perpendicular, respectively, to the molecular chain axis and b is the distance between
two adjacent fold planes, Hf is the enthalpy of fusion for unit volume of the crystalline
component and Tm it is equilibrium melting point.
Due to the presence of the amorphous component, the crystallisation process in the
blends differs from that of pure crystalline polymer and depends on the miscibility between
the two components in the melt at Tc.
For a miscible blend, the presence of an amorphous component influences the spherulite
growth rate, G, because not only Tg and Tm but also Kg and U* are dependent on blend
composition. The temperature-composition dependence of the growth rate in crystalline/
amorphous polymer has been formulated by the HLM theory taking into account the
melting point depression and the variation of the chain mobility by mixing [37-40]. In
particular Boon and Azcue [37] showed that for miscible blends the expression of the
free energy required for the formation of a critical size nucleus, has to contain an entropy
contribution due to the probability of selecting crystalline sequences. This contribution
given by:
(2kTmTc ln 2 / bHfT)
For a miscible blend the transport term [11] is the sum of two terms U1* and U2*, where
U1* is the energy required for the rejection of the amorphous component from the crystal
428
Crystallisation, Morphology and Melting in Polymer Blends
growth front to the amorphous regions and U2* the energy required for the transport of
the crystallisable molecules from the melt to growing crystals. Both U1* and U2* are
dependent on glass transition. The equation describing the dependence of G on
crystallisation temperature and composition can be written as:
This relationship has been used with success and demonstrated to be valid when the
spherulite growth rate is constant throughout the crystallisation.
On the basis of the theory, and knowing the glass transition temperature of the blend
and of the pure crystallisable components, Tg2, it is possible, with some approximation,
i.e., neglecting the effect of composition on Tm, to predict the influence of an amorphous
diluent on the spherulite growth rate of a crystallisable polymer, G2:
Tg2 < Tgblend Gblend < G2, because both U* and Kg influence G negatively;
Tg2 > Tgblend Gblend > G2 or Gblend < G2 if the favourable effect of the transport term
overcomes or not, the unfavourable effect of Kg, respectively.
This prediction has been demonstrated to be valid for most of the systems with some
exceptions, for examples for polymer blends where the difference in viscosity between
the components changes drastically with the temperature [41].
The HLM theory has been also modified by Saito and co-workers [39] to comprise the
exclusion effect of the amorphous material from the growth front. This theory was
successful applied to system showing non-linear time variation of the spherulite growth.
The following equations were proposed:
Kg
G 1 D M exp -
Tc ( T)f
Kg
G 11 / 2 (DM DS)1/2 exp -
Tc ( T)f
Where DM and DS are the mutual diffusion coefficient for the secondary nucleation process
and the self-diffusion coefficient for the surface spreading process, respectively. The
formulation of DM and DS can be found in the works of Brochard and co-workers [42]
and Skolnick and co-workers [43].
429
Handbook of Polymer Blends and Composites
Figure 11.3 a) Sperulite radius versus time for four blends whose components are
strongly interacting, Tc = 46.5 C. : PEO/ ethylene-methacrylic acid copolymer
(EMAA), : PEO/styrene-hydroxystyrene (SHS) copolymer, X: 70/30 PEO/EMAA,
: 70/30 PEO/SHS [50]
Reprinted with permission from L. Wu, M. Lisowski, S. Talibuddin and J. Runt,
Macromolecules, 1999, 32, 5, 1576, Figure 3. Copyright 1999, American Chemical Society
430
Crystallisation, Morphology and Melting in Polymer Blends
Figure 11.3 b) A map of the spherulite growth mode as a function of molecular weight
of iPS (Mc) and aPS (Ma) for 60/40 iPS/aPS mixtures. : Okada and co-workers,
: Yeh and Lambert, : Keith and Padden. The open symbols represent non-linear
growth and the closed symbols represent linear growth [44]
With permission from T. Okada, H. Saito and T. Inoue, Polymer, 1994, 35, 5705,
Figure 10. Copyright 1994, Elsevier Science
The first studies in the area of crystallisation of polymer blends were by Keith and Padden
on atactic polypropylene/iPP and atactic polystyrene (aPS)/iPS blends [45, 46].
Several other studies followed those studies and in the literature plenty of examples of
the crystallisation behaviour of crystallisable polymer blends can be found.
As most of the crystallisation behaviours of polymer blend are connected with the mutual
miscibility of the component in the melt at Tc, the systems reported are assembled into
two categories: miscible blends and phase separated blends.
With few exceptions, the literature reports have indicated a depression of the growth
kinetics of the crystallisable component, upon addition of a miscible non-crystallisable
material. For several PEO based blends [47-54], PVDF blended with acrylates [55-57],
431
Handbook of Polymer Blends and Composites
In the systems reported above, with the only exception of sPS/aPS blends, in fact, the
Tg(blend) is higher than Tg2. The entity of the depression is dependent on the nature of the
material used, on the composition and on a complex balance among the degree of
miscibility, influence of the non-crystallisable material on the nucleation and on the
crystallisation regime. Very high depression of the spherulite growth rate due to the
432
Crystallisation, Morphology and Melting in Polymer Blends
presence of a polymeric diluent are often observed for low values of Tc, whereas increasing
Tc, the decrease becomes smaller. For the sPS/aPS blends, the variation of Tg with
composition is not discernible because of the proximity of the Tgs of the two components.
Therefore it is evident that the growth rate reduction of sPS/aPS is not related to changes
in transport process. Thus the decrease in spherulite growth rate is due to an increase of
Kg [62]. The role of the intermolecular interactions on spherulite growth rate was studied
in detail by Runt and others [49]. Generally, at a given Tc, growth rates for blends with
strong interacting polymers are considerably lower than those with weakly interacting
polymers with comparable Tg, (see Figure 11.5).
Figure 11.5 Log spherulite growth rate versus Tc for PEO and PEO based blends with
high Tg amorphous polymer: X PEO; PEO/PMMA 90/10=, * 80/20 PEO/PMMA;
70/30 PEO/PMMA, 90/10 PEO/SHS, 80/20 PEO/SHS, 70/30 PEO/SHS
Reproduced with permission from L. Wu, M. Lisowski, S. Talibuddin and J. Runt,
Macromolecules, 1999, 32, 5, 1576, Figure 1a. Copyright 1999, American Chemical Society
433
Handbook of Polymer Blends and Composites
Such a decrease is explained by supposing that the presence of the amorphous component
produces an increase of the folding entropy and leads to the formation of crystals with
more disordered folding surface. Conversely when the melt viscosity decreases with
composition an increase of e is observed as in the case of iPP/poly (-pinene) (PP), see
Table 11.2. The observed increase of e is probably due to a reduction of the entropy of
folding and therefore the formation of more homogeneous and regular folding surfaces.
In the blends, the iPP chains, having in the melt a lower viscosity can fold down more
regularly [41].
Inoue and co-workers [56] and Maiti and Nandi [72] on the other hand believed that the
blending can not cause significant changes in e in absence of regime transitions.
The addition of a miscible polymer to a semicrystalline one can have an important effect
on the regime under which crystallisation occurs. The regime transition takes place due
to the change in the relative magnitude of i versus g, where i is the rate of deposition of
the first stem on the growth front and g is the rate of attachment of the following stems
in the chain on the crystal surface. The regime changes from III to II and then to I as i/g
434
Folding surface free energy
decreases. Saito and co-workers [39] investigated the regime transition for PVDF/aPMMA
blends as function of aPMMA molecular mass and composition, Figure 11.7. They
suggested that the regime transition III>II>I takes places as the composition and/or the
Table 11.2 Free energy of folding surface (e) for iPP/PP blends
iPP/PP(wt %) e (N/m)
100/0 0.18 9
95/5 0.226
90/10 0.23 0
80/20 0.29 1
70/30 0.30 1
Figure 11.7 Log G versus volume fraction of aPMMA: : molecular mass of aPMMA2 =
110 000, M2 = 390 000. Solid and broken lines are calculated by equation 11.2
(regime I and III, respectively), dashed lines by equation 11.3 (regime II) [39]
Reprinted with permission from H. Saito, T. Okada, T. Hamane and T. Inoue,
Macromolecules, 1991, 24, 4446, Figure 4. Copyright 1991, American Chemical Society
435
Handbook of Polymer Blends and Composites
molecular mass of the amorphous material increases. Maiti and Nandi [72] investigated
the regime transition of PVDF/polymethyl acrylate (PMA) as function of PVDF head-to-head
defect and blend composition. They found that the regime transition temperature decreases
with the increase in PMA concentration and at a given composition with the increase of
structural defects of PVDF.
It is reported that the addition of polyvinylacetate (PVA) to PEO causes a change from
regime I to regime II [53] of crystallisation. De Juana and Cortazar [69] studying the
crystallisation of PCL/polyhydroxy ether of bisphenol A blends and Greco and Martuscelli
working on PHB/PVAc blends [63] both also report a study of regime transition in blends.
They found that the crystallisation of pure PCL in the explored temperature range takes
place in regime II, whereas the crystallisation of blends shows a regime II-III transition.
The growth process of plain PHB crystals conforms to regime III, whereas that of crystals
grown from PHB/PVA blends to regime II. In the case of PHB/aPHB blends the regime II
- III transition temperature remains constant with blends, whereas the ratio between the
nucleation constants increases with composition.
For the iPP/PaP blends the temperature of regime II to regime III transition decreases moving
from 138 C for plain iPP to 118 for the 70/30 blends. This decrease was related to the
higher mobility of iPP chain in the blends compared to pure iPP that favours the lateral
spreading of the attached segment on the substrate instead of nucleation [70].
436
Crystallisation, Morphology and Melting in Polymer Blends
consideration. The diluent effect is constant with blend composition as the composition
of the crystallisable component rich phase is independent of nominal composition. The
presence of the other phase as well as the entanglements can reduce the diffusion rate of
the crystallisable chain segments and hence the crystallisation rate.
In the case of sPS/PVME [59] and iPS/PVME [75, 76] and poly(p-phenylene sulfide)
PPS/amorphous polyamide (PA) blends [77], the addition of the amorphous material
causes an increase of the spherulite growth rate. Note that for these blends Tgblend is
lower than Tg2, see Figure 11.9.
437
Handbook of Polymer Blends and Composites
Figure 11.9 Spherulite growth rate of iPS and iPS/PVME blends as a function of
crystallisation temperature for different blend compositions. iPS; *iPS/PVME 90/10;
iPS/PVME 70/30; iPS/PVME 50/50 [75]
Silvestre and co-workers [84-86] reported that in the case of the immiscible system PEO/
polydicyclo exylitaconate (PDCHI), the presence of the non-crystallisable materials causes
438
Crystallisation, Morphology and Melting in Polymer Blends
a decrease in the crystallisation rate, increasing the energy related to the motion of the
crystallisable molecules, towards the growing spherulites in the melt at Tc.
In several other systems with component immiscible in the melt no influence of the non-
crystallisable polymer has been found. The PHB/ethylene-propylene rubber (EPR) system
can be cited as an example [63].
T(t) = Ti - (t)
where Ti is the initial temperature and (t) is a function of time. In the simplest case,
(t) = t, where = dT/dt is the constant cooling rate.
Ziabicki [90], in his first attempt to model dynamic solidification of polymers, proposed
to the non-isothermal process is considered as a sequence of isothermal steps. The proposed
equation is a series expansion of Avrami equation. The drawback of the Ziabicki theory
is that it can be applied only in the range of temperatures where isothermal crystallisation
data are available. Another method was proposed by Nakamura and co-workers [91,
439
Handbook of Polymer Blends and Composites
92] who assumed isokinetic conditions. Ozawa [93] accounted for the effect of cooling
rate, , on dynamic crystallisation, by properly modifying the Avrami equation. This
method can be used when crystallisation occurs at constant cooling rate. The most
interesting feature of the Ozawa method is the possibility of comparing the results of
crystallisation during continuous cooling with those obtained by means of the Avrami
equation in isothermal conditions. As this treatment requires values of relative crystallinity
at a given temperature for different cooling rates, the temperature range over which the
analysis can be applied can be very narrow. Moreover, the assumption of constant cooling
rates may cause problems in modelling the development of crystallisation in polymer
processing. Recently, Ziabicki presented a new model for the analysis of dynamic
solidification of polymers [94, 95]. Compared to earlier models, this treatment includes
transient thermal and athermal effects.
440
Crystallisation, Morphology and Melting in Polymer Blends
increase of crystallisation rate of the components in the blends was explained in terms of
mutual influence exhibited by the single components with respect to each other. The
inner stresses in UHMWPE aggregates, due to the physical network entangled tie
molecules, lead to an increase in HDPE crystallisation rate in the blends. Partial co-
crystallisation of HDPE molecules with a part of UHMWPE was probably the reason for
the increase of UHMWPE crystallisation rate.
Minkova and co-workers [102] investigated the effect of a liquid crystal polymer (LCP),
namely the sebacic acid, 4,4-dihydroxybiphenyl 4-hydroxybenzoic acid (SBH), on the
crystallisation process of polyethylene in polyethylene - LCP graft copolymers, prepared
by reactive blending and by polycondensation. The crystallisation temperature of the PE
component of the copolymer decreased upon increasing the concentration of SBH. It was
found that the former crystallised at slightly higher temperatures than the latter. This was
attributed to the fact that the PE crystallisable segments and the SBH grafts of the copolymers
obtained by reactive blending were longer than those of the copolymers prepared by
polycondensation. The analysis of the non-isothermal crystallisation kinetics showed that
the mechanism of crystallisation of PE changed for SBH content higher than approximately
50%, due to the decrease of both nucleation and crystal growth rates.
441
Handbook of Polymer Blends and Composites
The conditions at which iPP solidifies from the melt play a major role in crystalline
structure and morphology, and thus on the properties, of the polymer. Above a high
critical cooling rate, only the irregular smectic phase is formed. Aggregates with nanoscale
size have been observed, and these were found to coarsen during annealing [103].
Lim and Lloyd [106, 107] studied the overall non-isothermal crystallisation kinetics for
nucleated and non nucleated iPP, as it crystallised from the melt in blends with
dotriacontane (C32H66). The nucleating agent used was adipic acid. The crystallisation
peak temperatures for filled samples were higher than those of unfilled samples, indicating
that adipic acid was a good nucleating agent for iPP. The difference between the onset
and the peak temperatures was taken as a measure of the overall crystallisation rate. The
difference was smaller for all the nucleated samples [107]. The non-isothermal
crystallisation process was analysed by Ozawa and Ziabicki methods. The value of Ozawa
exponent indicated that the mechanism for crystallisation was instantaneous nucleation
with three-dimensional growth of the crystallites. At lower crystallisation temperatures,
sporadic nucleation occurred. The cooling crystallisation function had a negative
exponential dependence on temperature, and showed that the addition of the diluent did
not modify the mechanism of crystallisation of nucleated and non-nucleated iPP. The
analysis of the non-isothermal crystallisation process with Ziabicki theory indicated a
three-dimensional growth of the crystallites with instantaneous nucleation, in agreement
with Ozawa analysis.
The influence of the PP, obtained from a natural terpene on the growth rate and bulk
non-isothermal crystallisation of iPP was investigated [70, 97]. It was shown that the
solidification process depends on cooling rate, composition and miscibility of the
components. For the blends with PP content equal or lower that 30% w/w, whose
amorphous phase remained homogeneous with cooling rate used, the overall non-
isothermal crystallisation rate was depressed with respect to plain iPP, because of the
diluent effect of polyterpene. For the 50/50 blend, which presented two amorphous phases,
solidification started at temperature higher than that of both plain iPP and blends
containing a lower PP content 30%, since the diluting effect of PP in the iPP-rich
phase was outweighed by an increased number of nuclei that originated from the
polyterpene-rich phase domains. The addition of PP to iPP also lowers the radial growth
rate of iPP spherulites grown during non-isothermal cooling. The presence of PP has a
great influence on the regime II-III transition, which is shifted to lower temperatures
with increasing PP content. The authors show the use of a constant cooling rate allows
reliable growth data to be obtained and permits one to enlarge the range of temperatures
where crystallisation can be studied.
442
Crystallisation, Morphology and Melting in Polymer Blends
Non-isothermal crystallisation process of isotactic PB-1 and of its blends with an amorphous
oligomer, HOCP, was investigated by Silvestre, Cimmino and Di Lorenzo [109], see
Figure 11.10. The samples were melted at 160 C for ten minutes and cooled to room
temperature at different rates: 0.5, 1, 2 and 4 C/minute. The morphology of crystallised
samples was characterised by the presence of small and large spherulites, nucleated at
different times and temperatures. The two kinds of spherulites grew from heterogeneous
nuclei having different activation energies. The temperature at which the two different
kinds of nuclei became visible when the samples were cooled from the melt at 4 C/minute,
depended on blend composition, as reported in Table 11.3. There was clear evidence that
nucleation of both nuclei was delayed by addition of HOCP, suggesting that, in the blends,
the heterogeneous nuclei needed a higher undercooling to become active.
443
Optical microscopy
0 89 69
10 85 60
20 80 54
40 70 42
60 59 28
The induction time and the time required for the completion of crystallisation decreased
with cooling rate. The addition of HOCP to PB-1 shifted the crystallisation onset and
peak temperatures to lower values and increased the broadness of the transition. These
results suggested that HOCP lessened the crystallisation rate of PB-1 when the process
was performed under continuous cooling. Similar results were obtained in isothermal
conditions [109]. The decrease of crystallisation rate was attributed to the diluent effect
of HOCP in the PB-1 rich phase, to the presence of the HOCP rich phase domains, and
to the occurrence of entanglements between the phases.
The experimental data were analysed with Ozawa and Ziabicki theories. The Ozawa
equation was satisfactorily used to describe the dynamic solidification of all the samples.
The Avrami exponent, calculated with the Ozawa method, was close to 3, for both plain
PB-1 and blends, in quite good agreement with the value obtained in isothermal conditions
[109]. Conversely, the use of Ziabicki theory did not give good agreement with the
experimental results. It was found that the zero-order approximation did not describe
the non-isothermal crystallisation process of PB-1, probably indicating that athermal
nucleation was not negligible.
The crystallisation rate coefficient, CRC [96], decreased with the addition of HOCP,
from 18 h-1 of plain PB-1 to 6 h-1 of the PB-1/HOCP 60/40 blend, confirming that the
addition of HOCP caused a decrease of the non-isothermal crystallisation rate of PB-1.
A study of the effect of the addition of aPS on the non-isothermal crystallisation of sPS
was presented by Park, Lee and Park [110]. The cooling rates used ranged from 2 to
40 C/minute. The increasing of aPS content caused the decrease of dynamic crystallisation
444
Crystallisation, Morphology and Melting in Polymer Blends
peak of sPS, but did not affect the enthalpy of crystallisation. The crystalline form of
sPS, , obtained by solidification during the cooling from the melt, was not altered by
the addition of aPS and by increasing cooling rate (1-5 C/minute).
11.4.1.2 Polyoxyolefins
The non-isothermal crystallisation process of PEO, plain and blended with aPMMA and
PVA, was investigated by Silvestre and co-workers [111, 112].
Crystallisation during cooling of PEO/aPMMA blends was studied using DSC with
different cooling rates.
In quasi-static conditions, for a cooling rate of 0.625 C/min, nucleation and growth of
PEO crystals were governed by thermal mechanisms. The experimental data for plain
PEO and for the blends with up to 20% (wt/wt) of aPMMA, conformed to the theoretical
values calculated with Ziabicki treatment. The zero-order approximation is valid only
to describe the non- isothermal phase transition of plain PEO. At higher cooling rates, in
the case of the blends, athermal nucleation could not be neglected, and the same
approximation did not fit the experimental results. For the blends, experimental data
showed no agreement with Ozawa theoretical predictions, whereas crystallisation data
of plain PEO could be fitted to straight lines only for temperatures T 42 C.
For PEO/PVAc blends with up to 30% PVAc content, for a given cooling rate, the
broadness of transition, t, was almost constant with composition, whereas for higher
blend composition t increased with PVAc content.
Herrero and Acosta [113, 114] studied the effect of polyepichlorhydrin (PECH) on non-
isothermal crystallisation of PEO/polyphosphazene blends (PPz), by applying Ozawa
theory and Avrami method extended to non-isothermal conditions Herrero and Acosta
investigated also the effect of PECH on the non-isothermal crystallisation of (PEO/PPz)
blends. For the PEO/PECH system, the Avrami constant, K, was found to increase with
increasing concentration, as it usually happens for compatible systems, whereas in the
445
Handbook of Polymer Blends and Composites
case of the PEO/PPz system, K did not vary with composition, showing incompatibility
of the components. For the ternary blends, K diminished with PECH composition. The
Avrami exponent was between 1 and 2, when the data were analysed with Avrami theory,
and between 2 and 3 with the Ozawa equation.
11.4.1.3 Polyesters
Non-isothermal crystallisation of PET blended with the LCP, Rodrun 3000, at different
composition was studied by Incarnato and co-workers [115]. The results indicated that
the thermal behaviour of the system was strongly influenced by blend composition,
viscosity ratio and conditions used to treat the samples. It was found that a very low
amount (2%) of LCP determined an increase in the overall crystallisation rate, whereas
increasing the amount of LCP resulted in a progressive decrease of solidification rate.
This behaviour was ascribed to a competition between the nucleating effect of LCP and
a diminished diffusion rate of PET chains towards the growing crystals, due to the presence
of LCP droplets. At low LCP concentration, the nucleating effect prevailed, whereas, for
high LCP amounts, the diffusion of PET to the growth front became the rate-controlling
step in the crystallisation process. The crystallinity of PET in the blends was found to
increase on increasing the LCP content. The experimental data were analysed with Ozawa
method and the calculated Avrami exponent, n = 3, did not change with composition,
indicating that the nucleation and growth mechanism of PET spherulites were not
influenced by the addition of Rodrun. The cooling crystallisation function of PET
decreased with increasing temperature, in agreement with the findings of Ozawa.
446
Crystallisation, Morphology and Melting in Polymer Blends
11.4.1.4 Polyamides
Brucato and co-workers discussed the morphology and crystal structure of nucleated
polyamide 6 in relation to thermal history [118]. The crystallinity fraction decreased
with cooling rate. Under very high cooling rates the sample remained essentially
amorphous, with no crystalline entities detectable by optical microscopy. As scanning
rate decreased (in the range 200-70 C/s), crystallites started to appear; they were first
isolated and small, then contiguous and larger. For these cooling rates, only the crystalline
form was seen. At scanning rate below 70 C/s, larger polymer fractions could undergo
crystallisation in the form and the crystalline form gradually disappeared. These
results fitted the general framework that crystallisation in and forms takes place at
high and low temperatures, respectively [119, 120]. If the cooling rate from the melt was
relatively low, the polymer was able to crystallise in form (at high temperatures) and
solidification in form was inhibited. Conversely, if the scanning rate was too large for
a significant crystallisation in the form, the polymer remained available to undergo
crystallisation, which seemed to be characterised by faster kinetics.
The dynamic crystallisation of PPS based blends was studied by several authors [121-
128]. The addition of a LCP to PPS can appreciably reduce the melt viscosity of the
thermoplastic material, thus enhancing its processability. Blends of PPS with several LCP
have been studied and the effect of LCP on non-isothermal crystallisation of PPS has
been the subject of several investigations.
Minkova and Magagnini analysed the dynamic solidification of blends of PPS with an
aromatic copolyester-amide Vectra-B950 (Hoechst-Celanese) (VB) [121]. PPS
crystallisation temperature and the crystallisation rate coefficient were markedly increased
with the addition of 2%-50% VB. CRC was 52 h-1 for plain PPS, and 150 h-1 for the
blends. The dependence of Tp on cooling rate was much stronger for neat PPS than for
the blends, suggesting that the addition of VB to PPS increased the nucleation density of
PPS, in agreement with isothermal crystallisation studies [122-125].
Hong and co-workers [126] studied the non-isothermal crystallisation of PPS blended
with a LCP composed of 2,6-hydroxynaphthoic acid (60%), terephthalic acid (20%)
and aminophenol (20%). Also in this case the increase of the crystallisation temperature
and a decrease of Avrami index with the addition of the LCP suggested that LCP can
promote the crystallisation of PPS, by acting as nucleating agents.
447
Handbook of Polymer Blends and Composites
Shibata and co-workers studied the effect of polyether sulfone (PES) on the non-isothermal
crystallisation of PPS. They found that the onset crystallisation temperature and the
peak crystallisation temperature were little affected by the PES content, whereas the
completion crystallisation temperature increased with increasing PES content [127]. The
presence of polyamide 6 (PA6) modified the crystallisation parameters of PPS, promoting
a faster crystallisation rate. This enhancement was attributed to the possible interfacial
interaction between the two polymers that can induce heterogeneous nucleation [128].
11.5 Morphology
The morphology of crystallisable blends depends strictly on the composition, chemical
nature of the components, crystallisation conditions, and interaction between the
components [11, 19].
When a crystalline polymer is blended with a miscible amorphous diluent, during the
crystallisation from the melt three different morphologies can occur, assuming that the
diluent is rejected from the crystal and that the crystallisable component crystallises
according to a spherulitic morphology: (1) the amorphous material can be trapped within
the lamellar stacks (interlamellar); (2) it can be excluded from the lamellar stacks and
resides within the spherulite (interfibrillar); and (3) it can be totally excluded from the
spherulite region (interspherulitic). The first morphology is considered as the inclusion
morphology, whereas the other two are considered as the exclusion morphology. Different
types of morphology may coexist, leading to multiple locations for the amorphous diluent
[129-133]. Segregation of amorphous diluent is natural since the driving force of
crystallisation tends to separate the two components from being mixed. However the
principle dictating the location of diluent is not well understood. It has been suggested
that the diluent molecules confined in the interlamellar regions are deformed by the
crystals and have lower conformational entropy. An entropic driving force is thus
developed, tending to pull the diluent molecules out of the interlamellar regions [134]. In
addition to this entropic force, the crystallisation driving force of the crystallisable
segments in the interlamellar zones should act as another force to push the diluent out of
448
Crystallisation, Morphology and Melting in Polymer Blends
the interlamellar regions [135]. Two driving forces then exist to reject the amorphous
diluent out of the interlamellar zones: the entropic force associated with the tendency to
resume random-coiled conformation and the crystallisation driving force of crystallisable
segments in the interlamellar regions. These two forces compete against the favourable
interaction between the diluent and the amorphous portion of the crystalline polymer in
the interlamellar regions. Exclusion of the amorphous diluent out of the interlamellar
region is consequently governed by the magnitude of the interaction (12 parameter) and
the degree of undercooling. In addition to the thermodynamic factors an important role
is played by the kinetic chain diffusivity.
Keith and Padden [45], suggested that the distance over which non-crystallisable impurity
may be segregated is determined by the rejection parameter , defined as the ratio between
the diffusion rate of the non-crystallisable component in the melt and the spherulite
growth rate of the crystallisable component. This parameter has unit of length and provides
a measure of the segregation distance. Qualitatively a value of smaller than few hundreds
of A (about the value of the interlamellar region thickness) implies that the amorphous
material is trapped between the crystallisable lamellae. If is larger than the interlamellar
amorphous regions but smaller than the interfibrillar zones (few thousand of A), the
amorphous material is present mainly in interfibrillar regions. For very high values, the
non-crystallisable component segregates in domains larger than the interfibrillar regions
in intraspherulitic and/or interspherulitic zones.
The location of the amorphous material during the crystallisation can be hence defined
taking into consideration, the segmental interaction between the components together
with the diffusion of the non-crystallisable component and the growth rate of the
crystallisable components. It was in fact observed that the inclusion versus exclusion
model depends on three factors: (1) segmental interaction between the components; (2)
diffusion of the non-crystallisable component and (3) growth rate of the crystal
component. The crystal growth depends on (1) and (2) and mobility and growth can be
changed by diluent concentration and diluent glass transition.
Several papers on the morphology of polymer blends can be found in literature. Most
work has focused on revealing the morphology in specific system, whereas limited efforts
have be made to systematically evaluate the variable that could effect the formation of
segregation. Groenickx and co-workers studied the effect of the Tg of the diluent on the
segregation distance in PCL/PVC and PCL/chlorinated polyethylene (CPE) blends [136,
137]. Runt and co-workers considered the roles of both intermolecular interaction and
Tg of the diluent on the segregation distance [48, 49]. The molecular mass-crystal growth
rate-segregation distance relationships were critically discussed by Chen and co-workers
[136], see Figures 11.11 and 11.12.
449
Handbook of Polymer Blends and Composites
a)
b)
c)
450
Crystallisation, Morphology and Melting in Polymer Blends
a) b)
30 m
c)
451
Handbook of Polymer Blends and Composites
1) Very low crystal growth rates lead mainly to lamellar exclusion regardless of the
interaction strength between the components. Low crystal rates can be obtained
decreasing the undercooling and increasing the composition.
2) Miscible diluents having similar interaction strength are completely excluded from
lamellar stacks if their glass transition is low. In contrast they are enclosed in the
lamellar stack if their Tg is high.
The presence of a second component can alter also the structural and morphological
parameters of the crystallisable component. The addition of the amorphous component
can influence the structure of the crystals of semicrystalline polymer. In the case of iPP
blended with a hydrogenated oligocyclopentadiene resin or with rubber material, the
results of the X-ray show the formation of the smectic phase of iPP at temperatures
where bulk iPP crystallises only in the monoclinic form. The amount of smectic phase
present in the blend is dependent on the blend composition [138-140]. The modification
of the crystalline form of PVDF with polymethyl methacrylate, polyethyl methacrylate,
polymethyl acrylate and polyvinyl fluoride has been reported for samples quenched from
the melt and then annealed above the glass transition temperature [141-144]. PVDF
crystallises from the melt in three different polymorphs (, and ) depending on the
temperature and the time of crystallisation. At the lower crystallisation temperature (150-
170 C) and at lower crystallisation times the phase is obtained for all fractions; at
higher Tc (170-180 C) and (T higher than 180 C) are observed. The and phases
are less abundant with increasing blend composition for each fraction and phases
decreases with increasing blend composition for each fraction.
The modification of the polymorphic behaviour of sPS with PPO was also reported
[145]. For melt crystallised samples the formation of the form in these blends requires
much lower temperature and shorter melting times than for pure sPS. Guerra and co-
workers [146] suggested that the phenomenon was due to a more rapid loss of memory
of the form in the presence of PPO molecules.
For samples crystallised from the quenched amorphous phase a more perfect modification
of the form and the thermodynamically more stable form are obtained instead of the
disordered modification of the form.
452
Crystallisation, Morphology and Melting in Polymer Blends
Also Woo and Wu [147] evaluated the phenomenon of alteration of the polymorphism
of SPS in presence of PPO and aPS. They found that only form crystals were found in
the blends, whereas both and crystal forms were observed with aPS or PPO. They
concluded that the greater randomness in the melt state of the blends might be
unfavourable for formation of form crystals from the melt.
Wide and small angle x-ray experiments were used to investigate the crystallisation time
dependence of the small angle scattering spectroscopy (SAXS) microstructural parameters
for the PEO based blends by Lisowski and co-workers [150]. In particular several amorphous
polymeric diluents were used presenting different intermolecular interaction with PEO
and mobility in the melt. It was found in agreement with the studies of Silvestre and co-
workers [11] that the microstructural parameters for the blends was generally similar to
that observed for PEO with the exception that the magnitude of the initial decrease in L
and lc was significantly larger for strong interacting mixtures. This result for the weakly
interacting blend was associated with interlamellar diluent placement, whereas for strong
interacting blends it was associated with the reduction in the degree of supercooling.
The effect of miscible polymer diluent on the development of morphology of POM and
PEO based blends was investigated by Yeh and co-workers [151] and Silvestre and co-
workers [11], respectively. Blending a low Tg miscible diluent does not alter the lamellar
parameters of POM, whereas the blending with a high Tg diluent causes an increase in
amorphous thickness and a decrease of lamellar thickness with time in the secondary
crystallisation regions. In PEO/PVAc blends the hedritic character (hexagonal shaped
cracks, indicating step-like growth and lack of Maltese cross shape) of the PEO spherulites
disappears following the addition of the PVA [152]. Banded spherulites of PCL were
catalysed by the presence in the melt of atactic styrene-acrylonitrile copolymer (SAN)
[153, 154]. Banded spherulites were also found in some other blend systems [155-159].
Wang and co-workers reported that the periodic distance of extinction rings of PCL
453
Handbook of Polymer Blends and Composites
decreases with increasing SAN content and increases with crystallisation temperature.
They found that the periodic distances of the ring-banded spherulites with different blends
ratio fall on one master curve, when plotted as a function of the overall mobility of the
mixtures. This suggests that the mobility of the chain segments mainly controls the
formation of the ring-banded spherulites, see Figure 11.13.
454
Crystallisation, Morphology and Melting in Polymer Blends
Okabe and co-workers observed spiral crystals in PVDF/PVA blends [160]. They found
that spiral crystals form predominantly at high Tc. The microfibrils at the core are partially
aligned with their long axes parallel to the spiral axis during the initial growth stage. As
the spiral wave propagates, the fibre axes gradually change their orientation in the
perpendicular direction. In these blends PVA chains are reject into interspiral regions,
see Figure 11.14.
Addition of the aPMMA [161] and PMA [157] makes the PVDF spherulites more open
and causes the banding periodicity to decrease. The conditions for dendrite crystallisation
in various blends of PVDF have been studied by Hellmann and co-workers [157]. In
blends with aPMMA or styrene-co-methylmethacrylate, the modification of PVDF
forms dendrites, that appear at moderate undercooling and high concentration of the
crystallisable material, whereas the modification forms more or less perfect spherulites.
In the case of sPS/PVME blends the presence of PVME influences also the birefringence
sign of the sPS spherulite [162]. Silvestre and co-workers [162] have, in fact, observed
sPS spherulites having negative birefringence. For the same crystallisation temperature
and composition, the positive spherulites grow faster than the negative ones.
Figure 11.14 Atomic force microscopy pictures of the 90/10 PVF2/PVAc blend at
different magnifications, displaying spiral ridges and valleys consisting of self-
assembled microfibrils. The blend specimen was crystallised at 170 C, but the picture
was obtained after quenching to room temperature [160]
Reprinted with permission from Y. Okabe, T. Kyu, H. Saito, T. Inoue, Macromolecules,
1998, 31, 17, 5823, Figure 11. Copyright 1998, American Chemical Society
455
Handbook of Polymer Blends and Composites
The morphology of numerous phase-separated blends has been recently reviewed by Silvestre
and co-workers [11]. They observed small domains in the melt phase, indication of the
liquid-liquid phase separation for the PEO/E(1-y)-co-VAc(y) (y, molar fraction, .43), iPS/
PVME, iPP/HOCP and sPS/PVME blends. During crystallisation, the growing spherulites
occlude droplets of the phase rich in non-crystallisable component mainly in their
intraspherulitic regions. The condensed phase is characterised by one crystalline phase and
two amorphous phases, one rich of the crystallisable component and the second one rich
of the non-crystallisable material. For these blends it was found that the long period and
the amorphous layer thickness are almost independent of composition and the phase rich
in the non-crystallisable component is mainly segregated in interfibrillar regions.
For the PVDF/aPMMA blend, the presence of a region of PVDF high concentration region
in the amorphous phase between lamella was observed, suggesting a phase separation in
the amorphous region or a gradient of concentration with the highest PVDF concentration
occurring close to the PVDF crystals [163-166]. Jungnickel and co-workers [167] proved
the existence of a composition profile dependent on Tc around the growing spherulite. In
particular they found for the PVDF/PEA blends that the composition of the amorphous
component increases remarkably upon crystallisation. The enrichment of the non-
crystallisable material at the growth front causes the spherulite growth rate to slow down.
The kinetics of mixing or demixing can interfere and compete with that of crystallisation,
when the blend system presents a miscibility gap. Which of these phase transitions
dominates, depends on composition and on the thermal boundary conditions, as first
reported by Tanaka [168, 169], Li and co-workers [170-172], Endres and co-workers
[173], Cimmino and co-workers [174] and Jungnickel and co-workers [167]. The situation
was clearly illustrated in the case of poly-caprolactone/aPS and PVDF/polyethylacrylate
blends that present a demixing curve (UCST and LCST type, respectively), intersected by
the melting point depression curve and in the case of iPP/HOCP when the phase diagram
presents both UCST and LCST [174]. In the PCL/aPS system crystallisation from an
homogeneous melt causes release of the non-crystallisable material, that at lower and
intermediate temperature remains in the amorphous phase between lamellae spherulites,
whereas at higher crystallisation temperature is rejected into the remaining melt. In both
cases the aPS increases its concentration and the binodal concentration can be reached,
giving rise to new and interesting supermolecular structure. At a lower crystallisation
temperature the amorphous phase remains between the lamella and does not decompose,
when reaching the binodal composition, due to the limited diffusion rate. Increasing the
temperature, the diffusion to a limited extend becomes possible and a phase separation
inside the interlamellar regions occur. The phase rich in crystallisable material can post
crystallise, giving rise to the formation of new lamella within the spherulites. For very
high crystallisation temperatures, the non-crystallisable material diffuses into the melt
and a concentration gradient profile arises. As soon as the binodal composition is reached
456
Crystallisation, Morphology and Melting in Polymer Blends
in the melt at the spherulite surface, demixing occurs with the formation of droplets of
the non-crystallisable component rich phase. For the iPP/HOCP system, which is
dependent on temperature, blend composition and cooling rate from the melt, it is possible
to obtain one or two phases with iPP in smectic or monoclic form. Moreover for this
system, a new crystallisation process induced by demixing takes place following the
phase separation in the iPP rich phase [175].
For the system aPMMA/PHB [17, 18] the phase structure is dependent on cooling rate
and composition. In particular the crystallisation and separation process are in competition
with each other. In the melt the system is homogeneous. This homogeneity can be preserved
if the sample is fast quenched at 50 C. Phase separation is found at temperature below
170 C. Slow cooling the material from the melt at Tc, a liquid-liquid phase separation
occurred before crystallisation. After crystallisation, three phase were present. A crystalline
PHB phase, and two amorphous phases. If crystallisation occurred from the homogeneous
glass state, the crystallisation preserves the homogeneity, indicating an homogeneity
induced by crystallisation.
A kind of phase separation can be induced in some miscible blends, where the crystallisable
component crystallises forming primary and secondary lamellae, as shown by Dreezen
[176]. A model was proposed for describing the morphology of PEO/PESF at two different
temperatures. In this model the secondary PEO lamellae are located next to the primary
ones. After the melting of the secondary lamellae, a three phase-system forms, consisting
of primary PEO lamellae, an amorphous PEO and a miscible amorphous PEO/PESF, see
Figure 11.15.
Many reports can be found in the literature on the morphology of blend systems whose
components are immiscible above and below the melting temperature. In the melt and
after crystallisation the minor component forms its own domains. The morphology of
immiscible blends can be divided into three classes: dispersed, stratified and co-continuous
morphologies. Dispersions of droplets of the minor phase in a matrix of the major phase
are the most common. These type of blends are often found in brittle polymers modified
by rubber. This minor phase can also be dispersed as fibres, for examples in self-reinforcing
polymer blends, for example in self-reinforcing polymer blends. The effect of interfacial
tension, viscosity and physical crosslinks on the coarsening of co-continuous morphologies
in polymer blends was studied by Veenstra and co-workers [177].
For blends made with a thermoplastic polymer and an elastomer,the domains of the
disperse phase are generally spherical in the melt. In the condensed state, after
crystallisation, the material is characterised by the presence of spherulites, that include
the elastomeric domains in intra- and/or inter-spherulitic regions. The state and mode of
dispersion of the rubber particles is strongly dependent on the molecular structure as
well as on the molecular mass of the rubber component.
457
Handbook of Polymer Blends and Composites
458
Crystallisation, Morphology and Melting in Polymer Blends
whose thickness and perfection are dependent on crystallisation temperature and time of
annealing at Tc. The thermodynamic factors are a consequence of the thermodynamic of
mixing. In fact thermodynamic considerations predict that the chemical potential of a
polymer decreases following the addition of a miscible diluent, and in the case of a
crystallisable polymer such a decrease results in a depression of the equilibrium melting
temperature. Nishi and Wang [178] using the Flory-Huggins approximation, derived the
following equation to describe the dependence of the equilibrium melting temperature
on composition:
1 1 RV2 u 1 1
= (ln 2 + ( )1 + 1212) (11.4)
Tm Tm Hf V1u m 2 m 2 m 1
1 1
A plot of against 12 should give a straight line passing through the zero of the
Tm Tm
axes and from the slope 12 can be calculated. A list of 12, for several miscible blends,
obtained by using Equation 11.4 is reported in Table 11.1.
Many authors have experienced difficulties in fitting Equation 11.4 to their data because
generally the plot of the left-hand side versus the right-hand side has an intercept with
the axis different from zero, due to the limits of the Flory-Huggins theory, that considers
the 12 to be independent of composition and it is unable to predict the lower critical
solution temperature and the dependence of the melting point depression to morphological
factors. These factors can be very relevant to some blends. The equation of the melting
point depression as function of composition derived by Kwei and Frisch [179] takes into
account also morphological effects on Tm:
Tm _ Tm Tm T
Hf ( ) 1 m = C1 12RTm 12
Tm m1 2m2
Other expressions were derived by Chow [180] and Alfonso and Russell [38] to describe
the dependence of equilibrium melting temperature on composition, morphology and
diffusion rate.
The practical difficulty in using this equation is the determination of the equilibrium
melting points. The simplest way is to use the Hoffmann and Weeks equation [181]:
1 T
Tm' = Tm (1 )+ c
459
Handbook of Polymer Blends and Composites
where is a morphological factor and Tm the observed melting point of the sample
crystallised at Tc. The equilibrium melting point can be determined from a plot of Tm
versus Tc. The limit of this method is that the measured Tm depends not only on the
crystallisation temperature, but also on crystallisation time and in the case of blend on
composition [182-184].
A most reliable way to calculate the equilibrium melting point is through the equation
that relates the melting point to the lamellar thickness:
2 e
Tm' = Tm 1 (11.5)
Hf lc
where e is the fold free energy, Hf is the heat of fusion per unit volume of crystal and
lc the crystalline thickness. Plot of Tm versus 1/lc should result in a straight line. The
extrapolated intercept of Tm at 1/lc = 0, (i.e., at infinite crystal thickness), gives the value
of the equilibrium melting point. Phillips and Rensch [185], modified the Equation 11.5
to take into account the maximum extended chain value, that is dependent on molecular
mass. As reported by Runt and Martynowicz [186], another difficulty is the determination
of the true melting temperature of the crystals that were actually formed at Tc. In fact
semicrystalline polymers often undergo lamellar thickening upon the heating of the thermal
analysis experiment. In case of blends the lamellar thickening is resulted dependent not
only on the heating rate but also on composition.
Equation 11.4 has also been reformulated [187] in terms of the equation of state theory
[188] to obtain the interaction energy term 12, which is concentration independent and
denotes the energy change for the formation of contacts between species 1 and 2 compared
to contacts between like species.
Acknowledgement
We express our thanks to Dr. M.L. Di Lorenzo for help in the preparation of the non-
isothermal crystallisation section and to Mr. Giuseppe Narciso for the preparation of
part of the figures of the paper.
References
460
Crystallisation, Morphology and Melting in Polymer Blends
2. D.R. Paul and J.O. Altamirano in Copolymers, Polyblends and Composites, Ed.,
N.A.J. Platzer, American Chemical Society, Washington, DC, USA, 1975.
3. S.R. Murft, J.W. Barlow and D.R. Paul in Multicomponent Polymer Materials,
Eds., D.R. Paul and L.H. Sperling, Advances in Chemistry Series, American
Chemical Society, Washington, DC, USA, 1986, Chapter 18.
5. D.R. Paul and J.W. Barlow, Polymer Science and Technology, 1980, 11, 239.
6. D.R. Paul in Multicomponent Polymer Materials, Eds., D.R. Paul and L.H.
Sperling, Advances in Chemistry Series, American Chemical Society, Washington,
DC, USA, 1986, Chapter 1.
7. D.R. Paul and J.W. Barlow, Journal of Macromolecular Science C, 1980, 18, 1, 109.
14. S. Cimmino, E. DAlma, C. Silvestre and M.L. Di Lorenzo, inventors; CNR Italy,
assignee; NA 99A000063, 1999.
15. Thermoplastic Elastomers from Rubber-Plastic Blends, Eds., S.K. De and A.K.
Bhowmick, Ellis Horwood Ltd., New York, NY, USA, 1990.
461
Handbook of Polymer Blends and Composites
18. S. Cimmino, P. Iodice, C. Silvestre and F.E. Karasz, Journal of Applied Polymer
Science, 2000, 75, 6, 746.
20. P.J. Flory, Principles of Polymer Chemistry, Cornell University Press, Ithaca, NY,
USA, 1953.
21. W.J. MacKnight, F.E. Karasz and J.R. Fried in Polymer Blends, Eds., D.R. Paul
and S. Newman, Academic Press, New York, NY, USA, 1978, Chapter 5.
22. J.M.G. Cowie, Journal of Macromolecular Science, Physics Edition, 1980, B18,
4, 569.
24. L.A. Utracki and J.A. Jukes, Journal of Vinyl Technology, 1984, 6,285.
25. M. Gordon and J.S. Taylor, Journal of Applied Chemistry, 1952, 2, 493.
27. T.K. Kwei, Journal of Polymer Science, Polymer Letters Edition, 1984, 22, 307.
28. T.G. Fox, Bulletin of the American Physical Society, 1956, 2,153.
30. C. Silvestre, S. Cimmino, F.E. Karasz and W.J. MacKnight, Journal of Polymer
Science, Polymer Physics Edition, 1987, 25, 12, 2531.
462
Crystallisation, Morphology and Melting in Polymer Blends
34. B. Wunderlich, Macromolecular Physics, Academic Press, New York, NY, USA,
1976.
35. J.D. Hoffman and R.L. Miller, Polymer, 1997, 38, 13, 3151.
36. J.D. Hoffman and R.L. Miller, Macromolecules, 1988, 21, 10, 3038.
37. J. Boon and J.M. Azcue, Journal of Polymer Science, Part A-2,1968, 6, 885.
38. G.C. Alfonso and T.P. Russell, Macromolecules, 1986, 19, 4, 1143.
39. H. Saito, T. Okada, T. Hamane and T. Inoue, Macromolecules, 1991, 24, 15, 4446.
43. J. Skolnick, R. Yaris and A. Kolinski, Journal of Chemical Physics, 1988, 88, 1407.
44. T. Okada, H. Saito and T. Inoue, Polymer, 1994, 35, 26, 5699.
45. H.D. Keith and F.J. Padden, Journal of Applied Physics, 1964, 35, 1270.
46. H.D. Keith and F.J. Padden, Journal of Applied Physics, 1964, 35, 1286.
48. S. Talibuddin, J. Runt, L-Z. Lin, B. Chu, Macromolecules, 1998, 31, 5, 1627.
49. S. Talibuddin, J. Runt and J.S. Lin, Macromolecules, 1996, 29, 7527.
50. L. Wu, M. Lisowski, S. Talibuddin and J. Runt, Macromolecules, 1999, 32, 5, 1576.
463
Handbook of Polymer Blends and Composites
57. T.K. Kwei, G.D. Patterson and T.T. Wang, Macromolecules, 1976, 9, 5, 780.
59. L.M. Robeson, Journal of Applied Polymer Science, 1973, 17, 12, 3607.
60. S. Cimmino, E. Di Pace, E. Martuscelli and C. Silvestre, Polymer, 1993, 34, 13,
2799.
61. G.S.Y. Yeh and S.L. Lambert, Journal of Polymer Science, Part A-2, 1972, 10, 7,
1183.
62. B.K. Hong,W.H. Jo and J. Kim, Polymer, 1998, 39, 16, 3753.
64. N. Lotti, M. Pizzoli, G. Ceccorulli and M. Scandola, Polymer, 1993, 34, 23,
4935.
65. R. Pearce, G.R. Braun and R.H. Marchessault, Polymer, 1994, 35, 3984.
66. H. Verhoogt, B.A. Ramsay and B.D. Favis, Polymer, 1994, 35, 24, 5155.
67. Y. Kumagai and Y. Doi, Polymer Degradation and Stability, 1992, 35, 1, 87.
68. M. Avella, E. Martuscelli and M. Raimo, Journal of Materials Science, 2000, 35,
3, 523.
70. M.L. Di Lorenzo, S. Cimmino and C. Silvestre, Macromolecules, 2000, 33, 10,
3828.
464
Crystallisation, Morphology and Melting in Polymer Blends
75. L. Amelino, E. Martuscelli, C. Sellitti and C. Silvestre, Polymer, 1990, 31, 6, 1051.
77. N. Bulakh and J.P. Jog, Journal of Macromolecular Science Physics, 1999, 38, 277.
81. S.N. Omenyi, A.W. Newmann, W.W. Martin, G.M. Lespinard and R.P. Smith,
Journal of Applied Physics, 1981, 52, 789.
83. E. Martuscelli, C. Silvestre and L. Bianchi, Polymer, 1983, 24, 11, 1458.
465
Handbook of Polymer Blends and Composites
95. A. Ziabicki and A. Sajkiewicz, Colloid and Polymer Science, 1998, 276, 8, 680.
96. P. Khanna, Polymer Engineering and Science, 1990, 30, 124, 1615.
98. M.L. Di Lorenzo and C. Silvestre, Progress in Polymer Science, 1999, 24, 917.
99. L. Minkova and M. Mihailov, Colloid and Polymer Science, 1989, 267, 7, 577.
100. A.K. Gupta, S.K. Rana and B.L. Deopura, Journal of Applied Polymer Science,
1994, 51, 2, 231.
101. A.K. Gupta, S.K. Rana and B.L. Deopura, Journal of Applied Polymer Science,
1992, 44, 74, 19.
102. L. Minkova, T. Miteva and P.L. Magagnini, Colloid and Polymer Science, 1997,
275, 6, 520.
103. C.C. Hsu, P.H. Geil, H. Miyaji and K. Asai, Journal of Polymer Science: Part B
Polymer Physics, 1986, 24, 10, 2379.
106. G.B.A. Lim and D.R. Lloyd, Polymer Engineering and Science, 1993, 33, 9, 529.
107. G.B.A. Lim, K.S. McGuire and D.R. Lloyd, Polymer Engineering and Science,
1993, 33, 9, 537.
466
Crystallisation, Morphology and Melting in Polymer Blends
108. Y. Feng, X. Jin and J.N. Hay, Journal of Applied Polymer Science, 1998, 69, 10,
2089.
110. J.Y. Park, H.S. Lee and O.O. Park, SPS 13th International Meeting, Seacus, NJ,
USA, 1997, p.6-B.
111. M.L. Addonizio, E. Martuscelli and C.Silvestre, Polymer 1987, 28, 2, 183.
113. C.R. Herrero and J.L. Acosta, Polymer Journal (Japan), 1994, 26, 7, 786.
114. C.R. Herrero and J.L. Acosta, Polymer International, 1993, 32, 4, 349.
115. L. Incarnato, O. Motta and D. Acierno, Polymer, 1998, 39, 21, 5085.
116. R. de Juana, A. Jauregui, E. Calahorra and M. Cortzar, Polymer, 1996, 15, 37,
3339.
119. L.G. Roldan and H.S. Kaufman, Journal of Polymer Science, Polymer Letters,
1963, 1, 603.
121. L.I. Minkova and P.L. Magagnini, Polymer, 1995, 36, 10, 2059.
122. T.G. Gopakumar, R.S. Ghadage, S. Ponrathnam, C.R. Rajan and A. Fradet,
Polymer, 1997, 38, 9, 2209.
123. L.I. Minkova, M. Paci, M. Pracella and P.L. Magagnini, Polymer Engineering and
Science, 1992, 32, 1, 57.
124. M. Pracella, P.L. Magagnini and L.I. Minkova, Polymer Networks and Blends,
1992, 2, 225.
467
Handbook of Polymer Blends and Composites
125. L.I. Minkova, S. De Petris, M. Paci, M. Pracella and P.L. Magagnini in Processing
and Properties of Liquid Crystalline Polymers and LCP Based Blends, ChemTec
Publishing, Ontario, Canada, 1993, 153.
126. S.M. Hong, B.C. Kim, K.U. Kim and I.J. Chung, Polymer Journal (Japan), 1992,
24, 8, 727.
128. K. Mai, S. Zhang, H. Zeng, Journal of Applied Polymer Science, 1999, 74, 13,
3033.
129. R.S. Stein, F.B. Kambatta, F.P Warner, T. Russell, E. Balizer, Journal of Polymer
Science, Polymer Symposia, 1978, 63, 313.
132. S.D. Hudson, D.D. Davis and A.J. Lovinger, Macromolecules, 1992, 25, 6, 1759.
133. B.B. Sauer and B.S. Hsiao, Journal of Polymer Science, Polymer Physics Edition,
1993, 31, 81, 901.
134. T.P. Russell, H. Ito and G.D. Wignall, Macromolecules, 1988, 21, 6, 1703.
135. H-L. Chen, L-J. Li and T-L. Lin, Macromolecules, 1998, 31, 7, 2255.
136. G. Defieuw, G. Groeninckx and H. Reynaers, Polymer, 1989, 30, 12, 2158.
137. G. Defieuw, G. Groeninckx and H. Reynaers, Polymer, 1989, 30, 12, 2164.
138. S. Cimmino, P. Guarrata, E. Martuscelli and C. Silvestre, Polymer, 1994, 32, 3299.
140. C.W. Yip, Proceedings of Plasticon 81, Symposium 4, Warwick, UK, 1981, Paper
No.30.
468
Crystallisation, Morphology and Melting in Polymer Blends
143. W.W. Doll and J.B. Lando, Journal of Macromolecular Science and Physics,
1970, B4, 897.
144. G. Guerra, F.E. Karasz and W.J. MacKnight, Macromolecules, 1986, 19, 7, 1935.
147. E.M. Woo and F.S. Wu, Journal of Polymer Science: Polymer Physics Edition,
1998, 36, 15, 2725.
148. J-C. Ho and K-H. Wei, Macromolecules, 2000, 33, 14, 5181.
150. M.S. Lisowski, Q. Liu, J. Cho, J. Runt, F. Yeh and B.S. Hsiao, Macromolecules,
2000, 33, 4842.
151. F. Yeh, B.S. Hsiao, B. Chu, B.B. Sauer and E.A. Flexan, Journal of Polymer
Science: Polymer Physics Edition, 1999, 37, 3115.
152. C. Silvestre, F.E. Karasz, W.J. MacKnight and E. Martuscelli, European Polymer
Journal, 1987, 23, 10, 745.
155. Z. Wang, X. Wang, D. Yu and B. Jiang, Polymer, 1997, 38, 23, 5897.
157. D. Braun, M. Jacobs and G.P. Hellmann, Polymer, 1994, 35, 706.
469
Handbook of Polymer Blends and Composites
159. K. Schulze, J. Kressler, and H.W. Kammer, Polymer, 1993, 34, 3704.
160. Y. Okabe, T. Kyu, H. Saito and T. Inoue, Macromolecules, 1998, 31, 17, 5823.
161. B.S. Morra and R.S. Stein, Journal of Polymer Science: Polymer Physics Edition,
1982, 20, 2243.
164. B. Hahn, J. Wendorff and D.Y. Yoon, Macromolecules, 1985, 18, 718.
165. Y. Ando and D.Y. Yoon, ACS Polymer Preprints, 1987, 28, 26.
166. Y. Ando and D.Y. Yoon, ACS Polymer Preprints, 1988, 29, 381.
167. R. Wellscheid, J. Wust and B-J. Jungnickel, Journal of Polymer Science: Polymer
Physics Edition, 1996, 34, 2, 267.
168. H. Tanaka and T. Nishi, Physics Review, Letters, 1985, 55, 1102.
170. Y. Li, M. Stein and B-J. Jungnickel, Colloid Polymer Science, 1991, 269, 772.
172. Y. Li, L. Schneider and B-J. Jungnickel, Polymer Networks and Blends, 1992, 2,
135.
173. B. Endres, R.W. Garbella and J.H. Wendorff, Colloid Polymer Science, 1985,
263, 361.
470
Crystallisation, Morphology and Melting in Polymer Blends
177. H. Veenstra, J. Van Dam and A.P. de Boer, Polymer, 2000, 41, 8, 3037.
181. J.D. Hoffman, G.T. Davis and J. Lauritzen in Treatise on Solid State Chemistry,
Volume 3, Ed., N.B. Hanay, Plenum Press, New York, NY, USA, 1976, Chapter 6.
185. P.J. Phillips and G.J. Rensch, Journal of Polymer Science, Polymer Physics
Edition, 1989, 27, 155.
187. D.J. Walsh, S. Rostami and V.B. Singh, Macromolecular Chemistry, 1985, 186,
145.
188. P.J. Flory, R.A. Orwoll and A. Vrij, Journal of the American Chemical Society,
1964, 86, 3515.
189. R.E. Bernstein, D.R. Paul and J.W. Barlow, Polymer Engineering Science, 1978,
18, 683.
192. D.R. Paul, J.W. Barlow, R.E. Bernstein and D.C. Wahrmund, Polymer
Engineering Science, 1978, 18, 1225.
471
Handbook of Polymer Blends and Composites
193. N.K. Kalfoglou, Journal of Polymer Science: Polymer Physics Edition, 1982, 20,
1259.
194. I.A. Katime, M.S. Anasagashi and M.C. Peletiero, European Polymer Journal,
1987, 23, 907.
195. T.K. Mandal, J.F. Kuo and E.M. Woo, Journal of Polymer Science: Polymer
Physics, 2000, 38, 4, 562.
197. M.K. Neo and S.H. Goh, Polymer Networks and Blends, 1993, 3, 131.
198. M.K. Neo and S.H. Goh, European Polymer Journal, 1991, 27, 927.
199. M.K. Neo and S.H. Goh, Polymer, 1992, 33, 3203.
200. C.A. Cruz, J.W. Barlow and D.R. Paul, Macromolecules, 1979, 12, 726.
203. J.E. Harris, S.H. Goh, D.R. Paul and J.W. Barlow, Journal of Applied Polymer
Science, 1982, 27, 3, 839.
204. M.M. Matzner, D.L. Schoher, R.N. Johnson, L.M. Robeson and J.E. McGrath in
Permeability of Plastic Films and Coating to Gases, Vapors and Liquids, Ed.,
H.B. Hopfenburg, Plenum Press, New York, NY, USA, 1973.
472
Radiation Effects on Polymer Blends
12
Radiation-Effects on Polymer Blends
Mihaela C Pascu
Introduction
In order to overcome the difficulty that most polymers are thermodynamically immiscible
and form multiphase systems, with weak physical and chemical interactions between
phases [1], pretreatment of one or both components of the blends may be used [215],
causing their properties to be modified and thus obtaining new materials. In other cases,
surface or bulk treatment of the blend like a whole is necessary for further enhancement
of its properties [16-46].
Controlled modification of the properties of the polymers and of their blends is very
important for either obtaining materials with a prolonged lifetime (thus enhancing their
performance in current applications), or for obtaining environmental friendly materials,
with a high biodegradation rate (decreasing, in this way, the negative impact of their
degradation products on the environment).
This chapter presents the most used irradiation techniques for modifying the properties
of the polymer blends, i.e., plasma and corona treatment, particle beam bombardment,
-ray irradiation.
In the case of cold plasmas, the feedback gases are broken into positive ions, electrons,
and chemically reactive species, which flow to the exposed surface and physically or
chemically interact with it. Besides these species, radiations with a large spread of
wavelength, from bulk plasma, arrive at the substrate surface. An anomalous discharge,
shown in Figure 12.1a, consists of a vacuum chamber containing a plane cathode placed
at one edge of the discharge tube and of a cylindrical anode, with almost the same diameter
as that corresponding to the vacuum chamber and found at a small distance from the
cathode. If the polymer blends are exposed at the very limit of the negative glow, then it
is possible the simultaneous treatment of one side of the sample with an electron beam
(EB) from the cathodic fall, while the other side is exposed in the negative glow plasma.
Ions can be extracted from a low pressure discharge plasma. The treatment system plasma-
ion beam can be a double plasma device (Figure 12.1b). This one consists of two vacuum
473
Handbook of Polymer Blends and Composites
cylindrical chambers (A the source chamber and B the target chamber), separated through
a transversal grid, negatively polarised. In this way, the ionising electrons are not allowed
to pass from one chamber to the other one. The anode of chamber A, being positive with
respect to the ground, the ions from chamber B are reflected, while a beam of positive
ions passes from the source plasma towards the target plasma, where the polymer samples
have been placed.
Corona discharge is a widely used plasma technique, because it has the advantage of
operating at atmospheric pressure. This type of electrical discharge occurs in regions of
high electric fields near sharp points or wires, in electrically stressed gases (Figure 12.1c).
Modifications of polymer surfaces by plasma treatments (both corona and low pressure
glow discharges) give many important advantages. This is why, plasma processes have
been widely accepted over the years in various industrial applications, such as:
microelectronics, aerospace, automotive, biomaterials, and packaging, to give only a
few examples.
474
Radiation Effects on Polymer Blends
One of the most important results obtained by plasma treatment is that, during this process,
the energy is transmitted not only on the sunshine part of the substrate, but also to the
shade part of it [3]. This fact leads to the production of a thin plasma polymer coating on
all the surfaces of the treated substrate, behaviour which distinguishes the plasma technique
from all the other applied methods, i.e., UV irradiation.
Because the plasma technique allows the modifications within a very limited small region
at the surface, it is considered to be a very promising method for appropriate surface
modifications, without having a damaging effect on the mechanical performance of the
material.
Plasma pretreatment has been applied to obtain a maximum adhesion energy at the matrix-
fibre interface [8, 47].
Over the years, plasma treatment proved that it represents an useful tool for improving the
adhesion between fibres and rubber materials, by grafting reactive substrates on the fibre
surface. By varying the plasma treatment conditions, one can practically tailor the surface
structure and properties of the polymeric materials.
Oxygen plasma treatment [48, 49] is known to introduce oxygen containing chemical
functional groups and micro-pitting on the fibre surface. The oxygen containing groups
enhance the wetting of the fibre by the matrix and improve the interfacial adhesion between
them. Micro-pittings are responsible for inducing mechanical locking between the fibre
and the matrix, also leading to an improved interfacial adhesion.
Main changes in the properties of different blends and composites, obtained by plasma
pre-irradiation technique are presented in Table 12.1.
475
476
Figure 12.2 Variation of: a) tensile modulus and b) tensile strength for unidirectional
composites based on epoxy, reinforced with corona-treated UHMPE fibres and, with
original UHMPE fibres, respectively, in function of the filler volume fraction
(Redrawn from the data of [4])
that corona-treatment modifies the surface of UHMPE fibres, thus obtaining materials with
reduced strength, but leading, in the meantime, to an increased adhesive bonding.
Argon plasma treatment is usually used to improve the printability, adhesion, and other
properties of the polymeric surface. A 13.56 MHz RF argon plasma pretreatment was
applied by Moon and Jang to an UHMPE fibre [5]. The increase of the ILSS with exposure
478
Radiation Effects on Polymer Blends
Improvement of the interfacial adhesion was also evidenced for the case of UHMPE fibre/
epoxy resin composite, the UHMPE fibre being firstly pre-treated, for 1 minute, in oxygen
plasma [6]. After plasma treatment, the UHMPE fibre was impregnated with PB. ILSS
dependence on the concentration of PB coating solution shows a significant decrease of
ILSS of UHMPE fibre/epoxy composite with increasing the PB concentration, behaviour
which Moon explained by the existence of low modulus, low strength, and high viscous PB
interlayer in the interfacial region. For a PB coating higher than 3 wt%, the ILSS of the
studied composite is a little higher than that of the untreated UHMPE fibre/epoxy composite.
This could be due to the fact that PB coating on the oxygen plasma treated UHMPE fibre
makes the fibre surface smooth and decreases the mechanical interlocking. Another possible
explanation is the incomplete coating of the UHMPE fibre by PB, the effective mechanical
interlocking, occurring through the uncoated micro-pittings.
PP-gas plasma pretreatment of Kevlar and graphite fibres introduced in a PP matrix [8]
tends to decrease the surface tension of the fibre, in contrast to the results obtained with
air, oxygen, and nitrogen plasma treatment. This maybe due to a layer of plasma-polymerised
PP, deposited on the fibre surface. This fact was shown by an increase in the contact angle
between the fibre surface and the distilled water, with increasing the treatment time
(Figure 12.3). It was found that PP-gas plasma treatment improved the interfacial adhesion
by up to 40% for PP/Kevlar composites and by 20% for PP/graphite composites.
Scanning electron microscopy (SEM) examination of both composite types evidenced several
debrils on the plasma treated Kevlar fibre surface, while the physical structure of graphite
fibre did not change appreciable after the same treatment.
Tensile strength was also improved after PP-gas plasma treatment (Figure 12.4), its maximum
value being obtained for a 3 minute exposure time.
For nitrogen and oxygen plasma pretreatment of graphite fibres from an epoxy matrix [8],
the tensile strength presents a minimum at an intermediate flow range of 30-50 cm3/min
(Figure 12.5). This behaviour correlated well with the theoretical expectation that plasma
etching is maximum at intermediate flow rates, at which the effect of plasma is highest
for the fibre surface.
479
Handbook of Polymer Blends and Composites
Figure 12.3 Contact angle between the fibre surface (Kevlar and graphite) and the
distilled water versus the PP-gas plasma exposure time of the fibres (Redrawn from the
data of [8])
Figure 12.4 Transversal tensile strength for PP/Kevlar and PP/graphite composites
versus the PP-gas plasma exposure time of the fibres (Redrawn from the data of [8])
480
Radiation Effects on Polymer Blends
Figure 12.5 Tensile strength of graphite fibres from an epoxy matrix versus the gas
flow rate; nitrogen and oxygen plasma pretreatment (Redrawn from the data of [8])
Figure 12.6 Tensile strength of graphite fibres from an epoxy matrix versus the exposure
time in nitrogen and oxygen plasma, for a constant gas flow (Redrawn from the data of [8])
481
Handbook of Polymer Blends and Composites
For a constant flow rate, the tensile strength significantly decreases with increasing the
treatment time, whatever the gas used for the plasma discharge (Figure 12.6). The
treatment time was not higher than 15 minutes, for 30 minutes of plasma exposure,
burning spots appearing on the fibre surface.
Figure 12.7 Interlaminar shear strength (ILSS) (a), and Youngs modulus (b) for
graphite fibre/epoxy composite versus the treatment time of the fibres in an oxygen
plasma (Redrawn from the data of [9])
482
Radiation Effects on Polymer Blends
Jang and co-workers [8] consider that the activated gases could be controlled in order to
obtain: the removal of a weak boundary layer, the reshape of the surface profile, chemically
active sites, grafting of functional groups to the fibre surface, a layer of plasma polymer
on the fibre surface. All of these possible effects of the plasma treatment could lead to
the improvement of a fibre/matrix interface.
The energy of the UV photons emitted by the plasma (especially in the case of an inert gas)
is high enough to lead to the breakdown of the chemical bonds, e.g., C-C, C-H, giving rise
to free radicals that can migrate along the chain, recombining between them. As a function
of the treatment conditions and of the nature of the treated polymer, chain scission, formation
of double bonds, and crosslinking are also possible [51-53].
Modifications of polymer systems after plasma exposure are listed in Table 12.2.
Treatment of iPP/LER/PP-g-GMA blends, having a LER content between 3.96 and 13.06
wt%, in an anomalous discharge in argon (13 Pa; discharge tension: 600800 V; exposure
time: 1, 5, 10 min) [18-20] showed an increase in the degree of crystallinity (Figure 12.8) and
in the free surface energy (Figure 12.9) compared to the untreated samples. Changes in the
crystallinity are also clearly shown by the results of optical microscopy in polarised light for
the treated samples, when compared to the untreated ones (Figure 12.10). Modification of
crystallinity and of crystallographic forms could arise from: LER action as a nucleating
agent, phase separation, chemcrystallisation, or local heating during plasma exposure.
It was also obvious that a basic character (high values of the electron donor parameter of
the free surface energy) was obtained for all the treated samples. This result correlates
well with the significant increase of the intensity of the absorption band corresponding
to the hydroxyl groups in infra red spectra (groups with a basic character [55]) after
plasma treatment.
Air-corona discharge treatment with different intensities of the electrode current was
applied for LDPE, LMDPE and their blend (LDPE/LMDPE; 80:20) [17]. A significant
increase of oxygen concentration can be observed in the surface layer after the corona
exposure, a higher current intensity leading to a higher oxygen concentration
(Figure 12.11).
483
484
Plasma exposure
concentrations) in LMDPE and
LDPE/LMDPE treated blend.
3 Isotactic polypropylene (iPP) (melt Argon plasma of an anomalous - Partial crosslinking of iPP; [19]
flow index (MFI): 2.46 g/10 min; discharge (13 Pa; 600, 700, 800 V; - Appearance of double bonds;
intrinsic viscosity in decaline, at 135 10 mA; 1, 5, 10 min). - Increase of the crystallinity index;
C: 0.909 g/cm3)/epoxy modified - Increase of the relative proportion
lignin (LER) (epoxy equivalent: between and crystallographic
0.065; dynamic viscosity: 17.500 cP; forms;
low molecular weight)/glycidyl - Increase of the refractive index;
methacrylate grafted polypropylene - Increase of the free surface energy;
(PP-g-GMA) (PP Himont 6823; flow - The surfaces have a basic character;
index: 0.5 g/10 min; density: 0.90 - Decrease of the carbon concentration
g/cm3) from C-C bond, and increase of
[O-C=O] and [C-O].
Plasma exposure
300 W; 27 Pa; 60 s) or
485
Handbook of Polymer Blends and Composites
Figure 12.8 Crystallinity degree of iPP and iPP/LER blends before and after the
treatment in an argon plasma (t = 5 min) versus the LER content in the blends
(Redrawn from the data of [18])
Figure 12.9 Surface free energy of iPP and iPP/LER blends before and after the
treatment in an argon plasma (t = 5 min) versus the LER content in the blends
(Redrawn from the data of [19])
486
Radiation Effects on Polymer Blends
untreated treated
iPP
(0 wt% LER)
Figure 12.10 Results of optical microscopy in polarised light for: iPP, and an iPP-based
blend with 9.08 and 13.05 wt%LER, respectively; x 600 magnification
487
Handbook of Polymer Blends and Composites
Figure 12.11 Oxygen percentage versus the intensity of the electrode current for:
LDPE, LMDPE, and an 80 LDPE/20 LMDPE blend (Redrawn from the data of [17])
Figure 12.12 Carbonyl (C4) percentage versus the intensity of the electrode current for
LDPE, LMDPE, and an 80 LDPE/20 LMDPE blend (Redrawn from the data of [17])
488
Table 12.3 Effects of ion beam bombardment on some polymeric materials
to 1 x 1015 ions/cm2, at room temperature, from both foil sides (in order to amplify the
optical changes) [10]. Immediately after the treatment, the irradiated PE was exposed, at
room temperature, for 12 hours, to a 2 wt% aqueous solution of alanine, in order to
obtain prospective materials for biological experiments. The concentration of the free
radicals produced by irradiation decreases after alanine exposure, for intermediate
irradiation fluences (Figure 12.13). This fact could be explained by the diffusion of alanine
into the radiation-damaged PE layer, where it is chemically bonded onto the double
bonds and onto the free radicals.
Under the influence of radiation such as fast electrons, - and X-rays, the energy transferred
to the molecules of the medium results in primary ionisations and excitations. After
absorbing energy, polyatomic molecules give rise to free radicals, responsible for most
chemical transformations observed in polymers [76]. The main effects of electron beam
irradiation [77] are, chain scission, oxidation, and unsaturation, depending on dose rate,
and oxygen content. Under irradiation, polymers generally undergo structural changes.
Figure 12.13 Dependence of the free radical concentration in the PE and PE/alanine
surface layer on the Xe+ ion fluence (Redrawn from the data of [10])
490
Radiation Effects on Polymer Blends
Depending on their chemical structures, some polymers crosslink (PE, PP, polystyrene
(PS), phosphazene polymers), while others degrade (polymethacrylates) [21, 24, 78].
Crosslinking of polymeric materials is followed by an increase in the tensile strength,
and hardness, while chain scission leads to a decrease in these properties [79, 80].
Radiation-induced crosslinking of PVC/epoxidised natural rubber (ENR) blends under
electron beam bombardment gives rise to the enhancement of the mechanical properties
of the blends [25].
High energy electron beams were first used in the 1950s, their applications ranging from
sterilisation of medical equipment to curing of coating or composites. Curing processes
can be better controlled using electron beams, than with thermal curing. Electron beam
curing is a fast non-thermal process which utilises highly energetic electrons at controlled
doses, to polymerise and crosslink polymeric materials. Electron beam curing can take
place at both ambient and sub-ambient temperatures. The advantages of this technology
[82] over conventional autoclave curing include significantly reduced curing time, leading
to order of magnitude improvements in throughput, ability to utilise low cost tooling
materials, reduced volatiles emissions, reduced energy consumption, and control over
curing energy absorption profile. Electron beam cured materials have been shown to
possess excellent mechanical properties, with low moulded-in stresses, high glass transition
temperatures and low void content. Foster Corporation from Dayville, Connecticut, USA
[83], uses electron beam irradiation of elastomer/polymer blends in order to improve
elongation, tensile and heat resistance, with applications in peripheral and angioplasty
balloons, catheter shafts, hubs and tips.
Electron beam pretreatment of polymers (such as PE, iPP) proved to be a very useful
technique for obtaining a functionalised polymeric surface, it being possible to bond the
treated surface with inorganic fillers and other plastic materials [84]. Stronger and tougher
PE materials were obtained by functionalising PE under electron beam bombardment,
without adding any monomers and auxiliaries in air; significant enhancement of the
compatibility of PE with inorganic filler was reached using this technique [84].
In Table 12.4, most important changes in the properties of the polymer blends, obtained
after EB irradiation are listed.
491
492
Electron beam
- PE/poly(dodecyl methacrylate-co- pronounced for PE/DMA-co-EMA
ethyl methacrylate (DMA-co- than for PE/BMA-co-MMA;
EMA); DMA:EMA 1:4, 16.5 - Significant increase of the
wt% PE crosslinked PE content;
- Changes of the morphology of the
PE/BMA-co-MMA with the
irradiation dose (from spherical
particles embedded in a matrix to a
sponge-like porous structure).
3 Plasticised blends of potato starch EB irradiation under nitrogen - Allylurea consumption; [22]
with urea or allylurea; brittle films flow (8 mA, 100 kGy per pass) - Increase in the rigidity;
of 100-200 m thickness; 20, 30, - Increase of Tg at higher irradiation
40, and 50 parts of additive per doses;
hundred parts of starch - Decrease of the soluble fraction
with the irradiation dose.
Table 12.4 Continued
No. Polymer blend/Composite Treatment conditions Observed changes after the treatment Ref.
4 Fluorocarbon rubber EB irradiation in air - Increase of the static and dynamic modulus with [23]
(vinylidene-fluoride-co-hexa- (0-500 kGy) the irradiation dose;
fluoropropylene-co-tetrafluo- - Increase of Tg with the dose;
Electron beam
roethylene (FKM) (F content - Decrease of the elongation at break;
68%, H 1.4%)/filler (HAF - Increase of the volume fraction of rubber with the
(particle size 29 nm) and irradiation dose.
SRF (60 nm) CB, silica
(precipitated, 30 nm), and
clay (soft, 1500 nm))
5 PE (HDPE 2480)/35 wt% EB irradiation (20 and - Decrease of the switch temperature corresponding [24]
CB (carbon black) 200 kGy, at 25 C to the heating and to the cooling phase;
and 170 C - Degradation of the polymer structure; much
damage and many defects at 170 C
- Increase of the crosslinking density.
6 70/30 PVC (grade MH66, EB irradiation (2 - Increase of the gel fraction with the irradiation [25]
Electron beam
11 CPP (2.5 mol% ethylene units, EB irradiation in air, at room - The tensile strength and the [80]
MFR 10)/Bionolle (grade 1010)/ temperature (2 MeV, 1 mA, 10 elongation at break: quite stable up
pelletised PP grafted maleic kGy/pass) to 80 kGy of irradiation dose;
anhydride (Modic) (as - Increase of the gel fraction with the
compatibiliser); 25/75, 50/50, and irradiation dose.
75/25 CPP/Bionolle
Electron beam
495
Handbook of Polymer Blends and Composites
Figure 12.14 Residual unsaturation versus the EB irradiation dose for acrylated and
methacrylated resins, in thick and thin geometry (Redrawn from the data of [84])
Figure 12.15 Glass transition temperature versus the EB irradiation dose for acrylated
and methacrylated resins, in thick and thin geometry (Redrawn from the data of [84])
496
Radiation Effects on Polymer Blends
of the gel content in IPN based on PE and methacrylate (MA) copolymers with increasing
irradiation dose (Figure 12.16) is attributed mainly to the decomposition of polymethyl
methacrylate (PMA). The fact that PS from a series of PS/poly(vinyl methyl ether) blends
does not offer any protection from radiation crosslinking in miscible blends was also
proved by the study on the effect of an electron beam irradiation on the gelation behaviour
of these blends [86]. By irradiation of the IPN, the content of the crosslinked PE
significantly increases with the dose (Figure 12.17), even at 200 kGy, almost 80 wt% of
the PE being already crosslinked. At higher doses, the degree of PE crosslinking remains
constant ( 90 wt%).
SEM micrographs of these IPN systems showed that PE forms the matrix, while the MA
copolymers represent the dispersed particles, which decompose under irradiation.
CB loaded semi-crystalline PE composites have been electron beam irradiated (20 and
200 kGy), at 25 C and 170 C [24]. These composites can exhibit a positive temperature
coefficient of resistance (ptc) (which means a network formation of the conductive CB
fillers), usually followed by a negative temperature coefficient (ntc) (the rearrangement
of the fillers within the low-viscosity polymer matrix above its melting point). For the
two irradiated samples, both the switch temperature around which a low ptc effect
switches to a higher one (TSh) or vice versa (TSc) shift to lower temperatures, when
Figure 12.16. Dependence of gel content on the irradiation dose for PE/BMA-co-
MMA IPN after extraction (Redrawn from the data of [21])
497
Handbook of Polymer Blends and Composites
Figure 12.17 PE content after irradiation and extraction versus the irradiation dose for
PE/BMA-co-MMA IPN (Redrawn from the data of [21])
compared with the non-irradiated system (Figure 12.18a and 12.18b). It was also observed
that the radiation-induced crosslinking occurs not only between the polymer chains, but
also between the chains and CB, and between the CB particles, to form CB aggregates.
At ambient temperature, the CB particles concentrate in the amorphous region,
corresponding to a crystal/amorphous polymer and CB particle three-phase system.
The radiation attacks the polymer structure, leading to degradation and increasing the
crosslinking density. The high dose irradiation at a temperature of 170 C caused much
damage and many defects in the polymer structure.
498
Radiation Effects on Polymer Blends
499
Handbook of Polymer Blends and Composites
Figure 12.19 Free surface energy of EB irradiated iPP/LER blends versus the LER
content, compared to the untreated samples (Redrawn from the data of [19])
Ionising radiation induces chemical reactions in polymer blends, that result in changes
both in molecular structure [87], and in macroscopic properties (such as: mechanical
properties [14, 32, 34, 39, 87-89], and interfacial adhesion [12, 14, 32]).
When oxygen is present, radiation-induced changes are often quite different compared with
irradiation under inert atmosphere [87]. Depending on transferred energy, dose rate, diffusion
coefficient of the oxygen in the components, significant differences in the radiation resistance
of blended polymers, the crosslinking rate can exceed the damaging one [32, 87].
500
Radiation Effects on Polymer Blends
a vacuum, inert fluid, or air (post-irradiation graft polymerisation). Gupta and co-workers
[94] grafted 2-hydroxyethyl methacrylate onto a PP monofilament by a simultaneous -
ray radiation technique, obtaining a biocompatible suture material.
Fibres resulted from graft polymerisation of acrylic acid onto polyamide 6 yarn were
additionally modified with penicillin, neomycin, or gentamycin, to obtain antibacterial
fibres to be used against Gram-positive and Gram-negative microorganisms [95]. In this
way, the release of antibiotics into solution proceeded for a quite long time, the antibiotic
remaining on the fibres still being enough to give them antibacterial properties.
Tg values of different PS/PMMA blends, for various irradiation doses, are presented in
Figure 12.20. The depression in Tg associated with increasing irradiation dose, whatever
the PS content in the blend could be attributed to -degradation of the PMMA phase.
However, the presence of higher concentrations of PS protects the blend against radiation-
induced degradation, which is shown by the increase in thermal stability over the 200-
340 C temperature range with increasing PS content, for a 50 kGy irradiation dose. In
this case, radiation crosslinking of PMMA polymer is predominant.
For higher doses (200 kGy), the PMMA component of the blend is subjected to oxidative
degradation, the thermal stability decreasing to higher extents.
On the basis of the kinetic analysis of the thermal decomposition of -irradiated PS/
PMMA blends (Figure 12.21), it could be concluded that the thermal stability of pure
PMMA decreases with increasing irradiation dose, while for PS, it increases with increasing
the irradiation dose. This behaviour is attributed to the different sensitivity of these
polymers towards -irradiation. Moreover, the fact that the activation energies of
decomposition of the blends are higher than those of the pure PMMA is a proof that PS
protects PMMA against oxidative degradation, at low doses.
501
502
506
Radiation Effects on Polymer Blends
The stability of PS against -irradiation and the pronounced oxidative degradation for
pure PMMA is also sustained by the variation in the intensity of the absorption bands
for a high PS content and for a high PMMA content (Figure 12.22).
Figure 12.22 Intensity of: a) C=O and b) C=C characteristic bands as a function of the
irradiation dose for PS/PMMA blends (Redrawn from the data of [31])
507
Handbook of Polymer Blends and Composites
Two types of PPO derivatives (p-t-BB-PPO, and B-PPO), blended with P2VN have been
60
Co -irradiated, at a dose of 50 or 200 kGy [88]. Tensile properties of the samples
drastically decreased with increasing irradiation dose, while their thermal stability was
not significantly affected for a dose value below 200 kGy.
The colour difference for PPO derivatives and their blends with P2VN before and after
-irradiation (Figure 12.23) showed, on one hand, the tendency to turn all of the surfaces
yellow and, on the other hand, a more stable light resistance for B-PPO than for p-t-BB-
PPO, for the blends having the same P2VN content (5 wt%).
508
Radiation Effects on Polymer Blends
Center for Irradiation of Materials from Huntsville, Alabama, USA [97], used Ion Beam
Assisted Deposition (IBAD) to create a compositional gradient at the coating-base material
interface, which increases the adhesion of the coating. IBAD is a combination of two
distinct physical operations, namely an ordinary physical vapour deposition on a substrate
and a simultaneous bombardment of the surface with a low energy ion beam.
During the last years, a considerable effort has been made by the members of the
Cooperative Research and Development Agreement (CRADA) [99] to develop new
materials. CRADA participants successfully developed and evaluated hundreds of new,
toughened and non-toughened EB-curable epoxy resin systems. Carbon fibre composites
manufactured using these systems were found to match favourably the performance of
conventional thermally-cured composites and were characterised by low void contents,
low water absorption values, high glass transition temperatures, and superior property
retention upon cryogenic and thermal cycling. Several toughened EB-curable resin
formulations have been evaluated for fibre polymer matrix composites.
Blends suitable for load-bearing applications have been obtained by -curing of 60/40 wt%
thermoplastic medium density polyethylene (MDPE) and elastomeric ethylene propylene
diene monomer (EPDM), used either in their gum form or loaded with two different
reinforcing fillers (HAF carbon black or precipitated SiO2 (Hi Sil) of concentration of
25-100 phr on the blend mass basis) [100].
Greater control of the crosslinking and chain scission reactions were achieved by varying
the -radiolysis temperature. The optimum temperature for the radiation crosslinking of
poly(tetrafluoro-ethylene-co-perfluoromethyl vinyl ether) (TFE/PMVE) was 10 C, at a
dose of 150 kGy [101].
509
Handbook of Polymer Blends and Composites
A new technique (ultrasonic irradiation) has been employed [104] to prepare poly(n-
butyl(methacrylate)) (PBMA/-Al2O3 nanocomposites, taking advantages of some of the
effects of ultrasounds, such as: dispersion, pulverisation, and activation. When subjected
to ultrasonic irradiation, n-butyl methacrylate (BMA) is polymerised to form poly(n-
butyl methacrylate (PBMA) latex without any chemical initiators, and the monomer
conversion reaches 72.5% in 25 minutes. Wizel and co-workers [105] also reported on a
composite material made of polystyrene and amorphous iron nanoparticles, under
ultrasound irradiation.
The effects of radiation utilisation in controlling the surface properties and obtaining
new materials is a very creative field, in a continuous progress.
References
4. A.G. Andreopoulos and P.A. Tarantili, European Polymer Journal, 1999, 35, 6,
1123.
5. S.I. Moon and J. Jang, Journal of Materials Science, 1999, 34, 17, 4219.
510
Radiation Effects on Polymer Blends
6. S.I. Moon and J. Jang, Composites Part A: Applied Science and Manufacturing,
1999, 30, 9, 1039.
7. J. Jang and S.I. Moon, Composites Science and Technology, 1999, 59, 4, 487.
8. B.Z. Jang, H. Das, L.R. Hwang and T.C. Chang in Interfaces in Polymer,
Ceramic, and Metal Matrix Composites, Ed., H. Ishida, Elsevier, New York, NY,
USA, 1988, 319.
9. J. Su, X. Tao, Y. Wei, Z. Zhang, L. Liu in Interfaces in Polymer, Ceramic, and Metal
Matrix Composites, Ed., H. Ishida, Elsevier, New York, NY, USA, 1988, 269.
11. S. Tan, G. Li and J. Shen, Journal of Applied Polymer Science, 2000, 77, 9, 1861.
12. Z. Yudong, C. Daitao, S. Nai, Z. Weijin and X. Duanfu, Gaofenzi Cailiao Kexue
Yu Gongcheng, 1999, 15, 5, 79.
13. A.E. Akinay and T. Tincer, Journal of Applied Polymer Science, 1999, 74, 4, 877.
14. A.E. Akinay and T. Tincer, Journal of Applied Polymer Science, 1999, 74, 4, 866.
511
Handbook of Polymer Blends and Composites
21. J. Hu, U. Schulze and J. Pionteck, Polymer, 1999, 40, 19, 5279.
23. I. Banik and A.K. Bhowmick, Journal of Applied Polymer Science, 2000, 76, 14,
2016.
24. X.S. Yi, J. Zhang, Q. Zheng and Y. Pan, Journal of Materials Science Letters,
1999, 18, 15, 1213.
25. C.T. Ratnam and K. Zaman, Polymer Degradation and Stability, 1999, 65, 3, 481.
26. C.T. Ratnam and K. Zaman, Die Angewandte Makromolekulare Chemie, 1999,
269, 42.
27. D. Zuchowska and Z.P. Zagorski, Polimery, 1999, 44, 7/8, 514.
28. Y.I. Matusevich, A.P. Polikarpov and L.P. Krul, High Energy Chemistry, 1999,
33, 4, 224.
29. Y.I. Matusevich and L.P. Krul, High Energy Chemistry, 1999, 33, 2, 72.
30. M.M. Nasef, H. Saidi and H.M. Nor, Journal of Applied Polymer Science, 2000,
77, 9, 1877.
31. K. El-Salmawi, M.M. Abu Zeid, A.M. El-Naggar and M. Mamdouh, Journal of
Applied Polymer Science, 1999, 72, 4, 509.
32. M.R. Ismail, M.A.M. Ali, A.A. El-Milligy and M.S. Afifi, Journal of Applied
Polymer Science, 1999, 72, 8, 1031.
33. N.I. El-Awady, Journal of Reinforced Plastics and Composites, 1999, 18, 15,
1367.
34. S.J.C. Miguez, E. B. Mano and C.M.C. Bonelli, Polymer Engineering and
Science, 1999, 39, 8, 1398.
36. Y. Hiroki and R. Toshihiko, inventors; Shin Etsu Chemical Company, assignee;
JP11,181,357, 1999.
512
Radiation Effects on Polymer Blends
39. M.A. Khan, M. Hossain and K.M.I. Ali, Journal of Applied Polymer Science,
1999, 74, 4, 900.
41. R.W. Venderbosch, H.E.H. Meijer and P.J. Lemstra, Polymer, 1994, 35, 20, 4349.
43. S. Zulfiqar and S. Ahmad, Polymer Degradation and Stability, 1999, 65, 2, 243.
44. S. Choe, Y.J. Cha, H.S. Lee, J.S. Yoon and H.J. Choi, Polymer, 1995, 36, 26, 4977.
45. S.Q. Gao, Y.L. Zhang, J.Z. Li and S.Y. Yang, High Performance Polymers, 1999,
11, 2, 167.
46. N.A. Mohamed and M.W. Sabaa, European Polymer Journal, 1999, 35, 9, 1731.
47. M.R. Wertheimer and H.P. Schreiber, Journal of Applied Polymer Science, 1981,
26, 6, 2087.
48. D.W. Woods and I.M. Ward, Surface and Interface Analysis, 1993, 20, 385.
49. B. Tissington, G. Pollard and I.M. Ward, Journal of Materials Science, 1991, 26,
1, 82.
50. C-M. Chan, Polymer Surface Modification and Characterization, Carl Hanser
Verlag, Munich, Germany, 1993.
52. E.M. Liston, L. Martinu and M.R. Wertheimer, Journal of Adhesion Science and
Technology, 1993, 7, 10, 1091.
513
Handbook of Polymer Blends and Composites
56. G. Marletta, Nuclear Instruments and Methods in Physics Research B, 1990, 46,
295.
58. J.W. Lee, T.H. Kim, S.H. Kim, C.Y. Kim, Y.H. Yoon, J.S. Lee and J.G. Han,
Nuclear Instruments and Methods in Physics Research B, 1997, 121, 474.
63. M. Gheorghiu, G. Popa and O.C. Mungiu, Journal of Bioactive and Compatible
Polymers, 1991, 6, 2, 164.
64. M. Gheorghiu, G. Popa, M. Pascu and C. Vasile in Metallized Plastics, Ed., K.L.
Mittal, Marcel Dekker, New York, NY, USA, 1998, 269.
514
Radiation Effects on Polymer Blends
67. E.C. Rangel, C.M. Lepienski and N.C. Cruz, Proceedings of the 15th
International Symposium on Plasma Chemistry, Eds., A. Bouchoule, J-M.
Pouveste, A-L. Thomann, J.M. Bauchine and E. Robert, Orleans, France, 2001,
Volume 6, 2445.
69. V. Svorcik, V. Rybka, R. Endrst and V. Hnatowicz, Materials Letters, 1996, 28,
4-6, 441.
72. L. Calgano, Nuclear Instruments and Methods in Physics Research B, 1995, 105,
63.
74. J. Won, M.H. Kim, Y.S. Kang, H.C. Park, U.Y. Kim, S.C. Choi and S.K. Koh,
Journal of Applied Polymer Science, 2000, 75, 12, 1554.
75. A.V. Rajulu, R.L. Reddy, K.M. Raju, D.K. Avasthi and D.K. Asokan, Nuclear
Instruments and Methods in Physics Research B, 1999, 156, 1-4, 195.
78. F.F. Stewart, R.E. Singler, M.K. Harrup, E.S. Peterson and R.P. Lash, Journal of
Applied Polymer Science, 2000, 76, 1, 55.
79. R.J.Woods and A.K. Pikaev, Applied Radiation Chemistry: Radiation Processing,
John Wiley & Sons, New York, 1994.
80. Zainuddin, M.T. Razzak, F. Yoshii and K. Makuuchi, Polymer Degradation and
Stability, 1999, 63, 2, 311.
81. V.S. Ivanov, Radiation Chemistry of Polymers, VSV, Utrecht, The Netherlands, 1992.
515
Handbook of Polymer Blends and Composites
82. http://www.e-beamservices.com/ebeam_ice.html.
83. http://www.fostercomp.com/services.html.
85. T. Glauser, M. Johansson and A. Hult, Polymer, 1999, 40, 19, 5297.
86. D.C. Mcherron and G.L. Wilkes, Polymer, 1993, 34, 19, 3976.
88. S.J. Lee, S.K. Kwon, W.J. Cho and C.S. Ha, Journal of Applied Polymer Science,
1999, 73, 9, 1697.
89. V. Kalkis, R.D. Maksimov and J. Zicans, Polymer Engineering and Science, 1999,
39, 8, 1375.
91. H.P. Brack, F.N. Buchi, M. Rota and G.G. Scherer, Polymer Materials Science and
Engineering, 1997, 77, 368.
92. S. Tan, G. Li, J. Shen, Y. Liu and M. Zong, Journal of Applied Polymer Science,
2000, 77, 9, 1869.
93. M. Ratzsch, H. Bucka, S.S. Ivanchev, A.M. Mesh and Khaikine, Journal of
Applied Polymer Science, 2000, 77, 4, 711.
94. B.D. Gupta, P.K. Tyagi, A.R. Ray and H. Singh, Journal of Macromolecular
Science Chemistry A, 1990, 27, 7, 831.
96. J. Sadlo, J. Bojarski, G. Strzelczak and J. Michalik, Polimery, 1999, 44, 3, 226
97. http://cim2.aamu.edu/FrontPage/activites.html.
98. http://searchpdf.adobe.com/proxies/0/15/92/76.html.
99. http://www.e-beamservices.com/ebeam_ice.html.
516
Radiation Effects on Polymer Blends
100. N.A. Shaltout, M.M. Abou Zeid, M.M. Abdel-Aziz and A.A. El Miligy, Polymer
International, 2001, 50, 3, 263.
101. J.S. Forsythe, D.J.T. Hill, A.K. Whittaker and A.L. Logothetis, Polymer
International, 1999, 48, 10, 1004.
102. A.M. Dessouki, N.H. Taher and M.B. El-Arnaouty, Polymer International, 1999,
48, 2, 92.
103. http://www.temarex.com/graft.html.
104. Y. Liao, Q. Wang and H. Xia, Polymer International, 2001, 50, 2, 207.
105. S. Wizel, S. Margel and A. Gedanken, Polymer International, 2000, 49, 5, 445.
517
Ageing of Polymer Blends and Composites
13
Ageing of Polymer Blends and Composites
Mihaela C. Pascu
Introduction
At the end of their useful lives, most polymer systems used in packaging, construction,
transportation, are concentrated in waste areas and pose environmental problems, such
as: storage, partial combustibility, a source of pollution, etc. All the properties that were
favourable during their period of use (stability, resistance and durability) operate against
their destruction [1].
Polymers in the environment undergo gradual degradation, due to the effect of different
factors, such as [2]: light (photodegradation); moisture, air, temperature (chemical
degradation); wind (mechanical degradation), and macro- and/or microorganisms
(biodegradation), most of them being coupled. After the degradation of polymer systems,
induced by the factors mentioned above, the materials that are obtained have different
characteristics when compared with those of the starting material [3].
519
Handbook of Polymer Blends and Composites
During shelf and service life, low molecular weight products diffuse from the polymeric
materials [9]. The diffusion of these products changes the properties of the material and
shortens the life-time.
This chapter discusses the behaviour of the polymer blends and composites under the
action of the environmental factors, such as: temperature, microorganisms, humidity,
solar radiation (UV being the most aggressive), metallic ions, etc.
In Table 13.1, changes shown in different polymer systems, aged in various conditions,
are presented.
Ageing experiments on allylurea (AU)-grafted starch films has been followed [19], both
in dry conditions (22% RH) and in wet atmosphere (92% RH), for 1 week. In the first
case, films became brittle, but after reconditioning for 1 week in an atmosphere with
58% RH, they regained their initial suppleness. When aged in a wet atmosphere, films
became extremely flexible and soft. After reconditioning at 58% RH, they regained their
initial suppleness, the transparency of the films being unchanged. Thus, it could be said
that the thermomechanical properties of these samples have been stabilised.
Lu and Isacsson [8] studied the artificial ageing of polymer modified bitumens, by three
different ageing procedures (Thin Film Oven Test (TFOT) (ASTM D 1754) [20], 163 C,
5 h; Rolling Thin Film Oven Test (RTFOT) (ASTM D 2872) [21], 163 C, 75 min;
Modified RTFOT (MRTFOT), 163 C, 75 min). Contents of carbonyl compounds and
sulfoxides increase by ageing. Increased complex modulus and elastic response after
applying the accelerated ageing procedures have been explained mainly by the oxidative
hardening of the bitumens. In the meantime, changes in compatibility (molecular
interaction) due to the oxidation of the bitumens may also influence the rheology.
Blends containing 40% PE and 60% plasticised starch with 30 wt% glycerol (40PE-SG)
or 50% PP and 50% plasticised starch with glycerol (50PP-SG) [2] have been compared
after biological ageing (soil test) and, respectively, chemical ageing (NaOH test).
40PE-SG and 50PP-SG blends, aged in soil during 4 months, showed a very large weight
loss (Figure 13.1), due to the fact that, during the soil ageing, 92% and 85% of plasticised
starch was removed from 40PE-SG and, respectively, 50PP-SG samples. Samples of pure
PE and PP, aged in similar conditions, demonstrated no significant weight changes (in
accordance with the results obtained by Hamid and co-workers [22]). Further weight
loss during chemical degradation with aqueous NaOH solution has been also observed.
520
Table 13.1 Changes in the properties of different polymer systems, aged in various conditions
Figure 13.1 Weight loss of polyolefin-modified starch blends after soil and NaOH
degradation test. Redrawn from the data in [2]
Tensile strength of both blends after the two accelerated ageing procedures changed only
insignificantly when compared with the initial samples (Figure 13.2a), while the elongation
at break has been characterised by an important increase, especially for 40PE-60SG,
after the soil test (Figure 13.2b). This behaviour was partially attributed to the changes
of physical structure of PO-starch blends caused by ageing.
Properties of CPP/Bionolle blends, irradiated with a 2 MeV EB, have been followed for
a 7 month indoor storage period [23]. After the ageing period, when the irradiation dose
was of 20 kGy, the tensile strength and the elongation at break, for the blends with a low
CPP content, remained almost the same as for the unaged samples. For an irradiation
dose higher than 50 kGy, both characteristics started to decrease gradually after 5 months,
due to the post-irradiation oxidation during storage.
Increasing CPP content in the blends led to very unstable elongation at break and tensile
strength during indoor storage, particularly for high irradiation doses. This signifies that
CPP, as the major component in the samples, stimulates the degradation, due to post-
irradiation oxidation of irradiated PP during storage [23].
524
Ageing of Polymer Blends and Composites
Figure 13.2 Relative changes of: a) tensile strength and b) elongation at break for
polyolefin-modified bitumens, artificial aged by three ageing procedures. Redrawn
from the data in [2]
525
Handbook of Polymer Blends and Composites
Ageing of plasma-treated polymeric surfaces (this ageing being equivalent with the
disappearance of the free radicals and of the functional groups of O-OH type from the
respective surfaces) could be slowed down by the increase of crystallinity and the
crosslinking of the surfaces [24]. In this way, the degree of ordering increases and,
subsequently, the mobility of the polymeric chains is reduced, leading to a decrease of
the ageing rate. Taking into account all these things, it was normal to observe this
behaviour also for the case of iPP/LER blends exposed in an anomalous glow discharge
in argon, and subsequently stocked at room temperature.
Figure 13.3 Variation of Sab and of S for iPP-based blends with: a) 3.96 wt% LER
and b) 9.08 wt% LER with ageing time, after plasma and EB treatment. Redrawn
from the data in [14]
526
Ageing of Polymer Blends and Composites
Increasing the ageing time (from 1 to 7 days, after this period the samples, plasma exposed
or EB bombarded, became stable), a loss of their wettability has been shown [17].
Variation with the ageing time of Sab (acid-base component of the free surface energy)
and S (free surface energy) indicated different behaviours for the studied samples, as a
function of the LER content. For blends with a low LER content, both Sab and S decrease
with the storage time after 7 days of ageing, their values approaching those corresponding
to the untreated sample (Figure 13.3a). This variation is probably due to the rotation of
the polar groups in the sample bulk or to the migration of low molecular fragments
towards the surface, thus implying the reduction of the interfacial tension [24] between
the surface and the storage medium.
For the blends with a higher LER content (Figure 13.3b), S obtained after the treatment
remains almost constant during ageing. The decrease of the ageing rate could be explained
by a more pronounced crosslinking of these blends, the crosslinking efficiency being
known to decrease with increasing PP content in the mixture [25].
13.2 Photodegradation
Ozone-generating UV light has been used for many years to clean organic contaminants
from various surfaces [2628], to modify the mechanical properties [29], to increase the
wettability of different polymer surfaces [30], as well as their adhesive properties and
dyeability [31, 32] UV treatment of surfaces was also used [33-35], to accelerate ageing
of polymeric materials by increasing the rate of the chemical reactions (in this way the
polymers became more environmentally friendly materials).
Meligi and co-workers [36] found that butyl rubber (BR) and polyisoprene (PI) are effective
in enhancing oxidative degradation of PP during EB irradiation in air (1 MV; 1 mA) and
during Xe exposure (using a Sunshine Xenon Life Weather Meter, at 30 C) after EB
irradiation. Thus, the presence of a small amount of rubber enhances oxidation and
chain scission of PP during and after EB irradiation. It was therefore concluded that this
technique may be useful to prepare polymers that are degradable in the natural
environment, after use.
For a polymer blend, photo-oxidative ageing induces chemical changes in the blend matrix,
as well as modifications of the aspect of the material surface [37, 38]. Heterogeneous
materials under strictly controllable conditions are obtained [29].
UV irradiation leads to the formation of oxidised products (mainly carbonyl and hydroxyl
groups) [37, 39] that determine changes in the macromolecular polarity, as well as
interactions between functional groups (including hydrogen bonding) near the interface
527
Handbook of Polymer Blends and Composites
of both polymers. In the meantime, during UV action, a small amount of copolymer can
be formed. These ones reduce the interfacial tension between the components of the
polymer blend, a better adhesion between them being, in this way, achieved.
The presence of polystyrene (PS) prolonged the induction period of polyvinyl methyl
ether (PVME) oxidation. The molecular weight of PVME decreased more slowly as the
PS content was increased [40]. A possible explanation for this behaviour is that, during
the induction period, a reaction between PVME radicals and PS took place, through the
abstraction of the tertiary atoms of PS. The resulting PS radicals were less reactive in
chain propagation and led to the retardation of the oxidation reaction.
On the other hand, UV radiation can also have negative effects on the polymer surfaces,
leading to the degradation of their physical properties, especially in the case of polyolefins,
PVC, polycarbonate (PC) [41], due to the appearance, during polymer degradation, of
generally low concentrations of chemical groups, i.e., oxidised groups [42].
Photodegradation of polymers is, in most cases, initiated by abnormal groups and defects
which can be introduced during their manufacture and which, even in a very low
concentration, may provide initiation points for peroxidation.
The most important goal of the studies involving exposure of plastics to UV light is to
assess the stability or durability of the material under well-defined conditions. A reliable
lifetime prediction capability would be highly desirable for both assuring the durability
of commercialised products, as well as for selecting new materials or formulations under
current development and evaluation, for use in long-term applications [43]. The durability
is, usually, firstly assessed on artificial exposure, expecting that, from the obtained results,
it will be possible to predict the lifetime of the material in natural conditions. However,
the correlation between various artificial exposure tests, e.g., WeatherOMeter, Xenontester,
Suntester, UVCON, QUV, SEPAP, and natural weathering involves several important
difficulties [43-46]. The most obvious concerns the presence of short-wavelength radiation
in the artificial exposure device, i.e., radiation below the 295 nm cut-off of sunlight on
earth. Choosing an adequate light source and/or using appropriate filters, i.e., filtered
xenon arcs, this aspect can be easily removed.
A high temperature during artificial exposure is another factor which makes difficult to
satisfactory correlate this one with natural weathering. This can be, however, considerably
reduced by ventilation and/or cooling.
528
Ageing of Polymer Blends and Composites
On the other hand, photo-oxidation will dominate during natural weathering, while
photolysis reactions may be very important under artificial exposure, due to the different
oxygen diffusion rate for these two cases.
On natural exposure and with some artificial exposure devices, photo-oxidation involves
preferentially the superficial layers of the films and, only to some extent, the bulk of the
material. With some accelerated exposure devices, oxidation of the surface layers becomes
negligible when compared with the oxidation of the bulk.
In view of the poor reliability of lifetime prediction based on various accelerated test
methods, alternative methods must be developed.
The experimental observation that, very often, the polymer quality is much less influenced
by the weathering conditions, for well-stabilised films, it means that it can be expected
that, in the future, the main factor determining the lifetime will be the stabilisation systems
[44]. This one can enhance the physical and aesthetic properties and can improve the
usage characteristics and service life of the finished articles [47].
In Table 13.2, there are listed the changes, observed by several authors, in polymer
systems, subjected to light action.
Photo-oxidative degradation of blends of PET and PPO was studied [39], considering
the mutual influence of both components. The diffusion coefficient of the chain segments
carrying the reactive groups (D) and the capture radius for radical recombination (r0),
for different ratios between the components, in the temperature range of 30100 C, are
presented in Figures 13.4 and 13.5. While r0 decreases with increasing temperature, D
values are low at low temperatures, after that increasing rapidly for all the blends (at 85
C, D presents a sharp increase). For the blends with a higher PET content (> 60 wt%),
the increase in D values is smaller with increasing temperatures, comparing with the
blends with a low PET content.
529
530
Figure 13.4 Dependence of the chain segment carrying the reactive group of UV-
irradiated PPO/PET blends on the temperature. Redrawn from the data in [37]
Figure 13.5 Dependence of the capture radius for radical recombination of UV-
irradiated PPO/PET blends on the temperature. Redrawn from the data in [37]
532
Ageing of Polymer Blends and Composites
It is well known that most polyolefins are resistant polymers to photo-oxidation. That is
why, their degradation products accumulate, in time, as waste, leading to problems
concerning the pollution, especially for the soil. Even when photo- and biodegradable
materials are produced, based on blends of natural biodegradable polymers (such as
lignin, cellulose, starch)/synthetic non-biodegradable polymers (such as polyolefins), the
degradation is slowed down due to the protective action of the thin outer layer of
polyolefins, for the films obtained. However, Kosikova and co-workers [53, 54] mentioned
that the blends with a high lignin content (> 8-10 wt%) present an appropriate rate of
photo- and/or biodisintegration.
It could be observed that, after 200 hours of irradiation, all the films containing LER/PP-g-
GMA were totally disintegrated. In the meantime, their colour changed from transparent green-
yellow to yellow or dark yellow, with increasing the LER content and the exposure time.
Variation of the free surface energy with the irradiation time is shown (Figure 13.6), for
the blends with a low LER content, values only a little higher than those corresponding
to the unexposed films. After 100 hours of UV irradiation, S significantly increases. The
blends with a high LER content are characterised by high values of S for the entire UV
exposure time domain; this behaviour is always found at the polymeric surfaces, unstable
to photo-oxidation. The increase of S with the irradiation time could be explained by
the formation of hydroxyl, carbonyl, and carboxyl groups [58], that increase the
hydrophilicity of the photooxidised films. The appearance of these groups have been
also proved by IR spectroscopy results.
13.3 Biodegradation
Biodegradation is defined, by Albertsson and Karlsson [59], as an event which takes
place through the action of enzymes and/or chemical decomposition associated with
533
Handbook of Polymer Blends and Composites
Figure 13.6 Variation of the free surface energy for IPP and IPP/LER blends with UV
irradiation time. Redrawn from the data in [57]
living organisms (bacteria, fungi, etc.), and their secretions. Abiotic reactions like
photodegradation, oxidation, and hydrolysis may also alter the polymer before, during,
or instead of biodegradation, due to the environmental factors. According to FDA test
protocols, an organic compound can be presumed biodegradable if more than 50% of
its carbon content is converted to CO2 during the test exposure [60].
534
Ageing of Polymer Blends and Composites
Both soil burial and bioreactor exposure experiments [64] showed a decrease in percentage
elongation and considerable weight loss for starch-based plastics. Moreover, the presence
of starch particles as filler within the LDPE matrix enhances substantially the gas
permeability (very important in terms of food preservation) of the LDPE/starch composite
structure, because of their strong hydrophilic character. Used as a rigid filler in polymeric
systems [60], starch generally leads to an increase in stiffness with corresponding decrease
in tensile strength and elongation at break. Chitosan-PVA blends, blends based on
gelatinised starch and 1,4-trans-polyisoprene (Gutta Percha), polysaccharide-cellulosic
blends, as well as polysaccharide-protein (as collagen and its derivative, gelatin) also
proved to have biodegradability characteristics [64].
Though soil composition usually varies from place to place, the soil burial test reflects
more than any other test, the real-life conditions. Table 13.3 presents the main changes
in the properties of different polymer blends buried in various soils or subjected to other
biodegradation tests.
Composite films of PBSA (blown film grade 3001) and starch (mainly composed of
amylose and amylopectin) have been buried in test reactors containing 50 g of a standard
soil mix (10:10:1 w/w/w mix of potting soil, sand, and composted manure; pH ~ 7),
maintained at 60% water-holding capacity and incubated in a controlled environment
chamber, at 30 C [60].
The average molecular weight of PBSA/starch samples, after 28 days of soil test exposure
(Figure 13.7), showed that the addition of starch to PBSA resulted in a more rapid decrease
in the molecular weight of PBSA. In the meantime, increasing the starch content in the
blends to as much as 20% resulted in enhanced rates of mineralisation and, hence,
significant decreases in polymer half-time. After 25%30% starch, no further
enhancement in biodegradability was observed. The numerous interconnecting holes
and fractures throughout the film surfaces, with fractures radiating from the holes that
contained starch granules, observed after 110 days of soil burial, showed that starch is
preferentially degraded, its incorporation in the blends enhancing degradation of the
PBSA component (first in the amorphous zones).
535
536
1. CPP (2.5 mol% ethylene EB irradiation of the blends (2 - Significant decrease of the tensile [23]
units)/Bionolle (grade 1010)/Modic MeV), followed by soil burial test strength and elongation at break
(as compatibiliser). in a mixture of 1/3 fermented with the burial time.
leaves, 1/3 pound soil, 1/3 garden - Significant weight loss after 3
soil; 27 5 C; moisture content
3. LDPE (Borealis; MFI = 2 g/10 min, Exposure to activated sludge in a - Increase of the weight loss with the [65]
at 190 C)/fatty-acid esters waste water treatment facility of ester amount and the exposure time.
(octanoated (OC), dodecanoated Haitoglou Bros. Company
(DOD), octadecanoated (OCD)) of (8 weeks).
amylose and starch (19% amylose
+ 81% amylopectin); 10, 20, 30,
40, 50 w of esters; 130 m
Figure 13.7 Average molecular weight of PBSA/starch samples before and after the soil
burial test in function of starch content in the blend. Redrawn from the data in [60]
Takasu and co-workers [62] followed soil burial degradation (in a soil composted for
more than 10 years; pH = 7.17.4; 70%80% relative humidity, 27 C) of PVA/chitin
derivative (chitin-graft-poly(2-ethyl-2-oxazoline)) from crab shell miscible blends. The
weights of the recovered blends decreased with the burial time (Figure 13.8), in relation
to the chitin content in the blends.
Scanning electron microscope (SEM) micrographs of the blends studied, after soil burial,
showed the existence of a number of holes with round shape bacteria. Biodegradation
took place on the surface of the film, the inner part of the recovered sample not being
degraded by bacteria and Actinomycetes.
The observed depressions of both Tg and Tm were proof of an increased mobility of PVA
segments, which might promote the biodegradation.
The conclusion of the authors [62] was that molecularly miscible hybridisation with
biodegradable polymer (thus enhancing the decay of the bulk structure) represents an
effective strategy to improve the poor degradability of pure PVA.
538
Ageing of Polymer Blends and Composites
COP/LER sheets, of 0.85 mm thickness have been exposed for two months in sewage
sludge with a complex of bacteria, after UV irradiation, for 50 hours, at 30 W [68]. The
small increase of the mass of the blends with increasing the exposure time (from 0.5 to
1 wt%/month), as well as the increase of their thickness (with about 10%-12%) have
been attributed both to the loss of LER or of some degradation products, and also to the
absorption of water and of other components of the sewage sludge. The accumulation of
micro-organisms on the sample surfaces must also be taken into account. The water
absorption in the samples favours the degradability starting with the hydrophilic
539
Handbook of Polymer Blends and Composites
Figure 13.9 BOD of PHB/PVA blends, carried out for 30 days in river water, versus
the PHB content in the blend. Redrawn from the data in [67]
component (LER) in the blend, the absorption phenomena being, thus, more important
for higher LER contents.
Polymer composite materials for aerospace applications are based on carbon (obtained
by oxidation and carbonisation of poly-acrylonitrile organic fibres), organic (poly-n-
amido-benzimidazol) and glass (T-10 cloth) fibres and epoxy binders. They were exposed
for 10 years under the warm humid climatic conditions of the city of Batumi on the
Black Sea coast [66]. It was found that, after one year of exposure in these conditions,
the mechanical properties, (e.g., tensile strength, bending strength, compression strength,
Youngs modulus, dynamic shear modulus), differed from the initial values by no more
than 30%. This behaviour has been explained, by Startsev and co-workers, by the
superposition of the effect of plasticisation with water and post-curing of the binder
[66]. Mechanical properties of these materials are determined, during the exposure in
warm, humid climatic conditions, both by the change in the interaction at the binder-
540
Ageing of Polymer Blends and Composites
fibrous filler boundary, as well as by the structural changes of the binder. In the meantime,
the inhomogeneity of climatic ageing across the thickness of the samples was obvious;
during ageing, the outer layers of the material have been, primarily, affected, the properties
of the inner ones being closer to the initial ones.
Studies made on iPP/LER/PP-g-GMA blends showed [69] the dependence of the surface
properties, and of the hydrophilicity on the lignin content. The hydrophilicity increase
could lead to a tendency towards biodisintegration for the blends containing
lignosulfonates. In the case of sample disintegration, the natural polymer could act by
water absorption, but the most important role is to favour the activity of the
microorganisms.
It was shown that the microbial colony extends more on the surfaces of the films with a
higher LER content and a longer soil burial time.
Elongation at break decreases for all the biodegraded samples (iPP and its blends with
LER), while the tensile strength decreases only for the blends, that are, therefore, more
susceptible to degradation than iPP.
IR spectroscopy results showed additional absorption bands when compared with the
undegraded samples; these bands are attributed to the groups containing oxygen, that
appear during the environmental degradation of lignin and/or iPP.
Even if the biodegradability of the iPP-based films increased after lignin incorporation,
only natural polymer incorporation is not sufficient for obtaining materials characterised
by a satisfactorily degradation rate, due to the protective covering of the film obtained
by using polyolefins. Another solution that could eliminate this inconvenience might be
the UV-irradiation of the mixtures of synthetic polymer/natural polymer. This procedure
determines the superficial attack of the respective blends and thus induces their
biodesintegration. To prove this behaviour, iPP/LER/PP-g-GMA blends, UV irradiated
under the previously mentioned conditions (see Section 13.2), have been used to test the
terrestrial growth of plants, by soil burial, in greenhouse conditions [70].
The surface of the photooxidised films became, after soil burial, very rough and
heterogeneous, due to the modification of the functional groups of the components (shown
by IR spectroscopy). The decrease of the free surface energy after soil burial, compared
to samples that have only been UV irradiated, is due to the fact that a part of the
photooxidised fragments have been lost during soil burial, by the depletion of the
superficial layer.
The significant increase of the hydrophilic character of the polymer blends plasma exposed
or EB bombarded, compared to the situation of UV irradiation (Figure 13.10) could be a sign
541
Handbook of Polymer Blends and Composites
Figure 13.10 Values of the free surface energy in function of the LER content in
IPP/LER blends, after different applied treatments (personal unpublished data)
for the possibility of using these two types of treatment to increase the environmental
degradation rate of the polymeric materials. This supposition could be explained by the fact
that an increased hydrophilicity implies water absorption at the surfaces of the treated
materials, and therefore favours the development of microorganisms on the respective samples.
Evidence that the chemical reactions that take place in multi-component polymer systems
lead to different reaction patterns, with possible interactions between the macromolecular
components or with their degradation products, was proved by investigation of the thermal
degradation behaviour of polymer blends. To obtain rapid information on thermal
degradation, materials are subjected to accelerated thermal ageing, following the time
dependence of weight loss, mechanical or electrical properties. The results obtained may
offer information on the lifetime at the operating temperature. The lifetime is often predicted
[71] by the procedure recommended by IEC-60216-3 [72]. An alternative procedure used
for the study of the chemical stability of the polymers is the oxygen uptake, which provides
542
Ageing of Polymer Blends and Composites
information on the oxygen amount in a certain oxidation process [73]. During thermal
degradation, interactions between the components of the polymer blends, with low Mw
products or with free radicals may take place [3]. When no interactions are present, the
degradation behaviour of a blend is comparable with that evaluated by additivity rules
between those corresponding to the pure components [74], i.e., for PMMA/PS blends.
In some cases, the chemical interactions produce destabilising effects (leading to acceleration
of the degradation rates) (for example, in PVC/chlorinated rubber, PVC/PVA, PVC/ poly
ethylene-co-vinyl acetate (EVA) [3], PVC/polyamide 6 (PA 6) [75], PVC/polystyrene-
acrylonitrile copolymer (SAN), EVA/SAN [76], PVC/PMMA, PVC/high impact PS [77]),
while in other situations, the obtained result is a stabilising one (such as in PMMA/
polychloroprene, PVC/polybutadiene rubber (PBR), and blends of PS with PVC,
polyvinylidene chloride (PVDC), PVA, PI [3, 77]). The resulting behaviour [3, 74] depends
on the blend morphology and composition, as well as on temperature and on the difference
in degradation temperatures of the mixture components, such as in blends of PS with
PVDC, polybutadiene (PB), PPO. When degradation rates are slowed down or
decomposition temperatures are shifted to higher values for all the blend components, i.e.,
in PVC/PS blends [74], an optimum stabilising effect is obtained in the blend produced. If
one of the blend components is stabilised and the second one is not negatively affected (its
decomposition pattern remains the same in the blend as when it is alone), the blend can be
considered to be stabilised.
The degree of miscibility of the blends affects the degradation stability of the polymeric
systems, stabilising or destabilising effects being obtained. In the meantime, the presence
of compatibilisers influences the thermal stability of the blends, due to the different
interactions that may develop at the phase boundary between the components.
Varughese [78] found that in PVC/ENR blends, the thermal stability is much influenced by
the interaction between PVC and ENR. By varying the blend composition and with
appropriate selection of the blend components, the thermal stability and physicochemical
properties of the blends can be varied.
Blends of ethylene methyl acrylate copolymer and polydimethyl siloxane (PDMS) rubber
exhibit synergism in thermal and physicochemical properties [79].
Thermal stability of ternary blends of PVC, EVA and SAN is very good when compared to
pure PVC and binary blends [80]. In PVC/EVA/SAN blends, a thermodynamically miscible
region is formed, in which PVC acts as an interfacial agent for the immiscible SAN/EVA
system.
543
Handbook of Polymer Blends and Composites
In the thermal oxidation of macromolecules, new chemical groups are formed and changes
in molecular weights may occur at temperatures well below the onset of weight loss, as
a result of the reaction with oxygen [82].
Added at a level of a few parts per thousand, antioxidants can ensure the stabilisation of
a polymer during processing. In this way, degradation by crosslinking or by chain scission
of the macromolecules is inhibited, discolouration is reduced and materials are obtained
with well-controlled performances. The most used antioxidants are sterically hindered
phenols, phosphites and phosphonites, secondary aromatic amines and thioethers.
Decrease of the storage modulus [83-85], variation of the mechanical properties [63, 86-
88], changes in chemical nature and in morphology [3, 89], phase separation [89-91],
decrease of the molecular weight [71, 92] are some of the usually observed effects of
temperature on polymer blends (see also Table 13.4).
1) At 185 C (Tg + (253 C)), for 30 minutes, and then cooled down to room temperature
(samples denoted as refreshed);
2) At 145 C (Tg - (253 C)), and then slowly cooled to room temperature (samples labelled
as aged).
Study of the viscoelastic characteristics of the studied blends showed, besides the relaxations
associated with the dispersed and continuous phases, (i.e., - and -relaxation mode, respectively),
544
Table 13.4 Effect of temperature on the properties of polymer systems
No. Polymer blend/composite Accelerated ageing Observed changes after the ageing Ref.
1. LDPE/PP and LDPE/high density Thermomechanical degradation - Significant improvement of the [3]
[96]
[95]
Maurer and co-workers [97] as a
micromechanical transition. This
0.5 PVC/ENR/NBR
9.
547
Handbook of Polymer Blends and Composites
Figure 13.11 The ratio of the amplitudes for - and micromechanical transition for the
refreshed and aged samples in function of the frequency. Redrawn from the data in [86]
Activation energy for thermal degradation data for PVC-siloxane (PDPS and PDMDPS)
[98] increased with the siloxane content (Figure 13.12). This is due to the fact that
polysiloxanes are substantially more stable than PVC, so that increasing the quantity of
PDPS and PDMDPS in the blends gradually increase their thermal stability.
When used at temperatures higher than their Tg, a drastic reduction of the storage modulus
for PHB/PEO blends was observed. This was explained [92] by a long-range co-ordinated
motion of the molecules.
Formation of large membered rings in the ENR molecules in PVC/ENR blends led to an
increased rigidity of these molecules [96]. In ternary blends of PVC/ENR/NBR, at 80 C,
548
Ageing of Polymer Blends and Composites
Figure 13.12 Activation energy for thermal degradation for PVC/siloxane blends in
function of PVC content in the blend. Redrawn from the data in [98]
the formation of these ring opened structures is reduced, maybe due to the better miscibility
of NBR with PVC in this blend. Such protection was not observed at higher temperatures,
i.e., 140 C, as the ring opening is initiated by HCl.
13.5 Restabilisation/Recompatibilisation
At the end of its useful life, a thermoplastic material becomes waste. The main problem
in post-consumer plastics recycling is due to the chemical damage such as degradation
undergone by the polymers and by the stabilisation system used during processing steps
[99]. Newly formed functional groups during ageing enhance the sensitivity of the recyclate
to thermal and photodegradation [100]. Recycling of plastic waste includes several
technologies among them, reprocessing is preferred from economical and environmental
point of view. There are two main ways to reduce the negative effects of degradation in
various recycling steps [99]:
1. restabilisation;
549
Handbook of Polymer Blends and Composites
13.5.1 Restabilisation
As most plastics are degraded during ageing, they need restabilisation. Restabilisation of
post-consumer recyclates is necessary for effective protection against the detrimental
influence of shear forces and heat during processing, against UV light and oxygen to
modify the processing and ageing behaviour, while for outdoor service, besides this
protection, a supplementary quantity of stabiliser is required. The concept of restabilisation
and repair has been introduced to provide recyclates with similar properties to their
original use (closed loop) as a potential substitute of the virgin material [99].
In order to have a useful restabilisation process, a correct evaluation of the entire polymer
blend and of the selected stabilisers has to be made [101]. Properties of recyclates are
influenced by primary processing and application, as well as by the recovery source.
Due to the fact that recycled materials degrade faster than the equivalent virgin polymers,
special stabiliser systems are required. Addition of stabilisers to post-consumer plastics
improves the stability of the polymers during their second life, by maintaining their
mechanical properties, and thus contributes to the extension of their useful lifetime.
Several stabiliser systems used for recyclates are given in Table 13.5.
550
Restabilisation systems Polymer wastes
Table 13.5 Restabilisation systems for several mixed polymer wastes [102]
Waste type Stabiliser used
PE/EPDM bumper - Basic stabilisation + low and high
molecular weight hindered amine
stabiliser (HAS).
- Selected oxirane compounds.
PE/PP bottle fraction 0.2% Recyclostab 451
HDPE body + PP caps + PET and PVC as Recyclostab 451
impurities
Mixed PO + carbon black - Recyclostab 451 (9 antioxidants +
costabilisers).
- HAS + Chimasorb 944 or Tinuvin 783
(blend of antioxidants).
- Tinuvin 622.
- Recyclostab 550 costabilisers + HAS.
PE/PA film coextruded film from food HAS + selected antioxidants
packaging
LDPE/PA 6 Mixture Irganox 1098/Irganox
1078/Irgafos 168/Chimassorb 944
HDPE, PO, mixed plastics Recyclostab 411 (antioxidants +
costabilisers)
PP, PO blends Recyclostab 451
Mixed plastics: 55%-60% PO + 15%-20% - Recyclostab 811.
styrenics + 5%-8% PVC + 8% PET - Aromatic phosphite P-10.1% + 0.05%
AO-1 + 0.1%-0.2% (HAS-1 + HAS-2/LS-
1 + antioxidants (calcium salts of organic
acids and/or synthetic hydrocalcite +
octadecyl-3-(3,5-di-tert-butyl-4-
hydroxyphenyl)propionate.
Recyclostab a tailor-made stabiliser system, developed by Venture Group Additives
for Plastic Recycling from Ciba, that elaborated new solutions, new stabiliser
formulations for materials and products from recyclates. Irganox, Chimassorb, Tinuvin,
HAS light stabiliser systems [99]
551
non-reactive Compatibilisers recycling Plastic waste Twin-screw extruders
Hydrocarbon polymers (PE, - PE, PP, EPDM, PS grafted with: maleic [105]
PP, PS)/polar polymers anhydride, acrylic acid.
(PET, PA) - Copolymers of propylene grafted with acrylic
acid.
- PS grafted with: maleic anhydride; trimetallic
anhydride.
Concluding Remarks
Under the influence of different environment factors (light, moisture, air, temperature,
wind, macro- and/or micro-organisms), polymer systems undergo gradual degradation.
Thus, materials with different properties to those of the starting materials are thus,
obtained.
552
Ageing of Polymer Blends and Composites
In order to obtain information on the behaviour of polymer blends and composites during
natural ageing, studies for predicting the long-term effects of ageing from relatively short-
term experimental data are developed. NASA Langley Research Center [107] uses a
predictive methodology, the long-term response to physical ageing of some advanced
composites used in the aerospace industry being calculated based on the viscoelastic
parameters and ageing shift factors that were measured in the short-term laboratory
tests. Research carried out by Polycert Ltd. [108], is focused on merging of polymer
physics mathematical algorithms and computer software as well as non-destructive
standard tests according to ISO and ASTM techniques, in order to create expert systems,
for improving the speed and the quality of polymer formulation and polymer article
development. The main research objective of the Center for High Performance Polymeric
Adhesives and Composites [109] is to establish and integrate fundamental concepts for
the design, synthesis, processing, and manufacturing of polymeric adhesives and
composites in order to allow correlations and predictions of macroscopic physical
behaviour in engineering applications.
For prolonging the useful lives of plastic materials, these ones can be recycled, becoming
thus acceptable for second hand applications. Restabilisation/recompatibilisation is a very
effective approach to improve the quality of any recyclate, this one substituting the virgin
plastic, with comparable performances, in selected applications (non-pressure pipes, cable
ducts, waste bins, crates, bottles, drums, pallets, playground material, plastic lumber, etc).
References
553
Handbook of Polymer Blends and Composites
7. Y.S. Lipatov, V.F. Rosovitsky, T.T. Alekseeva and N.V. Babkina, Polymer
International, 2000, 49, 334.
10. A. Jha, B. Dutta and A.K. Bhowmick, Journal of Applied Polymer Science, 1999,
74, 1490.
11. F. Le Lay and J. Gutierrez, Polymer Degradation and Stability, 1999, 64, 397.
12. J. Niziol and J. Laska, Synthetic Materials, 1999, 101, 1-3, 720.
15. C.T. Ratnam and K. Zaman, Polymer Degradation and Stability, 1999, 65, 481.
20. ASTM D1754-97, Test Method for Effect of Heat and Air on Asphaltic Materials
(Thin Film Oven Test), 1997.
554
Ageing of Polymer Blends and Composites
21. ASTM D2872-97, Test Method for Effect of Heat and Air on Asphaltic Materials
(Rolling Thin Film Oven Test), 1997.
22. S.H. Hamid, M.B. Amin and A.G. Maadhah, Handbook of Polymer
Degradation, Marcel Dekker, New York, NY, USA, 1992.
23. Zainuddin, M.T. Razzak, F. Yoshii and K. Makuuchi, Polymer Degradation and
Stability, 1999, 63, 311.
26. M.J. Walzak, S. Flynn, R. Foerch, J.M. Hill, E. Karbashewski, A. Lin and M.
Strobel, Journal of Adhesion Science and Technology, 1995, 9, 1229.
27. J.M. Hill, E. Karbashewski, A. Lin, M. Strobel and M.J. Walzak, Journal of
Adhesion Science and Technology, 1995, 9, 1575.
30. P. Hedenberg and P. Gatenholm, Journal of Applied Polymer Science, 1996, 60, 2377.
33. L. Tong and J.R. White, Polymer Engineering Science, 1997, 37, 321.
34. R. Gallo, F. Severini, S. Ipsale and N. Del Fante, Polymer Degradation and
Stability, 1997, 55, 199.
36. G. Meligi, F. Yoshii, T. Sasaki, K. Makuuchi, A.M. Rabie and S.I. Nishimoto,
Polymer Degradation and Stability, 1997, 57, 241.
555
Handbook of Polymer Blends and Composites
38. B. Mailhot, S. Morlat and J.L. Gardette, Polymer, 2000, 41, 1981.
40. L. Stoeber, E.M. Pearce and T.K. Kwei in Polymer Durability. Degradation,
Stabilization, and Lifetime Prediction, Eds., R.L. Clough, N.C. Billingham and
K.T. Gillen, American Chemical Society, Washington, DC, USA, 1996, 503.
41. F. Macchi, Etude des Effets de lirradiation Electronique sur les Proprietes des
Copolymeres Ferroelectriques P(VDF-TrFE), 1990. [PhD Thesis]
47. I. Bachert and F. Werner, Plastics Additives and Compounding, 2000, 2, 9, 18.
48. A.L. Tisler, K.O. McElrath, D.S. Tracey and M.F. Tse, Proceedings of the ACS
Rubber Division Meeting, Cleveland, OH, USA, Fall 1997, Paper No.66.
50. A.K. Mubarak, H. Mojammel and A.K.M. Idriss, Journal of Applied Polymer
Science, 1999, 74, 4, 900.
556
Ageing of Polymer Blends and Composites
55. J. Lemaire, P. Dabin and R. Arnaud in Biodegradable Polymers and Plastics, Ed.,
M. Vert, I. Feijen, A. Albertsson, G. Scott and E. Chiellini, Royal Society of
Chemistry, Cambridge, UK, 1992, 32.
57. M. Pascu, M.M. Macoveanu, C. Vasile, A. Ioanid and R.C. Oghina, Cellulose
Chemistry and Technology, 2000, 34, 399.
58. C. Vasile in Handbook of Polyolefins, Ed., C. Vasile and R.B. Seymour, Marcel
Dekker Inc., New York, NY, USA, 1993, 491.
60. J.A. Ratto, P.J. Stenhouse, M. Auerbach, J. Mitchell and R. Farrell, Polymer,
1999, 40, 6777.
61. Zainuddin, M.T. Razzak, F. Yoshii and K. Makuuchi, Journal of Applied Polymer
Science, 1999, 72, 1283.
557
Handbook of Polymer Blends and Composites
66. O.V. Startsev, A.S. Krotov and L.T. Startseva, Polymer Degradation and Stability,
1999, 63, 183.
67. T. Ikejima, N. Yoshie and Y. Inoue, Polymer Degradation and Stability, 1999, 66,
263.
68. G. Cazacu, M.C. Pascu, G.E. Agafitei, A. Ioanid and C. Vasile, Macromolecular
Materials and Engineering, 2000, 283, 93.
70. M. Pascu, M.M. Macoveanu, A. Ioanid and C. Vasile, Cellulose Chemistry and
Technology, in press.
74. O. Chiantore, L. Trossarelli and M. Lazzari, Polymer, 1998, 39, 13, 2777.
75. F. Kubatovics and M. Blazso, Macromolecular Chemistry and Physics, 2000, 201,
349.
76. R. Asaletha, M.G. Kumaran and S. Thomas, Polymer Degradation and Stability,
1998, 61, 431.
77. W. Meissner and D. Zuchowska, Polymer Degradation and Stability, 1998, 60, 415.
78. K.T. Varughese, Kautschuk und Gummi Kunststoffe, 1998, 41, 11, 1114.
79. R.N. Santra, P.G. Mukundo, T.K. Chaki and G.B. Nando, Thermochemica Acta,
1993, 219.
80. P.P. Lizymol and S. Thomas, Thermochemica Acta, 1994, 233, 283.
81. B.A. Howell, Z. Ahmed and S.I. Ahmed, Thermochimica Acta, 2000, 357-358, 103.
558
Ageing of Polymer Blends and Composites
83. Z. Yudong, C. Daitao, S. Nai, Z. Weijin and X. Duanfu, Gaofenzi Cailiao Kexue
Yu Gongcheng, 1999, 15, 5, 79.
84. A.E. Akinay and T. Tincer, Journal of Applied Polymer Science, 1999, 74, 4, 877.
85. A.E. Akinay and T. Tincer, Journal of Applied Polymer Science, 1999, 74, 4, 866.
87. S. Choe, Y.J. Cha, H.S. Lee, J.S. Yoon and H.J. Choi, Polymer, 1995, 36, 26,
4977.
88. H.J. Choi, S.H. Park, J.S. Yoon, H.S. Lee and S.H. Choi, Polymer Engineering
and Science, 1995, 35, 20, 1636.
89. R.W. Venderbosch, H.E.H. Meijer and P.J. Lemstra, Polymer, 1994, 35, 20, 4349.
91. J. Choi, S. Lim, J. Kim and C.R. Choe, Polymer, 1997, 38, 4401.
92. M.M. Nasef, H. Saidi and H.M. Nor, Journal of Applied Polymer Science, 2000,
77, 1877.
93. T.K. Shin, J. Kim, H.J. Choi and M.S. Jhon, Journal of Applied Polymer Science,
2000, 77, 1348.
96. M.C.S. Perera, U.S. Ishiaku and Z.A.M. Ishak, Polymer Degradation and
Stability, 2000, 68, 393.
98. S. Zulfiqar and S. Ahmad, Polymer Degradation and Stability, 1999, 65, 243.
99. C. Vasile in Handbook of Polyolefins, Ed., C. Vasile, Marcel Dekker, Inc., New
York, NY, USA, 2000, 517.
559
Handbook of Polymer Blends and Composites
103. C-S. Ha, H-D. Park and W-J. Cho, Journal of Applied Polymer Science, 1999, 76,
1048.
104. P.S. Hope, J.G. Bonner and A.F. Milles, Plastics Rubber, Composite Process
Applications, 1994, 22, 147.
105. J.G. Bonner and P.S. Hope in Polymer Blends and Alloys, Eds., M.J. Fowkes and
P.S. Hope, Blackie Academic & Professional, Glasgow, UK, 1993, 46.
106. F.P. La Mantia, Polymer Degradation and Stability, 1993, 42, 213.
107. http://larcpubs.larc.nasa.gov/randt/1994
109. http://milieu.grads.vt.edu/interd/Nsf/Nsf-hpac.html
560
Degradation Behaviour of Polymer Blends and Thermal Treatment of Polymer Waste
14
Degradation Behaviour of Polymer Blends
and Thermal Treatment of Polymer Waste
14.1 Introduction
Depending on the lifetime stages of the polymeric materials, melt degradation (during
processing), acceleration of ageing, weathering (natural ageing) and decomposition may
be seen. The small structural polymeric inhomogeneities (oxygenated groups,
unsaturations, and other structural defects) and non-polymeric impurities (traces of
initiators, catalysts, metallic contaminants, photoactive pigments) can act as sensitive
points of the attack of the degrading agents. Usually the concentration of these active
impurities increases during the lifetime of the polymers.
561
Handbook of Polymer Blends and Composites
The kinetics and the mechanism of polymer degradation are very important aspects of
polymer science and engineering, with applications such as:
Improvement of the properties of polymeric materials by using the optimum type and
concentration of additives, e.g., antioxidants, flame retardant [3, 4];
As in all free radical mechanisms, the degradation of polymers involves the elementary
stages of initiation, propagation, eventually chain transfer and termination. Initiation is
the slowest reaction stage and generates macroradicals by statistical (chain scission) or
end-chain (depolymerisation) breaking of the macromolecular backbone or by scission
of the bonds with the pendant groups. The reactivity of the macroradicals, (e.g., the
R-CH2 radical has a higher reactivity than the R2CH) and the mobility of the hydrogen
atoms in the backbone, (e.g., the tertiary hydrogen is more reactive than the secondary
and primary one), strongly affects the mechanism of polymer degradation and
decomposition.
562
Degradation Behaviour of Polymer Blends and Thermal Treatment of Polymer Waste
The degradation of polymers by reactions of the side groups occurs in at least two stages.
In the first degradation stage the side groups can be eliminated from the main chain
forming diene macromolecular structures or can react in cyclisation or crosslinking
processes. The intermediate structure is further degraded at higher temperatures, leaving
solid residues. The main reactions that can be accompanied by other secondary reactions
during the first step of degradation of poly(vinyl chloride)(PVC) and
polyacrylonitrile(PAN) are shown in Figure 14.1:
For the conditions encountered in practical use the effect of temperature on the polymeric
materials is accompanied by the effect of atmospheric oxygen. The thermal oxidation
563
Handbook of Polymer Blends and Composites
depends on factors such as oxygen pressure, temperature, initiation rate, sample thickness,
supermolecular structure, polymer density and crystallinity, the manufacturing process
and the presence of catalyst residues. Small amounts of titanium-catalyst residues have a
slightly accelerating effect on oxidation while chromium-catalyst residues considerably
increase the diffusion rate. Different linear low-density polyethylenes (LLDPE) show
some similarities at oxidation but behave differently from low-density polyethylenes
(LDPE). Different Ziegler-type high-density polyethylenes (HDPE) also have similar
degradation behaviour but this is different from both Phillips-type HDPE and LLDPE
[8]. Iring and Tds [9] published a detailed discussion on the effect of chemical structure
and reaction conditions on the thermal oxidation of PE and polypropylene (PP).
564
Degradation Behaviour of Polymer Blends and Thermal Treatment of Polymer Waste
The kinetics of polymer degradation can be easily studied using thermal analysis
techniques, especially the non-isothermal ones that provide information related to a wide
temperature range and allow rapid determination of the kinetic parameters. However
the kinetic values obtained in isothermal conditions are more reliable.
The effect of all processes depending on the conversion () are included in a conversion
function that can be written in its integral (g()) or differential (f()) form. The general
differential form of the conversion function is:
f ( ) = m (1 ) n [ ln(1 )]p
Several solid-state kinetic models with the corresponding particular differential (f())
and integral (g()) forms of the conversion function were proposed to describe the
degradation mechanisms [14]. The variation of the conversion function in time gives the
reaction rate that depends on temperature by the rate constant (kT) containing the
activation energy (Ea) and the pre-exponential factor (A):
d()/dt=kTf() kT=Aexp(-Ea/RT)
Polymer degradation is kinetically described only by the complete set of the kinetic
parameters (Ea, A, n). The kinetic parameters are obtained using differential or integral
methods of solving the kinetic equation that gives the dependence of the reaction rate on
the time or temperature. The isoconversional methods that use experimental data obtained
at different heating rates are the best methods to determine the kinetic parameters.
For simple chemical reactions of the A B type, the kinetic studies are enough to
determine the reaction mechanism. Polymers and polymer blends have complex
degradation mechanism therefore the kinetic parameters have global values for which
physical significance is difficult to give. Coupled techniques for analysis such as evolved
565
Handbook of Polymer Blends and Composites
gas analysis techniques (EGA) based on the determination of reaction products, gives
useful information to establish the degradation mechanisms and about the interactions
between the primary degradation products or the interactions between the components
during the degradation of polymer blends.
On the basis of a proposed degradation mechanism one can develop kinetic models and
a concordance of the predicted kinetic parameters with the experimental ones is a proof
for the validity of the mechanism. Table 14.1 presents some proposed kinetic models
that consider the thermal decomposition of single polymers to occur by single, parallel
or consecutive reactions, each one with its reaction rate (k i). For example the
decomposition models for LDPE and PS consider the reactions of both the weak bonds
(branches or double bonds LDPE1, PS1) and the normal bonds (LDPE2, PS2). The kinetic
models are similar with the ones for the gas-phase pyrolysis of long chain hydrocarbons
but significant differences appear at polymer degradation due to the solid-gas or liquid
(melted polymer)-gas reactions. For example internal isomerisation can be neglected in
polymer bulk due to inhibited rotations of large C-C segments but is important at the
interface where the steric hindrance decreases. Table 14.2 presents some kinetic parameters
reported in literature for the thermal degradation of several polymers. These kinetic
parameters show a wide range for each polymer due to the different sample characteristics
and experimental conditions.
The kinetic models for the degradation of single polymers represent the first step for
developing models for the pyrolysis of mixed polymers in full-scale pyrolysis systems
where various processes overlap and interactions between the components can occur. A
general kinetic model for polymer pyrolysis and the corresponding kinetic equations,
proposed by Ramdoss and Tarrer [56] considers that the plastic (P) is converted to heavy
oil (H), light oil (L), and gases (G), and the coke (C) as degradation residue:
k1 k2 k3 k4
P H P L P G H L
k5 k6 k7 k8
L G H G H C L C
dP / dt = (k1 + k 2 + k 3)P dH / dt = k1P (k 4 + k 6 + k 7 )H
dL / dt = k 2 P + k 4 H (k 5 + k 8 )L dG / dt = k 3P + k 5 L + k 6 H
dC / dt = k 7 H + k 8 L
The kinetic models describe the pyrolysis process by simplified reaction pathways
considering each single reaction step as representative for a complex network of reactions.
For example very few PE degradation pathways contribute to the global kinetic or if
many parallel reaction occurs they have similar kinetic parameters.
566
Degradation Behaviour of Polymer Blends and Thermal Treatment of Polymer Waste
Table 14.1 Kinetic models used for single plastic thermal decomposition
Kinetic model Differential equations Ea N A (min -1) Reference
aPS1 V1 + R1
k1
d[PS1]/dt = -A1 e-Ea1/RT[PS1]n1 Ea1 = 120 n1 = 1.6 A1 = 1.06 x 108 Miranda,
bPS2 V2 + R2
k2
d[PS2]/dt = -A2 e-Ea2/RT[PS2]n2 Ea2 = 185 n2 = 0.8 A2 = 2.32 x 1013 2000 [15-18]
PVC aHCl + bI
k1
d[PVC]/dt = -A1 e-Ea1/RT[PVC]n1 Ea1 = 198 n1 = 1.0 A1 = 3.57 x 1018 Miranda,
bI cV1 + dR1
k2
d[I]/dt = b (A1 e-Ea1/RT[PVC]n1 - A2 e-Ea2/RT[I]n2) Ea2 = 143 n2 = 1.1 A2 = 9.95 x 1010 2000 [15-18]
dR1 eV2 + fR2
k3
d[R1]/dt = d (A2 e-Ea2/RT[I]n2 A3 e-Ea3/RT[R1]n3) Ea3 = 243 n3 = 1.6 A3 = 5.77 x 1016
d[R2]/dt = f (-A3 e-Ea3/RT[R1]n3)
V1
k1
HDPE k2 E1 = 214 A1 = 2.84 x 1012 Conesa and
HDPE I E2 = 239 A2 = 9.64 x 1015 co-workers,
I V2
k3
E3 = 200 A2 = 6.96 x 1012 1996 [19]
k1
PS k2 V1 E1 = 276 n1 = 1.0 A1 = 3.67 x 1019 Marcilla and
PS I E2 = 220 n2 = 0.5 A2 = 4.11 x 1016 Beltrn
I V2
k3
E3 = 187 n3 = 0.8 A2 = 8.58 x 1013 1995 [20]
567
568
M1: Only the M1 model, for which the calculated E450: E450: E450: Marcilla and
k1
PVC V1 results are presented, explains the asymmetry E a 1
= 159 n1 = 1.1 A1 = 1.77 x 1013 Beltrn 1996,
k2
PVC I of DTGA curves for two PVC commercial E a 2
= 151 n2 = 1.2 A2 = 3.74 x 1012 1997 [21, 22]
I V2
k3
resins E450 (Aiscondel) and H367 (Solvay) E a 3
= 253 n3 = 2.3 A3 = 5.5 x 1020
I R
k4
Ea4 = 263 n4 = 2.0 A4 = 9.75 x 1021
M2:
PVC aV1+bI
k1 H367: H367: H367:
k2
bI cV2+eR Ea1 = 158 n1 = 0.6 A1 = 2.61 x 1013
Ea2 = 109 n2 = 0.5 A2 = 2.85 x 109
M3:
Ea3 = 113 n3 = 2.8 A3 = 4.48 x 109
PVC V1
k1
Ea4 = 96 n4 = 1.4 A4 = 2.14 x 107
PVC R
k2
M4:
PVC aV+bR
k1
Plasticiser:
PL V
k
569
kinetic parameters Thermal degradation
570
kinetic parameters Thermal degradation
571
kinetic parameters Thermal degradation
572
Degradation Behaviour of Polymer Blends and Thermal Treatment of Polymer Waste
Mixing of the polymers with low molecular or other high molecular components is a
very important step for the preparation of polymer blends and materials. Mixing is usually
achieved by melt processing that strongly affects the properties of the resulting materials.
The main possible reactions during the mechanical degradation are the chain scissions
that are caused by high temperature and shear stress in twin-screw extruders and decrease
the molecular weight and the melt viscosity while increasing the melt index. Crosslinking
or chain branching have opposite effects to chain scission and are caused mainly by low
shear stress and high residence time in single screw extruders. For example during the
processing of HDPE at 140-210 C a concurrence occurs between the molecular
enlargement by chain branching and the decrease of molecular weight by chain scission,
the final properties of the material depending on the relative extent of the two processes
[57].
The initiation of the mechanical degradation of the polymers occurs by statistical breaking
of the C-C bonds (forming terminal C-radicals, with the possibility for chain scission, as
characteristic for PP and PS) or of the C-H bonds (forming in-chain C-radicals able to
chain branching). The weak bonds of the macromolecules are also sensitive to the action
of the mechanical forces, e.g., the susceptibility of Ziegler-Natta LLDPE to
thermomechanical degradation is correlated to the number of branches [58]. The
macroradicals are transformed by disproportionation (resulting in unsaturation of the
end-chain) and recombination (resulting in branching and crosslinking, as characteristic
for PE) or can attack a central double bond (forming another C-centered free radical, as
in HDPE). The macroradicals can also react with trace amounts of oxygen in melt forming
573
Handbook of Polymer Blends and Composites
If the interactions between the blend components are negligible, the cross-recombination
is not probable without the use of an efficient free-radical initiator (reactive processing)
and an additive behaviour in degradation is expected, as in the case of PE/PP blends. PE
degradation by mainly chain branching and crosslinking leads to increased viscosity and
torque while PP degradation by mainly chain scission leads to the opposite effects than
in PE. Therefore no changes in viscosity and rheological properties occur for PE/PP blends
of certain composition giving the false impression of no degradation.
The thermal stability of the melt processing blends depends strongly on the blending
technique as observed for the blends of polyurethane with chlorosulfonated polyethylene,
epichlorohyrin and polychloroprene elastomers using sulfur cure systems. For these
systems the blends prepared by mixing the curative-added polymers have lower thermal
stability than the curative-added mixed polymers and the curative-added thermally
pretreated mixed polymers due to the interchain crosslinks formed between the blend
components before the addition of curatives [60].
Thermal degradation of polymer blends is more complex than the degradation of single
polymers. In polymer blends with non-similar components similar reactions with different
rates and/or various types of reactions at different extents can occur. Generally, depending
on the blend composition and the reaction conditions many competitive and consecutive
reactions can take place and it is not possible to find only a certain reaction mechanism
operating. The thermal stability of polymer blends depends on the type of mixed polymers,
the morphology of the blend and the degradation environment.
Of much interest is the fact that interactions appear in the polymer bulk during thermal
degradation of the blends, between the blend components and the low molecular weight
products and free radicals that are formed by the scission of the polymeric chains and
migrate across the phase boundaries. The interactions are more important in miscible
systems than in blends in which phase separation occurs.
Because of the complex interactions the degradation of the polymers in blends occur
differently compared to separated polymers, as in the case of blends of PS and poly(-
methyl styrene) (PMeSt) with acrylate polymers or in poly(ethylene oxide)/PMMA
574
Degradation Behaviour of Polymer Blends and Thermal Treatment of Polymer Waste
blends. Direct interactions between the blend components were not reported for
degradation at high temperatures. Polymer blends have a similar thermal behaviour as
graft copolymers but it is different from random copolymers [61].
The polymers are diluted in blends and this fact can slightly increase the overall
degradation rate of the components, (e.g., PMMA/PMeSt, poly(butyl acrylate)/PMeSt),
or can drastically reduce the relative yields of oligomers, (e.g., PMMA/PS). Usually the
cross-termination stabilises both components, (e.g., PMMA/PS) and the intermolecular
transfer stabilises one component (that extracts the H atom) but destabilises the other,
(e.g., poly(butyl acrylate)/PS). In heterogeneous systems kinetic effects are possible mainly
at phase boundaries, e.g., PMMA/PVC, PMMA/PE [62].
The degradation products can either destabilise or stabilise the blend, depending on the
nature and the ratio of the components or on the temperature. The morphology and the
interfacial interactions also influence the thermal stability.
Literature reviews on degradation and ageing of polymer blends were presented by Wypych
[64] and Pospisil and co-workers [65]. Particular aspects for degradation of blends
containing polyolefins, PS, PVC, etc., are presented above.
In blends with recycled LDPE (partially oxidised during use) and polyamides the carbonyl
groups from the degraded polyolefin interact with the amide group increasing the phase
compatibility of the blend. In binary blends the thermooxidative degradation of LDPE is
accelerated by ethylene-acrylic acid copolymer (due to the easy H atom abstraction
from the carboxyl groups formed during degradation) and is inhibited by plasticised
starch. Ternary LDPE/ethylene-acrylic acid copolymer/plasticised starch blend degrades
faster than LDPE but slower than starch-free blends [67, 68].
575
Handbook of Polymer Blends and Composites
The vinyl polymers, e.g., PS, PMMA, accelerate PP degradation and form graft or block
copolymers by the attack of the vinyl radicals to the PP chain.
Grafting diethylmaleate onto PP increases the stability of the PP/polyamide 6 blends and
improves the dispersion and the processability of the blends. The best thermal stability
was obtained using 30 wt% PP-g-diethylmaleate [70]. Nitrile rubber (NBR) improves
the thermal properties of PP in blends increasing the initial degradation temperature.
Compatibilisation of the blends with phenolic- or maleic anhydride- modified PP also
improves the thermal stability [71].
Polystyrene is stabilised in blends with PVC, poly(vinyl alcohol) (PVA), PAN, polyisoprene
or polybutadiene (PB) due to the deactivation of PS macroradicals by intermolecular
reactions with products or structure elements of the second polymer in blend [72]. PS
depolymerisation (with formation of low molecular weight volatiles) and PVC
dehydrochlorination are retarded in PS/PVC blends because the chlorine radical abstracts
hydrogen atoms from PS and generates PS macroradicals that undergo chain scission
(with rapid decrease in molecular weight). PS is stabilised in PS/PVA blends due to the
reaction of a PS macroradical with the unsaturated bonds from the degraded PVA. PS is
destabilised in blends with PaMeSt and containing up to 10% PMMA. Cross-termination
reactions are prevalent during the degradation of PS/PMMA blends and decrease by a
factor of about 10 the rate constant of styrene and methyl methacrylate (MMA) monomer
evolution in blends compared to the individual polymers. Chain transfer is more important
than cross-termination in blends of PS with poly(alkyl acrylates) but no kinetic effects
appear in blends of PMeSt with poly(alkyl acrylates) because no tertiary hydrogen
atom is available for transfer [73]. The cross-termination is also less possible for the
radicals of PMeSt than for those of PS in blends with PMMA due to the additional
steric hindrance from the methyl group.
PS destabilises the polyolefins in PS/PE or PS/PP blends due to the chain transfer of PS
radicals to polyolefin chain. PMMA also destabilise PP in PMMA/PP blends on the basis
of the same mechanism. The decomposition of PS/poly(sodium 4-styrene sulfonate) (PSS)
blends occurs in two independent steps with no interactions between the components.
However PS exhibits enhanced thermal stability due to the thermal protection given by
the graphite-like char formed by PSS degradation. The PSS random copolymers degrade
in a single step with no char formation and no thermal stabilisation of PS [61].
PS/NBR blends are immiscible and incompatible but have higher thermal stability than
the separated polymers. The compatibility and the thermal stability are improved adding
NBR-g-PS but the blend still remains thermodynamically incompatible [74].
576
Degradation Behaviour of Polymer Blends and Thermal Treatment of Polymer Waste
HCl and Cl radicals promote the chain scission at the weak points and hinder the -
scission of the macroradicals. In PVC/PMMA blends HCl converts the ester side groups
of PMMA phase into anhydride rings which interfere with depolymerisation and stabilises
PMMA, while Cl radicals favour PMMA degradation. PMMA/PVC blends are less stable
than MMA/vinyl chloride copolymers and produce methyl chloride at degradation, as
proof for the reaction of Cl radicals from PVC with the PMMA phase. Dehydro-
chlorination of PVC is not affected in PVC/polycarbonate (PC) blends but the chlorine
radical migrates across the phase boundary and destabilises the PC phase decreasing its
molecular weight [72].
577
Handbook of Polymer Blends and Composites
poly(vinyl butyral) blends increases with the poly(vinyl butyral) content [80]. The PVC/
chlorinated polyurethane blends have improved thermal stability due to the interactions
between C=O groups of urethane segments and -H of the chlorinated polymer or due to
the dipole-dipole interactions between C=O and ClC functional groups [81]. Slight
destabilisation was observed when adding polydiphenylsiloxane or
polydimethyldiphenylsiloxane to PVC but when PVC amount is less than 50 wt% both
components are stabilised, especially in polydimethyldiphenylsiloxane/PVC blends [82].
Increasing the methacrylate-butadiene-styrene amount in blends with PVC the kinetics of
the degradation changes but the degradation mechanisms remain the same, with PVC
dehydrochlorination as a first step and simultaneous degradation of methacrylate-butadiene-
styrene and carbonisation of dehydrochlorinated segments of PVC as the second step [83].
The increased rubber content in PVC/epoxidised natural rubber retards the PVC
dehydrochlorination and favoured the second PVC degradation step [84].
PVC and its low molecular fractions affect the vapourisation-decomposition of additives,
e.g., di-2-ethylhexylphthalate plasticiser [85].
Table 14.3 shows the degradation behaviour of some polymer blends containing PE, PP,
PS and/or PVC.
PB phase decreases the resistance of ABS to thermal and photo-oxidation. When EPDM
replaces butadiene (acrylonitrile-EPDM-styrene; AES), the degraded EPDM and SAN
interact forming EPDM-g-SAN that acts as a compatibilising agent. The radical transfer
between the different polymeric structures is higher in EPDM-g-SAN copolymer than in
the EPDM/SAN segregated phase, increasing the chain scission in EPDM units and
decreasing SAN depolymerisation [94, 95].
In PC/PMMA mixture the PMMA macroradicals extract H atoms from the PC chains.
The zip length of depropagation is reduced, stabilising PMMA and chain scission occurs
destabilising PC [73].
578
degradation behaviour Commingled plastics
579
degradation behaviour Commingled plastics
580
degradation behaviour Commingled plastics
581
degradation behaviour Commingled plastics
Blending poly(aryl ether ether ketone) with VECTRA thermotropic liquid crystalline
polymer (LCP) changes only the rate of degradation (the LCP is destabilised at high
poly(aryl ether ether ketone) contents) while the mechanism remains identical to those
of individual polymers [100].
Plastic waste can come from industry or from municipal solid waste. According to their
origins, the plastic waste has different properties and qualities. Industrial plastic wastes
come from the plastic synthesis and processing industry. They are usually relatively free
of contaminants and available in large qualities. The amount of post-user plastic waste is
continuously increasing depending on the end-use sector, as shown in Figure 14.3 [101].
582
Degradation Behaviour of Polymer Blends and Thermal Treatment of Polymer Waste
Repelletisation and remoulding are the main ways of recycling the industrial homogeneous
plastic waste while thermal cracking is used to recycle the heterogeneous or mixed resins.
Municipal plastic waste represents 7-8 wt% of the municipal solid waste. It generally
consists of mixture of different kind of plastics: 46 wt% HDPE and LDPE, 16 wt% PP,
16 wt% PS, 7 wt% PVC, 5 wt% polyethylene terephthalate (PET), 5 wt% ABS and
5 wt% other polymers [102]. This waste and especially the plastic packaging is difficult
to treat or recycle due to its complex nature and composition, structural deterioration of
the polymeric components and the contamination with various organic, inorganic or
biological residues.
Landfill was the first method to solve the problem of the municipal solid waste but it
becomes expensive due to the lack of suitable sites and to the danger of leaching and soil
impregnation, with subsequent contamination of underground water. The process does
not allow the recovery of either the material or energy content of plastic waste and the
long-term effects of plastic degradation when landfilled are not well established.
583
Handbook of Polymer Blends and Composites
The main routes for recycling of plastic wastes (Figure 14.4) are the material recovery by
melt reprocessing of relatively pure polymers (back to polymer), the recovery of both
material and energy by thermal degradation (pyrolysis) of mixed, heavily filled plastics
and thermosets, to produce chemicals (feedstock recycling) or fuel oil and the recovery
of only energy (back to energy) by incineration and reduction in blast furnaces [105]:
Figure 14.4
The situation of plastic consumption and recovery in Western Europe in 1998 is presented
in Table 14.4. The aim in plastic recycling is to increase the mechanical and feedstock
recycling to the detriment of landfill and incineration. The situation of plastic recycling in
Western Europe and Japan is presented in Figure 14.5 and 14.6, respectively, [101, 106].
584
Degradation Behaviour of Polymer Blends and Thermal Treatment of Polymer Waste
585
Handbook of Polymer Blends and Composites
Closed loop recycling is limited to few polymeric materials, e.g., PVC windows being
used for windows, PET bottles to fibre applications, PP bumpers to plant pots, PE food
contact film to garbage bags, PE films to pipes, PP closures to non-critical car parts, PE
crates to pallets. At the end of the service life these materials retain most of the
characteristics of the initial polymeric resin but stabilisers must be added to assure the
re-stabilisation of the deteriorated structure [5]. Reprocessing can partially restore the
mechanical properties deteriorated during ageing or using of polymers; however for LDPE
and ABS the degradation was found to be more severe for alternating processing and
ageing than either processing or ageing alone [107].
The properties of mechanical recycled single plastics strongly depend on the presence of
traces of other plastics therefore waste has to be collected and sorted into different types
of plastics but this is difficult and/or expensive. Contamination is almost impossible to
be avoided even with the most sophisticated separation techniques. The PVC content in
PET cannot be decreased below 100 ppm, the level that induces PET degradation and
discoloration by hydrolytic degradation due to HCl that catalyses the reaction between
ester bonds and the retained moisture. For PET containing 2000 ppm PVC processed in
nitrogen atmosphere a competition occurs between the chain scission and the increase of
the molecular weight, the latest favoured by low processing temperatures and high mixing
speed [108].
Mixed (heterogeneous) plastics are normally unsuitable for melt reprocessing. The
incompatibility of the polymers in the mixture causes the recycled materials to have poor
mechanical properties and durability therefore it is necessary to re-generate the
morphology by re-compatibilisation and re-stabilisation [109].
The mechanical and rheological properties of the recycled polymers strongly depend on
the reprocessing device and the processing conditions. The properties are slightly affected
during the recycling of raw materials in devices with low residence time but worsening
of the properties occurs when increasing the number of recycling steps in processing
devices with a long residence time. The elongation at break and the impact strength are
very sensitive to the multiphase nature of the blends and they are the mechanical properties
most affected by the incompatibility of the components or by the degradation during
586
Degradation Behaviour of Polymer Blends and Thermal Treatment of Polymer Waste
recycling operation. On the basis of the experimental degradation curves Bernardo and
co-workers found a mathematical model that can predict the variation of the properties
of mixtures of virgin and recycled polymers with the number of reprocessing cycles [110].
Additives can be used to upgrade the mechanical properties of recycled polymer blends
but this procedure is expensive and energy-intensive for using for domestic mixed plastic
waste. Antioxidants can stop or slow the decrease of the mechanical properties due to
the thermomechanical degradation during reprocessing.
Pyrolysis breaks down the structure of polymers into smaller intermediate species that
can be used as fuel or raw material for petrochemical industry. The amount of gas is
decreased 5-20 times compared with incineration that leads to savings in the clean-up
installations. The liquid products (usually termed oils) are obtained with a high rate of
conversion and are a complex mixture of hydrocarbons depending on the plastic waste
composition and the pyrolysis conditions. In high temperature pyrolysis (>600 C) the
products are equivalent to the naphtha cracker fraction (mainly mixtures of aliphatic
hydrocarbons but also aromatics). Pyrolysis at low temperatures (<600 C) without using
catalysts produces waxes that can be feed in naphtha cracker. The waxes can condense
and stick at the outlets of the reactors or on the walls of the pipelines that might obstruct
the installation and force the pyrolysis process to stop. Similar problems arise with the
pyrolysis of PET when the produced terephthallic acid sublimates in the reactor and
condenses as yellow powder on the cold parts of the installation. Using catalysts allows
pyrolysis to occur at lower temperatures and offers the possibility to selectively obtain
valuable chemicals or to remove undesired degradation products.
587
Handbook of Polymer Blends and Composites
Char is produced when pyrolysing certain polymers such as PVC and PET due to secondary
repolymerisations of the primary degradation products. The char-forming tendency
increases with the aromatic character of the polymer and the substitution degree of the
aromatic nuclei with non aliphatic chains and if the polymer contains halogen atoms or
groups such as OH or =O, capable of interaction with the hydrogen atom of the aromatic
nuclei. The alkyl groups are a source of hydrogen and their presence near the aromatic
nucleus prevents the char formation.
Water is formed during the pyrolysis of polymers, that is formed in the process from the
oxygen-containing groups of the polymer, (e.g., COO, OH, COOH, etc.), or of the
catalysts, (e.g., using -FeOOH for ABS degradation [113, 114]), or may come from the
original moisture of the materials. The inorganic compounds (fillers, glass, steel belts,
etc.), and the non-volatile organic compounds (carbon black) of the polymeric materials
remain relatively unmodified after pyrolysis and they can be recycled separately.
The thermal processes for recycling of polymer waste are complicated by the fact that
the degradation of macromolecules requires large amounts of energy but plastics have
poor thermal conductivity. Polymers such as PVC and ABS cause problems during pyrolysis
due to the chlorine and nitrogen remaining in the oils. PVC can influence the extent of
degradation and the pyrolysis product distribution when mixed with other polymers
such as ABS, PC and polyoxymethylene [115]. The HCl is highly corrosive and the
chlorinated hydrocarbons produced can be precursors of toxic emissions such as
polychlorinated dibenzodioxins, dibenzofurans and biphenyls [116, 117]. Nitrogen can
cause the corrosion of engine parts and the formation of very harmful compounds such
as HCN or NOx when the oils are used as fuel [118, 119]. Nitrogen-containing 5- and 6-
membered heterocyclics compounds remain as coke inside the refinery catalyst making
its regeneration very difficult [120]. The amount of nitrogen-containing compounds in
the ABS degradation oil can be decreased if the process is conducted under suitable
condition [121].
The installations used for the pyrolysis of mixed plastic include melting vessels, blast
furnaces, autoclaves, tube reactors, rotary kilns, coking chambers or fluidised bed reactors
(conveyors, extruders, fluidised beds, molten salt reactors, rotating drum/furnaces, stirred
588
Degradation Behaviour of Polymer Blends and Thermal Treatment of Polymer Waste
tank reactors, vertical retorts) [122]. Rotary kiln processes have large temperature gradient
and long residence time (about 20 minutes or more) in the reactor that gives very diverse
product spectra. Fluidised bed reactors have good heat and mass transfer, constant
temperature in the reactor and short residence time of few seconds up to 1.5 minutes
that gives uniform product spectra [123, 124].
The Hamburg process uses a fluidised bed reactor with indirect heating and can produce
petrochemical oils containing mainly monomers, aliphatic hydrocarbons or aromatics
from plastic waste [123]. Using single polymers, it produces at 500 C, 98% monomer
from PMMA and poly(triflouro ethylene) and 75% monomer plus 10% oligomers from
PS [125]. For mixed plastics the process cannot be optimised to obtain only monomers
because benzene-toluene-xylene aromatics (BTX) are formed by Diels-Alder and
dehydrogenation reactions. From mixed plastic waste including PE, PP, polyesters, PS
and PVC the pyrolysis oil contains 30-40 wt% styrene and BTX fractions and less than
10 ppm of chlorine. The main part of fillers, impurities and heavy metals remains in the
residue [123, 126].
The Niigata Plastic-to-Oil Conversion Center (6,000 metric tons waste plastics per year)
is the first commercial plant in Japan for the thermal liquefaction process of waste plastics
into fuel oil. After removal of undesirable materials (metals, papers, wood pieces, PET
bottles) plastic waste is treated at 300 C in a dehydrochlorinator to produce light
hydrocarbon gases, further burned at 1000 C. Hydrogen chloride is absorbed in water
and recovered as 10 wt% HCl solution. The molten plastics are decomposed at 400 C
in a pyrolysis reactor to produce 34 wt% light oil and 17 wt% medium and heavy oil
containing 45 ppm chlorine. The medium oil has a Cetane Index of 46 and can be used
as diesel fuel [127].
The Kawatetsu Machinery Co. (KMC) waste plastic liquefaction plant (250 kg waste
per batch at 450 C) is based on the Sakata-Murata process and aims at the development
of small-scale treatment plants for the liquefaction of industrial waste plastics within the
industries and consumption of the products internally, as fuel oil. The oil yields for
liquefaction of PE, PP, PS and a mixture of PE/PP/PS (1/1/1) were 69, 72, 95 and 76 wt%,
respectively, [128]. The flow diagram of the KMC plant is shown in Figure 14.7.
Other systems such as Toshima Oil Reclamation System [129], Kuroki Process-6 [130]
were developed in Japan for converting plastic waste into fuel oil.
Stepwise pyrolysis of plastic waste using a cascade of reactors which were stirred well
was proposed by Bockhorn and co-workers [131]. The reactors use stainless steel spheres
that assure good mixing, good heat transfer and short retention time of the products to
avoid the secondary reaction to occur. A mixture of PE, PS and up to 15 wt% PVC was
decomposed in a cascade of three reactors at temperatures of 330, 380 and 440 C,
589
Handbook of Polymer Blends and Composites
Figure 14.7 The flow diagram of Kawatesu machinery waste plastic liquefaction plant
respectively, with a maximum feeding rate of 3 kg/h and retention time of 26 minutes in each
reactor. In the first reactor 99.6% of chlorine was removed with no pretreatment of the
waste. PS is decomposed in the second reactor to up to 93% monomer and monocyclic
aromatics and PE is decomposed in the third reactor to produce paraffins and olefins. No
HCl or gaseous pyrolysis products of PE were detected in the second reactor proving that the
polymers decompose independently [132, 133].
A vacuum pyrolysis technology, known as the pyrocycling process, was developed in Quebec
in the early 1980s. This was focused at the beginning on wood and related lignocellulosic
materials and now also applied for mixed polymers coming from municipal solid waste,
automobile shredding light fraction, used tyres or biomedical wastes [134]. The process
consists of thermal decomposition of organic materials at 450-550 C and total pressure of
2-15 kPa into primary fragments that are rapidly removed from the hot reaction zone by a
vacuum pump and recovered after condensation as pyrolysis oil [135]. The operation
conditions, milder than those for atmospheric pyrolysis and incineration enable the recovery
of larger oil quantities at the expense of gas and solid residue. Fast removing and cooling of
the products minimise the secondary reactions such as thermal cracking, repolymerisation
and recondensation reactions, gas phase collision, catalytic cracking and reduction and
oxidation reactions.
Williams and Williams pyrolysed a simulated municipal plastic waste containing LDPE,
HDPE, PP, PS, PVC and PET at various temperatures between 500 and 700 C and a residence
590
Degradation Behaviour of Polymer Blends and Thermal Treatment of Polymer Waste
time of about 15 seconds in a fluidised bed reactor using silica sand as bed material and
nitrogen as fluidising gas. Low temperature pyrolysis produces heavy wax similar to the
atmospheric residue cut from crude oil, that can be further catalytically cracked to produce
a gasoline type oil. Secondary reactions increase at high temperatures of 700 C when the
pyrolysis oil is similar to gasoline [136]. Polycyclic aromatic hydrocarbons with potential
adverse impact on the environment were determined by the same authors in the pyrolytic oil
of a 1:1 mixture of PS/PP in fixed bed reactor at temperatures from 500 to 650 C [137].
Mixtures of PE, PP and PS with compositions similar to those found in municipal plastic
waste were pyrolysed by Pinto and co-workers in an autoclave system at 430 C, 3.5 MPa,
20 minutes reaction time. For a 68 wt% PE, 16 wt% PP and 16 wt% PS composition, similar
to the Portuguese municipal plastic waste. The pyrolysis oil contains more than 30 wt%
aromatics and has a calorific value of about 45 MJ/kg. More aromatics and alkenes and
higher calorific value and octane number of the pyrolysis oil is obtained when increasing the
amount of PS or PP in the mixture [138]. Molecular sieve zeolites, ZnCl2, Fe2O3, Co-Mo and
Ni-Mo oxides laid on alumina have no effect on the total conversion, product yields or oil
composition [139].
The pyrolysis products from polymer waste can be upgraded to obtain synthetic crude oil by
hydrogenation in petrochemical plants. The synthetic crude oil has an aromatic character
when obtained from pyrolysed rubber waste containing styrene or leather compounds, and
a paraffinic character if obtained from pyrolysed plastic waste [141]. Polyolefins such as PP
can be dissolved at 160-170 C in medium oil and paraffinic oil from petrochemical processes
and further catalytically cracked to form 40-50 wt% light fraction (boiling below 250 C),
more that in thermal process (10-20 wt%) [142].
Electric cable waste containing polymers as a plastic portion of 90.8 wt% crosslinked LDPE
combined with a small quantity of EVA and a rubber portion of 4.6 wt% natural rubber and
4.6 wt% other polymers (chloroprene rubber, ethylene propylene dimer rubber and
chlorosulfonated polyethylene), additives and fillers (carbon black, paraffin and aromatic
oils, calcium and magnesium carbonates) were pyrolysed in a vacuum at 20 kPa and 450 C
by Chaala and co-workers [143]. The main pyrolysis products consist of 71 wt% wax with
small branched alkyl groups with different olefinic groups and 21 wt% carbon black with a
high absorption surface area of 60 m2/g and less than 5 wt% calcium carbonate. Minor
products were low-sulfur gas and liquids.
Table 14.5 presents the pyrolysis product yields of individual and mixed polymers.
591
592
/11.5/5.3
598
Catalysts are used in pyrolysis of plastics in two modes: a liquid phase contact (LPC)
when the catalyst is mixed with the polymer and act mainly on the partially degraded
oligomers and a vapour phase contact (VPC) when the catalyst acts on the hydrocarbon
vapours produced by the pyrolysis of the polymer [171, 172].
Fluid cracking catalysts such as silica-aluminas, zeolites and mordenites have acidic
properties. They decrease the pyrolysis temperatures and increase both the amounts of
iso-alkanes and aromatics in the C5-C10 range and the reaction rates. Reforming catalysts
based on transition metals supported on silica-alumina catalyse the isomerisation reactions
due to the acid sites of the support and the hydrogenation/dehydrogenation reactions
due to the metallic sites and can have higher selectivity for the formation of aromatics.
Activated carbon catalyses the cracking reactions of the polymers with formation of
normal alkanes to the detriment of isoalkanes. Metal oxides, metal complexes and alkali
or alkaline metal carbonates are also used for catalytic pyrolysis of plastics, with the
main aim of production of monomers [173].
TiCl3 produces mainly n-paraffins from the catalytic pyrolysis of polyolefin mixed plastic
at 435 C while HZSM-5 zeolite catalyst increases the amount of aromatics naphthenes
and isoparaffins, increasing the oil quality [174]. The TiO2/SiO2 catalyst converts the
PP/PET mixed plastic at 425 C into oils containing large amounts of hydrocarbons
heavier than C12 without formation of phthalic acids [175].
HZSM-5 zeolite and orthophosphoric acid modified zeolite catalysts (PZSM-5) increase
the gas yield, lower the liquid fraction amount and change the composition of the products
in catalytic decomposition of PE and PP compared to the uncatalytic decomposition
[176]. When these catalysts are used for the pyrolysis at 400-450 C of a HDPE/LDPE/
iPP/PS/ABS/PET polymer mixture similar in composition with the municipal plastic waste
without PVC, they increase the gas yield and aromatic content of the liquid fraction. The
zeolite catalyst produces a high number of isomers while PZSM-5 having increased
selectivity and decreased acidity promotes the formation of -alkyl substituted and
condensed ring aromatic hydrocarbons. The product yield of catalytic degradation is
close to the one obtained at non-catalytic degradation at 600-700 C [177, 166]. When
the polymer mixture also contains PVC, the interactions of different components in the
mixture during pyrolysis, for example the reaction of HCl with pyrolysis-derived
unsaturated compounds, is manifested by the formation of different chlorinated
hydrocarbons during pyrolysis of PVC and during pyrolysis of commingled plastic [178].
Uddin and Sakata have reported that for the catalytic degradation of PE and PP in vapour
phase contact over silica-alumina and ZSM zeolite the thermally degraded hydrocarbons
underwent further decomposition into gaseous products [171, 172]. The catalytic
599
Handbook of Polymer Blends and Composites
degradation in the liquid phase contact increased the yield of oil. Non-acidic mesoporous
silica catalysts promoted the degradation of PE and PP into lighter hydrocarbons and
suppressed the deposition of coke on the catalytic surface compared with acidic catalysts
such as silica-alumina and ZSM-5.
Cr2O3/Al2O3 catalysts increase by 10 times the amount of gas produced and upgrade the
decomposition products when they are used for the pyrolysis of medical polymer wastes.
The condensable products are light, yellow liquids containing 3-8 wt% aromatics while
at non-catalytic pyrolysis, waxes have resulted [147, 179].
Tyres are composed of very different components such as elastomers [100 parts per
hundred of rubber (phr)], vulcanising agents (5-10 phr), accelerators (1-3 phr), activators
and other additives (1-5 phr), processing oil (5-30 phr) carbon black (50-70 phr) fillers
and steel, heterogeneous distributed along the tyre [180]. The elastomers from tyres are
mainly SBR but natural rubber, synthetic cis-polyisoprene or cis-PB can be mixed with
SBR in different ratios. Used tyres represent a source of energy and valuable chemical
products but there are problems with their recovery by pyrolysis due to the technology
deficiency, financial constraints and limited market and/or value for the products.
The pyrolysis oil consists of the processing oil from the tyre formulation, organic additives
and rubber degradation products and unlike petroleum, the tyre-derived pyrolytic naphtha
fraction contains highly branched chain isomeric hydrocarbons. The pyrolysis oil may
be used as fuel, as feedstock in petroleum refinery or as source of chemicals such as
benzene, dl-limonene (dipentene), toluene and xylene. Microcrystalline waxes, lubricants
for plastics, surfactants, -olefins, plasticisers, coatings, sizing agents or monomers can
be also extracted from the pyrolytic oil obtained from the used tyres [181]. Nitrogen and
sulfur containing compounds have to be identified and removed or treated before the
catalytic upgrade of the pyrolysis oil [182]. The non-condensable gases may be used as
fuel in the process and the solid char may be used as smokeless fuel, carbon black or
activated carbon.
The Pyrovac process can pyrolyse the scrap tyres with continuous feeding of 200 kg/h at
415 C and 2 to 10 kPa pressure to obtain 55 wt% oil, 25 wt% carbon black, 9 wt%
steel, 5 wt% fibres and 6 wt% gas. The pyrolysis oil has a heating value of 43 MJ/kg, a
sulfur content of 0.8 wt% and is rich in benzene and other valuable petrochemical
compounds [134].
600
Degradation Behaviour of Polymer Blends and Thermal Treatment of Polymer Waste
The Onahama Smelting and Refining Co., plant in Iwaki City, Japan, decomposes 4.5 t/h
of whole tyres to obtain 25-30 wt% heating oil, 35-40 wt% carbon black and 10 wt%
steel [183].
Pyrolysis of used tyres was performed by Chang and co-workers in an argon plasma to produce
a syngas containing CO2, CO, CH4, C2H2, C2H4, H2 and NH3, without polluting gas such as
H2S, SO2 and NOx. The metal compounds are concentrated in the ash particles [184].
Conesa and co-workers [185] found that the tyre wastes decompose in three stages that
can be kinetically modelled considering the fraction of the tyre (Fi), the gases (Gi), the
solid residue (Si) and the yield coefficient of the solid fraction (si):
Fi (1 s i ) G i + s i Si
The mineral oil decomposes in the first stage to produce linear hydrocarbons and CO2.
The polymer matrix decomposes at higher temperature, the rubber producing benzene,
some hydrocarbons and CO2 in the second stage and SBR producing aromatic compounds
in the third stage).
Plastics are used in the automotive industry mainly in form of alloys, blends and
composites. The matrix resins may be PP (27 wt%), polyamide (14 wt%), PE (12 wt %),
PVC (12 wt%), thermoplastic olefinic elastomer (10 wt%), ABS (8 wt%), polycarbonate
of bis-phenol-A (4.5 wt%), thermoplastic polyesters, (i.e., PET, poly(butylene
terephthalate), poly(ethylene 2,6-naphthalene dicarboxylate)) (4.2 wt%), polyphenylene
ether (4 wt%), PC/ABS (1.8 wt%), others (acrylics, poly(vinylidene fluoride),
polyoxymethylene, thermoplastic urethanes, styrene-maleic anhydride, polyphenylsulfide)
(<1 wt%) [186].
The automotive industry favours the design of new cars for easy dismantling of plastic
body parts and the use of recyclable resins. Therefore recycling of specific plastic parts
from automotive industries such as bumpers or fenders causes fewer problems than
municipal plastic wastes. The best way for recycling of these parts is by reprocessing
with recovery of morphology (re-compatibilisation and re-stabilisation).
Beside the large plastic parts a large amount of plastics in automobiles cannot be easily
separated, i.e., the interior trim, and remains in the light fraction (known as fluff) of
automobile shredder residue (ASR). ASR is a heterogeneous mixture of plastics, rubber,
foam, textiles, glass and dirt that makes the recycling more complicated due to lower
value and high diversity of composition. Pyrolysis is the best way for recovery the auto
601
Handbook of Polymer Blends and Composites
fluff, producing gases, liquids and residual fractions depending on the process, the reaction
temperature and the residence time in the reactor.
Vacuum pyrolysis of ASR at 510 C and 10 kPa produces 55 wt% solid residue, 18 wt%
organic liquid phase, 16 wt% pyrolitic water and 7 wt% gas [187]. The gases contain
39 vol.% H2, 5 vol% CO, 9 vol% CH4, 29 vol% CO2 and if used as fuel can provide all
the heat required for the process. The pyrolysis aqueous phase is heavily contaminated
with soluble chemicals and the organic condensate is a low-sulfur liquid fuel containing
33 wt% mono aromatic, 9 wt% aliphatic, 4 wt% unsaturated and 25 wt% cyclic
compounds and heaving a high heating value. The oil also contains 3.5 wt% chlorinated,
9 wt% nitrogenated and 3 wt% oxygenated compounds and 4 wt% carboxylic acids.
From the pyrolysis of ASR, Day and co-workers [188] obtained gaseous and liquid
hydrocarbon products with the quantity and composition depending on the pyrolysis
temperature and the reaction time. A fast pyrolysis process at 700-850 C produces no
pyrolitic oil while screw-kiln pyrolysis at low temperatures (500 C) favours the formation
of liquids containing mainly alkanes, alkenes and mono-aromatics. Polycyclic aromatic
hydrocarbons were not detected in the liquid fraction but they were present in the solid
residue. While the gas fraction may be acceptable as fuel, the pyrolysis oil has to be
upgraded especially due to the presence of chlorine from PVC and of the flame retardants.
The auto fluff can also be recycled by injection moulding using pyrolysis products resulting
from mixed plastic wastes as melt flow improvers [189].
The flaming combustion of polymers involves heating the polymer, thermal decomposition/
pyrolysis (producing combustible and non-combustible gases, liquids, tars and solid
charred residues) and ignition in air of the volatile pyrolysis products [190]. The heating
source or the radiant energy from the flame causes the pyrolysis of the polymer. The
volatile pyrolysis products mix with air and burns in the combustion zone above the
surface producing CO2, H2O and incomplete combustion products, e.g., CO and soot.
602
Degradation Behaviour of Polymer Blends and Thermal Treatment of Polymer Waste
The pollutants in the gaseous emission are eliminated by various conventional and very
modern methods. The methods for removal of polyhalogenated aromatics and of dioxins
were recently reviewed [7]. The fly ash contains various toxic compounds such as organic
dioxins or inorganic metal oxides and heavy metals. A treatment method for the fly ash
can be plasma melting that decomposes the dioxins and produces a molten slag that can
go to landfill [191]. Treatment methods for both the gaseous emissions and the fly ash
are very costly and the incorporation of clean-up systems into the incinerators dramatically
increases the operation cost that makes incineration not economically.
References
3. J. Murphy, The Additives for Plastics Handbook, Elsevier, Oxford, UK, 1996.
603
Handbook of Polymer Blends and Composites
14. J. Llpiz, M.M. Romero, A. Jerez and Y. Laureiro, Thermochimica Acta, 1995,
256, 205.
16. R. Miranda, J. Yang, C. Roy, C. Vasile, Polymer Degradation and Stability, 1999,
64, 127.
18. R. Miranda, J. Yang, C. Roy, C. Vasile, Polymer Degradation and Stability, 2001,
72, 469.
19. J.A. Conesa, A. Marcilla, R. Font and J.A. Caballero, Journal of Analytical
Applied Pyrolysis, 1996, 36, 1.
20. A. Marcilla and M. Beltrn, Polymer Degradation and Stability, 1995, 50, 117.
21. A. Marcilla and M. Beltrn, Polymer Degradation and Stability, 1996, 53, 251.
22. M. Beltrn and A. Marcilla, Polymer Degradation and Stability, 1997, 55, 73.
25. C-H. Wu, C-Y. Chang, J-P. Lin and Y. Liang, Journal of Hazardous Materials,
1998, 58, 195.
26. W.G. Oakes and R.B. Richards, Journal of the Chemical Society, London, 1949,
619, 2929.
29. R. Inoue, S. Ouchi and S. Yasuhira, Chemistry of High Polymers, Japan, 1956,
13, 38.
30. L.A. Wall, S.L. Madorsky, D.W. Brown, S. Straus and R. Simha, Journal of The
American Chemical Society, 1954, 76, 3430.
604
Degradation Behaviour of Polymer Blends and Thermal Treatment of Polymer Waste
31. D.A. Anderson and E.S. Freeman, Journal of Polymer Science, 1961, 54, 253.
32. B. Dickens, Journal of Polymer Science: Polymer Chemistry Edition, 1982, 20,
1065.
34. R. Knmann and H. Bockhorn, Combustion Science and Technology, 1994, 101,
285.
35. C-H. Wu, C-Y. Chang and J-L. Hor, Waste Management, 1993, 13, 221.
37. R.W.J. Westerhout, J. Waanders, J.A.M. Kuiper and W.P.M. van Swaaij,
Industrial and Engineering Chemistry Research, 1997, 36, 1995.
38. J.W. Park, S.C. Oh, H.P. Lee, H.T.Kim and K.O. Yoo, Polymer Degradation and
Stability, 2000, 67, 535.
39. S.R. Urzendowski and A.H. Guenther, Journal of Thermal Analysis, 1971, 3, 379.
40. S.L. Madorsky and S. Strauss, Journal of Research of the National Bureau of
Standards, 1954, 53, 361.
41. L.A. Wall and S. Straus, Journal of Polymer Science, 1960, 44, 313.
42. T.E. Davies, R.L. Tobias and E.B. Deterli, Journal of Polymer Science, 1962, 56,
485.
43. N. Grassie and W.B.H. Leeming, European Polymer Journal, 1975, 11, 819.
46. M. Day, J.D. Cooney and C. Klein, Journal of Thermal Analysis, 1993, 40, 669.
47. M. Day, J.D. Cooney and J.L. Fox, Journal of Thermal Analysis, 1994, 42, 397.
48. C. Vovelle and H. Mellote, European Polymer Journal, 1983, 19, 387.
605
Handbook of Polymer Blends and Composites
51. G.G. Cameron and G.P. Kerr, European Polymer Journal, 1968, 4, 709.
52. G.G. Cameron and G.P. Kerr, European Polymer Journal, 1970, 6, 423.
53. K. Kishore, V.R. Pai Verneker and M.N.R. Nair, Journal of Applied Polymer
Science, 1976, 20, 2355.
54. A. Marcilla and M. Beltrn, Polymer Degradation and Stability, 1995, 48, 219.
55. R. Miranda, J. Yang, C. Roy and C. Vasile, Polymer Degradation and Stability,
1999, 64, 127.
56. P.K. Ramdoss and A.R. Tarrer, Fuel, 1998, 77, 293.
57. M.K. Loultcheva, M. Proietto, N. Jilov and F.P. La Mantia, Polymer Degradation
and Stability, 1997, 57, 77.
58. I.A. Hussein, H. Kam, S.K. Goyal, E. Karbashewski and M.C. Williams, Polymer
Degradation and Stability, 2000, 68, 381.
59. W.R. Waldman and M.A. De Paoli, Polymer Degradation and Stability, 1998, 60,
301.
60. B.B. Khatua and C.K. Das, Polymer Degradation and Stability, 2000, 69, 381.
61. Q. Yao and C.A. Wilkie, Polymer Degradation and Stability, 1999, 66, 379.
62. D.M. Bate and R.S. Lehrle, Polymer Degradation and Stability, 1998, 62, 57.
63. C. Albano and E. de Freitas, Polymer Degradation and Stability, 1998, 61, 289.
66. P. Ghosh, B. Chattopadhyay and A.K. Sen, European Polymer Journal, 1996, 32,
1015.
606
Degradation Behaviour of Polymer Blends and Thermal Treatment of Polymer Waste
69. C. Vasile, M. Brebu, H. Darie, R.D. Deanin, V. Dorneanu, D.M. Pantea and O.G.
Ciochina, International Journal of Polymeric Materials, 1997, 38, 219.
71. S. George, K.T. Varughese and S. Thomas, Polymer, 2000, 41, 5485.
73. D.M. Bate and R.S. Lehrle, Polymer Degradation and Stability, 1997, 55, 295.
74. R. Asaletha. M.G. Kumaran and S. Thomas, Polymer Degradation and Stability,
1998, 61, 431.
75. W. Meissner and D. Zuchowska, Polymer Degradation and Stability, 1998, 60, 415.
76. T. Kovacic, B. Baric and I. Klaric, Polymer Degradation and Stability, 1995, 49,
365.
77. K. Kuzelov and Z. Vymazal, European Polymer Journal, 1999, 35, 361.
78. J. Dong, P.M. Fredericks and G.A. George, Polymer Degradation and Stability,
1997, 58, 159.
79. P.P. Lizymol and S. Thomas, Thermochimica Acta, 1994, 233, 283.
80. N.A. Mohamed and M.W. Sabaa, European Polymer Journal, 1999, 35, 1731.
81. K. Pielichowski and I. Hamerton, European Polymer Journal, 1999, 36, 171.
82. S. Zulfiqar and S. Ahmad, Polymer Degradation and Stability, 1999, 65, 243.
83. B. T. Andricic and T. Kovacic, Polymer Degradation and Stability, 1999, 65, 59.
84. K.T. Varughese, Kautschuk und Gummi Kunststoffe, 1988, 41, 1114.
607
Handbook of Polymer Blends and Composites
85. M. Brebu, C. Vasile, S.R. Antonie, M. Chiriac, M. Precup, J. Yang and C. Roy,
Polymer Degradation and Stability, 2000, 67, 209.
86. C. Albano and E. De Freitas, Polymer Degradation and Stability, 1998, 61, 289.
87. K. Csomorov, J. Richly and A. Bcsi, Journal of Thermal Analysis, 1993, 35, 77.
88. K. Murata and M. Akimoto, Journal of the Chemical Society of Japan, Chemistry
and Chemical Industry, 1979, 6, 774.
89. B. Dodson and I.C. McNeill, Journal of Polymer Science: Polymer Chemistry
Edition, 1976, 14, 353.
91. R. Knmann and H. Bockhorn, Combustion Science and Technology, 1994, 101,
285.
92. M. Day, J.D. Cooney and C. Klein, Journal of Polymer Science: Polymer
Chemistry Edition, 1993, 40, 669.
93. M. Day, J.D. Cooney and J.L. Fox, Journal of Thermal Analysis, 1994, 42, 397.
96. S. Mohanty, P.G. Mukunda and G. B. Nando, Polymer Degradation and Stability,
1996, 52, 235.
97. P. Jansen and B.G. Soares Polymer Degradation and Stability, 1996, 52, 95.
98. L. Lamas, G.A. Mendez, A.J. Mller and M. Pracella, European Polymer Journal,
1998, 34, 1865.
99. A. Jha and A.K. Bhowmick, Polymer Degradation and Stability, 1998, 62, 575.
100. M. Naffakh, G. Ellis, M.A. Gmez and C. Marco, Polymer Degradation and
Stability, 1999, 66, 405.
101. N. Mayne, Proceedings of the Recycling 2000 Symposium, Jesenik, Czech Republic,
2000.
608
Degradation Behaviour of Polymer Blends and Thermal Treatment of Polymer Waste
102. Characterization of Municipal Solid Waste in the United States, 1990 Update,
EPA 530-SW-90-042, US Environmental Protection Agency, Office of Solid Waste
and Emergency Response, Washington, DC, USA, 1990.
104. Kaiser Foundation Health Plan Inc., EPRI Journal, 1994, 19, 33.
106. Annual Report on Plastic Products, Plastic Waste and Resource Recovery, Waste
Plastic Management Institute, Japan, Tokyo, 1998.
108. M. Paci and F.P. La Mantia, Polymer Degradation and Stability, 1999, 63, 11.
109. L.A. Utracki in Frontiers in the Science and Technology of Polymer Recycling,
Ed., G. Akovali, C.A. Bernardo, J. Leidner, L.A. Utracki and M. Xanthos, Kluwer
Academic Publishers, Dordrecht, The Netherlands, 1998, 343.
110. C.A. Bernardo, A.M. Cunha and M.J. Oliveira, Polymer Engineering and Science,
1996, 36, 511.
111. N.T. Dintcheva, N. Jilov and F.P. La Mantia, Polymer Degradation and Stability,
1997, 57, 191.
113. M. Brebu, M.A. Uddin, A. Muto, Y. Sakata and C. Vasile, Energy and Fuels,
2001, 15, 559.
114. M. Brebu, M.A. Uddin, A. Muto, Y. Sakata and C. Vasile, Energy and Fuels,
2001, 15, 565.
609
Handbook of Polymer Blends and Composites
118. F. Radtke, R.A. Koeppel and A. Baiker, Environmental Science and Technology,
1995, 29, 2703.
121. M. Brebu, M.A. Uddin, A. Muto and Y. Sakata and C. Vasile, Energy and Fuels,
2000, 14, 920.
122. M. Xanthos and J. Leidner in Frontiers in the Science and Technology of Polymer
Recycling, Ed., G. Akovali, C.A. Bernardo, J. Leidner, L.A. Utracki and M.
Xanthos, Kluwer Academic Publishers, Dordrecht, The Netherlands, 1998, 407.
124. W. Kaminsky and J.S. Kim, Journal of Analytical Applied Pyrolysis, 1999, 51, 127.
128. Final report of the MESO Project, Japan Eco-environment Products, Tokyo,
Japan, 2000.
610
Degradation Behaviour of Polymer Blends and Thermal Treatment of Polymer Waste
134. C. Roy, Proceedings of the Waste and Scrap in the Rubber Industry Treatment
and Legislation Conference, Brussels, Belgium, 1992.
136. E.A. Williams and P.T. Williams, Journal of Analytical Applied Pyrolysis, 1997,
40-41, 347.
137. P.T. Williams and E.A. Williams, Proceedings of the 1st International Symposium
on Feedstock Recycling of Plastics, Sendai, Japan, 1999, Paper PE-6.
141. C. Vasile, C. Savu and E. Voicu-Bosovei, La Rivista dei Combustibili, 1995, 49,
145.
142. C. Vasile, C. Savu and E. Voicu-Bosovei, La Rivista dei Combustibili, 1995, 49,
181.
146. Y. Sakata, M.A. Uddin, K. Koizumi and K. Murata, Polymer Degradation and
Stability, 1996, 53, 111.
147. M. Brebu, H. Darie, A. Spac and C. Vasile, Analele Stiintifice ale Universitatii Al.
I. Cuza Iasi, 1997, Volume 5, 25.
611
Handbook of Polymer Blends and Composites
148. W. Ding, J. Liang and L.L. Anderson, Fuel Processing Technology, 1997, 51, 47.
149. E.A. Williams and P.T. Williams, Journal of the Institute of Energy, 1998, 40,
347.
150. P.T. Williams and E.A. Williams, Journal of Analytical Applied Pyrolysis, 1999,
51, 107.
151. K. Takuma, Y. Uemichi and A. Ayame, Applied Catalysis, A: General, 2000, 192,
273.
152. G.E. Bertolini and J. Fontaine, Conservation & Recycling, 1990, 4, 203.
153. D.S. Scott, S.R. Czernik, J. Piskorz and D.St.A.G. Radlein, Energy and Fuels,
1990, 4, 407.
159. M. Brebu, M. A. Uddin, A. Muto, Y. Sakata and C. Vasile, Energy and Fuels,
2000, 14, 920.
160. M. Brebu, M. A. Uddin, A. Muto, Y. Sakata and C. Vasile, Energy and Fuels,
2001, 15, 559.
161. M. Brebu, M. A. Uddin, A. Muto, Y. Sakata and C. Vasile, Energy and Fuels,
2001, 15, 565.
612
Degradation Behaviour of Polymer Blends and Thermal Treatment of Polymer Waste
167. E.A. Williams and P.T. Williams, Journal of Analytical Applied Pyrolysis, 1997,
40-41, 347.
168. H.P. Wenning, Journal of Analytical Applied Pyrolysis, 1993, 25, 301.
170. L.L. Sharp, R.O. Ness and J.M. Sosa, Proceedings of the 45th Canadian
Chemical Engineering Conference, Qubec City, Qubec, Canada, 1995, 125.
171. Y. Sakata, M.A. Uddin and A. Muto, Journal of Analytical Applied Pyrolysis,
1999, 51, 135.
172. Y. Sakata, M.A. Uddin, K. Koizumi and K. Murata, Chemistry Letters, 1996, 3,
245.
173. A.G. Buekens and H. Huang, Resources, Conservation and Recycling, 1998, 23,
163.
174. W. Ding, J. Liang and L.L. Anderson, Fuel Processing Technology, 1997, 51, 47.
177. C. Vasile, H. Pakdel, M. Brebu, J.N. Murwanashyaka, P. Onu, H. Darie and S.M.
Cioclteu, Proceedings of the 1st International Symposium on Feedstock
Recycling of Plastics, Sendai, Japan, 1999, Paper PE-4.
613
Handbook of Polymer Blends and Composites
178. C. Vasile, M. Brebu, Y. Sakata, H. Pakdel, Ch. Roy and R. Miranda, Journal of
Environmental Protection and Ecology, submitted.
179. C. Vasile, M. Brebu and H. Darie, Proceedings of The XXII Scientific Session of
Communications, Research Center OLTCHIM S.A., Romania, 1996, Volume 3,
1362.
181. C. Vasile, A. Paun and C. Savu, La Rivista dei Combustibili, 1997, 51 161.
183. T. Kamijo, Kobe Steel Ltd., Chuo-ku, Kobe, Private communication, 1991.
184. J.S. Chang, Y. Takakura, S. Jamal, S. Ono and S. Teii, Proceedings of the 4th
International Thermal Plasma Process Conference, Japan, 1996, 403.
185. J.A. Conesa, R. Font and A. Marcilla, Journal of Analytical Applied Pyrolysis,
1997, 43, 83.
186. L.A. Utracki in Frontiers in the Science and Technology of Polymer Recycling,
Ed., G. Akovali, C.A. Bernardo, J. Leidner, L.A. Utracki and M. Xanthos, Kluwer
Academic Publishers, Dordrecht, The Netherlands, 1998, 341.
188. M. Day, Z. Shen and J.D. Cooney, Journal of Analytical Applied Pyrolysis, 1999,
51, 181.
190. R.E. Lyon, Polymer Degradation and Stability, 1998, 61, 201.
191. S. Suzuki and T. Minoura in Frontiers in the Science and Technology of Polymer
Recycling, Eds., G. Akovali, C.A. Bernardo, J. Leidner, L.A. Utracki and M.
Xanthos, Kluwer Academic Publishers, Dordrecht, The Netherlands, 1998, 437.
614
Singular Thermal Behaviour of Polystyrene/Polydimethylsiloxane Blends
15
Singular Thermal Behaviour of
Polystyrene/Polydimethylsiloxane Blends
Sergei M. Lomakin, Elena V. Koverzanova and Guennardi E. Zaikov
15.1 Introduction
During the preparation of this chapter it was assumed, that the title should start from
the word unpredictability. Meanwhile, we have decided to replace this word by the
word singularity because it is quite possible to predict the polymer blends behaviour if
one could know their mechanisms of thermal and thermo-oxidative degradation.
For the first time the singularity of thermo-oxidation of organic compounds has been
discovered by Emanuel and co-workers [1]. They studied the co-oxidation of decane (C10H22)
and ethylbenzene (C6H5C2H5). It was established that in liquid phase at a temperature of
about 150 C, decane can be easily oxidised to decyl hydroperoxide (C10H21OOH). Under
these conditions the ethylbenzene is easy oxidised in the appropriate hydroperoxide
(C6H5CH(OOH)CH3) as well. However, in a mixture with decane ethylbenzene it is oxidised
even faster and decane completely ceased to oxidise in a mixture.
The reason of this phenomena is a special mechanism of co-oxidation of these two organic
compounds. Let us designate the decane as R1H and the ethylbenzene as R2H. Then the
mechanism of co-oxidation process up to the first molecular product, hydroperoxide,
looks as follows:
R2 + O2 R2OO (15.4)
615
Handbook of Polymer Blends and Composites
The basic feature of the decane oxidation and acceleration of the process of ethylbenzene
oxidation in their mixture is described by reaction 15.1 when the alkyl radical of decane
reacts with ethylbenzene. Under this condition the molecule of decane is regenerated and
the molecule of ethylbenzene converts into the alkyl radical.
The rate of reaction 15.1 depends on the bond energy of the R2-H, bond. The less the
bond energy, the greater will be the rate of this process. Thus, nonadditivity during
oxidation processes of organic mixtures is determined by the bond energies of oxidising
components in a mixture.
One could imagine two inhibitors of oxidation process: amine (Am-H) and phenol (Ph-
OH). It is known that the mechanism of their inhibitive action consists of inhibitor
interaction with the peroxide radical (ROO.) of the oxidising organic compound (RH).
k
AmH + RO2
7
Am + ROOH (15.6)
k
PhOH + RO2
7
PhO + ROOH (15.7)
Let us consider the case where the amine is a much more effective inhibitor then phenol.
Kinetically this means that the rate constant of reaction, k7, is considerably higher than
the rate constant, k7, k7 >> k7, and the induction period of the oxidation process in the
presence of amine () is much higher than the period of induction upon the oxidation at
presence of phenol ():
>>
Figure 15.1 illustrates the dependence of the induction period during ethylbenzene
oxidation on a ratio of amine and phenol in a mixture at various concentration of
inhibitors.
616
Singular Thermal Behaviour of Polystyrene/Polydimethylsiloxane Blends
Such behaviour is determined by the reaction of regeneration of the amine by the phenol:
Thus, the amine is consuming the phenol. The rate of this reaction depends on the bond
energy of PhO-H. These two examples give an idea of realisation of a similar model for
organic polymer blend thermal oxidation and combustion processes.
In general, many polymers are blended together to enhance physical and mechanical
properties. In the present paper we evaluate changes in thermo-oxidative stability that
can be attributed to blending. On the other hand, we were interested to know whether
blending may provide a method to control thermal stability.
There are some publications which describe different polymer mixtures thermal
degradation behaviour. Richard and Salter studied the thermal degradation of polystyrene
(PS) blended with poly--methylstyrene [3]. They found an increased yield of styrene
617
Handbook of Polymer Blends and Composites
Mizutani and co-workers [4] suggested that polypropylene (PP) degradation is accelerated
by the presence of vinyl polymers, such as polystyrene or polymethylmethacrylate. During
the degradation process the polymers were incorporated into the polypropylene chain,
forming graft or block copolymers by a mechanism involving attack by a vinyl polymer
radical on the polypropylene chain.
Gardner and co-workers [5] studied the rates of degradation of polystyrene and poly--
methylstyrene with acrylate polymers and reported that the rates of degradation of blends
were significantly different from those measured for the polymers.
McNeill [7-9] studied a variety of blends and considered the degradation of many polymer
blends to fall into one of two categories:
In the work of Bate and Lehrle [10], an attempt has been made to determine to what extent
the following factors are important to evaluate the rates and mechanism of the degradation
of blends (PMMA/PS, PMMA/high density polyethylene (HDPE), PMMA/PVC):
3. Cross-reaction.
The results show that heterogeneous blends degrade predominantly within their phase-
separation regions, but may also give rise to some cross-products, which may form by
618
Singular Thermal Behaviour of Polystyrene/Polydimethylsiloxane Blends
small radical or molecule migration across these phase boundaries. For the systems studied
this cross-product formation is much a secondary effect, because the observed rate of
formation of methyl methacrylate (MMA) monomer from the heterogeneous blends was
not found to be significantly different from that shown by the homopolymer. Pyrolysis
of homogeneous blends has been shown to involve interacting mechanisms, which may
or may not lead to cross-product formation. However, rates of formation of major
pyrolysis products are found to change significantly, in some cases by one order of
magnitude, and these results have been interpreted in terms of degradation mechanism.
In particular, it has been suggested that the following effects are involved in the degradation
mechanism of homogeneous blends:
15.2 Experimental
Polymer blends were prepared as described previously [11]. The PS was the commercial
sample used in the earlier investigation. The PDMS was also a commercial product, with
ultrahigh molecular weight (106), methyl end groups and a gummy texture.
The pyrolysis of PS, PDMS and their blends (80 phr of PS:20 phr of PDMS) was done in
a pyrolysis tube cell at temperatures of 300 C, 400 C, 500 C or 600 C in air (flow
rate 30 cm3/min). The products of pyrolysis were dissolved in hexane at 0 C.
Extensive pyrolysis of samples over a one minute period of time was carried out in a
laboratory model pyrolyser in an air atmosphere at 400 C. The oven temperature was
monitored with a thermocouple and a stability of 5 C was attainable.
619
Handbook of Polymer Blends and Composites
A vertical TG balance Derivatograpf 950Q (nitrogen or air flow of 100 cm3/min) was
used for a kinetic study of thermal degradation of the polymer samples.
Samples of degradation products have been analysed by gas chromatography analysis with a
Zvet 500M using electron capture detector. A glass column (4 m x 3 mm) filled with OV-17
(phenylmethyl silicone) was used in the GC analysis. Temperature of column was 230 C.
GC/MS analysis of samples was performed using a Varian 3300 gas chromatograph
with mass spectrometry detection (ion trap) Finnigan MAT ITD 800. A DB-5 fused
capillary column (30 m x 0.32 mm), temperature programmed from 50 C to 270 C at
10 C per minute was used with MS detection. Mass spectra detection was accomplished
in the electron impact mode scanned from 40 to 650 Da with the energy of 70 eV. All
mass spectra were searched against the National Bureau mass spectral library.
A/ = E/RTmax2 exp(E/RTmax )
where A is the pre-exponential factor; = dT/dt is the heating rate and Tmax is the
temperature at which the mass flux from solid to gas is a maximum. When the rate of
change of mass flux (c) with time is zero:
620
Singular Thermal Behaviour of Polystyrene/Polydimethylsiloxane Blends
d2(1-c)/dt2 = 0
Using the equation A/ = E/RTmax2 exp(E/RTmax) the activation energies and pre-
exponential factors were determined [11].
It has been found that the values of activation energy and pre-exponential factors of PS,
as a part of blend and the thermal degradation in nitrogen are about 30% more than for
pure PS. On the other hand, the values of activation energy and pre-exponential factors
of PDMS, as a part of, thermal degradation in nitrogen are about 25% lower than for
pure PDMS [11, 12].
For PS as a part of blend, the products of degradation of PS diffuse through the phase
boundary, which cause interaction with PDMS polymers. Preliminary GC-MS analysis
of products of degradation from the blend were made at temperatures 500 C and 700 C
(flash pyrolysis). The detailed analysis of the mechanism of thermal degradation of blends
is discussed in [11]. Apparently, PDMS acts as an inert component slowing down the
termination reaction by dilution of macroradicals formed in random scission degradation
process of PS-component (see Figure 15.2). On the other hand, it stabilises the PS by
means of interpolymer recombination which leads to cross-products of thermal
degradation [11].
621
Handbook of Polymer Blends and Composites
This dilution leads to reduction of the rate of termination and increases the overall rate
of the thermal degradation of the PS-part. As a result of this phenomena, the energy of
activation of the PS-part can be lowered. However, the visual effect of stabilisation of
PS-part in blend can be explained by way of interpolymer recombination which caused
the stabilisation of the first degraded component, PS.
For the second component of the blend, PDMS, the values of activation energy and pre-
exponential factor are about 25% lower than for pure PDMS [11].
This means that the second component is becoming less thermally stable in the blend
than when heated alone. Since we have found cross-degradation products (see the
Section 15.2.3) for these components, the behaviour of the blend can be explained by
kinetic factors. During the decomposition of the second component of PDMS, the
interpolymer recombination caused the termination of radical depolymerisation
(propagation step) of PS. On the other hand, the process of PDMS-component thermal
degradation is accelerated by means of radical initiation of polystyrene macroradical
fragments. That is why it becomes less thermally stable.
PS: The analysis and identification of chromatograms and mass spectra showed that
the major pyrolysis product of pure PS at 300 C, 400 C, 500 C and 600 C in air
is styrene (54.9%, 76.2%, 67.0% and 57.9%, respectively, see Table 15.1). Obviously,
the process of radical depolymerisation (initiated by oxygen) remains the dominant
reaction upon the PS degradation at the temperature range of 300-600 C [13].
At 300 C the most abundant products of PS thermal oxidative degradation are styrene
(54.9%) and benzaldehyde (29.4%) (Table 15.1). At 400 C among the identified products
are: methylbenzene (8.4%), ethylbenzene (2.3%), allylbenzene (0.6%), -methylstyrene
(5.4%), etc., (see Table 15.1). Under pyrolysis conditions at 500 C an increase in the
degradation product concentrations has been noticed. The major amounts of the products
(except for styrene) are methylbenzene (3.2%), benzaldehyde (7.2%), benzacetaldehyde
(4.6%), ethylbenzene dimer (3.1%) etc., (see Table 15.1).
At 600 C a further increase of the degradation products is clearly seen. The condensed
aromatics occur except for the products of pyrolysis listed above (naphthalene (0.6%),
benzofuran (0.3%), biphenyl (0.1%)) (see Table 15.1).
PDMS: The most abundant products of PDMS thermal and thermal oxidative
degradation are cyclotrisiloxane hexamethyl (D3) and cyclotetrasiloxane octamethyl
(D4) followed by cyclopentasiloxane decamethyl (D6) (Table 15.2). Increase in
temperature causes an increase in the contribution of cyclic siloxane oligomers, such
as cyclohexasiloxane dodecamethyl, cycloheptasiloxane tetradecamethyl (D7) and
622
products of PS Thermal degradation
623
products of PS Thermal degradation
624
Singular Thermal Behaviour of Polystyrene/Polydimethylsiloxane Blends
At temperatures higher than 400 C a new product has been found which was supposedly
identified as 1,3,3,5,5-pentamethyl cyclotrisiloxane-1-heptamethyl pentasiloxane.
Characteristic peaks for this compound are m/e = 73, 102, 193, 341, 429 (see Figure 15.3).
The products of PDMS thermal oxidative degradation are given in Table 15.2.
PS/PDMS blends
At 300 C the most abundant products of the blend are the products of PS thermal oxidative
degradation: styrene (14%) and benzaldehyde (31%) (Table 15.3). The composition of the
products of blend pyrolysis at 400 C and 500 C does not differ from the basic product
distribution of the initial PS and PDMS pyrolysis (Tables 15.1, 15.2, 15.3):
625
products of PDMS Thermal degradation
626
products of PDMS and PS blend Thermal degradation
627
products of PDMS and PS blend Thermal degradation
628
products of PDMS and PS blend Thermal degradation
629
Handbook of Polymer Blends and Composites
630
Singular Thermal Behaviour of Polystyrene/Polydimethylsiloxane Blends
Apparently, these two compounds are formed by the interaction of radical PS-fragment
and the PDMS matrix.
Scheme 15.1 shows assumed radical chain kinetic model for the thermal degradation of
PS with the presence of PDMS fragments [11, 14].
In particular, in PS/PDMS blends the reaction steps of the main process are shown in
Scheme 15.2: chain initiation to form primary radicals: end chain initiation (1), random
scission (1a), b-scission (1b);
631
Handbook of Polymer Blends and Composites
632
free-radical mechanism
633
free-radical mechanism
634
free-radical mechanism
635
Handbook of Polymer Blends and Composites
Figures 15.8-15.12 illustrate the difference in the thermal degradation of PS, PDMS and
their blends at a heating rate of 10 C/min in nitrogen and in air. From these figures can
be seen the influence of oxygen on the degradation process. The thermal degradation of
PDMS was studied by Camino and co-workers [12]. The presence of two stages in PDMS
(Figure 15.8) in air was explained by chain-end oxygen-catalysed depolymerisation
followed by random-scission and crosslinking oxidative process [12]. The limiting stage
of this process is ultimately the diffusion and evaporation of monomer fragments.
In polymer blend of PS/PDMS the dilution of the basic component (PS) by the PDMS
leads to its stabilisation and destabilisation of the second one, PDMS. This phenomenon
is explained in terms of interpolymer recombination. Two of the degradation products:
2-phenyl-4(1,3,3,5,5-pentamethylcyclotrisiloxane)-butane and 2-phenyl-
4(1,3,3,5,5,7,7-heptamethylcyclotrisiloxane)-butane were assigned to the products
of cross-interpolymer recombination which can accelerate the process of PDMS
depolymerisation by means of radical initiation of PS* fragments [11].
636
Singular Thermal Behaviour of Polystyrene/Polydimethylsiloxane Blends
Figure 15.9 TG and DTG curves of PS in nitrogen at the heating rate of 10 C/min
Figure 15.10 TG and DTG curves of PS in air at the heating rate of 10 C/min
637
Handbook of Polymer Blends and Composites
Figure 15.11 TG and DTG curves of PS/PDMS blend (80:20) in nitrogen at the
heating rate of 10 C/min
Figure 15.12 TG and DTG curves of PS/PDMS blend (80:20) in air at the heating rate
of 10 C/min
638
Singular Thermal Behaviour of Polystyrene/Polydimethylsiloxane Blends
An extreme combustion behaviour for system PP-PP-co-PE [15] has been found, which
can be explained in terms of oxidative degradation of polymer materials. The features of
autooxidation and combustibility of blends of isotactic PP and (PP-co-PE) as thin films
(50-80 m) were studied. ASTM D2863 [16] {(ISO 4589 Part 2) [17]} was used to
determine the limiting oxygen index (LOI) for polymer films 140 by 52 mm.
The correlation between polymer thermal oxidative degradation and its combustion under
diffusion flame conditions may represent an interesting specific application. In general,
the solid phase polymer reaction can play a very important role in the reduction of
polymer combustibility. If we can decrease the reaction ability of a polymer relative to
oxygen, the critical conditions of the diffusion flame stability would change. These
polymers would have a different combustibility. It was shown that the composition of
PP/PP-co-PE (62:38) has the highest induction period of auto-oxidation [15, 18].
The combustibility tests confirm this hypothesis (Table 15.4) [15]. The sample with the
minimal reaction activity, PP/PP-co-PE (62:38), has the highest value of LOI (21) and the
char yield (3.4%).
The application of different polymer blends to alter the fire behaviour of polymeric
materials can be potentially useful in obtaining novel flame retardant materials. It can
be used by itself or can be used as basis to achieve ecologically-friendly flame resistant
plastics.
639
LOI PP/PP-co-PE
char yield PP/PP-co-PE
Table 15.4 LOI and char yield for PP/PP-co-PE compositions [15]
Composition LOI Char yield, %wt.
PP 17 0
PP/PP-co-PE (37.5/62.5) 19.5 1.5
PP/PP-co-PE (61.5/38.5) 21 3.4
PP-co-PE 18.5 0.5
References
2. V.A. Kritsman, G.E. Zaikov and N.M. Emanuel, Chemical Kinetics and Chain
Reactions. Historical Aspects, Nova Science Publishers, New York, NY, USA,
1995.
8. I.C. McNeill and M.A. Mohammed, European Polymer Journal, 1972, 8, 975.
9. I.C. McNeill and M.A. Mohammed, Polymer Degradation and Stability, 1995,
50, 285.
10. D.M. Bate and R.S. Lehrle, Polymer Degradation and Stability, 1998, 62, 57.
640
Singular Thermal Behaviour of Polystyrene/Polydimethylsiloxane Blends
11. S.M. Lomakin, E.V. Koverzanova, N.E. Shilkina and G.E. Zaikov,
Khimicheskaya Fizika, 2002, 21, 2971.
12. G. Camino, S.M. Lomakin and M. Lazzari, Polymer, 2001, 42, 6, 2395.
15. S.M. Lomakin, L.S. Shibryaeva and G.E. Zaikov, Chemical Physics Reports,
1999, 17, 12, 2337.
16. ASTM D2863, Standard Test Method for Measuring the Minimum Oxygen
Concentration to Support Candle-Like Combustion of Plastics (Oxygen Index),
2000.
18. L.S. Shibryaeva, A.A. Veretennikova, A.A. Popov, T.A. Gugueva and A.A.
Kanauzova, Polymer Science Series A, 1999, 41, 4, 472.
641
Abbreviations
643
Handbook of Polymer Blends and Composites
644
Abbreviations
CD Conjugated diene
CED Cohesive energy density
CER Cyanate ester resin
CMC Carboxymethylcellulose
CMc Critical micell concentration
Co Copolymer
COP Polyethylene terephthalate-co-isophthalate copolymer
CoPC Copolycarbonate
CoPES Copolyester
CoPO Copolyolefin
CPE Chlorinated PE
CPP Polypropylene-co-ethylene
CPU Computer processing unit
CPVC Chlorinated PVC
CRA Crosslinking agents
CRADA Cooperative Research and Development
CRC Crystallisation rate coefficient
CS Chitosan
CSC Critical surface concentration
CT-I Charge-transfer interaction
D Diffusion coefficient
DADPM Diaminodiphenyl methane
DAP Diallyl phthalate
DBS Dibenzylidene sorbitol
DBSA Dodecylbenzenesulfonic acid
DBTDL Dibutyl tin dilaurate
DCE 1,2-Dichloroethane
DCM Dichloromethane
DCP Dicumyl peroxide
DD Deacetylation degree
DDRM Deformed drop retraction method
645
Handbook of Polymer Blends and Composites
DDS Diaminodiphenylsulfone
DEM Diethyl maleate
DES Diethylsuccinate
DETA Diethylenetriamine
DGEBA Diglycidyl ether of bisphenol A
DHBP 2,5-Bis(tert-butylperoxy)-2,5-diethyl hexane
DIN Deutsche Insitute fr Normung
DMA Dynamic mechanical analysis
DMA-co-EMA Poly(dodecyl methacrylate-co-ethyl methacrylate
DMPAP 2,2-Dimethoxy-2-phenylacetophenone
DMS Dynamic mechanical spectroscopy
DMTA Dynamic mechanical thermal analysis
DNA Deoxyribonucleic acid
DOD Dodecanoated
DPC Differential photocalorimetry
dpf Denier per fibre
DRIFT Diffuse reflectance infrared spectroscopy
DSC Differential scanning calorimetry
DTA Differential thermal analysis
DTG Differential thermal gravimetric analysis
DTGA Derivative thermogravimetric analysis
DTM Differential thermo-mechanical analysis
DVB Divinyl benzene
E Ethylene
Ea Activation energy
EA Ethylacrylate
EB Electron beam
EBA Ethylene butyl acrylate
EC Ethylcellulose
EEA Ethylene-ethacrylate copolymer
EGA Evolved gas analysis
646
Abbreviations
647
Handbook of Polymer Blends and Composites
648
Abbreviations
649
Handbook of Polymer Blends and Composites
650
Abbreviations
651
Handbook of Polymer Blends and Composites
652
Abbreviations
653
Handbook of Polymer Blends and Composites
PGDO Polyglycidol
PH Phenoxy
PHA Polyhydroxy alkanoic acid
PHB Poly hydroxybutyrate
PHB-HV Poly(3-hydroxybutyrate-co-3-hydroxyvalerate)
PHEA Poly(hydroxyethyl acrylate)
PHE-Bis A Polyhydroxyether of bis phenol A
PHECA Poly([?-hydroxyethyl)-3,5-dinitrobenzoate]acrylate
PHECM-2 2-Poly([N-2-(hydroxyethyl)-carbazolyl]methacrylate)
PhOH Phenol
PhP Phenolic polymer
phr Parts per hundred rubber
PHSC Perturbed hard-sphere chain
PI Polyisoprene
PIB Polyisobutylene
Pi-PMA Polyisopropyl methacrylate
PISO2 Polyimidesulfone
PKVS Poly(potassium vinyl sulfate)
PLA Polylactic acid
PLA-co-GA Poly(lactide-co-glycolide)
PLC Polymer liquid crystal
PMA Polymethacrylate
PMAA Poly (methacrylic acid)
PMEA Polymethyl ethyl acrylate
PMMA Polymethyl methacrylate
PMMI Poly(itaconic acid monomethylester)
PMPS Polymethylphenylsiloxane
PmpSuc Poly(2,2-dimethyl-1,3-propylene succinate)
p-MSAN p-Methyl styrene-acrylonitrile
PMVE Perfluoromethyl vinyl ether
PNaA Poly(sodium acrylate)
654
Abbreviations
655
Handbook of Polymer Blends and Composites
PS(t-OH) Poly[styrene-co-p-(2-hydroxypropane-2-yl)styrene]
PSA Polysilicic acid
PSCS Polystyrene-stat-chloromethyl styrene
PSf Polysulfone
PSS Poly(styrene sulfonate)
PSSA Poly(styrenesulfonic acid)
p-t-BB-PPO p-t-Butylbenzoyl poly (phenylene oxide)
ptc Positive temperature coefficient of resistance
PTFE Polytetrafluorethylene
PTMEG Polytetramethylene ether glycol
PTT Polytrimethylene terephthalate
PU Polyurethane(s)
PUI Polyurethane ionomer
PUp Polyurethane prepolymer
PVA Polyvinyl alcohol(s)
PVAc Poly(vinyl acetate)
PVB Polyvinyl butyrate
PVBTAC Poly(vinyl benzyl trimethylammonium chloride)
PVC Polyvinyl chloride
PVCM Polyvinyl crosslinking monomer
PVDC Polyvinylidene chloride
PVDF Polyvinylidene fluoride
PVE Polyvinyl ester
PVEDEG Polyvinyl ether of diethylene glycol
PVEEG Poly(vinyl ether of ethylene glycol)
PVH Polyhydroxyvalerate
PVME Polyvinylmethyl ether
PVP Poly(N-vinyl pyrrolidone)
PVT Pressure-volume-temperature
r Random copolymer
r0 Capture radius for radical recombination
656
Abbreviations
RC Regenerated cellulose
RF Radio frequency
RH Relative humidity
RPE Radiation-crosslinked PE
rpm Revolutions per minute
RTFOT Rolling thin fim oven test
S Styrene
S/MA Styrene-maleic anhydride copolymer
S/S Side-by-side structure
SA Sebacic acid
SAD Succinic dimethacrylate
SAFT Static associated fluid theory
SAlg Sodium alginate
SAN Styrene-acrylonitrile copolymer
SANS Small angle neutron scattering
SAXS Small angle scattering of x-rays
SB Styrene-butadiene copolymer
SBH 4, 4-Dihydroxybiphenyl 4-hydroxy benzoic acid
SBR Styrene-butadiene-rubber
SBS Styrene-butadiene-styrene
S-co-MMA Styrene-MMA copolymer
SD Spinodal decomposition
SDMS Poly(styrene-co-dimethyl siloxane)
S-EB Styrene-(ethylene-butene)
SEBS Styrene-ethylene-butene-styrene copolymer
SEM Scanning electron microscope
Semi-IPN Semi interpenetrating polymer network
Seq-IPN Sequential interpenetrating polymer network
SG Plasticised starch with glycerol
SHS Styrene hydroxy styrene
Si Silicon
657
Handbook of Polymer Blends and Composites
658
Abbreviations
TG Thermogravimetry
TGA Thermogravimetric analysis
Thermid RTM Acetylene-endcapped polyimides for resin transfer moulding
THF Tetrahydrofuran
ti Oxidation time
Tm Melting temperature
TMAEAC 2-Trimethylammonioethylacrylate chloride
Tmax Maximum temperature
TMOS Tetramethoxysilane
TMP Trimethylolpropane
TMPC Tetramethyl polycarbonate
TMPTA Trimethylpropane
TNPP Tris-nonyl phenyl phosphite
TPE Thermoplastic elastomer
TPP Thermoplastic polymer
TPS Thermoplastic starch
TPU Thermoplastic PU
Ts Spinodal temperature
TVA Thermal volatilisation analysis
UAR Urethane acrylate resin
UBMI Urethane modified bismaleimide
UCST Upper critical solution temperature
UHMPE Ultra high modulus PE
UHMWPE Ultra high molecular weight polyethylenene
ULDPE Ultra low density PE
UOY Undrawn oriented yarn
UPE Unsaturated polyester
UR Urethane resin
UV Ultraviolet radiation
VA Vinyl acetate
VB Vectra B950
659
Handbook of Polymer Blends and Composites
660
Polyolefin Blends
1
Polyolefin Blends
1.1 Introduction
Polyolefins are a very important class of commercial polymers, being used in a wide
range of applications. Starting in 1950, new catalysts for olefin polymerisation were
discovered and the development lead both to an improvement of the polyolefins quality
and also to diversification of their applications.
The individual members of the polyolefin family offer a fairly broad spectrum of structures,
properties and applications. This spectrum can be broadened even further by blending
individual polyolefins with other polymers. Furthermore, many other polymers can be
improved by adding polyolefins to them and by compatibilisation.
Table 1.1 gives a list of the preparation methods and properties of several polyolefin
blends.
Several reviews on polyolefin blends have been published [202, 203]. The polyolefin
blends could be classified in two groups: blends containing only polyolefins and blends
of polyolefins with other polymers
Major commercial blends of polyolefins with each other include addition of low molecular
weight polyethylene (LMWPE) to ultra-high molecular weight polyethylene (UHMWPE)
to improve processability, addition of LDPE to LLDPE to improve processability, addition
of EPDM to PP to improve low temperature impact strength and addition of PP to
EPDM to produce thermoplastic elastomers.
Different blends of polyethylenes were studied, including HDPE, LDPE, LLDPE and
UHMWPE. The single-screw extruder or the twin-screw extruder were used for preparing
these blends which were characterised by morphology, mechanical properties, rheology,
1
2
methods of preparation
Polyolefin blends
viscosity [9];
- Improvement of the tensile strength and
elongation at break; small change in the
elasticity of the melt [1];
2. HDPE/LLDPE - melt; SEM, - The co-crystallisation effect upon the 10-14
- single screw Mechanical, increasing of flexibility, impact strength,
extruder; XRD, tensile strength and modulus [10, 11];
- twin-screw IR, - Improvement of cracking resistance [12];
extruder; DSC, - A network between crystallites and tie
- roll-mill; WAXS macromolecules which contain short
- solution; branches is obtained [12];
Table 1.1 Continued
No. Blend Compatibilising Methods of study Observations Ref.
agent and
preparation methods
2. Continued - The half-time of crystallisation increases until
30% HDPE and is constant in the 30%-70%
range, then increases suddenly [13];
- Several blending methods have been tested;
methods of preparation
Polyolefin blends
the blends obtained in twin-screw extruders
have the best mechanical properties [14];
3. UHMWPE/ - melt; Rheology, - Good mechanical properties [9, 15, 16]; 9,
LDPE, - roll-mill; Morphology, - Small amounts of PP add for adjusting of 15-18
LLDPE, HDPE - internal mixer; Mechanical, viscosity [9];
DSC, - Modification of the morphology on drawing
films from melt-state [17];
- LLDPE/HDPE blends co-crystallise [16];
- The influence of blending conditions upon
the compatibility [18];
- A blend of UHMWPE/LLDPE is more
homogenous when it is prepared in two
blending steps [18].
4. LDPE/LLDPE DCP [19]; Rheology, - Blends with 20%-30% have better 19-27
EPDM [20, 21]; SANS, processability [22];
- melt; DSC, - Homogeneous blends until MW = 105 and 4
- single screw Mechanical, branches/100 atoms C [23];
Polyolefin Blends
extruder; Dynamic - Distinct melting peaks [24]; In the 20%-40%
- twin-screw mechanical; range of composition the mechanical
extruder; properties are better than those obtained
- Brabender mixer; additively [25];
- The influence of the structure upon the
elasticity of melt [26];
3
4
methods of preparation
Polyolefin blends
with methanol at - Crystallinity blend [30];
100 C [29];
6. sPP/iPP - solution; WAXS, - Crystallisation; 32,
SAXS, - Phase separation; 33
SEM,
DSC;
7. PP/UHMWPE - solution in xylene; Mechanical, - Improved mechanical properties; 34
SEM, - A better compatibility of UHMWPE than
Thermal; HDPE due to a greater flexibility of the
chain;
Table 1.1 Continued
methods of preparation
Polyolefin blends
No. Blend Compatibilising Methods of study Observations Ref.
agent and
preparation methods
8. iPP/HDPE EPDM [35, 36]; DSC, - Incompatible [39]; 35-41
EVA [35, 36]; Mechanical, - The blends with 10%-20% have improved
EP block copolymer Dynamic, tensile strength and thermal properties [39];
[37]; Rheology, - Improved mechanical properties [35, 40];
Grafted co-polymers Microscopy, - EVA and EPDM are more efficient, all the
and EP, WAXS, properties increase and EVA does not
EPB block SAXS, influence the crystallinity of PP [35, 36];
copolymers [38]; SEM, - Incompatible in the 80%-20% range
- melt; compositions [37];
- twin-screw - Improvement of the properties for both iPP-
extruder; rich and HDPE-rich blends [37];
- Decrease of spherulite size [38];
- In EPB blends, the temperature of
crystallisation, the tensile strength and the
tensile modulus increase [38];
- Hydrogenated oligo(cyclopentadiene) [41];
- Crystallisation [41];
9. PP/LLDPE EP block SEM, - Kinetics of crystallisation [43-46]; 42-49
copolymers [42]; Mechanical, - The influence of the thermal treatment:
SEBS [43]; Rheology, temperature (190-280 C) and time (0-290
Polyolefin Blends
EP [44]; DSC, min) [45, 46];
SEBS [44]; TEM, - Mechanical stress-strain curves [47];
SEBS-g-MA [44]; SAXS, - Correlation between morphology and
- melt; DMTA, properties [48];
- single screw Microscopy; - The conditions used in preparation have
extruder; influence upon the morphology [49];
5
6
methods of preparation
Polyolefin blends
60 rpm, 15 min WAXS, EPDM [55];
[50]; SAXS, - The interfacial layer is formed by the
- premixing EPDM DMTA, migration of PP molecules in the elastomeric
and peroxide on SEM; phase [55];
roll-mill, then with - The improvement of impact strength [57];
PP in Brabender - Strain mechanism [57];
[51]; - Smaller diameter of EPDM particles
- internal mixer [52]; (~0.1 m) [57]
- Ionomer with Na+ has compatibiliser effect
at 5% concentration [50];
- A dynamic vulcanisation [51];
Table 1.1 Continued
methods of preparation
Polyolefin blends
No. Blend Compatibilising Methods of study Observations Ref.
agent and
preparation methods
10. Continued - Thermoplastic IPN [51];
- Dynamic crosslinking [58];
- The glass transition temperature decreases;
- The adhesion of phases increases [58]
- Dynamic vulcanisation [52];
- Better properties in presence of ionomer salt
Zn2+ [52];
11. PP/EPR Ethylene-olefin-1 SAXS, - EPR dissolves in amorphous PP [61, 62]; 57,
copolymer [59] TEM, - Improved impact strength [57, 61-63]; 59-66
agent/co-agent [60] Mechanical- - Above 40% EPR, a co-continuous structure
- melt; dynamic, is obtained as a result of a spinodal
- single screw SEM, decomposition [61, 62];
extruder; HVEM, - The influence of the ethylene content in
- twin-screw Morphology, copolymer [64];
extruder; NMR, - The influence of EPR structure [63];
Rheology, - Flow resistance and modulus decrease with
Thermal, the content of EPR [63];
Mechanical - The glass transition temperature of PP
changes very little [65];
- The effect of grafting MA [66];
Polyolefin Blends
- The viscosity exhibits a positive deviation
from the additive behaviour [67];
- The influence of viscosity ratio upon the size
of particles [67] and mechanical properties
[60];
- Crystallisation [67];
7
8
methods of preparation
Polyolefin blends
DSC;
14. iPP/PE REM, - Compared to EPDM, the elastomers of PE 69
elastomers SEM, improve the processability;
DSC;
15. PP/ethylene- DSC, - The blends with copolymers rich in butene-1 70
olefin-1 XRD, and hexene-1 are miscible with amorphous
elastomer TEM, PP;
Mechanical,
Dynamic;
methods of preparation
Polyolefin blends
Table 1.1 Continued
No. Blend Compatibilising Methods of study Observations Ref.
agent and
preparation methods
16. PP/PB OM, - Miscible; 71
SEM, - The thermal treatment induces a partial
SAXS phase separation of the components;
- Crystallisation;
17. LLDPE/ EP- DSC, - A negative deviation of viscosity from 72
butene-1 SEM, additive behaviour; Improved processability
Rheology, and mechanical properties for the
Mechanical, compositions having 20% terpolymer;
Dynamic;
18. LLDPE/ - Solution in xylene TEM, - Phase separation; 73
HPB-MMA at 120 C and SEM, - Morphology;
diblock precipitation with WAXS,
copolymers alcohol; DSC;
19. PE/SB, - melt, Rheology - The influence of different copolymers upon 74
SBS, - extruder; the rheological behaviour: diblock, triblock,
SIS starblock or statistical
20. EVA/LDPE XRD, - The blends 50/50 are miscible and the 75,
Mechanical, Gordon-Taylors equation is verified [75]; 76
DSC; - Improvement of the tensile strength [75, 76];
Polyolefin Blends
- Crystallisation [76];
21. PP/EVA - melt, Mechanical, - 10% EVA have a better effect than EPDM 77-79
- single screw SEM, [77, 78];
extruder, Rheology, - Improved mechanical properties [77-79];
TEM; - Morphology [79];
9
10
methods of preparation
Polyolefin blends
26. HDPE/EPDM- - melt; WAXD, - 20% functionalised EPDM improves impact 83
functionalised - roll-mill, 140 C, DSC, strength;
15 min; SEM;
27. f-HDPE/ Butyl peroxide; SEM, - Reactive extrusion in presence of carbon 84
EPDM - melt; Mechanical; black;
- reactive extrusion; - Very high tensile strength and modulus;
28. PP, EP block DMTA, - Stronger interactions in blends with PP-g- 85
copolymer/PP- SEM, MA;
g-MA
methods of preparation
Polyolefin blends
Table 1.1 Continued
No. Blend Compatibilising Methods of study Observations Ref.
agent and
preparation methods
29. EPDM/LLDPE Dicumyl peroxide; Mechanical, - Improvement of the tensile strength and 86
- melt; Thermal, modulus;
- internal mixer; Rheology; - The crystalline structure of LLDPE isnt
affected;
30. PE/PS SBS [87]; DMTA, - Testing of different types of SBS [87]; 87-93
Block and sequential SEM, - Improved dispersion [87, 91];
copolymers HPB-b- Multifractal - Waste recovery [92];
PS [88]; analysis [88], - The sequential copolymers are more efficient
Compatibiliser [89] XRD, [88];
SEBS [90]; Densimetry, - Crystallinity [93];
Poly(S-co-DMA); Mechanical, - Improved mechanical properties [89];
PE-g-MA [91] DSC, - The effect of thermal conductivity upon the
- melt; Dynamic, phase inversion [90];
- Haake-Buchler Morphology; - Shift of Tg [91];
mixer, 180 C, - Decrease of the particle size [91];
60 rpm;
- Brabender;
31. PP/ethylene- Mechanical, - The influence of copolymer composition; 94
Polyolefin Blends
hexene-1 Dynamic; - At 30% hexene-1, the blend has two Tg;
copolymers - At 57% hexene-1, the blend is homogeneous
with a single Tg;
11
12
methods of preparation
Polyolefin blends
SEBS [100-105]; Morphology, - The elongation at break increases [100];
SBS [106]; IR; - The rheology is not modified until 5%
- melt; compatibilising agent is added [101];
- Brabender, 200 C, - The morphology depends on the
90-100 rpm, 15 compatibilising agent [102];
min [100-102]; - 4% SEBS - co-continuous structure [104];
- twin-screw - The optimum concentration is 3% [105];
extruder; - Improved mechanical properties [105];
- solution; - A better stability [106];
- The interfacial tension decreases at 2% block
copolymer [106];
methods of preparation
Polyolefin blends
Table 1.1 Continued
No. Blend Compatibilising Methods of study Observations Ref.
agent and
preparation methods
33. Continued - The influence of the molecular weight of
copolymer [107];
- The interfacial interactions increase with the
content of SEBS having the low molecular
weight and less for that with high molecular
weight [107];
34. iPP/aPS PP-g-PS [109] DSC, - A uniform distribution of particles [109]; 108-
iPP/SEP[108] SEP [110, 111] SEM, - The decrease of sizes [108-112]; 112
SBS [110-112] TEM, - Core-shell morphology [108, 109];
SEBS [111] XRD, - Improved adhesion between phases [110,
SAXS, 111];
Mechanical, - Improved mechanical properties [108, 110,
Morphology, 111];
Microscopy; - Co-continuous morphology [112];
- Both yield stress and Youngs modulus were
reduced [108];
- SEP affected the heterogeneous nucleation
and solidification of the iPP matrix [108];
35. PP/PS SBS [113-115]; Microscopy, - The influence of PS viscosity on dispersion 113-
Polyolefin Blends
IPS-b-iPP 50:50 Mechanical, [118]; 120
[116]; Rheology, - The decrease of crystallinity degree with %
SBS, DSC, increase of PS [118];
SEBS [114]; Dynamic, - The effect of blending conditions [119];
SEP [117] SEM, - The influence of dispersion degree on the
- melt; DMTA, properties [120];
13
14
methods of preparation
Polyolefin blends
- Uniform distribution of domains;
- Tensile strength increases;
3 8. PP/NBR PP-g-MA [123]; Mechanical, - 1%-15% compatibiliser [123]; 123,
PP-g-Ph [123]; Morphology; - The size of domains decreases [123, 124]; 124
PP-g-Ph [124]; - Improvement of mechanical properties [123];
- The tear strength increases [124];
39. PP/SAM Polyolefins (EPR, Mechanical - Impact strength similar to PP; 125
iPP) functionalised - Surface reactivity similar to SAM;
with amino groups
methods of preparation
Polyolefin blends
Table 1.1 Continued
No. Blend Compatibilising Methods of study Observations Ref.
agent and
preparation methods
40. PP/ABS - melt; Mechanical - Improved mechanical properties; 126
- single screw
extruder;
41. PP/PMMA Poly(P-b-MMA); SEM, - Reduction of PMMA particles; 127
- solution in xylene TEM, - Reduction of PP crystallinity;
at 120 C followed Mechanical; - Improved mechanical properties;
by drying;
42. PP/PC SEBS [128]; Mechanical, - The compatibiliser interacts easier with PP 128-
PP-g-S [129]; SEM, [128]; 130
PP-g-S/AN [129]; DSC, - The SEBS surrounds the PP drops embedded
PP-g-S/BuMA [129]; DMTA, in the PP matrix [128];
SEBS [129]; Rheology, - 4% compatibilising agent [129];
PP-g-MA 2,5-20% WAXD, - The glass transition temperatures do not
[130]; Thermal; modify substantially [129];
- melt; - The best mechanical properties are obtained
- Brabender; with the PP-g-S/AN copolymer [129];
- twin-screw - The size of disperse particles decreases [130];
extruder; - The crystallisation degree of PP decreases
[130];
- Mechanical properties increase significantly
Polyolefin Blends
[130];
43. PVC/LLDPE Chlorinated PE Mechanical, static - Improved mechanical properties; 131
and dynamic,
SEM,
TEM;
15
16
methods of preparation
Polyolefin blends
3% compatibilising Microscopy, - Increase of the adhesion between phases
agent [136]; Rheology, [137];
SEBS-g-MA [137]; SEM, - The elongation at break increases for
PE functionalised Morphology; compositions containing 10%
through photo- compatibilising agent and decreases suddenly
oxidation [138]; after that [137];
- melt; - Improved mechanical properties [138];
- twin-screw
extruder;
- Brabender, 240 C,
150 rpm [138];
methods of preparation
Polyolefin blends
Table 1.1 Continued
No. Blend Compatibilising Methods of study Observations Ref.
agent and
preparation methods
48. PA6/LDPE Mechanical - Improved mechanical properties; 140
compatibilisation at
50-100 s-1 strain
rate;
- melt;
- Brabender;
49. PA/PE HDPE-g-MA; Mechanical, - Reduction of particle size; 141
LDPE-g-MA; SEM,
- melt; TEM;
- twin-screw
extruder;
50. PP/PA11 PP-g-AA; Rheology, - Reduction of particle size; 142
- melt; Mechanical,
- twin-screw Dynamic,
extruder; SEM;
51. PP/PA66 PP-g-MA [143-147]; SEM, - Improved mechanical properties [143]; 143--
- melt; DSC, - The influence of MA content [143, 146]; 147
- twin-screw Mechanical, - Improved dispersion [145];
extruder; Microscopy, - The size and the form of particles do not
Polyolefin Blends
SAXS, depend upon the temperature [145];
TEM, - The size of domain decreases [146, 147];
Rheology; - The study of phase inversion [147];
- Improved properties of the blends rich in PA
[147];
17
18
methods of preparation
Polyolefin blends
164]; in situ - Fine and regular structure [156];
compatibilisation - The crystallinity decreases [156];
[165] - The study of structure - processing
- melt; conditions - morphology - properties -
- reactive extrusion; relationships [163];
- twin-screw - Improved compatibility by the increase of
extruder; grafting percentage and PA content or the
- Brabender mixer; decrease of molecular weight of PA [149];
- The influence of the processing conditions
[157];
methods of preparation
Polyolefin blends
Table 1.1 Continued
No. Blend Compatibilising Methods of study Observations Ref.
agent and
preparation methods
52. Continued - Shift of the melting peaks [158];
- SEBS-g-MA is much efficient [161];
- EPR-g-MA decreases the ductile-brittle
transition temperature [161];
- Significant change of volume during of
deformation at the blends compatibilised
with EPR-g-MA [162];
- The influence of water absorbtion upon the
relaxation temperature and dynamic-
mechanical properties [164];
- Effects of accelerated ageing at 200 C [164];
53. PP/PA12 PP-g-MA [168-170] SEM, - Finer dispersion; 168-
PP-MA-g-PEO [170] DSC, 171
EP-g-MA [171] WAXS,
Mechanical
54. PA 10 10/PP PP-g-GMA [172- Morphology, - The improvement of the tensile strength and 172-
174] Mechanical, the impact strength [172, 173]; 174
- melt; SEM, - PP-g-GMA reduces the size of the disperse
- extruder; Thermal, domains [172-174];
Polyolefin Blends
Rheology, - The influence of the viscosity ratio upon the
Microscopy; phase inversion and size of particles [174];
- The study of particle properties upon the
phases inversion [174];
19
20
methods of preparation
Polyolefin blends
Compatibiliser
[182];
- melt;
5 8. PE/LCP PE-g-MA Rheology, - A slight increase of viscosity and mechanical 183
Mechanical, properties due to the hydrogen bonding
between components;
59. PE-AA/PS; PS-GMA Microscopy - Reduction of domain size; 184
PBT/PS
methods of preparation
Polyolefin blends
Table 1.1 Continued
No. Blend Compatibilising Methods of study Observations Ref.
agent and
preparation methods
60. PBT/HDPE Ionomer [185, 186]; DSC, - Fine dispersion [185]; 185,
- melt; DMTA, - Good mechanical properties [185, 186]; 186
- single-screw SEM, - Improved adhesion between phases [186];
extruder; Mechanical,
Rheology;
61. PP/PBT (30%) PP-g-oxazoline; SEM, - The improvement of the ultimate tensile 187
- melt; DMTA, strength and impact strength;
- twin-screw DSC;
extruder;
62. PP/PBT, LCP Compatibiliser with FTIR, - 5% improves the impact strength for blends 188
epoxy groups Mechanical; 80/20 and 20/80;
- Reduction of the dispersion phase size;
63. PP/PBT Emulsifying agent Mechanical, - The study of morphology; 189
(60/40) Microscopy - The influence of the emulsifying agent upon
the morphology and mechanical properties;
64. PP/LCP Mechanical, static - Modulus and tensile strength increase; 190
and dynamic, - E increases, tan decreases;
SEM, - The thermal stability of PP increases;
TGA;
Polyolefin Blends
65. PP/PBT, PP-g-MA; Rheology, - The mechanical properties are improved, 191
polyphenyl- PP-g-GMA; FTIR; especially the impact strength of the blend
ether PP-g-Ph- with PBT;
formaldehyde;
- melt;
- extruder;
21
22
methods of preparation
Polyolefin blends
7 0. PPS/PET LDPE-g-GMA; SEM, - Fine morphology; 195
- melt; Mechanical; - Ageing strength;
- twin-screw - Improved mechanical properties;
extruder;
71. HDPE, - melt; DSC, - Two distinct melting points; 196
PP, - twin-screw Mechanical; - The elongation at break increases;
CoPP, extruder; - Modulus and yield stress decrease with the
TerPP/M-PE M-PE content;
methods of preparation
Polyolefin blends
Table 1.1 Continued
No. Blend Compatibilising agent Methods of study Observations Ref.
and preparation
methods
72. iPP, - melt; Morphology, - Homogeneous distribution of the polyester 197
HDPE/ - twin-screw Mechanical, and the dye in the matrix of polyolefins;
hyperbranched extruder; DSC, - Strong decrease of the torque with increasing
polyester SEM, amount of polyester;
Resistance of the
dye-containing
blends against
solvents,
73. HDPE/PP/ - injection Soil burial test, - Two separate peaks; 198
biodegradable DSC, - The biodegradation does not start in the
additives Mechanical; crystalline zone;
- E increases;
74. PE/PA6 SEBS-g-OXA, Morphology, - More homogeneous blend morphologies; 199
PP/PBT EP-g-OXA, Mechanical; - Improved impact strength;
PP/PA6 PE-g-OXA (60/30/10); - Decreased tensile modulus and tensile
- melt; strength;
- twin-screw midi - The particles are much better embedded in
extruder; the matrix;
75. iPP/PE/EPDM iPP-g-BMI; Mechanical - Increased tensile strength and elongation at 200
- melt; break;
- twin-screw extruder;
Polyolefin Blends
76. iPP, EPDM; Mechanical - PCR is a ternary mixture of PP, HDPE and 201
HDPE/post- EVA; LLDPE in equal amounts;
consumer resin - melt; - Improvement of impact and tensile properties
(PCR) - extruder; at PCR contents less than 50%;
- EPDM and EVA improve the mechanical
properties;
23
24
methods of preparation
Polyolefin blends
EVA: ethylene-vinyl acetate copolymer SANS: small-angle neutron scattering
f-HDPE: styrene and maleic anhydride modified HDPE SAXS: small angle x-ray scattering
HDPE FTIR: Fourier transform infra red spectroscopy SB: styrene-butadiene diblock
HPB: hydrogenated polybutadiene SBS: styrene-butadiene-styrene block copolymer
IR: infra red spectroscopy SEBS: styrene-ethylene-butylene-styrene block
GMA: glycidylmethacrylate copolymer
HDPE: high density polyethylene SEM: scanning electron microscopy
HVEM: high voltage electron microscopy SEP: styrene-ethylene/propylene copolymer
IFR: imbedded fibre retraction method sPP: syndiotactic polypropylene
IPN: interpenetrating polymer network SIS: styrene-isoprene-styrene copolymer
Polyolefin blends
methods of preparation Polyolefin Blends
EPDM and EVA are more efficient when they are used
like compatibilising agents in iPP/HDPE, LDPE/PP
and PP/PE blends. They are prepared in a twin-screw
extruder and have improved mechanical properties
which is dependant on the effect of the blending
conditions. Generally, EVA is not as effective in
improving the mechanical properties as is EPDM.
MMA: methylmethacrylate
25
Handbook of Polymer Blends and Composites
SEP diblock copolymer is a very effective compatibiliser, too, for immiscible PP/PS blends.
Notched impact strength and ductility were significantly improved, especially for the PP/
PS blends weight ratio 90/10 and 70/30. Remarkable improvements were obtained when
10 wt% of SEP was added, converting the brittle PP/PS blend to quite impact resistant
polymeric material [110, 117]. SEP affected the heterogeneous nucleation and
solidification of the iPP matrix, changing the crystallite and spherulite sizes and partly
the crystallite orientation [108]. The particle size of the dispersed phase was reduced and
the interfacial adhesion was improved as well [110, 117].
The SEP diblock copolymer proved to be a superior compatibiliser compared to SBS and
SEBS triblock copolymers. Low-molecular weight SEBS appeared to be a more efficient
compatibiliser in PP/PS blend than high-molecular weight SEBS [111].
SBS makes migration of iPP chains from the melt easier during the solidification process of
the blend. It also diminishes coalescence and consequently reduces dispersed particle diameter
[112]. Incorporation of iPP functionalised with BMI [200] in the incompatible blends of
iPP/PE/EPDM leads to the increase of two times of the tensile strength (from 105 to
225.5 kgf/cm2), the elongation at break being 13-21%, close to that of a reference sample.
The other functionalised polyolefins (PO) are useful for compatibilisation of the immiscible
PO/polyamide or PO with various elastomers mixtures, obtaining enhanced mechanical
and thermal properties.
The blends of iPP and HDPE with a PCR were investigated [201]. These blends were
modified with EPDM and EVA copolymers. The PCR resin used is a ternary mixture of
PP, HDPE and LLDPE in equal amounts. The iPP/PCR blends had good impact properties
but poor tensile properties. Compatibilisation of these blends with EPDM and EVA was
effective at improving the impact properties. The tensile properties and impact strength
HDPE/PCR blends had poor impact and tensile properties. EPDM and EVA improved
these properties insufficiently for general recycling, except at lower 25% PCR contents.
Blends of polyolefins and olefin copolymers with other polymers include polyamide in
HDPE for impermeability, maleated EPDM in Nylon and other engineering thermoplastics
for impact strength, CPE and EVA in PVC as impact modifiers or plasticisers and the
growing use of CPE and maleated polyolefins as compatibilisers in other polymer blends.
The blends of polyolefins with different polyamides are very important thanks to their
improved mechanical properties, a finer domain and a better stability of phases. Different
26
Polyolefin Blends
PP grafted with MA is very used for compatibilising the blends of PP with PA. An improved
compatibility is obtained by increasing the grafting percentage, the PA content or the
decrease of PA molecular weight. PP-g-MA confers the finest phase morphology.
EPR functionalised with MA and SEBS functionalised with MA are used like
compatibilising agents in blends of PP with PA, SEBS-g-MA being more efficient. A
better dispersion and improved impact strength are obtained.
M-PE blends with HDPE, PP, CoPP and TerPP are thermodynamically immiscible, but
mechanically compatible. The degree of compatibility is the largest for the HDPE-M-PE
blend system compared to the other three blends system: PP-M-PE, CoPP-M-PE and
TerPP-M-PE. These results can be explained in terms of the similar chemical structure of
polyolefins [196].
The blends of polyolefins (PE and PP) with polyesters, PBT and PA6 are compatibilised
with OXA functionalised polyolefins and elastomers, e.g., SEBS-g-OXA, EP-g-OXA,
27
Handbook of Polymer Blends and Composites
PE-g-OXA [199]. Compatibilisation improved the toughness for all blends tested.
Morphological studies showed that the particles size was reduced and the adhesion of
the dispersed phase to the matrix improved by compatibilisation.
Current commercial practice has already made rapid progress in improving polyblend
compatibility and thus expanding the range of materials and properties which they offer
to the plastic industry.
On other area of serious interest is the recycling of mixed polymers in solid waste. Neither
consumers nor recyclers have been able to separate most of the polymers in solid waste and
reuse of these mixed materials has been far from satisfactory. Hopefully, development of
more successful compatibilisers and compatibilisation techniques should make these mixed
plastics more useful and thus contribute to solving the growing problems of solid state.
In all these areas, since polyolefins are the largest family of the commercial polymers,
their role in these developments will be a major factor in successful commercial practice.
Future development involves two competing trends. On the one hand, major polymer
producers will develop and sell standard grades of compatibilisers for major polyblend
markets. On the other hand, with growing understanding and experience in
compatibilisation, increasing numbers of polyblend processors will develop their own
proprietary ingredients and techniques for compatibilising the polyblends they sell.
References
3. F.P. La Mantia, D. Curto and D. Acierno, Acta Polymerica, 1984, 35, 71.
4. F.P. La Mantia and D. Acierno, Plastics and Rubber Processing and Applications,
1985, 5, 183.
5. A. Valenza, F.P. La Mantia and D. Acierno, Journal of Rheology, 1986, 30, 1085.
28
Polyolefin Blends
9. M. Grigorova and M. Mihailov, Journal of Applied Polymer Science, 1996, 62, 721.
10. A.K. Gupta, S.K. Rana and B.L. Deopura, Journal of Applied Polymer Science,
1992, 46, 99.
11. A.K. Gupta, S.K. Rana and B.L. Deopura, Journal of Applied Polymer Science,
1993, 49, 477.
12. Z.Q. Zhou, X.C. Lu and N. Brown, Polymer, 1993, 34, 2520.
13. S.K. Rana, Journal of Applied Polymer Science, 1996, 61, 951.
14. J.N. Hay and X.Q. Zhou, Polymer, 1993, 34, 2282.
16. T. Kyu and P. Vadhar, Journal of Applied Polymer Science, 1986, 32, 5575.
17. Y. Zhao, W.G. Zhang and D.C. Yang, Journal of Materials Science Letters, 1993,
12, 1309.
18. P. Vadhar and T. Kyu, Polymer Engineering and Science, 1987, 27, 202.
19. D. Abraham, K.E. George and D.J. Francis, Journal of Applied Polymer Science,
1998, 67, 789.
20. A.R. Plaza, E. Ramos, A. Manzur, R. Olayo and A. Escobar, Journal of Materials
Science, 1997, 32, 549.
21. A. Manzur, R. Olayo and E. Ramos, Journal of Applied Polymer Science, 1997,
65, 677.
22. D. Abraham, K.E. George and D.J. Francis, Journal of Applied Polymer Science,
1996, 62, 59.
23. R.G. Alamo, W.W. Graessley, R. Krishnamoorti, D.J. Lohse, J.D. Londono, L.
Mandelkern, F.C. Stehling and G.D. Wignall, Macromolecules, 1997, 30, 561.
24. A.C.Y. Wong and A.D.E. Sy, Journal of Materials Science, 1995, 30, 5672.
29
Handbook of Polymer Blends and Composites
27. L.A. Utracki and B. Schlund, Polymer Engineering and Science, 1987, 27, 1512.
28. B.K. Kim, M.S. Kim and K.J. Kim, Journal of Applied Polymer Science, 1993, 48,
1271.
31. J.J. Rajasekaran, J.G. Curro and J.D. Honeycutt, Macromolecules, 1995, 28,
6843.
32. R. Thomann, J. Kressler, S. Setz, C. Wang and R. Mulhaupt, Polymer, 1996, 37,
2627.
33. R. Thomann, J. Kressler, B. Rudolf and R. Mulhaupt, Polymer, 1996, 37, 2635.
34. W. Zhu, X.Q. Zhang, Z.L. Feng and B.T. Huang, Journal of Macromolecular
Science B, 1996, 35, 5, 795.
35. H.P. Blom, J.W. Teh and A. Rudin, Journal of Applied Polymer Science, 1996,
60, 1405.
36. H.P. Blom, J.W. Teh and A. Rudin, Journal of Applied Polymer Science, 1996,
61, 959.
37. S.J. Mahajan, B.L. Deopura and Y.M. Wang, Journal of Applied Polymer Science,
1996, 60, 1517.
38. W.N. Kim, S.I. Hong, J.S. Choi and K.H. Lee, Journal of Applied Polymer
Science, 1994, 54, 1741.
39. H.P. Blom, J.W. Teh and A. Rudin, Journal of Applied Polymer Science, 1995,
58, 995.
41. L.C. Mendes, E.B. Mano, E. Martuscelli and S. Cimmino, Polymer Bulletin,
1995, 35, 237.
42. V. Flaris and Z.H. Stachurski, Journal of Applied Polymer Science, 1992, 45, 1789.
43. U. Plawky, M. Schlabs and W. Wenig, Journal of Applied Polymer Science, 1996,
59, 1891.
30
Polyolefin Blends
44. V. Flaris, A. Wasiak and W. Wenig, Journal of Materials Science, 1993, 28, 1685.
45. Z.H. Stachurski, G.H. Edward, M. Yin and Y. Long, Macromolecules, 1996, 29,
2131.
46. Y. Liu and R.W. Truss, Journal of Applied Polymer Science, 1996, 60, 1461.
47. Y. Liu and R.W. Truss, Journal of Polymer Science: Polymer Physics Edition,
1995, 33, 813.
48. A.J. Muller, C. Latorre, G. Mendez, J. Rotino and J.L. Rojas, Proceedings of
ANTEC 94, San Francisco, CA, USA, 1994, Volume 2, 2418.
49. M.J. Hill, L. Oiarzabal and J.S. Higgins, Polymer, 1994, 35, 3332.
50. Y. Kim, C.S. Ha, T.K. Kang, Y. Kim and W.J. Cho, Journal of Applied Polymer
Science, 1994, 51, 1453.
51. Y. Kim, W.J. Cho and C.S. Ha, Polymer Engineering and Science, 1995, 35,
1592.
52. C.S. Ha, Y.K. Kim and W.J. Cho, Journal of Materials Science, 1996, 31, 2917.
54. W. Wenig and M. Asresahegn, Polymer Engineering and Science, 1993, 33, 14, 877.
55. W. Wenig and A. Wasiak, Colloid and Polymer Science, 1993, 271, 824.
56. C.Y. Chen, W.M.Z.W. Yunus, H.W. Chiu and T. Kyu, Polymer, 1997, 38, 4433.
57. B.Z. Jang, D.R. Uhlmann and J.B. Vander Sande, Journal of Applied Polymer
Science, 1985, 30, 2485.
58. T. Inoue and T. Suzuki, Journal of Applied Polymer Science, 1996, 59, 1443.
60. B.K. Kim and I.H. Do, Journal of Applied Polymer Science, 1996, 61, 439.
31
Handbook of Polymer Blends and Composites
64. G.M. Kim, G.H. Michler, M. Gehleitner and J. Fiebig, Journal of Applied
Polymer Science, 1996, 60, 1391.
66. J. Jancar, A. Dianselmo, A.T. Dibenedetto and J. Kucera, Polymer, 1993, 34,
1684.
68. N.M. Da Silva and M.I.B. Tavares, Journal of Applied Polymer Science, 1996,
60, 663.
69. A.L.N. Da Silva, M.I.B. Tavares, D.P. Politano, F.M.B. Coutinho and M.C.G.
Rocha, Journal of Applied Polymer Science, 1997, 66, 2005.
70. M. Yamaguchi, H. Miyata and K.H. Nitta, Journal of Applied Polymer Science,
1996, 62, 87.
72. K. Cho, T.K. Ahn, B.H. Lee and S. Choe, Journal of Applied Polymer Science,
1997, 63, 1265.
73. H. Zhao, Li Liu, T. Tang and B. Huang, Polymer Journal, 1998, 30, 775.
77. A.K. Gupta, B.K. Ratnam and K.R. Srinivasan, Journal of Applied Polymer
Science, 1992, 45, 1303.
32
Polyolefin Blends
78. A.K. Gupta, B.K. Ratnam and K.R. Srinivasan, Journal of Applied Polymer
Science, 1992, 46, 281.
82. I.H. Do, L.K. Yoon, B.K. Kim and H.M. Jeong, European Polymer Journal,
1996, 32, 1387.
83. X. Xu, X. Zeng and H. Li, Journal of Applied Polymer Science, 1992, 44, 2225.
84. Y.C. Ou, J. Zhu and Y.P. Feng, Journal of Applied Polymer Science, 1996, 59,
287.
85. P. Jarvela, L. Shucai and P. Jarvela, Journal of Applied Polymer Science, 1996, 62,
813.
86. K.H. Kim, W.L. Cho and C.S. Ha, Journal of Applied Polymer Science, 1996, 59,
407.
87. B. Brahimi, A. Ait-Kadi, A. Ajji and R. Fayt, Journal of Polymer Science, Polymer
Physics, 1991, 29, 945.
90. Y. Agari, A. Ueda and S. Nagai, Journal of Applied Polymer Science, 1992, 45,
1957.
91. J. Hu, B.H. Zhang, M.D. Song and Q.Y. Zhou, Chemical Journal of Chinese
Universities, 1996, 17, 1810.
92. V. Flaris, W.E. Baker and M. Lambla, Polymer Networks & Blends, 1996, 6, 29.
93. A. Lahrouni and J. Arman, European Polymer Journal, 1995, 31, 347.
33
Handbook of Polymer Blends and Composites
95. T. Li, V.A. Topolkaraev, A. Hiltner, E. Baer, X.Z. Ji and R. P. Quirk, Journal of
Polymer Science, Polymer Physics, 1995, 33, 667.
97. H.Q. Feng, J. Tian and C.H. Ye, Journal of Applied Polymer Science, 1996, 61,
2265.
98. J.W. Teh and A. Rudin, Polymer Engineering and Science, 1991, 31, 1033.
99. E. Kroeze, G. Tenbrinke and G. Hadziioannou, Polymer Bulletin, 1997, 38, 203.
100. K.U. Kim and D.J. Meier, Polymer (Korea), 1989, 13, 40.
101. K.U. Kim and D.J. Meier, Polymer (Korea), 1989, 13, 119.
103. S.A. Xu, M. Jiang and J.S. Shen, Polymer Journal, 1995, 27, 607.
104. S.A. Xu and C.M. Chan, Polymer Journal, 1998, 30, 552.
105. B. Brahimi, A. Ait-Kadi and A. Ajji, Polymer Engineering and Science, 1994, 34,
1025.
106. N. Mekhilef, B.D. Favis and P.J. Carreau, Journal of Polymer Science, Polymer
Physics, 1997, 35, 293.
108. G. Radonjic and I. Smit, Journal of Polymer Science, Polymer Physics, 2001, 39,
566
110. G. Radonjic, V. Musil and M. Makarovic, Acta Chimica Slovenica, 1997, 44, 29.
34
Polyolefin Blends
112. I. Smit and G. Radonjic, Polymer Engineering and Science, 2000, 40, 2144.
113. E. Navratilova and I. Fortelny, Polymer Networks & Blends, 1996, 6, 127.
114. D. Hlavata, Z. Horak and V. Fort, Polymer Networks & Blends, 1996, 6, 15.
115. Z. Horak, V. Fort, D. Hlavata, F. Lednicky and F. Vecerka, Polymer, 1996, 37,
65.
121. D.W. Park and R.J. Roe, Macromolecules, 1991, 24, 5324.
122. J. George, R. Joseph, S. Thomas and K.T. Varughese, Journal of Applied Polymer
Science, 1995, 57, 449.
123. S. George, R. Joseph, S. Thomas and K.T. Varughese, Polymer, 1995, 36, 4405.
125. N. Dharmarajan, S. Datta, G.V. Strate and L. Ban, Polymer, 1995, 36.
126. A.K. Gupta, A.K. Jain, B.K. Ratnam and S.N. Maiti, Journal of Applied Science,
1990, 39, 515.
127. S. Hosoda, H. Kihara, K. Hojima, Y. Satoh and Y. Doi, Polymer Journal, 1991,
23, 277.
128. K.R. Srinivasan and A.K. Gupta, Journal of Applied Polymer Science, 1994, 53, 1.
129. Z. Liang and H.L. Williams, Journal of Applied Polymer Science, 1991, 43, 379.
35
Handbook of Polymer Blends and Composites
130. Z.H. Yin, X.M. Zhang, Y.J. Zhang and J.H. Yin, Journal of Applied Polymer
Science, 1997, 63, 1857.
131. P.X. He, H. Huang, W.D. Xiao, S.Q. Huang and S.Y. Cheng, Journal of Applied
Polymer Science, 1997, 64, 2535.
132. B. Majumdar, H. Keskkula and D.R. Paul, Journal of Applied Polymer Science,
1994, 54, 339.
133. C.J. Wu, J.F. Kuo, C.Y. Chen and E. Woo, Journal of Applied Polymer Science,
1994, 52, 1695.
134. S. Fellahi, B.D. Favis and B. Fisa, Polymer, 1996, 37, 2615.
136. K. Chandramouli and S.A. Jabarin, Advances in Polymer Technology, 1995, 14, 35.
138. D. Curto, A. Valenza and F.P. LaMantia, Journal of Applied Polymer Science,
1990, 39, 865.
139. S. Lim and J.L. White, Polymer Engineering and Science, 1994, 34, 221.
140. B. Jurkowski and Y.A. Olkhov, Journal of Applied Polymer Science, 1997, 65,
1807.
142. Z. Liang and L. Williams, Journal of Applied Polymer Science, 1992, 44, 699.
145. J.S. Lin, E.Y. Sheu and Y.H.R. Jois, Journal of Applied Polymer Science, 1995,
55, 655.
36
Polyolefin Blends
147. P.T. Hietaoja, R.M. Holstimiettinen, J.V. Seppala and O.T. Ikkala, Journal of
Applied Polymer Science, 1994, 54, 1613.
148. M.L. Dilorenzo, L. Dimaio and D. Acierno, Polymer Networks & Blends, 1996,
6, 147.
150. S.N. Sathe, G.S.S. Rao and K.V. Rao, Journal of Macromolecular Science Pure
and Applied Chemistry, 1996, A33, 459.
153. J. Rosch and R. Mulhaupt, Journal of Applied Polymer Science, 1995, 56, 1599.
154. R.M. Holstimiettinen, K.P. Perttila, J.V. Seppala and M.T. Heino, Journal of
Applied Polymer Science, 1995, 58, 1551.
155. O.T. Ikkala, R.M. Holsti-Miettinen and J. Seppala, Journal of Applied Polymer
Science, 1993, 49, 1165.
156. S.N. Sathe, S. Devi, G.S.S. Rao and K.V. Rao, Journal of Applied Polymer
Science, 1996, 61, 97.
157. S.S. Dagli, M. Xanthos and J.A. Biesenberger, Polymer Engineering and Science,
1994, 34, 1720.
158. A. Valenza and D. Acierno, European Polymer Journal, 1994, 30, 1121.
159. A. Gonzalezmontiel, H. Keskkula and D.R. Paul, Polymer, 1995, 36, 4587.
161. A. Gonzalezmontiel, H. Keskkula and D.R. Paul, Polymer, 1995, 36, 4605.
162. A. Gonzalezmontiel, H. Keskkula and D.R. Paul, Polymer, 1995, 36, 4621.
37
Handbook of Polymer Blends and Composites
163. J. Rosch, R. Mulhaupt and G.H. Michler, Macromolecular Symposia, 1996, 112,
141.
165. S. Schlag, J. Rosch and C. Friedrich, Polymer Bulletin, 1993, 30, 603.
166. E.V. Konyukhova, A.I. Buzin, N.P. Bessonova and Y.K. Godovskii,
Vysokomolekulyarnye Soedineniya, Seriya B, 1996, 38, 286.
167. T. Tang and B.T. Huang, Journal of Applied Polymer Science, 1994, 53, 355.
168. T. Tang, Z.L. Lei and B.T. Huang, Polymer, 1996, 37, 3219.
169. T. Tang, Z.L. Lei, X.Q. Zhang, H. Chen and B.T. Huang, Polymer, 1995, 36,
5061.
170. T. Tang, H.Y. Li and B.T. Huang, Macromolecular Chemistry and Physics, 1994,
195, 2931.
171. Y. Long and R.A. Shanks, Journal of Materials Science, 1996, 31, 4033.
172. X.M. Zhang, Z.H. Yin, T.H. Na and J.H. Yin, Polymer, 1997, 38, 5905.
173. X.M. Zhang, X.L. Li, D.M. Wang, Z.H. Yin and J.H. Yin, Journal of Applied
Polymer Science, 1997, 64, 1489.
174. X.M. Zhang, Z.H. Yin and J.H. Yin, Journal of Applied Polymer Science, 1996,
62, 893.
175. H.G. Fritz, Q. Cai, M.E. Cagiao, L. Giri and F.J.B. Calleja, Journal of Materials
Science, 1995, 30, 3300.
176. J. Rosch and R. Mulhaupt, Journal of Applied Polymer Science, 1995, 56, 1607.
177. I. Campoy, J.M. Arribas, M.A.M. Zaporta, C. Marco, M.A. Gomez and J.G.
Fatou, European Polymer Journal, 1995, 31, 475.
178. J.Y. Wu, W.C. Lee, W.F. Kuo, H.C. Kao, M.S. Lee and J.L. Lin, Advances in
Polymer Technology, 1995, 14, 47.
38
Polyolefin Blends
180. J.M. Willis, B.D. Favis and C. Lavallee, Journal of Materials Science, 1993, 28,
1749.
181. S. Basu and D.D. Kale, Macromolecular Symposia, 1997, 118, 329.
182. G. Kumaravel and S.A. Jabarin, Advances in Polymer Technology, 1996, 15, 191.
183. F.P. Lamantia, R. Scaffaro, G. Plado, P.L. Magagnini and M. Paci, Polymer
Networks& Blends, 1996, 6, 171.
184. J.K. Kim, S. Kim and C.E. Park, Polymer, 1997, 38, 2155.
185. M. Joshi, A. Misra and S.N. Maiti, Journal of Applied Polymer Science, 1991,
43, 311.
186. M. Joshi, S.N. Maiti and A. Misra, Journal of Applied Polymer Science, 1992,
45, 1837.
187. T. Vainio, G.H. Hu, M. Lambla and J. Seppala, Journal of Applied Polymer
Science, 1997, 63, 883.
188. R.M. Holstimiettinen, M.T. Heino and J.V. Seppala, Journal of Applied Polymer
Science, 1995, 57, 573.
190. S.C. Tjong, S.L. Liu and R.K.Y. Li, Journal of Materials Science, 1996, 31, 479.
191. H.K. Kotlar and K.L. Borve, Proceedings of ANTEC 95, Boston, MA, USA,
1995, 1843.
192. N. Mekhilef, A. Aitkadi and A. Ajji, Advances in Polymer Technology, 1995, 14,
315.
193. L. Mascia and A. Valenza, Advances in Polymer Technology, 1995, 14, 327.
194. C.H. Chen, R.D. Wesson, J.R. Collier and Y.W. Lo, Journal of Applied Polymer
Science, 1995, 58, 1087.
195. C.M. Chen, H.R. Lee, S.Y. Wu, C.P. Chen and H.C. Kao, Proceedings of
ANTEC 94, San Francisco, CA, USA, 1994, Volume 2, 2002.
39
Handbook of Polymer Blends and Composites
196. D. Rana, C.H. Lee, K. Cho, B.H. Lee and S. Choe, Journal of Applied Polymer
Science, 1998, 69, 2441.
201. H.P. Blom, J.W. The and A. Rudin, Journal of Applied Polymer Science, 1998,
70, 2081.
202. R.D. Deanin in Handbook of Polyolefins, Second Edition, Ed., C. Vasile, Marcel
Dekker, Inc., New York, NY, USA, 2000, Chapter 22, 613.
203. R.D. Deanin and M.A. Manion in Handbook of Polyolefins, Second Edition, Ed.,
C. Vasile, Marcel Dekker, Inc., New York, NY, USA, 2000, Chapter 23, 633.
40
The Property Trends and Applications of Blends of Metallocene Plastics with Other Plastics
2
The Property Trends and Applications of
Blends of Metallocene Plastics with Other
Plastics
Anand K. Kulshreshtha
2.1 Introduction
Zirconium creates extremely high yields, as much as 40 million grams of PE polymer per
gram of zirconium. This meets the medical industrys concern for material cleanliness
and patient safety. This translates to 25 parts per billion of zirconium contaminated in
the final product if all of the catalyst was left in the product. If these yields are maintained
in full-scale production, they will be among the cleanest polymers available to the medical
industry.
Metallocenes do a few things very well [1]. They make a specific polymer chain with
great regularity which avoids impurities and low molecular weight volatiles. They reduce
low molecular weight impurities which produce PP copolymers with a higher rubber
content but without tackiness. They make a very narrow molecular weight distribution
(MWD) polymer (too narrow for some applications) without the use of chain breaking
additives. Moulding broader molecular weight distributions with mixed catalysts has
been demonstrated. Metallocenes seem to show much promise for the polyethylene and
the elastomer industries.
41
Handbook of Polymer Blends and Composites
Metallocenes are now used all over the world. Metallocene catalysts can be used:
Initially, there was a scare that the narrower MWD of metallocene would require a
redesign of processing equipment. These apprehensions were allayed with the development
of constrained geometry catalysts. Somewhere during this development process, blending
of metallocene resins with conventional resins was begun and plastomers, syndiotactic
polymers, co- and ter-polymers, covering the entire density range became commercially
available. Blended polymers give (a) synergistic and cost dilution benefits and (b)
improvement in processability.
A perceived deficiency in polymers produced from the metallocene catalyst is that their
narrow MWD makes processing more difficult. To overcome some of these problems,
metallocene catalysts are sometimes used in conjunction with conventional ZN catalysts
to produce polymers with broadened, but controllable MWD. Examples of such catalysts
to produce polymers are Bu2Mg and (n-BuCp)2ZrCl2-MAO cosupported on silica,
Ti(OBu)4-AlEt3- and Et(Ind)2ZrCl2-Al(i-Bu)3 on MgCl2. Lynx 100 and (n-BuCp)2ZrCl2
and MeAlEt3-TiCl4-Mg(OEt)2-dibutylphthalate catalyst is used for isotactic propylene
(iPP) polymerisation. It is not clear whether the titanocene is a polymerising species or is
simply an internal modifier for the conventional ZN olefin polymerisation catalyst.
Polymers with broad MWD or multiple phases can also be prepared by polymerising in
a multistage process. Porous polypropylene (PP) produced from a conventional ZN
catalyst was subsequently contacted with Et(Ind)2ZrCl 2 and MAO and used to
copolymerise ethylene and propylene, thus producing an impact copolymer. The same
42
The Property Trends and Applications of Blends of Metallocene Plastics with Other Plastics
metallocene catalyst was used to polymerise ethylene and 1-butene at 180 C in the first
reactor zone with a subsequent polymerisation taking place in the second zone at 200 C
using a conventional ZN catalyst.
Metallocene-catalysts can be integrated with the ZN catalysts, to obtain inside the same
granule, novel catalytic-systems and materials. For example, the high engineering
performances conferred by a monodisperse MWD obtained with metallocene-catalysts,
increase the polyolefins versatility, in comparison to the more expensive engineering
materials.
Moreover, the combination of polar Hivalloy resins with metallocenic polymers having
a low or high fluidity, could supply novel products for new and interesting applications.
Aiming at using metallocene-based catalysts (MBC) in existing plants and due to the
RGT, an original way has been found to obtain new materials endowed with outstanding
properties. The concept is to combine the traditional catalysis (MgCl2-based catalysts,
e.g., of PP), as a support for a MBC able to give a perfectly random ethylene-propylene
rubber (EPR).
43
mLLDPE LLDPE
Handbook of Polymer Blends and Composites
The resulting reactor blend constitutes a new composition with interesting properties
not achievable previously via mechanical blending.
2.3.1 Reactor Blends and Alloys (HDPE for Pipe Applications) [3]
Metallocene resins have a narrow MWD. However, processing of the melt becomes more
difficult with increasing average molecular mass or decreasing melt flow rate (MFR) for
polymers with a narrow MWD. To overcome these limitations in processing for higher
molecular weight (MW) compounds and therefore tougher materials, the MWD must be
considerably broadened for these extrusion or blow moulding grades to produce films,
pipes and large containers. For example, broadening of the molecular mass distribution
strongly decreases melt viscosity at high shear rates, considerably reduces extrusion defects
like melt fracture and increases melt strength. Products with broad molecular mass
distributions, both unimodal, and bimodal and their advantageous rheological behaviour
are well-known.
If the catalyst is designed properly the catalyst particle disintegrates at the beginning of
the polymerisation process, and the primary particles are spread evenly over the whole
and expanding polymer particle. Around each primary particle (approximately 5 to 15 nm)
the polymer is formed. In the course of polymerisation each polymer particle passes
through reactor 1 to get the low molecular weight fraction and reactor 2 get the high
44
The Property Trends and Applications of Blends of Metallocene Plastics with Other Plastics
molecular weight fraction or vice versa. This process is reproduced inside each polymer
particle leaving the cascade which means that around each primary particle layers of
both polymer fractions are built up. The different residence time of the particles has to
be taken into consideration. Consequently inside each polymer particle the low and high
molecular weight fraction are finely and evenly distributed in each other as shown
schematically in Figure 2.1.
The extremely low average molecular weight fraction improves flowability of the melt
and forms mainly the crystalline phase in the solid state which increases stiffness and
Youngs modulus. The high molecular weight fraction is responsible for good mechanical
properties. Furthermore, if the long chains are copolymers of ethylene and 1-olefins, the
stress crack behaviour can be greatly improved.
Although the density is higher for the bimodal material the comonomer in the long
chains is much more effective in terms of stress crack resistance. In comparison to unimodal
homo- or copolymers the failure time is higher by approximately two orders of magnitude.
Figure 2.1 Polymer particle morphology and molecular mass distribution using the
cascade technology
45
Handbook of Polymer Blends and Composites
Figure 2.2 Density of unimodal and bimodal copolymers at the same MFR
(MFR 190/5 = 0.45 g/10 min)
The most recent development of an innovative catalytic system represented by the mixed
catalysis, Multicatalyst Reactor Granule Technology (MRGT), and the latest Multizone
Circulating Reactor Technology (MZCR) have tremendously expanded the applications
of homopolymers, copolymers, and other polyolefin-based materials (Hivalloy) in all
the sectors where other stiff, tough, resilient, soft and flexible materials can be substituted.
The latest possibility [4] of the RGT has been in the so-called mixed catalysis or MRGT
where a porous polymeric granule obtained through ZN heterogeneous catalysis can be
impregnated by a metallocene catalyst. This generates a new range of properties not
achievable via a ZN catalyst alone. Catalloy is basically a two-step polymerisation. In
the first step a porous granule is produced. In the second step, this granule is saturated
with a different monomer and polymerised again. Hence these are called reactor blends
as distinct from extruder blends. The former are more intimately mixed polymers and
hence have superior property combinations. To obtain a new final polymeric alloy,
Catalloy would have either improved mechanical mixing, or generate a real novel material
with new expanded range of properties not even achievable via blending. Catalloy
combines the advantages of heterogeneous catalysis, like processability, with those of
metallocene catalysts such as physico-mechanical and elastic properties.
46
The Property Trends and Applications of Blends of Metallocene Plastics with Other Plastics
In this way, it is possible to have a real intimate compound in the reactor because it is
possible to control size and location of the separated polymeric phases within the same
polymer particle which allows one to achieve very homogeneous mixing.
Various physico-mechanical grades of unreinforced and glass fibre reinforced grades are
shown in Figures 2.3 and 2.4, respectively.
Figure 2.3 Comparison of Hivalloy and PP properties with other engineering plastics
47
Handbook of Polymer Blends and Composites
Figure 2.4 Physical properties comparison of Hivalloy and PP glass reinforced grades
versus competitive products (glass content 30% by weight)
Figure 2.3 shows the effect of increasing impact modifier level on the Izod impact of
HPP, using both ethylene butylene (EB) plastomer and EPR modifiers. Flexural modulus
decreases in a linear fashion. The increase in impact strength follows an S curve. The
effect is only slightly seen below about 15% modifier. From 15% to about 40% modifier,
the impact strength increases sharply. Impact strength eventually plateaus above 40%
modifier.
Table 2.2 shows a property comparison of blends of PP modified with m-plastomer and
conventional EP rubber for two different PP grades.
When used to modify HPP, conventional EPR causes stress-induced whitening or blush.
Substituting a metallocene plastomer for EPR dramatically diminishes the blush marks.
48
The Property Trends and Applications of Blends of Metallocene Plastics with Other Plastics
Blush appears during instrumental impact testing as a white ring formed on the test
plaque after penetration of the test dart. Metallocene plastomers are already widely used
as PP impact modifiers and are positioned to become the leading modifier type in the
near future, displacing EPR and ethylene-propylene diene monomer (EPDM). Plastomers
provide equivalent impact enhancement (also at low temperatures) with additional
performance advantages including easier dispersion, superior impact in high flow PP
resins, and low haze in clarified random copolymer resins (RCP) resins.
Downgauging is a continuing trend in the evolution of the films used for packaging. The
shipper, grocer and consumer have a common set of film requirements. Since the film is
49
Handbook of Polymer Blends and Composites
a primary means of advertising, film appearance as measured by its clarity, gloss and
high print quality are important. The film must also have puncture resistance but at the
same time be relatively easy to open. Each of these film requirements must be allowed
for in the design of new overwrap films. Film must have good processability in terms of
throughput combined with the ability to drawdown to thin gauges. The film must be
relatively stiff with low extensibility to permit good registration during printing and
good roll conformity.
The primary trend similar to almost all other packaging films is the constant drive for
cost reduction via film downgauging. There are four factors behind this trend. The obvious
one is cost savings through more packages per kilogram of film. Secondly thinner film
means improved packaging line efficiency due to fewer roll changes. Thinner film also
allows better utilisation of storage facilities. Lastly there is some consumer demand,
more prevalent in Europe, for reduced packaging as a method of source reduction/
recycling. Since film stiffness is a key attribute, HDPE is sometimes added to boost the
overall film stiffness even though the presence of HDPE tends to significantly degrade
clarity and toughness. Unfortunately many HDPE do not blend effectively with LDPE or
LLDPE and also significantly decrease film properties, especially the puncture
performance. With the advent of metallocene catalysed PE, an alternative blend
formulation route is now available in which high film stiffness can be achieved in a film
that possesses good toughness and clarity. This is achieved by blending together an
m-LLDPE and a new medium molecular weight (MMW) HDPE (Table 2.3). These
m-LLDPE-HDPE blends create new possibilities of film formulation where film toughness
and stiffness must be maintained while film thickness is reduced.
With LDPE, thinner films can be used, since it will tear and shred. With m-LLDPE and
its blends with LDPE, downgauging is possible, thus decreasing the material cost, yet
enhancing performance and properties.
To obtain a 15.2-17.8 m mLLDPE film with a modulus similar to a 22.9 m LDPE film
between 40%-45% HDPE should be used. Alternatively a slightly higher density mLLDPE
could be blended with a lower amount of HDPE.
Films of mLLDPE provide superior puncture performance at even one-half the thickness
of the conventional overwrap films. The puncture energy comparison, emphasises the
point that the use of mLLDPE yields very tough films.
Blends of mLLDPE can survive high speed dart impact testing and give comparably high
impacts even at thin gauges.
50
The Property Trends and Applications of Blends of Metallocene Plastics with Other Plastics
The use of an mLLDPE yields film clarity reasonably close to that of the current overwrap
film formulations. Adding a small amount (5%) of an LDPE to the mLLDPE/HDPE
blend would further improve the film clarity as well as adding to its bubble stability
during film extrusion.
The higher haze of the mLL/HD#3 films is most likely due to the higher HDPE content
of that blend (Figure 2.4).
The tensile strengths were significantly higher for the mLLDPE/HDPE films again
indicating a good potential for downgauging.
The higher puncture resistance and film strength of the mLLDPE films would tend to
reduce the chance of initiating a tear.
The results show that blends of mLDPE and MMW-HDPE provide a viable route for
downgauging paper overwrap film. These mLLDPE/HDPE films combine reasonable
51
Handbook of Polymer Blends and Composites
Figure 2.5 Processing performance in blown film processing HD blends @ 1.5 mm die
gap, LD blends @ 0.75 mm
stiffness and film clarity with superior toughness and puncture resistance. Heat sealing
performance of the blends could be further improved by adding a third component,
either an LDPE/ethylene-vinyl acetate (EVA) or even a metallocene plastomer. Processing
was found to be similar to conventional LLDPE. The use of coextrusion is an option to
further optimise the film properties.
The POE concentration effect for ENGAGE Polyolefin Elastomer DEG 8180 is shown in
Figure 2.6. These formulations were based on a 30 MFI impact copolymer of PP, 15%
52
The Property Trends and Applications of Blends of Metallocene Plastics with Other Plastics
Figure 2.6 Concentration effect: EG 8180-30 C notched Izod and falling dart impact
talc and varying weight percent of DEG 8180. As expected, higher levels of POE resulted
in greater impact performance. The ductile to brittle transition for falling dart impact
occurred at about 20 weight % POE while that for the notched Izod impact occurred at
25 weight % POE.
More rubber particles are needed to induce the ductile brittle transition temperature for
the notched Izod impact test compared with the dart impact test. Figure 2.7 is a comparison
of dart and Izod impact results of EG8100 and EG8180 blend systems demonstrating
the differences between the two impact test performances. As expected the Izod impact
test is more sensitive to the type of impact modifier.
Weld line strength was measured using an Izod impact test at room temperature. Results
are shown in Figure 2.8. Once again it is seen that the ethylene oxide (EO) based the
TPO have the best weld line strength measured using the above approaches.
Weld strengths measured using yield stress and Izod impact of the weld line establish
that ethylene-octene copolymers are superior to other olefins such as butene and
propylene.
The most efficient ENGAGE based impact modifiers are ENGAGE Polyolefin Elastomers
DEG 8180, EG 8150, and EG 8100, all of which are available in pellet form. For
applications in North America, EG 8100 offers the best balance of properties, while for
the European and Pacific areas, DEG offers the best balance of properties.
53
Handbook of Polymer Blends and Composites
54
The Property Trends and Applications of Blends of Metallocene Plastics with Other Plastics
Recent advances in single-site metallocene catalysts have created novel olefin compositions
and properties that yield cost/performance similar to that of flexible PVC. Many of these
materials are not sterilisable, but still offer compatibility with intravenous, lipophilic
drug solutions, better gas barrier properties, and the ability to meet environmental
requirements in various countries.
In PP, for example, novel propylene-based elastomers promise to provide both flexibility
and high-temperature properties. Polypropylene develops elastomeric properties when it
has an A-B-A-B-block copolymer structure. The A represents crystalline, high-melt-point,
isotactic domain crosslinks and B represents blocks of amorphous elastomeric atactic PP.
Above the glass transition temperature of PP, the crosslinks soften and the material
becomes shapable. Cooling reforms the crosslinks and restores elasticity.
In addition to key mechanical properties, materials for medical devices and packaging
should be highly transparent, resistant to high temperatures (121 C) and pressure, and
environmentally compatible. Transparency allows unimpeded visual inspection for
particulate matter contamination or complete dissolution of medications. Sterile delivery
of therapies makes compatibility with steam autoclaving (the preferred method of
sterilisation) a must.
PVC is prominent in medical applications due to its high performance/cost ratio. Its
modulus ranges from 10 MPa to greater than 1 GPa, optical clarity boasts measured
haze values under 5%, and its flexible formulations rival natural rubber.
Exxon has announced a family of medical grade polyolefin resin blends that utilise its
metallocene catalyst technology to combine outstanding radiation tolerance, clarity and
autoclavability in one formulation. The mouldable materials are blends of Exact plastomer
ethylene polymers (made using Exxpol metallocene catalyst technology) and conventional
polyethylene is typically used as a modifier in up to 30% ratios with conventional PP. In
general it is hard to simultaneously attain an optimised balance of properties for medical
moulding other than by using liquid modifiers which in turn can create quality penalties.
Exxon claims to offer a versatile, cost effective substrate materials for moulded parts,
e.g., syringes and connectors. Properties of the blends can be tailored to offer a wide
55
Handbook of Polymer Blends and Composites
range of ductility and resistance to softening. Materials can often be prepared by simple
pellet blending. The technology imparts a bluish tint to end-products, which cloaks the
undesirable yellowing that often affects PP when it is gamma sterilised.
Finacene PE is the new trade name introduced by Fina Chemicals for the marketing and
sales of metallocene catalysed PE.
Two experimental grades are available: Finacene PE ER 2225 and Finacene PE ER 2226.
Both grades have a density of 0.932 g/cm3; Finacene PE ER 2225 with a MI of 2.02 g/min
has a medium to narrow MWD, while Finacene PE ER 2226 with a MI of 2.25 g/min has
a narrow MWD.
Results were evaluated against the expected benefits of metallocene catalysed medium
density polyethylene (mMDPE), i.e., optics, sealability, toughness, downgauging/stiffness
and processability.
2.9.1.2 Optics
2.9.1.3 Sealability
When pure mMDPE is used it shows a beneficial melting point/softening point ratio,
allowing the film to be sealed at a lower temperature than conventional MDPE (5-10 C),
but keeping the same high temperature resistance (Figure 2.11).
In a blend of LDPE with a 0.3 MI and a density of 0.924, the crystallisation alteration
of mMDPE leads to a substantial drop of sealing temperature (> 15 C) for the same
welding force.
56
The Property Trends and Applications of Blends of Metallocene Plastics with Other Plastics
Figure 2.9 Finacene PE ER 2225 blended with LDPE (MFI of 0.7 and density of
0.924) compared to a conventional MDPE
2.9.1.4 Downgauging/Stiffness
Fine mMDPE are perfectly in line with conventional MDPE in terms of high secant
modulus leading to downgauging possibilities (see Figure 2.12).
2.9.1.5 Processability
From the multiple trials that were run, one can describe the global processability picture
of mMDPE as follows:
57
Handbook of Polymer Blends and Composites
Figure 2.10 Finacene PE ER 2226 blended with LDPE (MFI of 2.0 and density of
0.922)
no loss of output in comparison with the best facilities which used to manufacture
LLDPE. In the worst cases (LDPE screws), loss of a maximum of 10% of output (in
kg/h) largely compensated for by downgauging.
Although a limited number of experiments are available, it is seen that mMDPE can be
run on a cast line without showing the draw resonance effect.
58
The Property Trends and Applications of Blends of Metallocene Plastics with Other Plastics
Ferro has developed a series of unfilled polyolefin alloys called RxLOY, for the disposable
medical market. There are several grades developed for film, tubing, blow and injection
moulding applications.
The key physical properties of polypropylene homopolymer (control) and the blended
samples containing 10% by weight of the blend (impact) modifiers, i.e., EP rubber,
59
Handbook of Polymer Blends and Composites
plastomer and very low density polyethylene (VLDPE) are summarised in Table 2.4. The
room temperature notched Izod and elongation at break values are slightly higher for
PP/EP rubber blend, followed by PP/plastomer blend, while PP/VLDPE blend has the
lowest values in the series.
The key physical properties of mineral (talc, 20% by weight) filled blends along with PP
homopolymer control are summarised in Table 2.5. The flow properties of the unfilled
blend samples are similar to their filled counterparts. As expected, the HDT values are
higher for filled products as compared to unfilled blends. It is interesting to note that PP/
plastomer/talc compound is stiffer, stronger and has higher heat resistance, but at a reduced
toughness, i.e., impact resistance, compared to other two PP/EP rubber/talc and PP/
VLDPE/talc compounds.
60
The Property Trends and Applications of Blends of Metallocene Plastics with Other Plastics
Similar results and trends are noticed with instrument measured impact events. The talc
particles are uniformly dispersed in all the blended compounds, whose ash contents are
nearly identical, i.e., 20% by weight. Based on these observations, the impact strength
may be attributed to the talc particles tending to crystallise PP homopolymer in the
presence of elastomer or vice versa.
2.10.2 Stress-Whitening
The unfilled blend samples and PP homopolymer control were examined using the
polarising light microscope for their crystalline morphology. The skins of tiny crystalline
61
Unfilled blends properties
62
Impact modified PP blends properties of
The Property Trends and Applications of Blends of Metallocene Plastics with Other Plastics
is thinnest in the PP/plastomer blend, at around the 280 m. The skins on the other
samples are in the range 300-320 m. The transition zone (or band) of extremely bright
crystallites is thickest in the PP/plastomer, at about 160 m compared to the others at
130-140 m. The heat conductivity of the PP/plastomer blend sample may be different
enough that the skin and bright zone are crystallising differently. The minor phase
could not be discerned in the blends, probably due to very small domains, too small to be
resolved with the light microscope. The area of stress whitening was found to be the
smallest in the blend with the plastomer (Table 2.6), the colour also appears to be the
least intense. The diminution of stress-whitening in plastomer modifier is the reason why
it is more compatible with the PP homopolymer and exhibits favourable crystalline
morphology when compared to EP rubber or VLDPE modifier.
The property trends described previously may be explained by the near amorphous nature
of EP rubber as compared to the semicrystalline nature of the VLDPE while the plastomer
exhibits an intermediate behaviour.
Ferro observed some unique advantages in their RxLOY tubing compounds. The grades
containing metallocene resin, exhibit several performance improvements including contact
clarity, kink resistance and recovery, lower cost and higher extrusion rates with minimal
sacrifice in softness.
63
markets for Plastomers
markets for sPP
Projected markets for m-plastomers and m-syndiotactic PP are shown in Table 2.7. These
cover the vast range of compounds, blends and alloys, masterbatches, mouldings, film,
extrusion coating, foam, adhesives, etc.
64
The Property Trends and Applications of Blends of Metallocene Plastics with Other Plastics
Polyethylene blends have been widely studied to attempt to broaden the spectrum of
applications of polyolefins. Branch structure is the single parameter that determines the
miscibility behaviour of polyethylene blends, when the molecular weight and its
distribution are not much different between the blend pairs. For the blend of HDPE with
LLDPE it has been reported that the difference in the content of short chain branch
(SCB) between the two polymers governs both the miscibility of the blend in the melt
state and the cocrystallisation behaviour from the melt. Phase separation in the melt and
separate crystallisation was observed when the difference in SCB content is large.
Distribution of SCB, in addition to its content, may affect the miscibility and crystallisation
behaviour of polyethylene blends. In the separate studies, HDPE/LLDPE blend with
LLDPE containing 3.6 mol% SCB showed cocrystallisation, while the blend with LLDPE
containing 1.4 mol% SCB showed separate crystallisation when the content of LLDPE
was above 50 wt%. LLDPE prepared by Ziegler-Natta LLDPE catalyst (ZNLLD) was
used in the former, while LLDPE prepared by metallocene LLDPE catalyst (MLLD) was
used in the latter. As the SCB in LLDPE are introduced by the copolymerisations of -
olefin, the distribution of the SCB could be different depending on the catalyst used.
Two types of LLDPE with different comonomer sequence distributions were blended
with HDPE (Table 2.8) and the effect of the distribution on the miscibility behaviour in
the melt state was studied. Phase diagrams were constructed based on the thermal
behaviour of the blends quenched from the melt. LLDPE prepared with ZN catalyst,
which had a more uneven comonomer distribution, was found to be more miscible with
lower upper critical solution temperature compared to LLDPE prepared with metallocene
catalyst. Rheological behaviour and morphology of the LLDPE/HDPE blends were also
examined. The results showed good agreement with those of the phase behaviour
determined by another method, thermal ZNLLD cocrystallised with HDPE in a wider
range of blend compositions and with a smaller amount of separate-crystallised LLDPE
(than MLLD did). The result was considered to be due to the fact that MLLD had more
65
Handbook of Polymer Blends and Composites
even comonomer distribution than ZNLLD had. The etched-surface morphology exhibited
that banded spherulites were more uniformly formed when cocrystallisation only occurred.
An optimum ES resin for compatibilisation has a styrene content of 48-57 weight% and
a melt index equivalent to those of the matrix polymers.
66
The Property Trends and Applications of Blends of Metallocene Plastics with Other Plastics
It was found that, adequate mixing time in a mixing device is needed to provide distributive
and dispersive mixings. Examples of good mixing devices are a twin screw extruder, or a
single screw extruder equipped with a mixing section. When a mineral filler is added to
the TPO compound, the choice of mixing devices broadens to include both the batch
and continuous type of compounders.
The improved impact and melt flow performance which can be obtained by using
metallocene catalysed LLDPE in FR-HDPE.
67
Handbook of Polymer Blends and Composites
Since the research in the area of metallocene blends is in its nascent stage, some of the
recent developments are:
sPP-iPP Blends
sPP shows nonspherultic morphologies which are different from the typical iPP spherulites.
A bundle-like morphology at lower crystallisation temperatures and single-crystal like
entities at higher crystallisation temperatures are observed.
High-molecular weight blends of iPP and sPP are immiscible due to unfavourable enthalpic
interactions.
The end-use performance of Dow enhanced polyethylene (m-PE) resins in cable jacketing
applications are discussed. These m-PE resins are polyethylenes enhanced with INSITE*
Technology. m-PE resins have increased flexibility with equivalent abrasion resistance
relative to other existing LLDPE produced by a gas-phase polymerisation process. As a
result telecommunication cable with m-PE jackets are easier to install and bend around
corners. m-PE resins also have improved notch resistance and low temperature tensile
properties.
PP-Plastomer Blend
The dispersion of metallocene plastomers in HPP varying from low to high melt flow
rates were examined. A base polypropylene with a melt flow rate of up to 20 gave a
blend with an average particle size of 0.5 m. Only at high MFR, i.e., 35, did the base
polypropylene alter the plastomer dispersion. In this instance, the average particle size
was increased to 0.8 m, but still less than 1 m threshold necessary for cold temperature
impact modification of polypropylene. Low interfacial tensions between ethylene-butene
plastomer and polypropylene promoted submicron dispersion of plastomer in both low
and high melt flow polypropylenes.
68
The Property Trends and Applications of Blends of Metallocene Plastics with Other Plastics
P OE 5 0.870 -
SBS 11 - 28
SIS 12 - 30
SEBS <0.1 - 29
Toughness as a function of weathering time for the SBC with POE was studied. As
expected, the S-EB-S (which has no unsaturation in its elastomer phase) maintains its
toughness over a much longer period compared to SBS and SIS. The POE also shows
good toughness retention due to its saturated hydrocarbon structure. Surprisingly, the
50/50 POE/SBS and POE/SIS blends maintain their toughness compared to the SBS and
SIS blend components. Thus POE can be considered to enhance the weatherability of
SBS and SIS elastomers.
sPP/aPP Blends
Melting and crystallisation behaviour of sPP and atactic polypropylene (aPP) has been
studied [15-17] using differential scanning calorimetry (DSC) and X-ray scattering.
Multiple endotherms were observed which indicate sPP segregation within sPP/aPP blends.
When the metallocene LLDPE was initially produced in the early 1990s, it did suffer
from processability bottleneck and it became necessary to blend it with LDPE. It is no
longer today as with contained geometry catalyst (CGC) and Exxpol catalyst LLDPE
can be produced with many different permutations and combinations of properties and
processability. The incentive for blending metallocene and conventional LDPE is to bring
down the cost premium which new compounds demand. From the processors view the
situation is win-win. Existing machinery can be used. Plastomers (POP) and elastomers
(POE) can be used with no problems of compatibility cyclo olefins (COC) are new
additions, ethylene-styrene (E/S) interpolymers provide compatible LLDPE-PS blends.
69
Handbook of Polymer Blends and Composites
References
6. S.Y. Lee, J.Y. Jho and W. Huh, Journal of Industrial and Engineering Chemistry,
1998, 4, 258.
7. S.Y. Lee and J.Y. Jho, Journal of Industrial and Engineering Chemistry, 1998, 4,
170.
10. B. MacIver and D. Chundury, Proceedings of ANTEC 95, Boston, MA, USA,
1995, Volume 3, 3407.
11. Y.W. Cheung and M.J. Guest, Proceedings of ANTEC 96, Indianapolis, IN, USA,
1996, 1634.
13. CAL 133, Flammability Test Procedure for Seating Furniture for Use in Public
Occupancies, State of California, Department of Consumer Affairs, North
Highlands, CA, USA, 1991.
14. UL 94, Test for Flammability of Plastic Materials for Parts in Devices and
Appliances.
70
The Property Trends and Applications of Blends of Metallocene Plastics with Other Plastics
15. N. Kashiwa and J.I. Imuta, Catalysis Surveys from Japan, 1997, 1, 125.
16. D. Parikh, M. Guest, R. Patel, W. Ahmed, S. Betso, T. Ho, R. Guerra and J.D.
Allen, Proceedings of ANTEC 96, Indianapolis, IN, USA, 1996, Volume 2, 1638.
17. R.A. Phillips and R.L. Jones, Macromolecular Chemistry and Physics, 1999, 200,
1912.
71
Polyvinyl Chloride-Based Blends
3
Polyvinyl Chloride-Based Blends
3.1 Introduction
X-ray studies show that PVC is substantially amorphous, although a small percentage (< 5%)
of crystallinity is present. Studies using nuclear magnetic resonance (NMR) techniques
specify that conventional PVC is about 55% syndiotactic and the rest largely atactic in
structure. The presence of chlorine in great quantities in the polymer gives it flame
retardancy properties [2].
Many of these disadvantages can be reduced using the copolymerisation of vinyl chloride
with other monomers, frequently:
2. vinylidene chloride,
73
Handbook of Polymer Blends and Composites
Copolymers can be processed at much lower temperatures, and therefore are less affected
by the work out operations.
The improvement of the processing characteristics and of the articles manufactured from
PVC was possible through association of the polymer with:
1. thermal stabilisers,
2. lubricants,
3. impact modifiers,
4. plasticisers,
6. antioxidants.
These additive groups, with the exception of impact modifiers and processing aids, are
in most cases micromolecular substances [2]. However, with the emphasis on the
advantages offered by the polymer blends, many of the micromolecular additives used in
PVC-based compositions were replaced with macromolecular compounds.
The use of another polymer blended with PVC offers three advantages:
3. PVC has a higher heat distortion temperature (HDT) than systems using liquid
plasticisers [4].
Although the studies on PVC based blends focus mainly on the properties, there are also
numerous investigations aiming to establish the degree of compatibility between
components.
The range of macromolecular compounds for which compatibility was determined and/
or used for processability and utilisation characteristics modification of PVC-based
compositions is extremely wide and covers both general use polymers [polyethylene,
chemical modified polyethylene (PE) and ethylene-based copolymers, polypropylene (PP)
and propylene copolymers, polystyrene (PS) and styrene-based copolymers, acrylic
74
Polyvinyl Chloride-Based Blends
polymers and copolymers, polyvinyl acetate (PVAc) and its derivatives, different types of
general use elastomers] and engineering polymers (polyamides and copolyamides,
thermoplastic polyesters and copolyesters, polycarbonates and, polyacetals).
The difference between solubility parameters of PVC (PVC = 21.7 J1/2 cm3/2) and
polyethylene (PE = 16.4 J1/2 cm3/2) is an indicator of the incompatibility of these polymers,
which is also confirmed by the morphology and poor mechanical properties of the PVC
blends realised with different sorts of PE [57].
Although the PVC/PE blends are heterogeneous and with low mechanical properties,
they are interesting because:
A great interest for PVC/PE blends is in electrets production, characterised through long-
term stable polarisation (sometimes even several years [8]). Such electrets are used in
manufacturing air filters for air-conditioning devices and vacuum cleaners.
To enhance PVC/PE blend properties, Nakamura and co-workers [10, 11] studied the
possibility of components static co-crosslinking (30 minutes pressing at 100 C), to increase
phase separation area resistance. As a co-crosslinking agent they used a system based on
tri-allyl cyanurate and ,-bis(t-butyl peroxide)isopropyl benzene peroxide.
The results showed that using 0.25 wt% peroxide and 4 wt% tri-allyl cyanurate gave the
PVC/PE (2:1 parts) blend, a significant increase of tensile strength and elongation, higher
peroxide proportions having little effect on the previously specified characteristics.
The same attempt was made by Rudin and van Ballegooie [12], using reactive extrusion,
who obtained little difference from the Nakamura results that were related to the type of
75
Handbook of Polymer Blends and Composites
PE [linear low density polyethylene (LLDPE)] and the different molecular weights of
the polymers used.
A mixture of PVC, high-density polyethylene (HDPE) and CPE was found suitable for
manufacturing weather-snipping gasket for automobile windows.
Blends that comprised PVC, PE and a polymer derived from at least 80 wt% acrylic or
alkyl, aryl, alkaryl, or methyl methacrylate ester and not more than 20 wt% acrylic or
styrenic monomers, were used for extrusion or injection moulding of pipes or pipe
fittings, garden tools, electrical, automotive and business machines or appliance parts,
toys, sporting goods, footwear, battery cases, conducts, construction profiles for medical
applications, automotive or furniture seating, wall coverings and bottles [13].
CPE containing less than 25 wt% chlorine are incompatible with PVC and are generally
not used in PVC blends. Those with 2440 wt% chlorine are the best impact modifiers
because they have practical miscibility [19]. CPE with 45 wt% chlorine are miscible
with PVC with lower critical solution temperature behaviour (LCST).
The usual chlorine content in CPE from PVC/CPE blends varies between 2540 wt%
and a content of 36 wt% is known to provide the best impact strength [18].
76
Polyvinyl Chloride-Based Blends
Depending on the mixing ratio, the blend toughness varies from that of PVC to CPE, the
heat resistance and heat distortion characteristics are nearly independent of CPE content
and outdoor weathering characteristics are similar to those of PVC. The electrical
properties, chemical and flame resistance of those blends are equivalent to those of PVC.
The mixtures of PVC, CPE and coal tar pitch were used as weather resistant roofing
sheets. Inserting methyl methacrylate-butylacrylate copolymer in PVC/CPE blends
provides excellent tensile and impact strength.
Remarkable results regarding the impact strength of the PVC/CPE blends were also
obtained with addition of diamine (0.110 wt%) [13].
The impact strength of PVC/CPE blends was also increased considerably by adding
modifiers with elastomeric properties [copolymer ethylene with VA [21], various butadiene
and butadiene-acrylonitrile oligomers, acrylic resins [2224], ethylene-propylene-diene
terpolymer [18] and hydroxy-terminated polybutadiene (PB) [18], epoxidated natural
rubber [25, 26]]. The elastomeric additives produced better adhesion between the
components of the boundary interfacial surface. The introduction of epoxidated natural
rubber in to the PVC/CPE blends ensures the improvement of the antivibrational
characteristics [25].
It has also been proved to have good physical and mechanical properties and good
processability characteristics PVC/CPE blends charged with different filling materials
(wood flour, kaolin, calcium carbonate).
Applications of PVC/CPE blends include profiles for windows, guttering, street barriers,
gas pipes, bench slats, sheet for facades, balcony cladding, chemical plant, refrigerators,
77
Handbook of Polymer Blends and Composites
traffic signs, foil for identification and punch cards, flame retardant wall coverings, cases,
prams and automobile interiors. Injection moulded components include extractor hoods,
gullies in sewage systems, guttering parts, caps for road reflectors post and bench slats.
PVC/CPE blends are mainly used for outdoor applications, but are limited by the prevailing
climatic conditions. Performance has been satisfactory for more 30 years in countries
where the mean annual temperature during the hottest month is not greater than 24 C.
Under these conditions the blends show good impact resistance over the whole temperature
range encountered in the applications, as well as being resistant to ageing, UV and
weathering over the intended period of use. The upper temperature is 6065 C [20].
There have been many studies on the compatibility and properties of blends made of
PVC and PE copolymers. Most of them focus on PVC/ethylene-vinyl acetate copolymers
(EVA) [2737].
Gradually increasing the content of VA in EVA converts them from crystalline, nonpolar,
rigid materials, into amorphous, polar rubbery materials. In blends of PVC with EVA,
there may be separate phases from crystalline PVC, amorphous PVC, crystalline PE blocks,
amorphous EVA, and one or two phases for molecular blends of amorphous PVC with
amorphous EVA. The concrete morphology of PVC/EVA blends hangs on the VA content
in EVA, molecular weight and proportion of the components in blend, mixing conditions,
and subsequent treatments the blends undergo [2737].
The studies have confirmed that increasing the VA content in EVA increases the polarity
closer to that of PVC, which induces stronger interfacial bonding between phases and
better miscibility of the two polymers. Simultaneous, increasing the VA content in EVA
decreases the PE crystallinity, making the polymer softer and more flexible. These two
effects determine the efficiency of EVA as an impact modifier in rigid PVC, and as a
polymeric plasticiser in PVC.
These affirmations have been confirmed by many studies emphasising that EVA of 50
80 wt% VA content produced maximum miscibility with PVC [32]. From DMA and
differential scanning calorimetry (DSC) it has also been found that EVA with 45 wt%
VA tends to mix with PVC on molecular scale when the content of the sample is greater
than 25 wt% [33, 37].
78
Polyvinyl Chloride-Based Blends
When PVC was blended with 025 wt% EVA as impact modifier, synergistic peak
improvement occurred at 515 wt% EVA. At higher VA content in EVA, the peak tends
to move toward higher EVA content and becomes broader [38]. Electron microscopy
showed that increasing EVA content, gives rise to EVA domains whose size grows, until
phase inversion made the EVA continuous matrix phase and its properties transpose
from rigid high-impact plastics to thermoplastic elastomers [36]. For blends with better
impact strength, the optimum VA content seems to be about 45 wt% [34]. Copolymers
having over 65 wt% VA are good plasticisers.
Research shows that addition of lubricants to PVC/EVA blends causes shifting of the
maximum peak plotted in the curve showing the influence of EVA content on impact
strength [39].
When chlorine is introduced into the EVA copolymer, the compatibility with PVC is
significantly improved, but the rubber character of EVA is simultaneously strongly
influenced [34].
Commercial PVC blends with EVA copolymer grafted with vinyl chloride, (EVA-VC)
have been used since the 1970s as a high impact strength rigid formulation [13].
79
Handbook of Polymer Blends and Composites
The PVC/EVA blends are used for window profiles, roof gutters, drain pipes, profiles for
benches and fences, light panels, road sign posts, panels for facade cladding, ventilation
systems, for chemical waste, gas or draining pipes, for cable jacketing and cable conducts,
thermoforming sheets, fitting, covers, bottles, and containers [13].
PVC and chlorinated PVC (CPVC) are antagonistically immiscible with PP. Accordingly,
there are few studies on PVC/PP blends. These studies emphasised, however, that adding
a little PP to the rigid PVC compositions gives them better processability and impact
strength [18]. The PVC/PP blends show a layered, wood-like structure.
Replacing PP with PVC-PP copolymer, the blends obtained can be calendered into good
impact strength sheets.
Polyvinyl chloride is immiscible with PS, the blends of these polymers having a biphasic
structure and poor mechanical properties [4, 4249]. However, taking into account the
individual performances of each of two polymers, it has been proved that the PVC/PS
blends are interesting both theoretically and practically. For example, a reduction of
elastic properties of PVC melts was reported, coupled with improvement of certain optical
and physical properties on blending with PS [48].
The blends obtained as a result of styrene polymerisation in the presence of PVC are also
heterogeneous, but with superior mechanical properties compared to those made by
mechanical compounding [50].
80
Polyvinyl Chloride-Based Blends
There are many studies referring to blends comprising PVC, binary or tertiary copolymers
of styrene. Among binary and tertiary copolymers, the most used in PVC blends are
those with acrylonitrile (ACN) and maleic anhydride.
The addition of SAN copolymer in PVC based compositions gave a thermal stability
reduction [60, 61].
The partial miscibility between PVC and SMA copolymers provides for a family of heat
resistant, fire-retardant alloys spanning the heat resistance range of low-end engineering
thermoplastics. SMA copolymers systematically decrease the melt viscosity of PVC to
promote flow and processability in alloy compositions. This flow promotion is
accompanied by an increase in heat resistance rather than the usual heat sacrifice from
traditional flow additives for PVC.
The adhesion between phases is particularly evident in the exceptional impact strength
derived from alloys of SMA copolymers with PVC. The impact strength is superior to
any of the competitive engineering thermoplastic as well as injection moulding grades
of PVC.
Among the ternary copolymers with styrene units suitable for PVC mixing, acrylonitrile-
butadiene-styrene (ABS) and methyl methacrylate-butadiene-styrene (MBS) copolymers
are most commonly used.
There were several studies of PVC/ABS blends, the results being in detail discussed in
[13]. Up to this point, will be shortly emphasised some aspects regarding miscibility,
mainly properties and application fields of PVC/ABS blends.
81
Handbook of Polymer Blends and Composites
If at beginning the name ABS was assigned to copolymers comprising different proportions
of acrylonitrile, butadiene and styrene manufactured using several methods, now they
constitute a generic denomination for a wide class of such polymeric materials. Because
many studies referring to PVC/ABS blends do not present information about the
composition and manufacturing method of ABS, the various results are obtained.
Accordingly, while some authors treat PVC/ABS as being compatible, others assert only
a partial miscibility of PVC with ABS copolymers [4, 63-65].
The different opinions regarding the PVC miscibility with ABS copolymers did not
influence, however, the common consent that as a result of the mixing of these polymers,
materials with good tensile and impact strength, improved processability, suitable density
and high heat deformation temperature are obtained [13]. Adding 15 to 20 wt% ABS to
PVC composites gives optimal characteristics.
The PVC/ABS blends offer excellent processability and they are formable by all standard
methods. They have been used to produce housing for electrical tools, electrical and
electronic equipment housing, mouldings for domestic appliances, aircraft, computers,
business machines, impact resistant blow mouldings, divers automotive items, pipes and
profiles, industrial tools, plugs and receptacle covers.
Terpolymers of methyl methacrylate, butadiene and styrene (MBS) are similar to ABS,
having a rubber dispersed phase and a plastic matrix. The studies of Tg, impact strength
and morphology of PVC/MBS blends conclude that the blends are immiscible [66]. The
impact strength reaches a peak at 20 wt% MBS concentration, with a value much higher
than either MBS or PVC individually. MBS can be used as an impact modifier, despite its
immiscibility. This is explained by the strong adhesive bond between MBS and PVC
phases [3, 4, 67].
Newer type of MBS, designed as modifiers for PVC, have controlled size of elastomeric
particles that make it possible to prepare transparent PVC/MBS sheets for packaging.
For example, MBS with cluster-like structures is being offered. The individual particle
forming the cluster has a diameter of 5070 nm. They are held together by styrene-
methyl methacrylate graft copolymer or terpolymer.
The term acrylic describes a family of polymers or copolymers containing repeating units
that can be considered as derivatives of acrylic or substituted acrylic acid.
82
Polyvinyl Chloride-Based Blends
Aspects of PVC compatibility with acrylic polymers are discussed in Chapter 9. In the
following sections only the properties and application areas of the PVC/acrylics blends
will be presented.
A bulk analysis of PVC/acrylics blends denotes that by choosing the appropriate acrylic
component one can assure an extremely wide area of properties like tensile strength,
impact strength, low-temperature flexibility and blocking resistance, low-temperature
impact behaviour, high HDT and toughness, good transparency, flame resistance, weather
resistance, chemical and solvent resistance, craze resistance, thermal stability and good
processability [13].
The PVC/acrylic blends can be obtained using all classical polymer-processing techniques
(extrusion, injection moulding, calendering, pressing, thermoforming, etc.), with specific
parameters for each type of blend.
PVC/acrylic extruded sheets have been used for thermoforming applications. Ease of
thermoformability, toughness, resistance to cleaning solvents and flame retardancy
characteristics of the blend have been the primary features leading to its use. The inherent
flame retardancy and low smoke-generation characteristics of PVC/polymethyl
methacrylate (PMMA) blends meet the aircraft fire safety standards. These aspects,
coupled with low-cost, led to use in aircraft and mass-transport vehicles.
PVC/acrylic blends have been used to form industrial, commercial and consumer goods,
(e.g., as wall coverings, corner guards, column covers, shelving, counter laminates, chairs,
seats, trays, ceiling tiles, window frames, weather boards, roof gutter systems, pipes,
swimming pools, ladders, claddings), in medical, electrical (cable ducts and conduits)
and diverse engineering equipment, for food or beverage equipment, as aircraft or mass
transit interior components, for powder tool housing, razor bodies, vacuum cleaner parts,
protective helmets, steering wheels, records, toys, sport articles, aircraft components
such as toilet bowls, floor pans, air diffusers, emerging respirator enclosures, office
equipment and computer parts, and other applications requiring good resistance to
weathering [13].
Although the classical definition of a polymer blend states that the components are
chemically different [68], there are, however, several PVC blends affiliated to this polymeric
materials group.
83
Handbook of Polymer Blends and Composites
Casy and Okano [69] showed that the 50/50 blend of quite higher molecular weight
PVC (HMW-PVC) and relatively lower molecular weight PVC (LMW-PVC), of which
the mathematical viscosity is equal to the medium molecular weight PVC, has a better
melt flow index than the medium molecular weight PVC for injection moulding.
Yamamoto and co-workers [70-71], and Kawakami and co-workers [72] find the same
result, and furthermore, they established that the tensile strength and tensile modulus of
blended PVC are enhanced with increasing low molecular weight (LMW)-PVC content
and decreasing molecular weight of LMW-PVC. Hence, tensile properties, are affected
by molecular weight distribution (Mw/Mn) of blended PVC. The impact strength diminishes
with increasing LMW-PVC content, and, especially with decreasing molecular weight of
LMW-PVC. They have also found that the blends of HMW-PVC and LMW-PVC have
improved thermal stability and processability [70-71].
Plasticised PVC membranes have been widely used as sensitive layers in potentiometric
and fibre-optic ion sensors. More recently, carboxylated PVC (PVC-COOH) which
contains 12 wt% carboxyl groups in the side chain of the PVC has attracted much
attention as a novel class of material for chemical sensor membranes [72, 73]. Given this
information, Babu and Gaikar [74] studied the possibility of making new ultrafiltration
membranes based on PVC/PVC-COOH blends.
Using several methods, the authors first underlined the good compatibility of PVC and
PVC-COOH, related to the same chemical nature and/or because of specific interactions
between C=O and Cl-CH groups.
The PVC/PVC-COOH (10/1) blend membranes, additivated with various solvents (methanol,
ethanol, butanol, cyclohexanol, ethyl acetate), gave a higher flux (8595 l/m2h) and analogous
rejection efficiency (94%) compared to PVC and CPVC membranes alone. The pore size
of blend membranes in the range of 2 to 21 was also found to be suitable for
ultrafiltration.
Although there are contrary opinions, many studies highlighted the incompatibility of
PVC with PVAc and the fact that the degree of miscibility also depends upon the method
of manufactory the blend [7577].
Even if they are characterised by good processing properties, PVC/PVAc blends have
poor mechanical properties and, as follows, limited use. This inconvenience may be
avoided if, in making PVC/PVAc blends, the mechanical mixing of the constituents is
replaced with VA polymerisation in the presence of PVC. A similar effect can also be
achieved if the blend with mainly components is associated with elastomers as:
polyisobutylene (PIB), polyethersulfide, polychloroprene (CR), polyisoprene (IIR),
butadiene-butylacrylate-styrene or vinyl chloride-ethylacrylate copolymers. The blends
84
Polyvinyl Chloride-Based Blends
McNeill and Jameison [80] have studied the immiscible PVC/PVAc blends by
thermogravimetry and thermal volatilisation analysis and have concluded that HCl
produced by the dehydrochlorination of PVC migrates into PVAc phase and causes
accelerated deacetylation of PVAc. Other studies [81] confirmed this conclusion.
Studies referring to PVC/polyvinyl alcohol (PVA) underlined that these two polymers
have a limited miscibility, a better degree of miscibility being observed to outermost
parts of the composition domain, 90/10 and 10/90 PVC/PVA, respectively.
Showing thermal stability and good processability, PVC/PVA blends can be used for
covers, synthetic fibres and blow moulding.
Because of the great number of PVC/engineering polymer blends, in the following pages
only compositions that include polyesters, polycarbonates, polyalkylene oxide, and liquid
crystal polymers (LCP) are discussed.
PVC/polyester blend morphology and properties are mainly subjected to type (aliphatic,
aromatic or aliphatic-aromatic), structure (CH2:COO ratio) and molecular weight of
polyesteric component.
If only PVC/aliphatic polyester (PEST) compositions are taken into consideration, theoretic
calculations demonstrated that the blends show a miscibility window between CH2:COO
of 4 to 10 [82]. There are two types of hydrogen bonding governing the miscibility
85
Handbook of Polymer Blends and Composites
between PVC and PEST; one is the interaction between hydrogen of the methylene and
chlorine of PVC, and the other is the interaction between a hydrogen of PVC and carbonyl
group of PEST. It is difficult to separate quantitatively two attractive interactions, which
increase an approach to zero as CH2:COO increases [83].
Upon overall consideration, at low values of the CH2:COO ratio, there is a strong
attractive interaction within polar PEST molecules and self-association results in blend
miscibility. As the value of the CH2:COO ratio is increased, the intermolecular interaction
between PVC and aliphatic PEST begins to overcome the self-association energy of
PEST. Although the energy should be definitely weak, because the intramolecular
interaction in PEST is not usually classified as strong self-association, the energy should
be stronger at lower CH2:COO ratio. The blend with a CH2:COO ratio value between
5 and 7 shows an optimum in intermolecular interactions. The maximum miscibility
appears at a CH2:COO ratio of 6, where binary intermolecular parameter (B from
equation Hmix/V = B12, where Hmix is the enthalpy of mixing; V is the total volume
of the blend; I is the volume fraction of component 1) was equal to 3.85 cal/cm3. At
higher values of the CH2:COO ratio, the interaction, and correspondingly begins to
show immiscibility. It has been reported for PVC/aliphatic PEST blends that the
immiscibility at high carbonyl content may be due to an unfavourable balance of
intramolecular interactions mostly occurring between methylene and carbonyl groups.
In the immiscible region, because of the lack of favourable interactions, entropy effects
such as unfavourable free volume effect should become more prominent, which were
not at all considered in the calculation.
Coleman and co-workers [84] used a solubility parameter method, modified with an
added term to account for the presence of favourable intermolecular interactions in
polymer blends. For a PVC/aliphatic PEST blend the optimum miscibility is located at a
CH2:COO ratio of 3.5. In the framework of the miscibility guide they suggested, the
miscibility is dependent on the balance between the favourable strong interactions, like
hydrogen bonds, and the physical interactions, which in general are unfavourable.
86
Polyvinyl Chloride-Based Blends
The position held by PCL in the aliphatic PEST series justifies the great number of studies
referring to the system PVC/PCL [8696]. From these studies, the blend system has been
found to be miscible, in the amorphous or molten state, throughout the entire composition
range 1090 wt% PCL. These studies have also shown that some PCL becomes crystalline
upon solidification in blends with 40 to 50 wt% or more PCL present.
Thus, PVC/PCL blends having less than 40 wt% PCL are homogenous, given to the full
compatibility of the system components. In this kind of blend PCL acts as macromolecular
plasticiser. The blends where the PEST component ratio exceeds 40 wt% have a complex
structure, being constituted from a mixture of homogenous PVC/PCL blend, with
amorphous structure, associated with crystalline and amorphous domains of PCL. PVC
is excluded from PCL crystallites, but remains in the interstitial areas of semicrystalline
materials [86]. The proportion of crystalline domains, the shape and size of PCL crystallites
from PVC/PCL blends having more than 40 wt% PCL depends on the molecular weight
of the components, the real proportion of PEST component in the blend, of the method
obtaining the blend, thermal treatment and mechanical stress that the blend undergoes
[89, 9498]. Blends achieved by solution casting exhibit a greater degree of crystallinity
than those obtained by melting. Stretched films have a greater degree of crystallinity
than unstretched films. The stretching primarily involves the structural transformation
of crystalline PCL from lamellae to microfibrils [94].
In solution, it was found that PVC and PCL are compatible over the whole range of
composition [9193]. The characteristics of the solutions containing PVC/PCL blends
depend on the nature of solvent [93].
PVC/PCL blends can be readily prepared by a hot compounding process. The compositions
are flexible, transparent, plasticised products [87, 99]. PVC partially destroys PCL
crystallinity. Thus, blends of two rigid polymers become soft and pliable compositions.
However, at high PCL content (>40 wt%), the blends become translucent and more rigid
because PCL crystallisation at time elapses. This effect is dependent on blend composition.
All PVC/polyneopentyl glycol adipate blends are transparent and showed one Tg. The
optical clarity of the blends and the existence of one composition-dependent Tg were
evidence that polyneopentyl glycol adipate is miscible with PVC. All blends remained
transparent when heated to 220 C. Added in 5 wt% proportions to PVC compositions,
polyneopentyl glycol adipate also acts as a thermal stabiliser [82]. The same effect was
87
Handbook of Polymer Blends and Composites
For PVC/polyethylene adipate blends, it was found that at reduced proportions of polyesteric
component, an improvement of thermal stability is also established. Over a wider range of
blend composition, however, the thermogravimetric analysis results show complex
behaviour, with initial stabilisation of the polymer, seen at lower polyethylene adipate
contents, being replaced by marked destabilisation at a higher polyester content [101].
Hourston and Hughes [102-103] and Nishi and co-workers [104] studied acoustic damping
of PVC/polytetramethylene terephthalate polytetramethylene ether glycol terephthalate
block copolymer. This segmented polyether ester (Hytrel 4055) is a block copolymer with
crystallisable tetramethylene terephthalate hard segments and polytetramethylene ether
glycol terephthalate soft segments. Dynamic mechanical studies indicated that blends
containing 2550 wt% Hytrel were completely miscible, a single glass transition point was
observed; but, as the Hytrel level was increased to 60 and 65 wt%, a shoulder becomes
apparent on the low-temperature side of the glass transition peak. Finally, at 80 wt%
Hytrel, two peaks were observed, indicating immiscibility of the blend. The Tg of PVC/
Hytrel blends was found to decrease with the amount of added Hytrel [105, 106].
PVC/Hytrel blends are used in flexible hose manufacture and electrical cable insulation.
Polyethylene terephthalate caprolactone copolyesters are miscible with PVC only if the
polyethylene terephthalate (PET) content is between 35 and 44 wt% [108].
The blends of PVC with aliphatic polyester aromatic polyester block copolymer (Perprene
1002 from Toyobo Corp.), with physically crosslinked structures and service temperature
of 125 C, were found useful for manufacturing heat-resistant insulation for electric
cables [13].
PVC blends with semi-aromatic copolymers obtained by reacting PET with polybutylene
glycol and 1,4-butanediol are homogenous and have good low-temperature impact
properties [13].
88
Polyvinyl Chloride-Based Blends
plasticisers [85]. Nishi and co-workers [109] attributed miscibility to the soft segment,
without clarification as to the specific group causing it. Robeson [110] ascribed
compatibility to the proton-acceptor potential of the ester carbonyl from PBT segment
by analogy to other polyester PVC miscible blends.
Mixtures of PVC and PBT polytetrahydrofuran (PTHF) block copolymers have been
reported to exhibit at least partial miscibility [111]. These block copolymers are excellent
permanent plasticisers for use in more demanding plasticised applications. These include
the examples previously cited for plasticised PVC with particular emphasis on shoe
components, where abrasion resistance is important. PVC/PBT PTHF copolymer blends
have been used in agricultural insecticide hose jackets to obtain better low-temperature
flexibility and abrasion resistance than typical plasticised PVC [112]. These blends have
been used in interior automotive upholstery, food packaging film, wall coverings, flooring,
and wire and cable coatings. PBT PTHF block copolymer, is also used as a permanent
plasticiser for PVC with better thermal stability, better release from highly polished
chrome-plated processing rolls, and the ability to use higher shear mixing screws in
extrusion processing.
PVC/PBT blends with a small ratio of polyesteric component are homogenous, while
those with great amounts of PBT induce biphasic mixtures, where a phase is homogenous,
amorphous, arising from PVC/PBT blend and the other crystalline, from PBT [110].
With the commercial demand for plasticised PVC compositions of greater permanency
for more demanding applications as well as for freedom from ecological implication of
plasticiser migration into the environment, recent emphasis has been placed on readily
available intermediate molecular weight plasticisers (Mn = 20004000). These oligomers
are generally polyester types, e.g., adipate or adipic acid and ethylene glycol. Examples
of those oligomers are cited in reference [113] and will not be discussed in detail because
the molecular weights are generally lower than the systems covered in this chapter.
With relatively similar structure as aromatic PEST, bisphenol A polycarbonate (PC) proved
to be incompatible with PVC [114]. However, to make these blends useful, system
compatibilisation using polydimethylsiloxane (PDMS) multiblock copolymer or
bis(hydroxyphenyl)-hexafluoropropane PC (6F-PC) has been tested.
89
Handbook of Polymer Blends and Composites
Gorelova and co-workers [115] studied the surface composition of PVC/PC/PDMS blends
and observed the formation of transparent solvent cast films from some of them. Papkov
and co-workers [116, 117] discovered the same. This finding is interesting because the
reflective indices of PVC, PC and PDMS are different and, consequently, transparency of
the blend films may suggest miscibility of their constituents, despite the fact that PVC is
individually incompatible with both PC and PDMS. Another possible explanation of
this event is that these blends are pseudocompatible and their transparency is caused by
the very small size of the dispersed block copolymer phase. As is already known, such an
optical effect can occur if the size of the dispersed phase particles does not exceed about
100 nm. If this is the case, at certain blend compositions, specific conditions could arise
that lead to a very high level of dispersed PC/PDMS blend in the PVC matrix.
Adding at least 15 wt% 6F-PC to the PVC/PC blend, a homogenous mixture with a
single Tg is formed. These blends are flame-resistant, and are also resistant to acids,
bases and many organic solvents. These mouldable, transparent and fireproof alloys had
the stability and strength of PC and processability of PVC. They were used for packaging,
bottles, parts of medical and chemical instruments, and for hot water pipes [13].
90
Polyvinyl Chloride-Based Blends
disappeared in PVC/LCP blends with an LPC content above 15 wt% owing to the
decomposition of the LCP during injection moulding. In this case, the mechanical strength
of the PVC/LCP blends with the higher LCP content tend to decrease sharply with
increasing LCP content.
Epoxidated natural rubber with 50 mol% epoxidation is polar and has similar properties
to synthetic elastomers [128]. These qualities generated interest in assessing the
compatibility of ENR with other polymers, particularly PVC [129139].
It has been reported [129134] that blends of rigid PVC with ENR-50 form a miscible
system in which ENR-50 reduces Brabender torque [135] and melt viscosity of rigid
PVC [131]. Both melt-mixed [131] and solution-cast [135] blends have been reported to
exhibit a single Tg lying between that of PVC and ENR-50 [136]. The miscibility of PVC
and ENR solid state has been established by the confirmation of hydrogen bonding from
Fourier transform infrared spectroscopy (FTIR) studies [136] and synergism in mechanical
and dynamic properties [133, 134]. Rheological studies revealed the pseudoplastic nature
of PVC/ENR blends [137]. Fracture studies also revealed that ENR-50 blends are capable
of acting as impact modifiers for PVC [135].
Earlier work on PVC/ENR blends by Nasir and Ratnam [138] indicated the need to use
suitable mixing conditions to attain optimum properties. Similar observations have also
been reported by Ishiaku and co-workers [136, 137]. They found that using a Brabender
Plasticorder coupled with a mixture/measuring head for a PVC-rich blend, optimum
properties are obtained by mixing either at high temperature, say 150 C, at low rotor
speed, or at higher rotor speed and lower temperature. On the contrary, for an ENR-rich
blend, optimum conditions are rather restricted to high temperature and low rotor speed
only. They also indicate that the choice of optimum blending conditions is not limited to
91
Handbook of Polymer Blends and Composites
any particular temperature, mixing time or rotor speed. Rather, optimum blends can be
obtained by selecting any set of conditions within a processing window (130150 C, 2
9 minutes, 2050 rpm).
The clarity of the film or sheets made from the PVC/ENR-50 blend support the idea that
the miscibility inferred from the melt state is equally true for the solid state. This statement
is further supported by the observation of synergism in the tensile properties. The tensile
strength of the PVC-rich blend, i.e., 4080 wt% PVC, exceeds the value expected if the
properties are merely additive [140]. A similar trend is also observed for elongation at
break, where this exceeds the additivity line at all compositions [136]. Maximum
synergistic effects in tensile strength occurred around the 70 wt% PVC region, which
also coincide with the observed increase in density. The sudden deviation from the observed
trends of such properties around the 70 wt% PVC composition range could also be a
reflection of improved packing.
PVC/ENR blends are prone to thermo-oxidative ageing. This has been attributed to
initial PVC degradation, which releases HCl that catalyses the ring-opening reaction of
the epoxided group. The combination of an optimum amount of antioxidant (2,2,4-
trimethyl-1,2-dihydroquinoline (TMQ)), and has been proven to be effective in curing
the thermooxidative ageing and irradiation stability of the blends [138, 141].
Varughese and co-workers [132] concluded that melt mixed blends of plasticised PVC
and ENR-50 have a phase morphology close to a single phase system. These blends have
a single Tg lying between the Tg of the blend components as shown from DMA and DSC
thermograms. Plasticised PVC has the maximum broadening in glass-rubber transition
of all the samples. The drastic reduction in Tg width of plasticised PVC with the
incorporation of ENR may be due to uniform distribution of dibutyl phthalate among
ENR molecules, which has joined with PVC particles. This reduction in Tg width from
plasticised PVC to the PVC/ENR blend suggests a more defined single-phase morphology
in the latter. Since the Tg of plasticised PVC/ENR blends is much lower than that of rigid
PVC/ENR blends [131], the former have greater low temperature flexibility.
92
Polyvinyl Chloride-Based Blends
From the previous discussion, it can be inferred that each type of plasticiser has its own
potential and limitations. For instance, the incorporation of DOP generally decreases
the mechanical properties of the PVC/ENR blends, although it simultaneously stabilises
the blend against thermooxidative degradation.
Based on the narrowing of the glass transition region observed by Varughese and co-
workers [132] in the DOP-plasticised PVC/ENR blend, it can be inferred that DOP is
intimately miscible with ENR, as is the case with PVC.
93
Handbook of Polymer Blends and Composites
Several compositions of PVC/ENR/CPE blends using various types of CPE (22, 35 and
45% chlorine contents), ENR-25, and ENR-50 were studied [140, 141]. The results
obtained indicate that some compositions using amorphous medium chlorine content
CPE and ENR-50 have a good miscibility, mechanical properties, and a damping behaviour
suitable for high performance vibration absorbers. Furthermore, flake-shaped fillers
(graphite, mica) could be used to improve the damping behaviour of PVC/ENR/CPE
blends [140].
The investigation carried out on DOP plasticised PVC and epoxidated 1,2-polybutadiene
(1,2-EPB), epoxidated 1,4-polybutadiene (1,4-EPB) mixtures mainly tracked the co-
stabiliser effect of the rubber in combination with metal soaps [Zn/Ca (2/1) stearates]
[142, 143]. It has been reported that 1,2-EPB exhibits good compatibility and finely
dispersed into compound PVC as compared to 1,4-EPB and that epoxidated PB build up
synergistic blends with metallic soaps. The synergistic effect of 1,2-EPB is superior to
1,4-EPB effect.
This type of blend historically represents the initial observation that miscibility of polymer
mixtures is possible [144]. The investigations on this particular blend outnumber any
other miscible polymer blend. It is also the result of scientific and commercial interest in
this blend over the past five decades [144-204].
The compatibility of the PVC/NBR blends was studied by numerous methods [dynamic
mechanical, dynamic viscoelastic behaviour, viscosimetric measurements, dynamic thermal
analysis (DTA), DSC, light scattering, gas permeability, density, stress-strain measurements,
mechanical properties, optical and transmission electron miscibility].
Dynamic mechanical tests have been a common method of investigation PVC/NBR blends
compatibility [105, 106, 145, 148, 154, 205-209]. Nielsen [145] and Breuers and co-
workers [148] found for PVC/NBR blends that a single maximum in damping occurs at
a temperature between the corresponding temperatures of the pure components. They
reasoned that the polymers were soluble each in other. Nielsen [154] concluded that the
94
Polyvinyl Chloride-Based Blends
broad damping peak found for a PVC/NBR blend indicated that the polymers were
soluble in each other, but the intermolecular attraction is so weak that considerable
association of like segments is taking place in the soluble mixture. Nielsen also pointed
out that in plasticised PVC, crystallites act as crosslinks preventing the rubber material
from flowing under load, even if it is not crosslinked by a chemical reaction, as in
vulcanised rubber.
Takayanagi and co-workers [155, 156, 163] measured the dynamic viscoelastic behaviour
of PVC/NBR blends prepared in solution and recovered by evaporation of the solvent.
They concluded that even though the two polymers mix molecularly there is some
microheterogeneity shown because of the different segmented environments. Matsuo
and co-workers [161, 165] examined a mill-mixed blend of PVC with NBR containing
8%40% ACN. The results of the dynamic viscosity measurement were interpreted to
indicate that a 20% ACN elastomer gave almost homogenous blends. Electron microscopy
pictures were interpreted to indicate that microheterogeneity existed even in the blends
with 40% ACN elastomer. Using DTA, Oganesov and co-workers [164] investigated the
homogeneity of mill-mixed blend of PVC with NBR containing 40% ACN. They came
to the conclusion that blends can be single or two phases depending upon the polymer
ratio. Feldman and Rusu [166, 168] examined the compatibility of PVC with NBR
containing 40% ACN by viscosimetric measurements of a blend solution in
cyclohexanone. They found that the polymers are compatible. Vasile and co-workers
arrived at the same conclusion [196, 209], using the same viscosimetric method for PVC/
NBR blends with 38% ACN elastomer and a Mooney plasticity of 45. Krause [170], in
her review on polymer compatibility, lists the combination of PVC with NBR as
compatible.
Using DTA and DMA, Jorgensen and co-workers [175] demonstrated the presence of
two Tg in commercial non-crosslinked NBR with less than 35% ACN content. They
concluded that for PVC/NBR blends containing 70% NBR, PVC acts as selective solvent
for only that portion of NBR which is highest in ACN content. Zakrzewski [176] used
phase contrast microscopy, DSC and torsion pendulum analysis to examine the
compatibility of mill-mixed PVC/NBR blends. He concluded that nitrile elastomers
containing 25%45% ACN are compatible with PVC at all levels of PVC.
Landi [171, 172, 177] reviewed previous studies on PVC/NBR blends compatibility and
used DSC to measure Tg. When a 29% ACN elastomer with two Tg was mixed with
PVC, only the higher Tg was affected. Only a nitrile elastomer containing 34% ACN
exhibited a single Tg in its PVC blends. Ranby and co-workers [183, 184, 190] examined
two properties of mill-mixed PVC/NBR blends with 2234% ACN using light scattering,
95
Handbook of Polymer Blends and Composites
gas permeability, density and DMA. Increasing the amount of ACN in the elastomers
resulted in better compatibility with PVC.
Wang and Cooper [191] investigated the morphology of PVC/NBR blends spincast from
tetrahydrofuran. Their studies were carried out at various temperatures using DSC, DMA,
transmission electron microscopy (TEM) and infrared dichroism technique. Nitrile
elastomer containing 44% ACN was found to be incompatible with PVC used under
conditions of their blend preparation. Nitrile elastomer containing 31% ACN was found
to be compatible with the PVC with some indications of heterogeneity in form of
microdomains which are rich in either PVC or NBR.
Reznikova and co-workers [146, 147, 149, 150] studied the effects of varying the ACN
content in the nitrile elastomer and the amount of NBR in the compositions, on PVC/
NBR blends properties. Nitrile elastomers containing 10%50% ACN were mill-mixed
with PVC over a range of 0%100% PVC. Nitrile elastomers containing 28% or less
ACN had poor compatibility with PVC giving blends which were cloudy and had weak
and poor elasticity. Lelckuk and Sedlis [151, 152] studied PVC/NBR blends prepared by
latex blending, followed by coagulation and then milling at 130140 C. Nitrile elastomers
with 19%40% ACN were used. The effect of varying the molecular weight from
40,000 to 390,000 of a 35% ACN elastomer was investigated. Nitrile elastomers
containing 35%40% ACN with low molecular weight gave the best processability and
properties.
The general PVC/NBR blends properties analysis emphasised that the rubber acts as a
plasticiser for PVC and PVC reinforces the rubber. Originally NBR was used as plasticiser
for PVC and PVC was used to improve ozone resistance and weathering of nitrile
elastomer. Today, PVC/NBR blends form a class of polymeric materials in their own
right and should be considered as such.
96
Polyvinyl Chloride-Based Blends
When used as thermoplastic materials, the PVC proportion of the blend is frequently
50 wt% or above and the properties are nearer those of a plasticised PVC. The nitrile
elastomer behaves as a special type of plasticiser, because it is:
1. Non-extractable;
2. Non-migratory;
3. Non-volatile.
Replacement of normal type PVC plasticisers with nitrile rubber gives improved:
2. Abrasion resistance;
3. Heat ageing;
4. Brittle temperature;
5. Impact strength;
7. Retention of mould patterns or embossing [146, 147, 149, 150, 158, 159, 167, 180,
181, 187].
Blends intended for use as rubbery type polymers regularly contain a major fraction of
nitrile elastomer and are compounded and crosslinked to give rubbery end products.
The most significant advantages obtained by addition of PVC into nitrile elastomer are:
97
Handbook of Polymer Blends and Composites
5. Flame resistance;
6. Volume resistivity;
The extent to which the tracked properties for PVC/NBR blends are achieved is influenced
by a great number of factors:
1. PVC/NBR ratio;
6. Mixing technique;
Reed [210] stated that nitrile elastomers plasticise PVC by a solvation mechanism in the
same way as monomeric plasticisers. He pointed out that when powdered PVC is added
to NBR and the batch is hot milled, the blend becomes clear and behaves as a normal
plasticised compound. The use of nitrile elastomer as a plasticiser resulted in a blend
with excellent resistance to swelling and to loss plasticiser due to extraction and volatility
[88]. Increasing levels of ACN will improve the oil resistance but the brittleness
temperature also increases. The optimum range of addition of plasticised PVC to nitrile
elastomers is 2040 wt% NBR with 29%39% ACN content [88].
98
Polyvinyl Chloride-Based Blends
Kronman and Karghin [160] found that the impact resistance of PVC was markedly
improved by low levels of NBR containing 18% ACN. Higher levels of 26 wt%
acrylonitrile copolymer were required for comparable improvement and the addition of
40 wt% acrylonitrile copolymer had no appreciable effect.
Mann and Williamson [181] stated that a controlled degree of compatibility is required
for reinforcement. They also refer to the work of Davenport and co-workers [154] who
found that the impact strength decreased somewhat with an addition of up to 5 phr and
then increased with rubber content.
Khne [169] established that impact resistance was appreciably greater if the NBR was
highly branched and partially crosslinked rather then linear. A suitable modifier was
9:1 NBR polymerised to 99% conversion and latex blended with PVC. The optimum
particle size was 0.10.3 m and larger particles were thought to act as notched.
Schwartz and Edwards [182] reported that most of the processing and vulcanisation
properties of compounds based on NBR/PVC blends are affected, often dramatically, by
one or more of the fundamental properties of the constituents of the blends, as well as by
the blend ratio.
In the preparation of blends from PVC and NBR there are two very important conditions,
which must always be satisfied if the optimum properties are to be obtained. These are:
(1) good dispersion of one polymer into the other, and (2) satisfactory fluxing of the
rubber into the PVC [167].
Mechanical blends of the solid polymers must also be fluxed at higher temperatures (150
160 C) to cause the complete homogeneity of the two polymers and to obtain the optimum
properties from the blend. A too low temperature will result in underfluxed material with
less than optimum properties. Too high a temperature may result in decomposition of
PVC, and possible gellation of the rubber. This results in inconsistency of the product, and
discoloured, high viscosity materials with a poor processing properties [167].
99
Handbook of Polymer Blends and Composites
It can be appreciated that the formation of blends of PVC and NMR from solid polymers
can be a difficult process, and certain one that requires very careful control to obtain
consistently good products.
The availability of pre-fluxed blend, i.e., a blend made via a latex blending stage, based
on specially developed polymers, offers many advantages:
2. As the blends are fluxed during manufacture, further high temperature mixing is not
required;
3. Because of the advantages described in (1) and (2) conventional rubber processing
equipment may be used to handle these blends;
6. Convenience of polymer from, i.e. thin sheet easy for handling and weighing or chip
for solution work; automatic weighing systems [167].
When powdered nitrile elastomers became available, they were combined with powdered
PVC compounds to make possible the processing of PVC/NBR blends in powder form.
The advantages of powder processing are discussed in several publications which show
typical compound formulations useful for thermoplastic applications [171, 172]. The
use of pre-crosslinked and powdered nitrile elastomer is reported to be effective in reducing
the elastic memory (nerve) of PVC/NBR blends [167].
In the last years, the investigations referring to PVC/NBR blends mainly centred on their
property enhancement. With that end in view, they realised and characterised more types
of interpenetrating polymer structures, giving PVC/NBR compositions with substantially
improved impact strength, compression set, permanent set, resiliency and high temperature
shape retention [192, 194, 195, 196200, 203].
Because PVC reinforces NBR, a blend has high gum strength. This means that mixing
the blend with fillers such as whiting and clay may produce cheap light coloured
compounds. For cheap black compounds, a similar mixture may be used with the addition
of a small amount of carbon black.
100
Polyvinyl Chloride-Based Blends
The use of a small amount of a highly reinforcing filler is permissible together with the
cheaper filler or of moderate amount of less reinforcing fillers. In this way, many
specifications may be matched without the loss of weathering properties. In areas where
the end product is not going to be used under strained conditions, then highly reinforcing
fillers will produce compounds with increased tensile, modulus, tear and abrasion resistance.
Plasticisers for PVC/NBR blends are the ones normally used in nitrile elastomers or PVC
to modify viscosity, tensile properties or to improve resilience and low temperature
properties.
For the compositions based on NBR/PVC, two classes of additives play a significant
role: the vulcanisation agents and those belonging to the protection group, which are
intended to provide an increase of end-product resilience. In order to accomplish pure
thermal destruction, thermooxidative and ozonant protection, one must make allowances
for both the elastomer part and PVC part protection [182, 188]. Among antidegradants
are suggested diphenylamine derivates, oligomers of dihydroquinoline in combination
with mercaptobenzimidazole or bis(4-hydroxyphenyl)propionate derivates [106, 179].
The protection of some mixtures was accomplished using alkyl-aryl and diaryl derivates
of N,N-disubstitute p-phenylendiamine, giving good performances until 150 C [174,
181, 189].
The PVC/NBR blends are primarily designed for extrusion and calendering, but injection,
blow, compression and transfer moulding are feasible as well. The blends are used in
manufacture of hoses, rolls, wire and cable insulation, footwear, sheets for expanded
insulation, foamed extruded profiles for door and window seals, automotive accessories,
diaphragms, conveyor belts, protective clothing, shoe soles, jacketing, roll covers, cable
and wire jackets [13, 166].
101
Handbook of Polymer Blends and Composites
The blends with epoxy functional groups content less than the minimum required are
biphasic; a phase represents homogeneous blend PVC/ESB and the other PVC [215
217]. SBR star copolymers are less miscible than linear copolymers, because of steric
restrains of epoxidated elastomer segments (miscible in PVC) from nonpolar styrenic
segments.
The most important commercial PVC-based blends are presented in Table 3.1.
102
Commercial PVC blends
Polyvinyl Chloride-Based Blends
103
Commercial PVC blends
Handbook of Polymer Blends and Composites
104
Commercial PVC blends Polyvinyl Chloride-Based Blends
105
Commercial PVC blends
Handbook of Polymer Blends and Composites
References
1. J.A. Brydson, Plastics Materials, 5th Edition, Butterworths, London, UK, 1989.
2. J. Frissell in Encyclopaedia of PVC, Ed., L.I. Nass, Marcel Dekker, Inc., New
York, NY, USA, 1976, 1, 258.
106
Polyvinyl Chloride-Based Blends
3. Polymer Data Handbook, Ed., J.E. Mark, Oxford University Press, Oxford, UK,
1999.
4. J.C. Huang, D.C. Chang and R.D. Deanin, Advances in Polymer Technology,
1993, 12, 1, 81.
6. D.R. Paul in Polymer Blends, Eds., R.D. Paul and S. Newman, Academic Press,
New York, NY, USA, 1978, 2, 35.
7. H. Ling and B.D. Favis, Industrial Engineering & Chemistry Research, 1997, 36,
1211.
8. A. Tripathi, H.K. Tripathi and P.C.K. Rallai, Journal of Applied Physics, 1988,
64, 4, 2081.
10. Y. Nakamura, Polymeric Materials: Science and Engineering, 1987, 57, 775.
12. A. Rudin and P. van Ballegooie, Journal of Applied Polymer Science, 1990, 39,
10, 2097.
13. L.A. Utracki, Commercial Polymer Blends, Chapman and Hall, London, UK, 1998.
15. Y.J. Shurand and B. Ranby, Journal of Applied Polymer Science, 1976, 20, 3105.
16. J.J. del Val, C. Lacabanne and A. Hiltner, Journal of Applied Physics, 1988, 63,
5312.
17. A. Tse, R. Laakso, E. Baer and A. Hiltner, Journal of Applied Polymer Science,
1991, 42, 1205.
107
Handbook of Polymer Blends and Composites
19. X. Xu, L. Zang and H. Li, Polymer Engineering and Science, 1987, 27, 398.
20. M.F. Champagne and R.E. Prudhomme, Journal of Polymer Science: Part B
Polymer Physics, 1994, 34, 615.
21. G.F. Ignatova, V.L. Trizno, A.F. Nicolaev and N.V. Daniel, Plasticheskie Massy
(USSR), 1994, 31, 1, 136.
22. H. Xi, Z. Liwn and L. Huilin, Polymer Engineering and Science, 1987, 27, 2,
328.
26. N.S. Koklas, D.D. Sotiropouloer, J.K. Kollitsis and N.K. Kolfoglu, Polymer,
1991, 32, 1, 26.
28. D. Feldman and M. Rusu, European Polymer Journal, 1974, 10, 1, 41.
29. C. Elmqvist and S.E. Svanson, Kolloid Zeitschrift und Zeitschrift fur Polymere,
1975, 253, 327.
30. C. Elmqvist and S.E. Svanson, European Polymer Journal, 1975, 11, 789.
31. J. Shur and B. Ranby, Journal of Applied Polymer Science, 1975, 19, 5, 1337.
32. B.G. Ranbt, Journal of Polymer Science: Symposium, 1975, 51, 1973.
33. C. Elmqvist and B. Ranby, European Polymer Journal, 1976, 12, 559.
35. E. Nolley, D.R. Paul and J.W. Barlow, Journal of Applied Polymer Science, 1979,
23, 623.
36. R.D. Deanin and N.A. Shah, Journal of Vinyl Technology, 1983, 4, 2, 167.
108
Polyvinyl Chloride-Based Blends
37. C.A. Cruz-Ramas and R.D. Paul, Macromolecules, 1989, 22, 1289.
38. R.D. Deanin and N.A. Shah, Organic Coatings and Plastics Chemistry, 1981, 45,
2, 290.
39. G. F. Hofman, R.J. Statz and R.B. Case, Journal of Vinyl Technology, 1994, 16,
1, 16.
41. J.J. Hickman and R.M. Ikeda, Journal of Polymer Science, Part A2, 1973, 11,
1713.
42. R.D. Paul, C.E. Viston and C.E. Locke, Polymer Engineering and Science, 1972,
12, 157.
43. R.D. Paul, C.E. Viston and C.E. Locke, Polymer Engineering and Science, 1973,
11, 1713.
44. R.D. Paul, C.E. Viston and C.E. Locke, Polymer Engineering and Science, 1972,
12, 157.
46. L.G. Boulard and D. Brountstein, Journal of Applied Polymer Science, 1986, 32,
6131.
48. J.S. Orville, The Science and Technology of Polymer Films, Wiley, New York,
NY, USA, 1971, 2.
49. Y. Haiyang, Z.P. Pingping, L. Guofen, W. Peng and R. Feng, Polymer Journal,
1999, 35, 345.
51. J. Heushen, J.M. Vion, R.Y. Jerome and P. Teyssie, Polymer, 1990, 31, 1473.
52. G.T. Carreba and S. Kandiraju, Advances in Polymer Technology, 1990, 10, 237.
53. J.K. Kallitsis and N.K. Kalfoglou, Polymer, 1989, 30, 2258.
109
Handbook of Polymer Blends and Composites
54. T.P. Russel, A.M. Mayers, V.R. Daline and T.C. Chung, Macromolecules, 1992,
25, 5783.
55. B. Boutevin, Y. Pietrosonta, J.J. Robin and T. Pallat, European Polymer Journal,
1988, 24, 953.
57. M. Rink, L. Monfrini, P. Gasparini and A. Paven, Plastics and Rubber Processing
and Applications, 1983, 3, 2, 145
58. R.Y. Jerome, B. Albert and P. Teyssie, Polymer Preprints, 1986, 27, 1, 72.
61. K. Kuzelova and Z. Vymazol, European Polymer Journal, 1999, 35, 351.
63. A.K. Kulshreshtha, B.P. Sing and Y.N. Sharma, European Polymer Journal, 1988,
24, 1, 33.
64. A.K. Kulshreshtha, B.P. Sing and Y.N. Sharma, European Polymer Journal, 1988,
24, 2, 191.
65. A.K. Kulshreshtha, B.P. Sing and Y.N. Sharma, European Polymer Journal, 1988,
24, 2, 191.
66. D.M. Fann, H.Y. Shih, C.M. Hsiao and S.K. Huang, Journal of Vinyl Technology,
1989, 11, 129.
67. S. Hawriliak and T.J. Shortridge, Journal of Polymer Science: Part B Polymer
Physics, 1990, 28, 1987.
69. W.J. Casey and K. Okano, Journal of Vinyl Technology, 1986, 1, 1, 37.
110
Polyvinyl Chloride-Based Blends
72. E. Linder, E Graf, Z. Niegreisz, K. Toth, E. Pungor and R.P. Buch, Analytical
Chemistry, 1988, 60, 295.
73. J. Anzai and C.C. Li, Polymer Communications, 1999, 31, 10, 375.
74. P.R. Babu and V.G. Gaikar, Journal of Applied Polymer Science, 1999, 73, 1117.
75. D.E. Bhagwager, C.I. Serman, P.C. Painter and M.M. Coleman, Macromolecules,
1989, 22, 4654.
77. C. Vasile, A. Stoleru, M. Esanu and G. Adamescu, Materiale Plastice, 1990, 27,
3, 117.
78. C. Vasile and I.A. Schneider, European Polymer Journal, 1971, 7, 1205.
79. M. Sabliovschi and G. Grigoriu, Journal of Thermal Analysis, 1983, 26, 1, 23.
80. I.C. McNeill and A. Jamieson, Journal of Polymer Science, Part A: Polymer
Chemistry Edition, 1974, 12, 387.
82. R.E. Prudhomme, Polymer Engineering and Science, 1982, 22, 1138.
83. S. Lee, J.G. Lee, H. Lee and S.J. Mumby, Polymer, 1999, 40, 5137.
84. M.M. Coleman, C.I. Serman, D.E. Bhagwager and P.C. Painter, Polymer, 1990,
31, 1187.
85. B.R. Riedl and R.E. Prudhomme, Journal of Polymer Science: Part B Polymer
Physics, 1986, 24, 2565.
86. R.S. Stein, A. Mirva, T. Yuasa and F.B. Khambatta, Pure and Applied Chemistry,
1977, 49, 915.
87. J.V. Kaleske in Polymer Blends, Eds., D.R. Paul and S. Newman, Academic Press,
New York, NY, USA, 1978, 369.
111
Handbook of Polymer Blends and Composites
89. P.T. Gilmore, R. Falabella and R.L. Laurence, Macromolecules, 1980, 13, 4, 880.
91. P. Zhy, H. Yang and S. Wang, European Polymer Journal, 1998, 34, 1, 91.
92. H. Yang, P. Zhy, S. Wang and Q. Guo, European Polymer Journal, 1998, 34, 3/4,
91.
93. H. Yang, P. Zhy, S. Wang and Q. Guo, European Polymer Journal, 1998, 34, 9,
1303.
94. D. Keroack, Y. Zhao and R.E. Prudhomme, Polymer, 1999, 40, 243.
95. H.L. Chen, J.L. Li and T.L. Lin, Macromolecules, 1998, 31, 2255.
96. D. Ma, H. Cai, J. Zang and X. Liaolie, Macromolecular Chemistry and Physics,
1999, 200, 2040.
97. J.R. Runt and P.B. Rim, Macromolecules, 1983, 16, 762.
98. S. Nojima and A. Peterlin, Journal of Polymer Science, Part A2, 1971, 9, 1191.
99. C.F. Hammer in Polymer Blends, Eds., D.R. Paul and S. Newman, Academic
Press, New York, NY, USA, 1978, 2.
100. I.C. McNeill and L.T. Memetea, Polymer Degradation and Stability, 1991, 33,
263.
101. I.C. McNeill and S. Basan, Polymer Degradation and Stability, 1993, 41, 311.
102. D.J. Hourston and D.I. Hughes, Journal of Applied Polymer Science, 1977, 21,
3099.
103. D.J. Hourston and D.I. Hughes, Polymer, 1979, 20, 829.
104. T. Nishi, T.K. Kwei and T.T. Wang, Journal of Applied Physics, 1975, 46, 4157.
106. N. Nakajima and S.Y. Kwak, International Polymer Processing, 1995, 10, 1, 1.
107. I.H. Park, J.S. Jang and Y. Chujo, Angewandte Makromolecular Chemie, 1995,
226, 1.
112
Polyvinyl Chloride-Based Blends
109. T. Nishi and T.K. Kwei, Journal of Applied Polymer Science, 1976, 20, 1331.
110. L.M. Robeson, Journal of Polymer Science, Part A: Polymer Letters, 1978, 16, 261.
111. D.J. Hourston and I.D. Hughes, Journal of Applied Polymer Science, 1977, 21,
3093.
114. S. Krause in Polymer Blends, Eds., D.R. Paul and S. Newman, Academic Press,
New York, NY, USA, 1978, 1.
115. M.M. Gorelova, A.J. Pertsin, V.Y. Levin, L.I. Makarova and L.V. Filimonova,
Journal of Applied Polymer Science, 1992, 45, 2075.
116. V.S. Papkov, G.G. Nikiforova, I.M. Raygorodsky and I.P. Storozlurk,
Vyskomolekulyarnye Soedineniya, Series B, 1975, 37, 428.
117. V.S. Papkov, G.G. Nikiforova, V.G. Nikalski, I.A. Krasotkin and
E.S.Obolonkova, Polymer, 1998, 39, 3, 631.
119. C. Marco, M.A. Gmez, J.G. Fatou, A. Etxeberria, M.M. Elorza and J.J. Iruin,
European Polymer Journal, 1993, 29, 11, 1477.
120. A.G. Margaritis and N.K. Kalfoglou, Journal of Polymer Science: Part B Polymer
Physics, 1988, 26, 1595.
121. A.G. Margaritis and N.K. Kalfoglou, Journal of Polymer Science: Part B Polymer
Physics, 1989, 27, 1767.
122. Y.Z. Meng and S.C. Tjong, Polymer, 1999, 40, 2711.
113
Handbook of Polymer Blends and Composites
125. N.K. Naqvi and A.N. Sen, Polymer Degradation and Stability, 1991, 93, 367.
126. E.V. Kolawale and P.O. Olugbeni, European Polymer Journal, 1985, 21, 8, 187.
128. I.R. Gelling and M. Porter in Natural Rubber Science and Technology, Ed., A.D.
Roberts, Oxford University Press, Oxford, UK, 1988.
129. A.G. Margaritis, J.K. Kallitsis and N.K. Kalfoglou, Journal of Applied Polymer
Science, 1986, 32, 2867.
130. A.G. Margaritis and N.K. Kalfoglou, European Polymer Journal, 1988, 24,
1043.
131. K.T. Varughese, G.B. Nando, P.P. De and S.K. De, Journal of Materials Science,
1988, 23, 3894.
132. K.T. Varughese, P.P. De, S.K. Sanyal and S.K. De, Journal of Applied Polymer
Science, 1989, 37, 2537.
133. U.S. Ishiaku, M. Nasir and Z.A.M. Ishak, Journal of Vinyl Technology, 1994, 16,
219.
134. U.S. Ishiaku, M. Nasir and Z.A.M. Ishak, Journal of Vinyl Technology, 1994, 16,
226.
135. K.T. Varughese, P.P. De, G.B. Nando and S.K. De, Journal of Vinyl Technology,
1987, 9, 161.
136. U.S. Ishiaku and Z.A.M. Ishak in Polymer Blends and Alloys, Eds., G.O.
Shanaike and G.P. Simon, Marcel Dekker, Inc., New York, NY, USA, 1999, 663.
137. U.S. Ishiaku, M. Nasir and Z.A.M. Ishak, Journal of Vinyl Additive Technology,
1995, 1, 142.
138. Z.A. Nasir and C.T. Ratnam, Journal of Applied Polymer Science, 1989, 38,
1219.
139. L.A. Utracki, Polymer Engineering and Science, 1982, 22, 1166.
140. S.N. Koklas, D.D. Sotiropoulou, K.K. Kallitsis and N.K. Kalfoglou, Polymer,
1991, 32, 66.
114
Polyvinyl Chloride-Based Blends
143. T. Iida, J. Kawato, S. Tanie and K. Goto, Journal of Applied Polymer Science,
1989, 37, 1685.
145. L.E. Nielsen, Journal of the American Chemical Society, 1953, 75, 1435.
146. R.A. Reznikova, A.D. Zaionchkovsky and S.S. Voyutsky, Kolloidnyi Zhurnal,
1953, 15, 111.
147. R.A. Reznikova, S.S. Voyutsky and A.D. Zaionchkovsky, Kolloidnyi Zhurnal,
1954, 16, 207.
149. R.A. Reznikova, A.D. Zaionchkovsky and S.S. Voyutsky, Zhurnal Teknicheskoi
Fiziki, 1955, 25, 6, 1045.
150. S.S. Voyutsky, A.D. Zaionchkovsky and R.A. Reznikova, Kolloidnyi Zhurnal,
1956, 18, 511.
151. S.L. Lelchuk and V.I. Sedlis, Zhurnal Prikladnoi Khimii, 1957, 30, 1041.
152. S.L. Lelchuk and V.I. Sedlis, Zhurnal Prikladnoi Khimii, 1958, 31, 790.
153. N.E.Devenport, L.W. Hubbard and N.P. Pettit, British Plastics, 1959, 32, 12,
549.
115
Handbook of Polymer Blends and Composites
164. Y.G. Oganesov, V.S. Ospchiuk, K.G. Mindyarov, V.G. Rayevsky and S.S.
Voyntski, Vyskomolekulyarnye Soedineniya, Series A, 1969, 11, 4, 896.
165. M. Matsuo, C. Nozaki and Y. Jyo, Polymer Engineering and Science, 1969, 9, 197.
168. D. Feldman and M. Rusu, Revue Generale des Caoutchoucs et Plastiques, 1971,
48, 6, 687.
173. R.D. DeMarco, M.E. Woods and L.F. Arnold, Rubber Chemistry and
Technology, 1972, 45, 1111.
116
Polyvinyl Chloride-Based Blends
175. A.H. Jorgensen, A.L. Chandler and E.A. Collins, Rubber Chemistry and
Technology, 1973, 46, 1087.
178. D. Feldman and M. Rusu, Plaste und Kautschuk, 1973, 20, 4, 259.
179. M.H. Abbelkader and S.M. Yussif, Plaste und Kautschuk, 1974, 21, 3, 521.
180. P.J. Corish and B.D.W. Powell, Rubber Chemistry and Technology, 1974, 47, 3, 481.
181. J. Mann and R. Williamson in Physics of Glassy Polymers, Ed., N.R. Haward,
John Wiley & Sons, New York, NY, USA, 1973, Chapter 8.
182. H.F. Schwartz and W.S. Edwards, Applied Polymer Symposia, 1974, 25, 243.
183. B.G. Ranby, Journal of Polymer Science: Polymer Symposia, 1975, 51, 89.
184. Y.J. Shur and B.G. Ranby, Journal of Applied Polymer Science, 1975, 19, 2143.
186. J.R. Dunn, Rubber Chemistry and Technology, 1976, 49, 4, 978.
187. J.R. Dunn, D.C. Coulthard and H.A. Pfisterer, Rubber Chemistry and
Technology, 1978, 61, 389.
188. J.R. Dunn, H.A. Pfisterer and J.J. Ridhend, Rubber Chemistry and Technology,
1979, 52, 2, 331.
189. P.H. McKinstay, International Plastic and Rubber, 1980, 30, 1, 10.
190. B. Terselius and B.G. Ranby, Pure and Applied Chemistry, 1981, 53, 421.
191. C.B. Wang and S.L. Cooper, Journal of Polymer Science: Part B Polymer Physics,
1983, 21, 11.
192. K.E. George, R. Joseph and D.J. Francis, Journal of Applied Polymer Science,
1986, 32, 2867.
117
Handbook of Polymer Blends and Composites
193. J.R. Wolfe Jr., in Thermoplastic Elastomers, A Comprehensive Review, Eds., N.R.
Legge, G. Holden and H.E. Schroeder, Hanser Publishers, Munich, Germany,
1987, 117.
194. G. Ivan, E. Bugan, R. Plesa and P. Bujenita, Industrie Usoara (Bucharest), 1988,
35, 2, 64.
195. P. Ramesh and S.K. De, Polymer Communications, 1990, 31, 466.
198. K. Fukumori, N. Sato and T. Kurauchi, Rubber Chemistry and Technology, 1991,
64, 4, 522.
199. C. Vasile, A. Stoleru, E. Costea and M. Sava, Bulletin of the Polytechnic Institute
of Iasi, Section II, 1992, 36, 1-2, 75.
200. A. Mathew, B.C. Charobathy and P.C. Deb, Journal of Applied Polymer Science,
1994, 53, 8, 1107.
201. N. Nakajima and S.Y. Kwak, Journal of Macromolecular Science Part B: Physics,
1995, 34, 1-2, 137.
202. S.Y. Kwak and N. Nakajima, Journal of Macromolecular Science Part B: Physics,
1995, 34, 1-2, 153.
203. M. Giugiuca and G. Ivan, Materiale Plastice, 1995, 32, 3-4, 217.
204. B.E. Krisyuk, A.A. Popov, N.M. Livanova and A.P. Farmakovskaya,
Vyskomolekulyarnye Soedineniya, Series A, 1999, 41, 1, 102.
205. D.C. Andrei and C. Andrei, Materiale Plastice, 1990, 27, 1, 23.
206. D.C. Andrei and C. Andrei, Materiale Plastice, 1990, 27, 2, 79.
118
Polyvinyl Chloride-Based Blends
212. P. Zhou and Z. Sun, Plastics and Rubber Processing and Applications, 1987, 7,
229.
214. S. Sha, T. Zhang and Z. Xie, Xiangjao Gongye, 1992, 40, 15, 297.
215. A. G. Mongavitis, I.K. Kollitsis and N.K. Kalfoglou, Polymer, 1989, 30, 2, 2253.
216. S.N. Koklas and N.K. Kalfoglou, Polymer, 1992, 33, 1, 75.
217. S.N. Koklas, N.G. Gavralas and N.K. Kalfoglou, Polymer, 1994, 35, 7, 1425.
218. L.A. Utracki, Polymer Alloys and Blends, Thermodynamics and Rheology,
Hanser Publishers, Munich, Germany, 1989.
119
Polystyrene and Styrene Copolymer Based Blends
4
Polystyrene and Styrene Copolymer Based
Blends
4.1 Introduction
Styrene polymers include homopolymer styrene (PS) and a series of copolymers of styrene
(S) with other monomers. Styrene polymers account for 10%-11% of plastics production
and consumption in North and South America and 8%-9% in Europe. The remarkable
growth of styrene plastics is, a consequence of their properties [1] that make them useful
in many fields such as electronics, automotive, building, household goods, etc.
121
122
properties of Styrene-containing
(Izod)
Vicat softening 66-90 100 90-121 80-85 110-130
copolymers
Temperature (C)
Heat deflection temperature 151-194 ~100 85-111 70-86 95-105 132-160 100-130
at 1.8 N/mm2 (C)
Linear thermal expansion 6-8 7 7-12 8.3 7-8
coefficient (10-5 K-1)
Dielectric constant 2.5 2.9 3.2 2.9-3.0 3.1-3.3
ABS: acrylonitrile-butadiene-styrene terpolymer; SAN: styrene acrylonitrile random copolymer; PVC:
polyvinylchloride; PC: polycarbonate; PSU: polysulfone; PPO: polyphenylene oxide; PPE: polyphenylene ether
Polystyrene and Styrene Copolymer Based Blends
For example, compared with normal impact resistant PS and ABS, PS modified with
ethylene-propylene-diene monomer (EPDM) has a weathering stability which is 5 to
20 times greater. ABS is used as an impact modifier additive, increasing Vicat softening
point, e.g., PVC, polymethylmethacrylate (PMMA), PC or thermoplastic polyurethanes
(TPU). In this connection rapid development in the recent years of polymer blends is of
particular importance. By their use is possible to correct the operation of the mixing and
reaction processes during the manufacture of polymer blends. Appropriate conditions of
blends manufacture hold great potential of the creation of new polymeric materials.
Compounding permits the preparation of copolymers and blends with the desired
properties by convenient choice within certain limits of the ratio of the elastomeric phase
and the hard phase.
PS is a rigid amorphous thermoplastic, with a low price but for many applications it has
some deficiencies: relatively low impact strength, low glass transition temperature (Tg),
poor resistance to chemicals (soluble in many aromatic solvents) and physical agents
(degradation at ultraviolet radiation (UV)). Many efforts have been made to overcome
these problems by blending PS with other polymeric materials. PS containing blends
follow the general rules of polymer blends with thermal and mechanical properties of
the material depending on blend composition (chemical nature, structure and amount of
components), temperature, morphology (homogeneous, co-continues or droplet-matrix
phases, size of dispersed particles and interfacial tension) and preparation procedure
(melt mixing, solvent casting).
Several groups of the most known and studied PS and styrene copolymers containing
blends are described in the following sections, taking into account their relative importance
at this very moment. The most important are impact PS, SAN, ABS, acrylate-styrene-
acrylonitrile terpolymers (ASA).
It is well known that the addition of an elastomer to the glassy PS can improve the
impact properties of the material. The dispersed elastomer particles are able to initiate
the formation of high densities of crazes (microscopic yielding) and shear bonds (phases
of step at 45 to the direction of stress), enhancing the energy absorption (toughness) of
the material and preventing the formation of cracks (macroscopically fracture of the
material). Also the rubbery particles are craze terminators, preventing the growth of
very large crazes. Thus, the crack propagation in the material is slowed down [3].
For best impact strength the amount of elastomer added should be about 5%-20% and
the size of the rubber particles (droplets) should be in the range of 0.5-10 m but tensile
123
Handbook of Polymer Blends and Composites
strength, softening point and modulus decreases [3-5]. Too small elastomer particles
(< 0.5 mm) dispersed in PS matrix are ineffective craze terminators and therefore do not
provide optimum toughness [6, 7].
A wide range of elastomers can be considered for blending with PS, their cost being an
important factor in the choice. The most used elastomers are PB [7-9], polyisoprene
(PIR) [8, 10], ethylene-propylene rubbers (EPR) [6, 11].
The PS/elastomer blends have an upper critical solution temperature (UCST) behaviour,
with phase separation when the temperature is decreased [12-15].
For a blend of low molecular weight (LMW) PS and PB (weight average molecular weight
(Mw) ~ 2400) the critical temperature and critical composition for miscibility are 150 C
and 50 wt.% PS, respectively, increasing with the Mw of the components, while replacing
PB with a styrene-butadiene random or block copolymer (PS/styrene-butadiene (SBR)
random copolymer, PS/poly(styrene-b-butadiene) copolymer (S-b-B)) the critical
composition is shifted towards higher PS amount of 70-80 wt.% [15].
At room temperature, PS/elastomer blends with the elastomer as minor component, are
inhomogeneous systems, with a phase separated morphology, in which the rubbery
component is dispersed as discrete spherical droplets having a diameter less than 20 m
(depending on blend composition) in the PS matrix [4-6, 8, 11, 16-18].
Within the larger PB droplets, submicron sized microinclusions of PS phase were observed
in a PS (matrix)/PB (dispersed phase) blend [9, 16]. Co-continuous morphology was
observed in a PS/Natural Rubber (NR) blend for 40-60 wt.% NR [5].
Cavanaugh and Nauman [8] have studied the particle coarsening mechanism produced
by compositional quenching (a homogeneous mixture of the dissolved components is
heated and flashed into a vacuum chamber at a pressure of approximately 665 Pa), in
binary blends of PS with different rubbers (PB, PIR, SBR) and PS/PB/S-b-B ternary blend
(23 vol% rubber, 8 wt.% S-b-B) concluding, on the basis of experimental results, that
the Brownian coalescence model of particle coarsening (Binder-Staufer mechanism)
describes the particle size distribution better at long ripening times while the bulk diffusion
model (Lifshitz-Slyozov-Wagner mechanism) provides a reasonable fit at short ripening
times. Both Brownian coalescence and bulk diffusion mechanism show that the average
particle radius grows with time (t) as t1/3. For the ternary blends, the particle growth rate
exhibits a lower dependence on time (as t0.14), which can be interpreted as an interfacial-
agent action of the S-b-B copolymer in PS/PB blends.
The droplets of uncrosslinked elastomer in PS matrix can grow upon moulding by Ostwald
ripening or by coalescence mechanism and the material will be properly toughening only
124
Polystyrene and Styrene Copolymer Based Blends
when the particles have ripened to the optimum size. The morphology of the blend (the
dimensions of the dispersed phase) can be controlled by the moulding time [4].
For a PS/NR blend, it was found that solvent casting gives better overall mechanical
properties than a melt mixing procedure. Thus by solution casting, the tensile impact
strength varies from 8-34 kJ/m depending on the solvent and blend composition while
by melt mixing these values are in range of 1.4-3.7 kJ/m. Minimum polydispersity (P)
and domain size can be obtained for any blend composition by solvent casting from
benzene when compared with carbon tetrachloride (CCl4) and chloroform (CHCl3) [5].
The interaction energy for PS/PB pair is about 2.5-3.1 J/cm3, increasing with the Mw of
components [12].
The PS/elastomer blends are usually compatibilised with various types of styrene-containing
copolymers. Adding a copolymer as compatibilising agent for a blend, the phase interface
becomes saturated at a critical concentration of copolymer (Ccrit), above which no further
decrease in the average particle size of dispersed phase is observed. Thus, in an 80/20 volume
ratio (v/v) PS/EPR blend (PS: number average molecular weight (Mn) = 125,000,
polydispersity (P) = 2.20; EPR: Mn = 69,000, P = 2.51) mixed in a Brabender mixer, the
average diameter of EPR dispersed particles in PS matrix decreases from 1.8 m to 0.5 m
when adding up to 0.04 g/cm3 (weight/volume of blend) of polystyrene-b-hydrogenated
polybutadiene (PS-b-HPB) block copolymer (Mn = 67,000, 30/Wt% PS) and remains almost
constant at higher concentrations of added copolymer [6].
The mechanical properties are not affected by the moulding process for commercial
HIPS (elastomer modified PS), as was proved by Cavanaugh and co-workers, the impact
strength (80 J/m) remaining constant over long moulding times (125 min) but they can
be improved for proper compatibilised PS/PB (Mn = 82000 (P = 2.4)/Mn = 130000 (P = 2.4);
67/23 v/v) blend. High molecular weight (HMW) poly(styrene-b-ethylene) (S-b-E)
copolymer was found to be the most effective in the improvement of impact strength of
the PS/PB blend, being capable of chain-entanglement in both homopolymer phases.
Thus a blend compatibilised with 8 wt.% S-b-E diblock-copolymer (Mn = 148000 (P = 1.2),
MnPS = 35000) has an impact strength comparative to HIPS, but by moulding the blend
for appreciatively 125 minutes, the impact strength can be increased about 3 times (315 J/m),
further moulding decreasing the impact strength. In this process, the impact strength
increases approximately linearly with the rubber particle diameter (from 0.45 to 0.85 m).
A shorter S-b-E diblock-copolymer (Mn = 31000 (P = 1.1), MnPS = 7000) improves the
interfacial adhesion, but the impact strength is only of 36 J/m, similar to the
uncompatibilised blends, due to the chain pull-out of the shorter PS block from the PS
phase. Styrene-ethylene random copolymers (SE) are less effective than block-copolymers,
however maximum improvement of the impact strength (129 J/m) was obtained for a
125
Handbook of Polymer Blends and Composites
random copolymer with symmetric composition (Mn = 140000 (P = 2.9), 48% bound
PS) [4].
For PS/EPR (Mn = 125000 (P = 2.2)/Mn = 69000 (P = 2.5); 80/20 v/v; melt blending at
200 C) blend, no major differences were found between emulsifying effect of poly(styrene-
b-(ethylene-co-butylene)) copolymer (S-b-EB) and poly(styrene-b-(ethylene-co-propylene))
copolymer (S-b-EP). In the very high Mw range of these interfacial agents there is little
effect of the Mw of the modifier on interfacial tension reduction. Using LMW S-b-EB
diblock-copolymer (Mn = 67000, 30% PS) and poly(styrene-b-(ethylene-co-butylene)-b-
styrene) (S-b-EB-b-S) triblock-copolymer (Kraton G 1652, Mn = 50000, 30% PS) a brittle
to ductile transition is observed in the Charpy test when the copolymer concentration
reaches the Ccrit value in the range of 3-4 vol%, while for HMW S-b-EB (Mn = 187 000,
26% PS) and S-b-EP (Kraton G 1702, Mn = 140 000, 27% PS) only brittle fracture
behaviour is observed which persists above the corresponding Ccrit of 2 vol%, despite
their excellent emulsification capability. The brittle to ductile transition is attributed to
the high number of joints (0.18 and 0.15 joints/nm2, respectively), at interface exhibited
by the LMW S-b-EB diblock and S-b-EB-b-S triblock-copolymers, respectively, since the
particle dispersed size for all compatibilisers are identical. Therefore in these systems the
number of joints across the interface and the saturation of the interface are the most
important parameters for the compatibilisation. The best option is to use copolymers
with the Mw high enough to entangle in both phases of the blend, but low enough to have
a significantly high area density at the interface. Lower Mw of the copolymers will also
decrease the possibility of micellar aggregation of the copolymer below Ccrit value [11].
126
Polystyrene and Styrene Copolymer Based Blends
In blends of LMW PS (Mn = 2200 (P = 1.07)) and PIR (Mn = 15000 (P = 1.08)), the
symmetric poly(styrene-b-isoprene) (S-b-I) (Mn = 141000 (P = 1.07), 59% PS) diblock-
copolymer and styrene-rich one S-b-I (Mn = 162000 (P = 1.02), 78% PS) have similar
efficiencies as compatibilisers, but lower than isoprene-rich S-b-I (Mn = 210000 (P = 1.15),
15% PS) that decreases the interfacial tension and shows smaller domain size and narrower
particles distribution. At 145 C the interfacial tension of 0.9 mN/m of PS/PIR blend
decreases when adding S-b-I. The decrease is stronger with increasing the PIR amount in
copolymer and levels off at 0.3 mN/m for 0.5 wt.% isoprene-rich S-b-I [17].
TC( f ) TC( f = 0)
= tC
TC( f = 0)
was found to decrease linearly with increasing diblock-copolymer content (f) below
12.5 wt.% and shifts the critical composition towards low PSD content. This is consistent
with the expected compatibilisation effects of adding copolymers into the binary polymer
blend. Also, it was found that the kinetics of the phase separation follows the Cahn-
Hillard-Cook model at early separation stages of the spinodal decomposition (SD), the
growth rates of the spinodal peak and position slowing down for all stages in the presence
of diblock-copolymer. In the intermediate and late stages of the SD, the scaling relationship
between spinodal peak intensity-position and time have been well established for binary
blends, with peak intensity Im t, peak position qm t- and the ratio between the two
empirical constants / 3 for the late stage of the SD. Experimentally, for most polymeric
blends in the intermediate stage were found the values of = 1 and = 1/3. For the PSD/
PB/PSD-PB system the exponent varies from 0.5 to 2.3 and varies from 0.2 to 1.1 with
no obvious scaling relationship between and . The time dependence of the peak
127
Handbook of Polymer Blends and Composites
position, qm, and the scattered intensity maximum Im, cannot be scaled into a universal
function as expected in binary blends without block-copolymer added, so the time
dependence of the concentration fluctuation growth do not follow a universal function
form with diblock-copolymer added [13].
Compatible blends can also be obtained by solving the elastomer in styrene monomer
followed by usual polymerisation of styrene [7]. Thus is possible to graft the elastomer
particles onto the PS chain. These grafted chains enhance the interfacial bonding between
the elastomer phase and the PS matrix. PS sub-inclusions in the elastomer droplets were
observed for a PS/PB blend containing 36 vol% PB. Increasing the degree of grafting, the
sub-inclusions becomes larger and the elastomer/sub-inclusion interface becomes obscure
(Figure 4.1).
For blends containing 36 to 47 vol% PB the yield stress decreases when the degree of
graft (degree of graft (%) = 100 (gel%-PBD%)/PBD%; where gel% is the weight fraction
of the methyl-ethyl-ketone non-soluble part in the sample and PBD% is the weight fraction
of polybutadiene in the sample) increases up to about 120% then increases. The blends
with larger elastomer particles have lower yield stress (the minimum value of yield stress
for 46 vol% PB blend is around 14 MPa, this value increasing to 20 MPa for 28 vol% PB
Figure 4.1 Transmission electron micrographs of the fracture surface for a styrene
monomer/PB mixture polymerised in situ. With increasing degree of graft (from left to
right) the PS-subinclusions in the rubber particles becomes larger and the
subinclusion/rubber interface becomes obscure [7]
Reprinted with permission from H.H. Choi, K.H. Ahn and S.Y. Chim, Polymer, 2000, 41,
5229. Copyright 2000, Elsevier Science Ltd.
128
Polystyrene and Styrene Copolymer Based Blends
blend). The decrease of the yield stress arises from the increase in the effective volume
fraction of elastomer phase and/or from the increase in the stress transferred to the
rubber particles while the decrease is due to the larger occlusions, which are not enclosed
by rubber layers and do not act as good craze initiators [7].
These blends combine the low mould shrinkage (the shrinkage of the material as it cools
after moulding) and great hardness of the thermoplastic amorphous PS with the chemical
resistance, good ductility and stress crack resistance of the semicrystalline polyolefins (PO).
PS/PE and PS/PP blends are typically immiscible systems [19-25] with mainly droplet-
matrix morphology (with diameters ranging from 1 to about 50 m depending on the
blend composition) and a sharp interface between the phases [19, 20, 22, 26, 27]. PS has
a higher viscosity than PE therefore the stresses acting on the minor phase is higher in
PS-rich than in PE-rich PS/PE blends and causes an easier break down of the minor phase
particle in melt blending process resulting in better phase dispersion [24].
The blends containing LMW PS shows a network-like structure with particle sizes of about
40 nm, which is believed to give more effective interactions between the phases [29].
The blend morphology depends on the composition, the blends with extreme PS contents
showing mainly droplet-matrix morphology while for some blends with PS content from
27 up to 78 vol% even co-continuous morphologies have been reported [28, 31, 32].
The blends with co-continuous morphology show synergistic effects, with isotropic
Youngs moduli values, which can exceed the prediction of those calculated from additivity
rule from the values of pure components over a wide composition range. Thus, in a
blend of 30% PS in PE, changing the dropletmatrix structure into a co-continuous
morphology will raise the modulus from approximately 400 MPa to approximately
750 MPa, resulting in an almost doubling of the stiffness of the material [31].
129
Handbook of Polymer Blends and Composites
the expressions for dynamic shear moduli G and G as a function of the linear viscoelastic
behaviour of the phases, the size distribution of the dispersed droplets and the interfacial
tension) by fitting the rheological experimental data with the theoretical curves, the
value obtained being about 5.2 mN/m.
The conventional melt mixing PS/PE blends combine over a broad composition range
the low ultimate tensile strength (B) of PE and the low elongation at break (B) of PS,
having an B, yield stress (Y) and B smaller than that of either separated components
[25]. This is due mainly to the poor interfacial adhesion between the two immiscible
components. Upon deformation, the two phases debond from one another, which lowers
the load-bearing cross-sectional area. The dispersed phase domains act as stress
concentrators, further weakening the overall material [32]. However, a synergistic
behaviour for a high density polyethylene (HDPE)/HIPS 8/2 weight ratio (wt./wt.) blend
was observed at temperatures between 150 C and 0 C, the shear modulus being higher
than that of each individual component due to the higher thermal expansion coefficient
for HDPE when compared to HIPS. Upon cooling, the HDPE matrix phase will shrink
more rapidly than the HIPS phase and the dispersed PS spheres will be compressed. Thus
strong mechanical interactions between the two phases are developed [24].
Having poor mechanical properties, the immiscible PS/PE or PS/PP blends are required
to be compatibilised to obtain blends with improved mechanical and thermal properties.
130
Polystyrene and Styrene Copolymer Based Blends
For a PS/2.5 mol% acrylic-modified low density polyethylene (acid-LDPE) blend the
degree of miscibility was slightly increased by the presence of acid groups in the LDPE
chain, which can interact with the electron-rich phenyl rings of PS through quasi acid-
base complexes. The dispersed phase is smaller than the PS/LDPE blends therefore the
acid groups decrease the PS/LDPE interfacial tension. For all composition ranges of the
PS/acid-LDPE blends the deviation of the storage moduli from the additivity level is
smaller than for PS/LDPE blends [20].
A new method has been developed to compatibilise PS/LDPE blends by a two-step cross-
linking process [21]. First LDPE is partially crosslinked in melt at high temperatures by
adding a small amount (0.05 wt.% from total PS/LDPE weight) of dicumyl peroxide
(DCP) and mixing for 5 min at 165 C and 30 rpm roller blades in a Haake Rheocord
mixer. The crosslinked LDPE was pelletised and blended with PS (PS/LDPE 80/20 wt./wt.)
at 165 C then adding poly(styrene-b-butadiene-b-styrene) diblock copolymer (S-b-B-b-S)
(10 wt.% from total PS/LDPE weight) after 5 minutes of mixing and further mixing for
another 5 minutes. S-b-B-b-S acts as strong coupling agent between the PS and LDPE
phases by crosslinking the residual free radicals in LDPE phase onto its butadiene
component and intermixing its styrene component with the PS matrix. These reactions
enhance the adhesion between the PS and LDPE phase, leading to significant improvement
of the mechanical properties of the blend (impact strength of 115.7 J/m which is higher
than some commercial HIPS, tensile strength at break of 35 MPa and B of 42%) [21].
One of the most frequently compatibilising method uses diblock and triblock copolymers
such as PS-b-PHB [24, 25], (S-b-B-b-S) [35], S-b-EB and S-b-EB-b-S [19, 23], S-b-EP and
[19, 36], poly(styrene-b-(ethylene-co-propylene)-b-styrene) (S-b-EP-b-S) [37], block or
graft PS-PE copolymers [25], which are added in small amounts of 1-5 wt.% (up to Ccrit)
[24, 25], as compatibilising agents to PS/PO blends (Table 4.2).
When the interface is saturated, the copolymer in excess either randomly disperses in
one phase or aggregates to form micelles (at concentrations above the critical micelle
concentration (CMC)), fibres or discrete droplets depending on the copolymer
concentration (Figure 4.2) [24, 38].
The blends of styrene acrylonitrile random copolymer with other polymers are miscible
only in a certain copolymer composition (window miscibility) and temperature range,
when there is a strong repulsive interaction between comonomer units of copolymer,
though all segmental interactions are positive [42-44].
131
PS/PE blends
Handbook of Polymer Blends and Composites PS/PP blends
Table 4.2 Compatibilised PS/PE and PS/PP blends using various styrene
copolymers
Type of blend Properties Ref.
PS/LLDPE - decreased dispersed particle size from 9-12 [26]
(GI5026, Novacor/Polymers to 2.1-2.3 m;
Scott Ltd.; 70/30 wt./wt.; melt - decreased interfacial tension from 5.2 to
mixing, 180 C, 50 rpm); 1 mN/m;
Compatibiliser: S-b-EB-b-S - unmodified interfacial tension/dispersed
(Kraton G 1650) 1 wt.%; particle radius ratio;
PS/LLDPE and PS/HDPE - in 70/30 PS/HDPE blend S-b-E and [39]
Compatibiliser: S-b-EB-b-S are more efficient than S-EP in
S-E, S-b-EB-b-S, S-b-EP; decreasing particle dispersed size, while in
30/70 blends all copolymers have
comparable emulsifying effect;
- in 30/70 PS/LDPE all copolymers exhibit
little effect on particle dispersed size;
- 70/30 PS/LDPE blend exhibits co-
continuous morphology which does not
change with 1 or 2 wt.% copolymer
addition;
PS/LLDPE - both copolymers (especially S-b-EB-b-S) [19]
(Styron 680, Dow Chemical tend to interlock the dispersed particles
Co./SCLAIR 2107, DuPont with the matrix forming a co-continuous
Co.; interlocking (interpenetrating) network, in
Haake batch mixer with the which a continuous phase fills the voids in
roller blades 180 C, 50 rpm); the continuous other phase;
- S-b-EB-b-S more efficient compatibiliser
Compatibiliser: than S-b-EP;
S-b-EB-b-S (Kraton G1652); - the domain size exhibits a minimum value
S-b-EP: (KratonG1702); in respect of mixing time, when is achieved
an equilibrium between break-up and
coalescence;
- a minimum droplet size and a more
uniform size distribution are obtained when
the compatibiliser is added after the PS
dispersed phase breaks up completely
within the LLDPE matrix (after 5 min of
mixing PS in LDPE);
132
PS/PE blends
PS/PP blends Polystyrene and Styrene Copolymer Based Blends
133
PS/PE blends
Handbook of Polymer Blends and Composites PS/PP blends
134
PS/PE blends
PS/PP blends Polystyrene and Styrene Copolymer Based Blends
For a blend containing a homopolymer (A) and a random copolymer with B and C
monomer units (BC) containing an mole fraction of B component, the interaction
parameter can be evaluated with a simple expression that considers the segmental
interactions parameters AB, AC, and BC. Thus [43, 44]:
blend = AB + AC (1-)-(1-) BC
Negative interaction parameter (blend) can be obtained for certain range of copolymer
compositions even all individual interaction parameters between the different segments
135
Handbook of Polymer Blends and Composites
are positive. This can occur if the mutual repulsion between the constituents of the
copolymer (B and C units) prevails over the interactions between either of these units
and the homopolymer units (copolymer effect: 0<AB< BC and 0<AC< BC) [44]. However,
other authors [43] suggested that the miscibility is not a consequence of the copolymer
repulsion effect but is rather due to specific interactions between the copolymer and
homopolymer units.
Recently Woo and co-workers proposed a new explanation for the miscibility phenomena
based on intermolecular forces that can appear between the chain segments of the
copolymer and those of counter homopolymer [42]. Thus miscibility can appear if the
polarity of chain segments in a SAN copolymer (which is a statistical average of the
polarity of the components) is close to that of the chain segments in the other component
of the blend. Similar polarities occur only for certain composition of SAN that explain
the existence of a miscibility window.
SAN and PC belong to the class of partly miscible polymers over the entire range of SAN/
PC compositions, when acrylonitrile (ACN) content in SAN is in the range of 23-70 wt.%
forming heterogeneous blends with a good interface adhesion, which ensures a good stress
transmission between constituents during deformation up to fracture [45, 46].
In SAN/PC blends the interfacial thickness and the interaction parameter (SAN/PC) depends
on both the temperature and ACN content of SAN. For blends of SAN containing 11.3,
136
Polystyrene and Styrene Copolymer Based Blends
25, 40, 50 or 59.3 wt.% ACN (Mw = 269000, 128000, 75800, 63600, 49500; P = 2.1,
2.1, 1.9, 1.8, 1.8, respectively, Mitsubishi Monsanto) with HMW and LMW PC (Mw =
37500, P = 1.4 and Mw = 6000, P = 1.3, respectively, Mitsubishi Gas Chemicals) the
interfacial thickness has a maximum value and the interaction parameter (SAN/PC) has a
minimum value at a temperature of about 200 C and a 30% to 40% ACN content of
SAN. This behaviour suggests the coexistent behaviour of UCST and lower critical solution
temperature (LCST) in the phase diagram of the mixture consisting of polymers with
suitable Mw. From the dependences SAN/PC with temperature and ACN and considering
the condition for the existence of a single phase that SAN/PC to be smaller than crit (the
value of the interaction parameter at the critical point) one can draw the miscibility
diagram for PC/SAN blends (Figure 4.3).
A loop-type miscibility region, e.g., a miscibility egg, appears in the SAN/PC system instead
of the usual miscibility window. crit decreases when increasing the Mw of SAN and PC
therefore at certain Mw the miscibility egg disappears and the blend is totally immiscible.
The phase behaviour of SAN/PC blends determined by cloud point measurements shows
an hour-glass type phase behaviour appearing for blends containing SAN with an ACN
Figure 4.3 Miscibility diagram of the PC/SAN system for two crit values: (- - -) 0.003,
() 0.004 [47]
Reproduced with permission from H. Li, Y. Yang, R. Fujitsuka, T. Ougizawa and T. Inoue,
Polymer, 1996, 40, 927, Figure 5. Copyright 1996, Elsevier Science
137
Handbook of Polymer Blends and Composites
amount of 11% and an UCST/LCST behaviour for 25% and 40% ACN in SAN, in
accordance with the measurements of the interaction parameter (Figure 4.4) [47].
Figure 4.4 Phase diagram of the PC/SAN(25 wt% ACN) mixture measured by the
cloud point method. (X) Cloud samples; () transparent samples [47]
Reprinted with permission from H. Li, Y. Yang, R. Fujitsuka, T. Ougizawa and T. Inoue,
Polymer, 1999, 40, 927, Figure 8. Copyright 1999, Elsevier Science
138
Polystyrene and Styrene Copolymer Based Blends
The size and shape of PC and SAN-rich domains depends on blend composition. The
two-phase domain structure appears to be coarser for the 60/40; 50/50 and 70/30 samples
while the structure is finer for the 80/20; 20/80; 30/70 and 40/60 PC/SAN blends [45] in
agreement with other authors results [48].
The mean diameter of domains reaches a maximum value of about 1.1 m in 50/50
blends while for extreme compositions the values are of about 0.45-0.50 m [45].
Schmidt and co-workers reported that the phase inversion interval, is in the range of 50-
60 wt.% PC/SAN composition [45], while Lednicky and co-workers reported a range of
20%-80% PC [48]. In the co-continuous morphology both phases contain particles of
both components of the blend [48]. In the PC/polystyrene acrylonitrile, 60/40 blends the
PC-rich and SAN-rich large domains were found containing 83% and 16% of PC,
respectively, [45].
Owing to very good interfacial adhesion in the PC/SAN blends, the plastic deformation
of both conjugate phases proceeds simultaneously. On the other hand, the orientation of
PC chains is much higher than that of the SAN chains. The latter effect can be attributed
to a high flexibility of the PC backbone, whereas the SAN chains are rather stiff [45].
During straining, pure PC exhibited plastic deformation and can be drawn up to about
170% while the attained elongation decreased when increasing the SAN percentage in
blend: the PC/SAN 30/70 specimen broke after yielding, the 20/80 blend and SAN is
brittle [45].
SAN/PMMA blends are miscible for 9-33 wt.% ACN in SAN [44, 50], with LCST
behaviour [51-53]. The critical composition was found to be in the range of 40%-60%
SAN content, with a critical temperature of about 195 C [51, 52].
These blends are characterised by dynamic symmetry, i.e., the Tg and the Vogel
temperatures of pure components are closely matched, leading to thermorheological
simplicity (time-temperature superposition (tTS) principle) for temperatures with 30 C
higher than Tg, due to the fact that composition fluctuations do not create regions with
appreciable different local dynamics [50]. However, thermorheological complexity can
be observed closer to the Tg of the blends, at high frequencies. Also the tTS principle can
139
Handbook of Polymer Blends and Composites
be applied in the miscible region and but not near the phase separation temperature
when the master curves of the storage modulus (G) shows a section of plateau, associated
with a structure formation in the blend [53].
SAN/SMA blends have higher miscibility, with narrower breadth of the Tg region and
highest volume contraction on mixing when ACN and maleic anhydride (MA) comonomer
contents in SAN and SMA, respectively, are not too different [55-57]. This behaviour is
associated mainly with a decrease of intramolecular repulsions upon mixing, opposed to
specific exothermic binary interactions [58].
Blends of amorphous SAN with a semi-crystalline polymer (PCL) are miscible for
8-28 wt.% ACN content in SAN [59, 60] exhibiting LCST behaviour with critical point
at 20 wt.% SAN composition and 100 C critical temperature (PCL: Mw = 40400 (P =
2.61); SAN 27.5 wt.% ACN, Mw = 16900 (P = 2.09)) [60].
The Tg increases with SAN concentration in the miscible SAN/PCL blends [59] while the
melting temperature of crystalline fraction of PCL was found to be 60 C for all
composition range in the blends [60].
The crystallisation of PCL occurs more rapidly when casting from solution than in melt
mixing [59]. For solution cast blends the crystallisation of PCL (Mw = 65000, Tg = 65 C,
Aldrich) develops at high concentrations of about 40 wt.% but there is always an
appreciable amount of PCL remaining amorphous and mixed with SAN (25 wt.% ACN,
Mw = 16500, Tg = 108 C) [59]. The degree of crystallisation in as-cast films remains
almost constant (50%) regardless of the blend composition but in the recrystallised films
it decreases with increasing SAN content and apparently drops to zero for blends
140
Polystyrene and Styrene Copolymer Based Blends
containing more than 40 wt.% SAN. Also, the as-cast films have higher Tg than the
recrystallised films, showing the greater crystallinity of PCL and smaller amount of the
amorphous PCL mixed with SAN [59].
Due to the miscibility of system, the crystallisation rate of PCL decreases with increasing
SAN content in blend [61].
The effect of SD on the crystallisation of PCL was also studied. By heating the blends
above LCST then cooling below the melting point of PCL (40 C) the longer the SD time,
the higher rate of crystallisation is, with an optimum SD time of 5 minutes yielding the
highest degree of crystallinity and melting temperature in the early stage of SD. In this
process, SAN (27.5 wt.% ACN, Mw = 169000 (P = 2.09)) segregates from PCL (Mw = 40400
(P = 2.61)) regions this being favourable for crystallisation because the PCL-rich regions
will have higher chain mobility due the exclusion of SAN having high Tg [60].
PS/PVME blends are miscible with single Tg and reversible LCST behaviour [62-70]
(Figure 4.5).
The critical composition is in the range of 20-40 wt.% PS content and the critical
temperature in the range of 95-130 C, depending on the molecular mass of the
components [64, 66-69]. When the molecular mass of PS is larger than that of PVME,
the critical composition is located at PVME-rich content [66].
In the early stages of the phase separation process, the composition of each phase changes
continuously with time but the fractions of the two phases remain unchanged. For a 50/50
PS/PVME (MV = 93000, Dow/MV = 87000, Aldrich) blend the ratio of PVME-rich domains
to PS-rich domains has constant value of about 0.5. In the first 2 minutes the PS content
of the PVME-rich domains decreases gradually from 50 to 20% while the PS content of
the PS-rich domains increases from 50 to 80%. These results clearly indicate that phase
separation is initiated by SD [62, 65]. After more than 2 minutes the composition of the
phases remains almost constant to 17/83 for the PVME-rich domains and 85/15 for the
PS-rich domains, close to the binodal composition [62].
The total reduction Q of the amount of a minor component in a given phase can be
described as:
ln Q = constant + 3 R t
141
Handbook of Polymer Blends and Composites
Figure 4.5 Phase diagram of PS/PVME blend as determinated by shear rheology from
inflection point in the G versus temperature curve (Trheo-o), by turbidity measurements
(Tcloud-x), and from intercept with 1/T axis of tangent (at inflection point) at {(G)2 /
(G T)}2/3 versus 1/T curve (Tspinodal-). Also is depicted the glass transition temperature
of the blend (Tg-) [68]
Reprinted with permission from Macromolecules, 1996, 29, 22, 7155, Figure 7. Copyright
1996, American Chemical Society
where R is the phase separation rate and t is the time of treatment by heating. A positive
value of R is characteristic for SD while negative values of R correspond to nucleation and
growth processes [71]. From the initial slope of ln Q versus t plot, Asano and co-workers
found an R = 0.5 min-1 value at 140C for both PVME and PS-rich domains [62].
When the molecular mass of PS is much higher than that of PVME the onset of the phase
separation process after a temperature jump is delayed to a longer time due to the
formation of a temporary network of the well-entangled polymers which need to relax
before the phase separation occurs. There is no difference in the onset of phase separation
process for linear or star-shaped low molecular PS [72].
The PS/PVME blends are miscible on a scale of 20-30 at temperatures lower than Tg of
the blend [62] due to the specific interactions between PS phenyl rings and the PVME
methyl groups [73].
142
Polystyrene and Styrene Copolymer Based Blends
From ESR studies on PS-labelled/PVME blends, Cameron and co-workers [63] have studied
the plasticising influence of the PVME on PS and showed that the appearance of the fast-
motion component in the ESR spectra is attributable to the onset of phase separation.
Comparing the temperature at which the fast-motion spectrum is first detected (Tf) in the
ESR spectra with the phase-separation or cloud-point temperatures (Tcloud) taken from
literature for comparable blends, they conclude that Tf and Tcloud show the same trends
with composition, but in all cases Tf is located about 65 C below Tcloud. This difference
highlights the sensitivity of the ESR technique in responding to compositional heterogeneities
on a molecular level before macroscopic phase separation occurs.
In the homogeneous region tTS principle holds on for both storage modulus (G) and
loss modulus (G) and typical low frequency () behaviour is observed (G 2 and
G ) [67].
Increasing the temperature above the Tg of the blend (T>Tg) the chain mobility increases
resulting in a decrease of G. As the LCST is approached and the concentration fluctuations
increase in amplitude, the formation of dynamic rich-PS domains having high modulus
occurs, leading to an increase of G. This upturn in G(T) curve (inflection point)
(Figure 4.6) can be assigned to the phase separation temperature. The rheological demixing
temperatures (Trheo) determined in this way are in very good agreement with those obtained
from turbidity measurements, Tcloud (Figure 4.5) [68, 69].
Assuming that near the phase boundary the interaction parameter is given by = A+B/T,
a linear dependence of {(G)2 / (G T)}2/3 versus 1/T curve was predicted, that leads to an
intercept with the 1/T axis denoting the spinodal temperature Tspinodal.(Figure 4.5) [68,
69]. This upturn in viscoelastic properties is related to the dynamic asymmetry of the
two blend components [68, 69]. Above Trheo the mobility effects control again the
viscoelastic behaviour of the phase separated blends (G decrease with increasing
temperature) [68, 69].
In the phase separated region, a regime of enhanced moduli, for the terminal slopes is
observed that is attributed to the PS-rich domains. The linear viscoelastic properties of
the phase separated PS/PVME blends can be, in a first approximation, adequately described
by a simple incompressible emulsion model considering a suspension of droplets of one
coexisting phase in the matrix of the other phase [68].
143
Handbook of Polymer Blends and Composites
Figure 4.6 Temperature dependence of G at = 0.5 rad/s for the 30/70 (___),
50/50 ( ) and 10/90 (.......) PS/PVME blends. The onset temperature for the
turbidity, is marked as the arrow [69]
Reprinted with permission from K.K. Kim and H.W. Son, Polymer, 1999, 40, 6789.
Copyright 1999, Elsevier Science Ltd
For the critical composition, log G versus log G plots show T dependence even in the
homogeneous region and become T independent in the homogeneous region for off-
critical compositions, due to the fact that the oscillatory shear-induced phaseseparation
is most evident for the critical composition [69].
144
Polystyrene and Styrene Copolymer Based Blends
For the blends of semicrystalline sPS with PVME, prepared by casting and precipitation
from solution, the phase behaviour and the morphology of the sPS/PVME and aPS/
PVME system is similarly independent of mixing method. By quenching the blends well
above the LCST temperature and slow cooling, the phase-separated morphology can be
reversed to the original state. Also the Tg dependence on blend composition shows that
the intermolecular forces between components in the PVME-rich blends are generally
lower than those in PS-rich blends. Thus, the less aggregation of PVME molecules in PS-
rich blends leads to a single Tg whose variation with blend composition approximates
better the intermolecular mixing, closer to prediction of the Fox and Gordon-Taylor
equations. In contrast, in PVME-rich blends the significant aggregation of PVME
molecules leads to a Tg that is relatively insensitive to the presence of minor sPS
component [70].
PS/PPO blends are miscible at the molecular level for any composition range [75] and
have a single Tg which increases with the increased PPO content of the blend [75-79].
The interaction parameter S/PPO was found to be of -0.52 [80]. These blends have excellent
melt flow and good dimensional stability, the presence of PS improving the low
processability of PPO (caused by high Tg = 200 C and Tm = 255 C) [74, 76].
In a blend of glass fibre reinforced syndiotactic PS (GFRsPS), 15 wt.% glass fibre with
PPO (blending in a single screw extruder), the Izod impact strength increases from 7.9 to
9.7 kJ/m2, flexural modulus melt flow index increases from 2636 to 3102 MPa and
tensile modulus increases from 4.9 to 21.4 g/10 min, and elongation at break (eB) decreases
from 13.6% to 10% when the GFRsPS content in the blend is increased from 25% to
50% [75].
145
Handbook of Polymer Blends and Composites
The PS/PMMS blends are immiscible [81-84] with LCST behaviour [85] and droplets-
matrix type morphology [82-85].
The blends having the components with comparable Mw (MPS = 52000 and MPMMA = 60000,
50/50 wt./wt. composition) can exhibit relatively good degrees of mixing (dispersion in
the range of 5 m) when they are cast at 50 C from a nonselective solvent (chloroform
or benzene) without the use of compatibiliser. This dispersion coming from the weak
repulsion between components in solution, is maintained due to the high viscosity of the
components, in the solid state. For very low molecular components (Mw in the range of
4-6000) the lower degree of dispersion in solid state could be attributed to the elevated
mobility of the short chains [83, 84].
Upon rapid precipitation with methanol from toluene solution at room temperature, the
process can be considered similar to the early stage of SD via temperature quenching, the
blends (MwPS = 40-80000 and MwPMMA = 100-320000) exhibit a periodically modulated
co-continuous morphology (Figure 4.7). By subsequent isothermal annealing at 170 C,
process that can be assimilated to the later stage of SD, a dispersed morphology for the
asymmetric compositions and a dual mode of dispersed morphology (regions of PS
discrete phase dispersed in PMMA matrix and of PMMA discrete phase dispersed in PS
Figure 4.7 TEM images of rapidly precipitated 30/70 PS/PMMA blend showing a co-
continuous modulated morphology. The white and dark areas represent the PMMA
and PS phase, respectively [86]
Reprinted with permission from J.K. Lee and C.D. Han, Polymer, 1999, 40, 2521.
Copyright 1999, Elsevier Science Ltd
146
Polystyrene and Styrene Copolymer Based Blends
matrix) for the 50/50 blend is formed. The morphology development is slower in 70/30
PS/PMMA blends than in 30/70 PS/PMMA blends [86].
In PS/PET blends the high melt viscosity of PET decreases, thus increasing the processability
of the material [88]. PS/PET are immiscible blends with droplet-matrix morphology [89,
90] with poor adhesion between PS spherical droplets and PET matrix, droplets being pulled
out during a fracture test. The PET (Mw = 38000) droplets in PS matrix (Mw = 273500) do
not have a well-defined spherical shape.
Tensile strength at yield is lower than the additive values, with a minimum around 270 kg/cm2
for a PS/PET (Mw = 273500/Mw = 38000; 40/60 wt/wt; melt blending 280C, 30 minutes)
blend (compared with 550 kg/cm2 and 500 kg/cm2 for pure PS and PET, respectively).
The modulus (at 0.2% elongation) does not change significantly with composition (little
147
Handbook of Polymer Blends and Composites
higher - about 15 GPa - than pure components, at 60 wt.% PET composition). In the
reactive SMA/PET (7 wt.% MA, Mw = 236000/Mw = 38800; melt blending at 280 C,
30 minutes) system due to the reaction between ester groups of PET and MA of SMA,
the SMA droplets (at 30 wt.% SMA composition) dispersed in continuous PET matrix
have a more uniform and smaller dimensions comparatively with the non-reactive system
(PS/PET 30/70 wt./wt.). Also, PET droplets in SMA/PET 70/30 wt./wt., blend were not
observed. Tensile strength at yield and modulus (at 0.2% elongation) exhibit the same
trend as the non-reactive PS/PET blends, with smaller values (especially at low PET
content). No effect of reaction on mechanical properties was observed for a PET content
higher than 70 wt.%. Both of PS/PET and SMA/PET blends show two Tg which do not
change with composition. The melting temperatures of PET in two blend systems are
independent of blend composition, with no depression of melting temperature. In
PS/PET blends, the melting temperatures of PET were observed in the whole blend
compositions, whereas, in SMA/PET reactive blends melting temperature was not detected
above 70 wt.% SMA content [90].
Recently PS/PET composites were prepared by the polymerisation of the styrene in the
carbon dioxide-swollen PET matrix (at 80 C for 4 h). Thus, part of the PS molecules
entangle with PET molecules, the blends having lower apparent viscosity a (from
approximately 1.9-2.3 Ns/m2 in pure PET to approximately 1.0-1.4 Ns/m2 for 22 wt.%
PS; at T = 270 C) and higher mobility than pure PET. Also, the Youngs modulus, tensile
strength and elongation at break increase with PS content in the blend, from 1.7 GPa,
39 MPa and 40% in pure PET to 2.9 GPa, 68 MPa and 51%, respectively, in a 33/100 (g/g)
PS/PET blend. The PS molecules are well distributed in the PET matrix and they cannot
be extracted by liquid solvents [88].
Another way to compatibilise PS/PET blends is using SMA (18 wt.% MA) and
poly(methylene(phenylene isocyanate)) (PMPI) (31.5% NCO groups, average functional
groups per chain of 2.7, BASF) as dual compatibilisers, SMA being totally miscible with
PS and PMPI reacting with PET and SMA simultaneously to form PET-g-PMPI-g-SMA
at interface, which give finer phase domain (Figure 4.8), higher viscosity, lower interfacial
tension and stabilization against coalescence, due to the coupling reaction between
terminal groups (hydroxyl) of PET, maleic anhydride of SMA and isocyanate of PMPI at
interface.
By addition of 0.5 wt.% PMPI in a 20/70/5 PS/PET/SMA blend, the tensile strength,
tensile elongation and the unnotched Izod impact strength increase from 48 to 58 MPa,
from 5.5 to 45% and from 150 to 200 J/m, respectively. SMA copolymer alone cannot
compatibilise the PS/PET blends, the morphology of PS/PET/SMA blends being as coarse
as pure PS/PET blends [89].
148
Polystyrene and Styrene Copolymer Based Blends
(a) (b)
Figure 4.8 Cryogenically fractured surface of 25/75 PS/PET (a) and 23/75/2/0.5 PS/
PET/SMA/PMPI (b) blends. PS was removed by etching with chloroform [89]
Reprinted with permission from M-Y. Ju and F-C. Change, Polymer, 2000, 41, 1719.
Copyright 2000, Elsevier Science Ltd
Owing to the intrinsically different polarities of PS and PA these blends are immiscible,
showing a two phase structure, a dispersed particles-matrix morphology, with sharp
interfaces, poor interfacial adhesion and poor mechanical properties [91-95]. Two Tg
were found (38 C for PA and 117 C for PS) [92].
The cold crystallisation of PA takes place at higher temperatures as compared with the
pure PA, because the PS/PA interface promotes the crystallisation process and allows it
to start at higher temperatures. No differences in the melting temperature and heat of
fusion of PA were observed in blends compared with pure PA [93].
The tensile strength and notched impact strength decreases (from 67 to 30 MPa and
from 6.8 to 1.8 kJ/m2, respectively), upon increasing the PS content up to 50 wt.% in
PS/PA6(Mn = 16000/RV-67 DuPont) blend [92].
Many studies on the compatibilisation of PS/PA blends have been carried out with
functionalised styrene copolymers capable of reacting with amide groups in the PA phase
during melt mixing, resulting in an in situ formed compatibiliser having PA grafts. Thus,
in a SMA/PA66 reactive system, the coalescence is eliminated and the particle dispersed
149
Handbook of Polymer Blends and Composites
sizes remains constant (0.2 m) for 0.1-20 wt.% SMA concentration range for both of
1.5% and 17% MA functionalised copolymers, appearing that a very little reaction is
required to produce a significant effect on the morphology [95].
Since the PEO segment has the same crystallisation behaviour as PA6, and forms hydrogen
bonds with the amorphous PA6, there is better affinity between compatibiliser and PA6.
The repulsion between PS segment and PA6 makes the polystyrene-b-poly(ethylene-oxide)
(PS-b-PEO) diblock-copolymer to penetrate the PS phase. The balance between these
two trends results in the enrichment of PS-b-PEO diblock at the interface when the PEO
segment is longer enough to anchor in the PA6 phase [92].
A PS-b-PEO block with an appropriate Mw (Mn about of 70000) having the PEO block
longer than PS block (70 wt.% PEO) was found to increase the notched impact strength
and tensile strength in a PS/PA6 (Mn = 160000/RV-67 DuPont; 20/80; melt blending at
225 C and compression moulding at 220 C and 19 MPa) blend from 3.1 to
approximately 5 kJ/m2 and from 39.5 to approximately 44.5 MPa, respectively, for a
copolymer concentration about 2 wt.%, higher concentrations leading to the micellisation
of copolymer in the PA6 matrix and weakening the blends. Also, the sizes of PS particles
were decreased and the thickness of the interface increased [92].
150
Polystyrene and Styrene Copolymer Based Blends
16
Figure 4.9 Scanning electron micrographs of fracture surfaces at room temperature in a PS/
PA6 blend uncompatibilised (a) and compatibilised with 5 wt% PS-P(MMA-co-MAA) [91]
Reprinted with permission from R. Fayt and P. Tessie, Journal of Polymer Science,
Polymer Letters, 1989, 27, 481. Copyright 1989, John Wiley & Sons Inc
be about 3 wt.% copolymer added. The presence of the copolymer retarded the cold
crystallisation of PA at lower temperatures. The unnotched and notched Charpy impact
strength increased in the 80/20/3 PS/PA12/PS-g-PEO blends comparatively with pure
PS/PA12 blends (from 4.2 to 6.9 and from 2.6 to 3.3 MPa, respectively), while the
elongation at break and the stress at break remained unmodified (5.4%-5.5% and 31
MPa, respectively), [93]. Macrophase separation of the copolymers in the PS phase and
a microphase separated graft copolymer interphase was observed at PS/PA6 interface for
copolymer concentrations higher than 3 wt.% [94].
151
Handbook of Polymer Blends and Composites
Blends of lightly sulfonated PS (SPS) and PA6 are partially miscible due to the strong
exothermic interactions involving ion-dipole or metal complex interactions between the
sulfonate and amide groups [96]. Thus, 8.1 Zn-SPS/PA6, 9.1 Mn-SPS/PA6 (8.1 and 9.1-mol%
sulfonate concentration) compression moulded blends, 5.4 and 9.8 Li-SPS/PA6 solution
casting blends [97] were found to exhibit LCST behaviour. Increasing the ionomer
concentration decreases the PA crystallinity. Above 40% ionomer, the 9.1 Mn-SPS/PA6
blends are completely amorphous [96]. However, miscibility was found to depend on the
cation type. Thus 5.4 and 9.8 Na-SPS/PA6 solution casting blends are not miscible [97].
The PS/PC blends are partially miscible exhibiting two Tg temperatures, one corresponding
to PS which does not noticeably vary with composition and another corresponding to PC
which increase with increasing PC content in the blends [98]. The morphology is a two phase
structure [99]: a pure PS phase and a PC-rich phase containing small amounts of PS [98].
Reversible UCST behaviour was reported for Zn-SPS/PC blends (zinc sulfonated
polystyrene ionomers) [100] with the cloud point temperatures decreasing (from 220 to
approximately 205 C for the critical composition, 60 wt.% PC) with increasing the
sulfonation level (from 8.7 to 13.7%) for a fixed Mw of the components (MnPC = 48000,
MnZn-SPS = 4000) and increasing (from 265 to 300 C for the critical composition, 40 wt.%
PC) with the Mw of the ionomer (MnPC = 48000, MnZn-SPS = 135-280000) for a fixed
sulfonation level (5.9%).
For these blends two Tg were found in the two-phase region, which are shifted towards
each other when the temperature is increased above the UCST. The Tg-composition
behaviour in the miscible region exhibits negative deviation from linear weighted average
of the components Tg values, which is consistent with a miscible blend with weak or no
specific interpolymer interactions. The miscibility of these blends comes from intrapolymer
repulsive interactions between the ionic and non-ionic species of the ionomer (copolymer
effect) [100].
152
Polystyrene and Styrene Copolymer Based Blends
The PS/TMPC blends are miscible in whole composition range, exhibiting LCST behaviour
and one Tg which increases with the TMPC content in blend [101-104].
The cloud-point curve is weakly dependent on blend composition but more strongly
dependent on the Mw of PS. Decreasing the Mw of PS from Mw = 330000 (P = 3.3) to Mw
= 100000 (P = 1.05) in a 50/50 PS/TMPC (MPCw = 37900 (P = 2.7)) blend the cloud-
point temperature increase from approximately 240 to 265 C [101].
Thermodynamic reversibility was found for the 50/50 composition. Time resolved light
scattering experiments have been performed with a 50/50 PS/TMPC blend (Mw = 330000,
P = 3.3 / Mw = 37900, P = 2.7, solvent casting) which was first preheated at 230 C and
then rapidly heated to 250 C. A time invariant maximum in the wave vector dimension
and an exponential growth of the scattering intensity were observed during the first
200 s of phase separation, both results being characteristic for early stages of SD as
predicted by the Cahn-Hilliard theory [101].
Also, miscibility of sPS with TMPC was reported, the blends exhibiting one Tg fitting the
Gordon-Taylor equation for k = 0.6 which means that the intermolecular interactions
are not so strong to restrict the chain mobility. The binary interaction energy for a sPS/
TMPC (Mn = 218000, P = 2.15, Dow Chemical/Mn = 4300, P = 3.90) blend was found B
= 0.92 J/cm3. Increasing the TMPC content increase the supercooling necessary for the
crystallisation of sPS and decrease the heat of crystallisation of sPS [105].
Table 4.3 presents a summary of the properties of some miscellaneous blends containing
PS or polymers containing styrene units.
153
154
polymers
viscosity [107].
S-b-B-b-S/LLDPE/LLDPE-g-PS - Solubilisation of LLDPE and PS segments Good compatibilisation of [108]
of LLDPE-g-PS into amorphous LLDPE LLDPE/S-b-B-b-S blend with
- precipitation from solution and PS domanins of S-b-B-b-S, connection more than 3 wt.% LLDPE-
then annealed at 160 C, by isomorphism of LLDPE free-branching g-PS
30 min; chains in LLDPE-g-PS with LLDPE
crystalline phase.
- Half of the total PS blocks located in
LLDPE-g-PS-rich interface between
S-b-B-b-S and LLDPE domains.
miscellaneous blends Styrene
polymers
Table 4.3 Continued
Mixture/composition Properties Observations Ref.
Blending method
Chlorinated - Compatibilisation effect of MMA on Improved mechanical [109]
polyethylene(CPE)/Poly(acryl- CPE/PAS blend, improved rheological properties of PAS/CPE
onitrile-styrene-methyl behaviour. blends.
methacrylate)(PASM) - Minimum melt viscosity for 5% MMA and
25% CPE.
Mixing CPE and PASM, - Increased impact strength, decreased tensile
polymers
SBR filled NR waste - Not miscible rubbers; NR reinforces SBR Reuse of waste and used [115]
matrix. rubber into processable
- two rolls mixing mill, curing - minimal thermal conductivity at 10 phr NR. form.
(150 C) in a press. - Increasing NR amount increases sulfur
migration from SBR matrix to NR phase
with decreasing crosslinking density of the
composite and increasing thermal
conductivity.
miscellaneous blends Styrene
Table 4.3 Continued
polymers
Mixture/composition Properties Observations Ref.
Blending method
SBR/NR - Immiscible blends, two phases structure, Good processability crack [116]
two Tg, co-continuous morphology at 1/1 and weather resistance,
Latex blending; NR latex with composition. preserving NR
60%dry rubber, SBR latex - mechanical properties improved with NR characteristics at low
with 38% dry rubber. content and prevulcanisation time due to temperatures.
increased degree of crosslinking and not
affected by surface active agents.
PS/S-b-B-b-S - Phase separation for both blends. [117]
in PSU matrix.
158
polymers
Modified SAN/Poly(methyl - Miscible one-phase blends if more than - Miscibility due to donor- [121]
acrylate-co-maleimide)(MA- 10 mol% DAT in SAN and more than acceptor-donor (DAD) and
MI) 22 mol% MI in MA-MI or more than ADA triple hydrogen
- ACN units in SAN partially 21 mol% DAT in SAN and 9-14 mol% MI bonds.
converted to 2,4-diamino- in MA-MI.
1,3,5-triazine (DAT) by
reaction with dicyandiamide;
Coprecipitation in water or
film casting from solutions;
miscellaneous blends Styrene
polymers
Table 4.3 Continued
Mixture/composition Properties Observations Ref.
Blending method
S-b-B-b-S/poly(styrene-co-met- - S-b-B-b-S/SMAA: pseudoplastic blends [122]
hacrylic acid) (SMAA) (torque decreases with increasing shear
S-b-B-b-S/poly(styrene-co- rate), interpenetrating phases.
sodium - S-b-B-b-S/SNaMA: phase separation
methacrylate)(SNaMA) increasing with NaMA amount and Mn in
SNaMA; minor degree of mixing at phase
Melt blending of S-b-B-b-S boundaries.
polymers
PS(Mn = 284000); PB-g-S(Mn for PBS/PEO at 75-85% PEO.
= 300000, 5.2 wt.% cisB); - PSB/PEO: less immiscible than PS/PEO due
PEO(Mn = 600000) to dipole-induced forces between relative
polar C-O in PEO and polarisable C=C in
PBS.
- PBS/PEO (17/83): miscibility parameter
increases linearly with decreasing
temperature, approaching zero at 20-21 C.
- reduced viscosity linear dependent on
concentration.
miscellaneous blends Styrene
Table 4.3 Continued
polymers
Mixture/composition Properties Observations Ref.
Blending method
Poly(styrene-b--capro- - PCL crystallinity increases with increasing [130]
lactone)(P(S-b-CL))/PVME PVME up to 50%.
- PVME mixes with only PCL block and not
Mixing solutions of with PS phase; segregation PVME/PCL
components in different ratio, occurs at low temperatures with no
casting on glass slides enhanced crystallinity of PCL block.
Poly(styrene-co-acrylic acid) - SAA/PEM, SAA/EM4VP, SAA/IBM4VP: - controlling blends [131]
(SAA)/poly(ethyl metacrylate) miscible blends in all proportions: single miscibility by steric effects;
(PEM) SAA/poly(isobutyl composition dependent Tg. - immiscible SAA/PIBMA
polymers
poly(styrene-co-methylmethac- to both CHMA/MMA and S/MMA from the S/CHMA and
rylate)(S-co-MMA), repulsive interactions. CHMA/MMA segmental
poly(styrene-co-cyclohexyl - S-co-MMA/S-co-CHMA, interaction parameters
methacrylate)(S-co-CHMA), S-co-CHMA/CHMA-co-MMA: narrow
CHMA-co-MMA miscibility window (S/CHMA repulsive
interactions).
Casting from 2 wt.% solutions - PCHMA/S90-co-MMA10/S50-co-CHMA50:
in THF experimental confirmation of the predicted
miscibility at 130 C for all composition
range and phase separation below 265 C.
miscellaneous blends Styrene
Table 4.3 Continued
polymers
Mixture/composition Properties Observations Ref.
Blending method
Poly(styrene-ran-t-butyl - Miscible blends in all compositions for 4.2- Ionomer blends: enhanced [80]
acrylate) (S-tBA) in acidic or 23 mol% tBA content in acid S-tBA; miscibility and mechanical
potassium salt form/PPO partial miscibility for high tBA contents properties due to ion-ion or
due to decreased interaction parameter. ion-dipole interactions and
Casting from CHCl3 solutions, - S-tBA ionomers: 0.5-2 mol% K+ - physical crosslinking
vacuum dried (45 C) completely miscible, 3-5 mol% K+ - between the ion-pairs.
partially miscible, over 5.6 mol% K+ -
immiscible with PPO due to the dipole-
polymers
- PC rich blends: SEBS-g-MA dispersed in
both PA6 domains and present in PC
matrix/PA6 domain boundary.
- PS phase: cylinder in hexagonal
arrangement in EB matrix; SEBS acts as
thermoplastic elastomer at domain
boundary.
- SEBS-g-MA also compatibiliser for PA6/PP
blends and toughen the Nylon 6,6/PP
blends [145, 146].
miscellaneous blends Styrene
Table 4.3 Continued
Mixture/composition Properties Observations Ref.
polymers
Blending method
PVC/CPE/ultrafine PS - Very good toughness and balanced - Low-price. [147]
particles (UPS) 100/10/2.5 mechanical properties. - Resistance to ageing and
- UPS promotes a small and uniform CPE corrosion.
PVC, CPE commercial, UPS network in PVC and improves the
(32% gel, 120 nm size) plasticising behaviour of PVC/CPE matrix
increasing the friction of PVC primary
particles
ABS/PC - PC Mw, processing conditions, type, size Moulding applications, [148,
Compatibilisers: and ABS content affect the morphology automotive industry. 149,
MA-grafted PP (MAP), epoxy and mechanical properties 150]
the electrolyte.
166
polymers
reaction in ABS/MGE, with increased blend
viscosity, no change in morphology but
decreased mechanical properties.
- PBT molecular weight and melt viscosity
affect the toughness and morphology but
not the tensile properties [154].
- MGE also compatibiliser for PBT/SAN
blends [155].
miscellaneous blends Styrene
Table 4.3 Continued
polymers
Mixture/composition Properties Observations Ref.
Blending method
PS/Poly(ethylene-ran-acrylic - Domain size of dispersed phase strongly PS-GMA: effective in situ [156-
acid)(PE-AA) PS/Poly(butylene decreased with increasing low amount of compatibiliser due to epoxy 158,
terephthalate)(PBT) PS-GMA, and depending on AA amount in groups/carboxylic groups 159,
Poly(styrene-ran-glycidyl PE-AA. reactions. 160]
methacrylate)(PS-GMA): in - morphology and rheological properties of
situ compatibiliser PP-g-AA/PS/PS-GMA affected by PAA
homopolymer not grafted on PP [160].
Melt mixing (200 C - 230 C)
polymers
SBR: Styrene-butadiene rubber
THF: Tetrahydrofuran
SIS: Styrene-isoprene-styrene
Polystyrene and Styrene Copolymer Based Blends
PS represents about 14 wt% of the municipal plastic waste (MPW) [169]. Expanded PS
(EPS) is present in MPW light fraction and forms immiscible blends with PE, PP and
other components of the waste if it is not properly separated. PS remains in MPW heavy
fraction, mixed with mainly PVC and PET. PS together with PVC and PET can be separated
from PE and PP by floating/sinking using water followed by PS separation from PVC
and PET using brine with specific gravity of 1.2 g/cm3. At laboratory scale, PS together
with PVC can also be separated from PP, PE and PET by dissolving in tetrahydrofuran at
room temperature followed by solvent volatilisation. Automatic lines such as Linpac
that use five successive hydrocyclones can separate EPS and PS from PO, PVC and PET
and recycle them as high performance grades. Commercially available recycled styrene-
containing resins such as Dylite EPS (EPS with 25% recyclates) and Retain (recycled PS/
PO blends) are competitively priced compared with virgin resins. HIPS can be identified
very fast (10 s) and with high accuracy (96%) for sorting from polymer mixtures by
flash pyrolysis coupled with FTIR.
The nature and the quality of the waste are limiting factors for the mechanical recycling
of polystyrene. PS and expanded (foamed) PS from packaging can be recycled by
reprocessing but additives such as sterically hindered phenols is to be added to avoid the
decrease of polymer molecular weight. Recycled PS can be used mainly in the non-food
sector as audio and video cases, offices requisites, insulation slabs in constructions or in
automotive industry. Expanded PS can be reused as sand substitute for insulation plaster,
as replacement of mineral aggregate in lightweight concrete and to increase the thermal
insulating while reducing the weight of light perforated bricks [170]. Foamed PS can be
used in receipt composition for thermo- and phono-insulating materials [171]. Waste PS
from electrical appliances was converted into sulfonated PS and used as polymer flocculant
for treating wastewater [172].
PS in complex polymer mixtures can be converted into styrene monomer and aromatic
hydrocarbons by thermal and catalytic degradation. Styrene monomer (75%) plus
oligomers (10%) are obtained at 500 C from PS by pyrolysis in a fluidised bed reactor
in the Hamburg process. The pyrolysis oil of a mixed plastic waste of PE, PP, different
content of polyesters, PS and PVC contains 30-40 wt% styrene and benzene-toluene-
xylene (BTX) fractions and less than 10 ppm of chlorine, the main part of fillers, impurities
and heavy metals remaining in the residue [174].
169
Handbook of Polymer Blends and Composites
4.14 Applications
In Table 4.4 shows industrial applications for some commercial blends containing styrene
polymers.
4.15 Conclusions
Despite its deficiencies (brittleness, low impact strength) which can be improved by
blending PS with other polymers, PS is a frequently used polymeric material being capable
to improve the mechanical and thermal properties of various polymers.
With few exceptions such as PS/PVME, PS/PPO, PS/TMPC systems, most of the blends
containing PS are immiscible and a compatibilisation is required in order to obtain an
improvement of the properties of the blend.
HIPS is one of the best known examples for improving the PS properties when a small
amount of elastomer blended with PS can rise the impact strength of PS dramatically.
Also, the low ductility and chemical resistance can be improved by blending the PS with
PE or PP, and the thermal stability by blending the PS with a miscible polymer with
higher Tg (such as PPO).
Due to its hardness and dimensional stability, PS can increase the modulus of PS/PE or
PS/PP blends. Also due to its excellent melt flow, PS can improve the processability of PS/
PPO and PS/PET blends.
170
industrial applications for
Table 4.4 Industrial applications for some commercial blends containing styrene polymers
blends
Styrene polymer
Blend Properties Applications
- SAN; - Crosslinking the elastomer leads to - Electrical and office equipment,
- ABS (SAN/NBR; PB/graft considerable improvement in impact thermoformed packagings [3], pipes and
polymers of styrene and strength [3]; fittings [2, 176], refrigerators,
acrylonitrile); - Improved weathering resistance in containers [2], household applications,
SAN/EPDM (acrylate rubber, CPE) multicoloured products, toys [2],
blends [3]; furniture [177].
- Foams of SAN exhibit good chemical
resistance (crude oil and mineral oil) [177]
blends
Styrene polymer
connectors, TV components)
automotive and transportation,
environmental engineering (pumps,
impellers, water purification equipment,
solar energy collectors) [3];
- PPO: automobiles, electrical
construction, cold and warm water
applications, sound-deadening
applications (foamable grades) [178]
industrial applications for
blends
Styrene polymer
Table 4.4 Industrial applications for some commercial blends containing styrene polymers
Blend Properties Applications
SMA/PC - Good retention of impact properties even - Automotive, surgical appliances, food
after extended heat ageing; good processors, cookwear [3];
weatherability [3];
(SMA, SAN)/PVC - Good weathering properties [3]. - Constructions (window profiles, pipes)
[3];
Many attempts have been made in order to improve the miscibility of PS with other
homopolymers. The addition of a styrene-copolymer is one of a widely used method of
compatibilising the various blend containing PS (PS/PO, PS/Elastomer, PS/PA, PS/PMMA
blends).
Reactive styrene copolymers with maleic anhydride and methacrylic acid also have been
used in reactive compatibilisation. The lightly SPS ionomers appear to be good candidates
for developing miscible blends with other polymers. On the one hand the ionic group
provides a rich chemistry for achieving specific interactions with polymers that contain
complementary polar groups, such as amine or amide groups (SPS/PA blends). On the
other hand, the strong repulsive interactions that occur between the ionic and non-ionic
species of the ionomer may be an effective mechanism for accomplishing miscibility
without specific interactions.
Taking into account the various available compatibilising methods, the PS containing
blends can have multiple practical applications and their study will be continued in order
to reveal new opportunities to obtain polymeric blends with improved properties.
References
4. T.J. Cavanaugh, K. Buttle, J.N. Turner and E.B. Nauman, Polymer, 1998, 39,
4191.
174
Polystyrene and Styrene Copolymer Based Blends
7. J.H. Choi, K.H. Ahn and S.Y. Chim, Polymer, 2000, 41, 5229.
8. T.J. Cavanaugh and E.B. Nauman, Journal of Polymer Science, Polymer Physics,
1998, 36, 2191.
10. K. Dalnoki-Veress, J.A. Forrest, J.R. Stevens and J.R. Dutcher, Journal of
Polymer Science, Polymer Physics, 1996, 34, 3017.
11. S. Polizu, B.D. Favis and T. Vu-Khanh, Macromolecules, 1999, 32, 3448.
12. R.J. Roe and W.C. Zin, Macromolecules, 1980, 13, 1221.
13. L. Sung and C. Han, Journal of Polymer Science, Polymer Physics, 1995, 33, 2405.
14. A. Karim, D.W. Liu, J.F. Douglas, A.I. Nakatani and E. J. Amis, Polymer, 2000,
41, 8455.
17. W.H. Jo, K.H. Nam and J.C. Cho, Journal of Polymer Science, Polymer Physics,
1996, 34, 2169.
18. B.K. Hong and W.H. Jo, Polymer, 2000, 41, 2069.
19. L.Y. Yang, T.G. Smith and D. Bigio, Proceedings of Antec 94, San Francisco, CA,
USA, 1994, Volume II, p.2428.
20. C. Caze and S.H. Pak, Journal of Materials Science Letters, 1997, 16, 53.
21. Z. Wang, C.M. Chan, S.H. Zhu and J. Shen, Polymer, 1998, 39, 6801.
23. M. Chiriac, B.S. Munteanu, G.G. Bumbu, M. Burlacel, A. Ioanid and C. Vasile,
Macromolecular Materials and Engineering, 2000, 283, 26.
175
Handbook of Polymer Blends and Composites
24. B. Brahimi, A. Ait-Kadi, A. Ajji and R. Fayt, Journal of Polymer Science, Polymer
Physics, 1991, 29, 945.
25. R. Fayt, R. Jerome and P. Teyssie, Journal of Polymer Science, Polymer Physics,
1989, 27, 775.
27. H. Veenstra, J.D. Dam and A. Posthuma de Boer, Polymer, 2000, 41, 3037.
29. C.Y. Han, M.D. Wolkowicz, D. Song and S.S. Dagli, Proceedings of Antec 96,
Indianapolis, IN, USA, 1996, Volume II, p.2606.
30. I. Smit and G. Radonjic, Polymer Engineering and Science, 2000, 40, 2144.
31. R.C. Willemse, A. Speijer, A.E. Langeraar and A. Posthuma de Boer, Polymer,
1999, 40, 6645.
34. E. Kung, A.J. Lesser and T.J. McCarthy, Macromolecules, 1998, 31, 4160.
35. G. Radonjic, V. Musil and M. Makarovic, Acta Chimica Slovenica, 1997, 44, 29.
37. Y. Changjiang and S. Xiaohua, China Synthetic Rubber Industry, 1995, 18, 363.
38. P.H.P. Macaubas and N.R. Demarquette, Polymer, 2001, 42, 2543.
40. G. Radonjic, V. Musil and I. Smit, Journal of Applied Polymer Science, 1998, 69,
2625.
41. M. Chiriac, G.G. Bumbu, V. Chiriac, C. Vasile, M. Burlacel and B.S. Munteanu,
International Journal of Polymeric Materials, 2001, 49, 419.
176
Polystyrene and Styrene Copolymer Based Blends
42. E.M. Woo, T.K. Mandal and L.L. Chang, Macromolecules, 2000, 33, 4186.
44. C.G. Robertson and G.L. Wilkes, Polymer, 2001, 42, 1581.
45. P. Schmidt, J. Kolarik, F. Lednicky, J. Dybal, J.M. Lagaron and J.M. Pastor,
Polymer, 2000, 41, 4267.
46. R.A. Mendelson, Journal of Polymer Science, Polymer Physics, 1985, 23, 1975.
47. H. Li, Y. Yang, R. Fujitsuka, T. Ougizaw and T. Inoue, Polymer, 1999, 40, 927.
48. F. Lednicky, J. Hromadkova and J. Kolarik, Polymer Testing, 1999, 18, 123.
49. G. Wildes, H. Keskkula and D.R. Paul, Polymer, 1999, 40, 5609.
50. J.A. Pathak, R.H. Colby, S.Y. Kamath, S.K. Kumar and R. Stadler,
Macromolecules, 1998, 31, 8988.
52. J.M. Piglowski and M. Bryjak, European Polymer Journal, 1998, 34, 11.
53. Q. Zheng, M. Du, B. Yang and G. Wu, Polymer, 2001, 42, 5743.
54. S. Zheng, J. Li, Q. Guo and Y. Mi, Journal of Materials Science, 1997, 32, 3463.
55. J.H. Kim, J.W. Barlow and D.R. Paul, Journal of Polymer Science, Polymer
Physics, 1989, 27, 223.
56. C. Wastlund, H. Berndtsson and F.H.J. Maurer, Macromolecules, 1998, 31, 3322.
57. Y. Takeda and D.R. Paul, Journal of Polymer Science, Polymer Physics, 1992, 30,
1273.
58. W. Heinen, C.B. Wenzel, C.H. Rosenmoeller, F.M. Mulder, G.J. Boender, J.
Lugtenburg, H.J.M. de Groot, M. Van Duin and B. Klumperman,
Macromolecules, 1998, 31, 7404.
59. D. Keroack, Y. Zhao and R.E. Prudhomme, Polymer, 1998, 40, 243.
177
Handbook of Polymer Blends and Composites
63. G.G. Cameron, M.Y. Quereshi and S.C. Tavern, Polymer International, 1998,
47, 5.
64. K. Tanaka, J-S. Yoon, A. Takahara and T. Kajiyama, Macromolecules, 1995, 28,
934.
67. A. Ajji, L. Choplin and R.E. Prudhomme, Journal of Polymer Science, Polymer
Physics, 1991, 29, 1573.
69. J.K. Kim and H.W. Son, Polymer, 1999, 40, 6789.
70. T.K. Mandal and E.M. Woo, Polymer, 1999, 40, 2813.
71. T. Nishi, T.T. Wang and T.K. Kwei, Macromolecules, 1975, 8, 227.
72. S. Pavawongsak, J.S. Higgins, N. Clarke, T.C.B. McLeish and D.G. Peiffer,
Polymer, 2000, 41, 757.
73. B. Mailhot, S. Morlat and J-L. Gardette, Polymer, 2000, 41, 1981.
75. S-H. Hwang, Y-S. Kim, H-C. Cha and J-C. Jung, Polymer, 1999, 40, 5957.
76. C. Creton, J.L. Halary and L. Monnerie, Polymer, 1998, 40, 199.
178
Polystyrene and Styrene Copolymer Based Blends
78. M. Song, D.J. Hourston, H.M. Pollock and A. Hammiche, Polymer, 1999, 40,
4763.
80. L.M. Leung, C.T. Lau, L. Chang, Y. Huang, B. Liao, M-C. Chen and G. Gong,
Polymer, 2001, 42, 539.
81. G. Ton-Tat, A.G. Shard, D.O.H. Teare and R.H. Bradley, Polymer, 2001, 42,
1121.
85. S.A. Madbouly, T. Chiba, T. Ougizawa and T. Inoue, Polymer, 2001, 42, 1743.
86. J.K. Lee and C.D. Han, Polymer, 1999, 40, 2521.
87. C. Koulic, Z. Yin, C. Pagnoulle, B. Gilbert and R. Jerome, Polymer, 2001, 42,
2947.
88. D. Li, B. Han, Z. Liu and D. Zhao, Polymer, 2001, 42, 2331.
90. K.H. Yoon, H.W. Lee and O.O. Park, Polymer, 2000, 41, 4445.
91. R. Fayt and P. Teyssie, Journal of Polymer Science, Polymer Letters, 1989, 27,
481.
92. T-Y. Guo, M-D. Song, G-J. Hao and B-H. Zhang, European Polymer Journal,
2001, 37, 241.
93. P. Jannach and B. Wesslen, Journal of Applied Polymer Science, 1995, 58, 753.
179
Handbook of Polymer Blends and Composites
96. Y. Ghebremeskel and R.A. Weiss, Proceedings of Antec 95, Boston, MA, USA,
1995, Volume II, p.2463.
97. P. Rajagopalan, J-S. Kim, H.P. Brack, X. Lu, A. Eisenberg, R.A. Weiss and W.M.
Risen, Jr., Journal of Polymer Science, Polymer Physics, 1995, 33, 495.
98. S. Kato, Y. Tsujita, H. Yashimizu, T. Kinoshita and J.S. Higgins, Polymer, 1997,
38, 2807.
99. L.W. Jang and D.C. Lee, Polymer, 2000, 41, 1749.
101. G.D. Merfeld and D.R. Paul, Polymer, 2000, 41, 649.
102. K.S. Jeon, K. Char, D.J. Walsh and E. Kim, Polymer, 2000, 41, 2839.
103. E. Kim, G. Krausch, E.J. Kramer and J.O. Osby, Macromolecules, 1994, 27,
5927.
104. E. Kim, E.J. Kramer and J.O. Osby, Macromolecules, 1995, 28, 1979.
105. K.A. Koh, J.H. Kim, D.H. Lee, M. Lee and H.M. Jeong, European Polymer
Journal, 1998, 34, 1229.
106. I. Luzinov, K. Xi, C. Pagnoulle and G. Huynh-Ba and R. Jrme, Polymer, 1999,
40, 2511.
108. H. Feng, C. Ye, J. Tian, Z. Feng and B. Huang, Polymer, 1998, 39, 1787.
109. J. Gao, D. Li, D. Wang and L. Yang, European Polymer Journal, 2000, 36, 2517.
111. Y. Yang, N. Otsuka, H. Saito, T. Inoue and Y. Takemura, Polymer, 1999, 40, 559.
112. H.T. Chiu and W.M. Chiu, Materials Chemistry and Physics, 1998, 56, 108.
180
Polystyrene and Styrene Copolymer Based Blends
113. H. Veenstra, B.J.J. Van Lent, J. Van Dam and A. Posthuma de Boer, Polymer,
1999, 40, 6661.
114. A.J. Marzocca, S. Cerveny and J.M. Mendez, Polymer International, 2000, 49,
216.
116. J.T. Varkey, S. Augustine, G. Groeninckx, S.S. Bhagawan, S.S. Rao and S.
Thomas, Journal of Polymer Science, Polymer Physics, 2000, 38, 2189.
117. E.V. Sanchez, J.L.G. Ribelles, M.M. Pradas, B.R. Figueroa and F.R. Colomer,
European Polymer Journal, 2000, 36, 1893.
119. T.O. Ahn, K. Kim, H.M. Park and H.M. Joeng, European Polymer Journal,
1997, 33, 781.
120. Y.S. Chun and W.N. Kim, Journal of Applied Polymer Science, 2000, 77, 673.
122. S.A.M. Ali and D.J. Hourston, European Polymer Journal, 1995, 31, 1.
128. V. Tanrattanakul, A. Hiltner, E. Baer, W.G. Perkins, F.L. Massey and A. Moet,
Polymer, 1997, 38, 2191.
129. E. Yilmaz, O. Yilmaz and H. Caner, European Polymer Journal, 1996, 32, 927.
130. W.Y. Yam, J. Ismail, H.W. Kammer, M.D. Lechner and C. Kummerlwe, Polymer,
2000, 41, 9073.
181
Handbook of Polymer Blends and Composites
131. F. Feraz, A.S. Hadji Hamou and S. Djadoun, European Polymer Journal, 1995,
31, 665.
132. H. Feng, C. Ye and Z. Feng, European Polymer Journal, 1997, 33, 743.
133. J. Prinos, C. Tselios, D. Bikiaris and C. Panayiotou, Polymer, 1997, 38, 5921.
138. J.A. Pomposo, A. Mgica, J. Arizaga and M. Cortzar, Acta Polymerica, 1998,
49, 301.
141. Y.H. Kim, M. Kikuchi, S. Akiyama, K. Sho and S. Izawa, Polymer, 1999, 40,
5273.
142. C. Park, W.H. Jo, H.C. Park and Y.S. Kang, Polymer, 2000, 41, 1765.
143. S. Horiuchi, N. Matchariyakult, K. Yase, T. Kitano, H.K. Choi and Y.M. Lee,
Polymer, 1997, 38, 59.
144. C.H. Lee, Y.M. Lee, H.K. Choi, S. Horiuchi and T. Kitano, Polymer, 1999, 40,
6321.
145. A.N. Wilkinson, L. Laugel, M.L. Clements, V.M. Harding and M. Marin,
Polymer, 1999, 40, 4971.
146. S.C. Wong and Y.W. Mai, Polymer, 2000, 41, 5471.
147. Q. Zhou, W. Yang, Q. Wu, B. Yang, J. Huang and J. Shen, European Polymer
Journal, 2000, 36, 1735.
148. S.C. Tjong and Y.Z. Meng, European Polymer Journal, 2000, 36, 123.
182
Polystyrene and Styrene Copolymer Based Blends
149. G.S. Wildes, T. Harada, H. Heskkula, D.R. Paul, V. Janarthanan and A.R.
Padwa, Polymer, 1999, 40, 3069.
150. G. Wildes, H. Keskkula and D.R. Paul, Polymer, 1999, 40, 7089.
152. W. Hale, H. Keskkula and D.R. Paul, Polymer, 1999, 40, 3665.
153. W.R. Hale, L.A. Pessan, H. Keskkula and D.R. Paul, Polymer, 1999, 40, 4237.
154. W. Hale, J.H. Lee, H. Keskkula and D.R. Paul, Polymer, 1999, 40, 3621.
155. W. Hale, H. Keskkula and D.R. Paul, Polymer, 1999, 40, 365.
156. J.K. Kim, S. Kim and C.E. Park, Polymer, 1997, 38, 2155.
157. S. Kim, J.K. Kim and C.E. Park, Polymer, 1997, 38, 2113.
158. S. Kim, J.K. Kim and C.E. Park, Polymer, 1997, 38, 1809.
159. H.K. Jeon and J.K. Kim, Polymer, 1998, 39, 6227.
160. J.K. Kim, D.K. Yi, H.K. Jeon and C.E. Park, Polymer, 1999, 40, 2737.
162. O. Urakawa, Y. Fuse, H. Hori, Q. Tran-Cong and O. Yano, Polymer, 2001, 42,
765.
166. C. Vasile and I.A. Schneider, European Polymer Journal, 1974, 10, 381.
167. C. Vasile, I.A. Schneider, A. Airinei and A. Chiriac, European Polymer Journal,
1976, 12, 749.
168. C. Vasile, I.A. Schneider and I. Gabe, European Polymer Journal, 1975, 11, 483.
183
Handbook of Polymer Blends and Composites
170. Recycling and Recovery of Plastics, Ed., J. Brandrup, M. Bittner, W. Michaeli and
G. Menges, Carl Hanser Verlag, Munich, Germany, 1996.
173. Frontiers in the Science and Technology of Polymer Recycling, Eds., G. Akovali,
C.A. Bernardo, J. Leidner, L.A. Utracki and M. Xanthos, Kluwer Academic
Publishers, Dordrecht, The Netherlands, 1998.
178. L.A. Utracki and G.L. Bata, Proceedings of CANPLAST-81, Montreal, Canada,
1981, p.25.
184
Ionomer Polyblends
5
Ionomer Polyblends
Rudolph D. Deanin
5.1 Introduction
Ionomers are organic polymers containing up to 20 mol% of ionic groups [1]. They may
be made by copolymerisation with ionic monomers, or by post-polymerisation reactions
to introduce ionic groups into existing polymers. The ionic groups are most often based
on carboxylic acids, less often on sulfonic acids, occasionally on amines, and sometimes
on more esoteric functional groups. While these have some ionic character in themselves,
they are generally neutralised, either partially or completely, to create highly ionised
groups which have much more pronounced effects on properties. Polymeric acids are
generally neutralised with alkali metal or divalent metal cations, while polymeric amines
are generally quaternised with halide or other negative anions.
The ionic groups are not miscible in the predominantly covalent organic polymer medium,
so they tend to segregate into purely ionic small multiplets and larger clusters, dispersed
in a continuous matrix of the covalent polymer [2]. This morphology is analogous to the
phase separation in most polymer blends, and many of the effects on properties are
similar.
Thus the clusters in ionomers provide thermoplastic processability, increased melt viscosity
[3], increased strength, and a rubbery plateau above the glass transition temperature
(Tg). The combination of strong electrical forces and cluster morphology has led to
practical applications in medical and industrial membranes [4], packaging [5], elastomers,
and adhesives [6].
When ionomers are used in polyblends, miscibility or compatibility depends upon the
attractive forces between the polymers and their effects upon each others morphology
[7]. The polyblend research literature may be organised according to (1) Intermolecular
attractions, and (2) Polymer backbones and properties.
185
Handbook of Polymer Blends and Composites
Polyblends of Nylon 6 or Nylon 2,10 with sulfonated polystyrene salts were miscible
due to coordination complex formation between the transition metal ions - zinc, copper,
manganese, and cadmium - and the amide groups of the Nylons [7, 11-16]. Miscibility
reduced the crystallinity of the polyamides (PA) and cluster size of the ionomers, decreased
water absorption of the Nylons, and increased Tg and melt viscosity.
186
Ionomer Polyblends
[7, 17-30]. These effects have been ascribed to a variety of mechanisms, but in all these
examples it is possible that coordination complex formation between the zinc ion and
nitrogen or oxygen atoms in the nonionic polymer may have contributed to the results.
Most ionomer polyblend research has focused on the blending of two ionomers whose ionic
groups are attracted to each other, thus increasing miscibility, viscosity, and strength [7]. The
strongest attractions are seen between polymeric pyridinium cations and polymeric sulfonate
anions. Thus, for example, pyridine has been copolymerised with isoprene, styrene, ethyl
acrylate, or methyl methacrylate, or grafted onto polyacetylene, and then quaternised to
form the polymeric cation. Sulfonic acid groups have been copolymerised or grafted with
EPDM, polystyrene, polytetrafluoroethylene, polyphenylene ether, or liquid crystal polyester,
and then neutralised with alkali metals to form the polymeric anion. Such pyridinium sulfonate
pairs have generally demonstrated high intermolecular attraction and miscibility [7, 31-36].
Some researchers have placed acyclic tertiary amine groups in polyisoprene (IR), poly-t-
butyl acrylate, polyalkylene oxide, polyurethane, or liquid crystal polyester, and then
quaternised them; and blended them with sulfonated polystyrene, poly--methyl styrene,
or polyacrylonitrile (PAN) to form equally effective polyblends [7, 37, 38].
Common effects of ion-ion attraction in these polyblends include some which are observed
generally, others which are specific to their particular polymer systems. Miscibility and
homogeneity were generally improved. Melt flow index increased in some systems, while
melt viscosity increased in others. Modulus increased in some, while flexibility increased
in others. Tensile strength, elongation, ductility, and impact strength were frequently
increased. Glass transition temperature increased in some, while low-temperature
flexibility and broader rubbery plateau were reported in others. Transparency and swelling
resistance were reported in some cases. Thus ion-ion attractions in ionomer polyblends
can offer a variety of practical property improvements.
187
Handbook of Polymer Blends and Composites
When cationic polymers are blended with anionic polymers, the strong attractions between
them generally favour miscibility. On the other hand, when two anionic polymers are
blended, the attractive forces between them are not as dependable. While positive effects
on miscibility and properties are often reported, some negative results have been published,
and others may well have gone unreported.
Polyblends of two sulfonate ionomers appear most reliable. Thus blends of sulfonated
EPDM, butyl rubber (IIR), polystyrene, and polyphenylene ether have all been reported
miscible [44, 45].
Polyblends of two carboxylated ionomers are often reported miscible. These include
carboxylated polyethylenes, EPDM, polypropylene (PP), polystyrene, polyvinylidene
fluoride (PVDF), PEAc, nitrile rubber, liquid crystal polymers (LCP), and polydiacetylene
[33, 46-51]. On the other hand, several studies have reported their attractive forces too
weak to produce miscibility; these included carboxylated polyethylene, polystyrene, and
PEAc [7, 52].
Polyblends of a sulfonated ionomer with a carboxylated ionomer have often been reported
miscible: polyethylene, EPDM, polystyrene, polyphenylene ether, and polydiacetylene
[33, 53, 54]; but blends of sulfonated and carboxylated fluoropolymers remained two-
phase systems [55].
Aside from miscibility, some of the property benefits which have been reported for these
polyblends include increased thermoplastic processability, melt viscosity, hardness,
modulus, tensile strength, tear strength, elongation, and creep resistance.
Thus a full understanding of polyblends of similar ion pairs still requires more intensive
study and understanding.
In practical research, development, and commercial practice, ionomers are often blended
with nonionic polymers, and the beneficial results are ascribed to attractive forces between
the ionic groups in the ionomers and any polar groups in the nonionic polymers. These
ion-dipole attractions are weaker than ion-ion and ion-coordination attractions, but
they are sufficient to be of considerable practical importance [7, 56].
188
Ionomer Polyblends
Thus, even though ion-dipole attractions may be weaker than ion-ion and ion-coordination
forces, they are of considerable practical importance, and extend the field of ionomer
polyblends over a much greater range of polymer pairs.
With or without ionic groups present, attractive forces between polymers range from
London Dispersion to dipole-dipole, hydrogen bonding, neutralisation of intramolecular
repulsions, phase separation in crystalline polymers, and self-association in block
copolymers. When ionomers are one or both members of the polyblend, these additional
attractive forces are still active and important, and they certainly contribute greatly to
the overall attractions in all polymer blends.
When researchers combine two polymers in a polyblend, they generally hope to combine
some of the best properties of each polymer in the blend, or achieve a better overall
balance of properties, or occasionally benefit from an unexpected synergism. Since most
polymer blends are too immiscible to give such benefits, they seek ways to increase degree
of dispersion into the range of practical compatibility.
As we have just seen, ionic groups in polymers can produce strong intermolecular
attractions, which can be a great help in increasing the degree of dispersion, often to the
extent of complete miscibility. Polymer pairs, in which one or both were ionomers which
produced such dispersion and/or miscibility, are listed in Table 5.1. When this resulted in
practical property benefits, these are listed as well.
189
properties Polymer pairs
Handbook of Polymer Blends and Composites
190
properties Polymer pairs
Ionomer Polyblends
191
properties Polymer pairs
Handbook of Polymer Blends and Composites
Impact Strength
Impermeability
Surface Conductivity
Processsability
Melt Viscosity
Hardness
Modulus
Tensile Strength
Tear Strength
Impact Strength
Tensile Strength
Elongation
Modulus
Tensile Strength
Tear Strength
PBD PS [7]
192
properties Polymer pairs
Ionomer Polyblends
193
Handbook of Polymer Blends and Composites properties Polymer pairs
References
194
Ionomer Polyblends
10. M.R. Tant and G.L. Wilkes in Ionomers: Synthesis, Structure, Properties and
Applications, Eds., M.R. Tant, K.A. Mauritz and G.L. Wilkes, Kluwer Academic
Publishers, Dordrecht, The Netherlands, 1997, 282.
11. R.A. Weiss and X. Lu, Proceedings of ANTEC 92, Detroit, MI, USA, 1992,
Volume 1, 1424.
12. R.A. Weiss and X. Lu, Proceedings of Polymat 94 - Polymer Blends II, London,
UK, 1994, 459.
14. R. Hagen and R.A. Weiss, Polymer, 1995, 36, 24, 4657.
15. Y. Feng, A. Schmidt and R.A. Weiss, Macromolecules, 1996, 29, 11, 3909.
16. Y. Feng, R.A. Weiss and C.C. Han, Macromolecules, 1996, 29, 11, 3925.
17. T. L. Boykin and R. B. Moore, Polymer Engineering and Science, 1998, 38, 10,
1658.
195
Handbook of Polymer Blends and Composites
18. S. Datta and D.J. Lohse, Polymeric Compatibilisers: Uses and Benefits in
Polymer Blends, Carl Hanser, Munich, Germany, 1996, 351.
19. R.D. Deanin and J. Jherwar, Proceedings of SPE Processing and Property
Enhancement Utilising Modifiers and Additives in Polymers, Newark, NJ, USA,
1985, 24.
20. R.D. Deanin and W-F. Liu, ACS Polymeric Materials, Science and Engineering,
1985, 53, 813.
21. R.D. Deanin, S.A. Orroth and R.I. Bhagat, Polymer Plastics Technology
Engineering, 1990, 29, 3, 289.
22. J. He and J. Liu, Journal of Applied Polymer Science, 1998, 67, 13, 2141.
23. H. Li and Z. Li, Journal of Applied Polymer Science, 1998, 67, 1, 61.
25. H. Li, Y. Wang and Z. Li, ACS Polymer Preprints, 1999, 40, 1, 173.
26. C.K. Samios and N.K. Kalfoglou, Polymer, 1998, 39, 16, 3863.
27. M.J. Sullivan and R.A. Weiss, Proceedings of ANTEC 91, Montreal, Canada,
1991, 964.
28. M.J. Sullivan and R.A. Weiss, Polymer Engineering and Science, 1992, 32, 8, 517.
29. R.A. Weiss, Y. Ghebremeskel and L.F. Charbonneau, Proceedings of ANTEC 99,
New York, NY, USA, 1999, Volume 2, 1595.
30. R. Xie and R.A. Weiss, Polymer, 1998, 39, 13, 2851.
31. S. Datta and D.J. Lohse, Polymeric Compatibilisers: Uses and Benefits in
Polymer Blends, Carl Hanser, Munich, Germany, 1996, 396.
32. E.P. Douglas, A.J. Waddon and W.J. MacKnight, Macromolecules, 1994, 27, 15,
4344.
34. Y. Pan, Y. Huang, M. Cheng, G. Cong and L.M. Leung, European Polymer
Journal, 1998, 34, 2, 213.
196
Ionomer Polyblends
35. R.A. Register and T.R. Bell, Journal of Polymer Science, 1992, 30, 6, 569.
37. C.G. Bazuin, M. Plante and S.K. Varshney, Macromolecules, 1997, 30, 9, 2618.
38. Y.S. Oh, Y.M. Lee and B.K. Kim, Journal of Macromolecular Science B, 1994,
33, 2, 243.
40. K.H. Hsieh and L.M. Chou, Journal of Applied Polymer Science, 1989, 38, 4, 645.
41. L.M. Robeson, J.A. Kuphal and M.S. Vratsanos, Journal of Applied Polymer
Science, 1996, 61, 9, 1561.
42. M. Rutkowska and A. Eisenberg, Journal of Applied Polymer Science, 1985, 30,
8, 3317.
43. M. Rutkowska and A. Eisenberg, Journal of Applied Polymer Science, 1987, 33,
8, 2833.
45. H-Q. Xie and B-Y. Ma in Multiphase Macromolecular Systems, Ed., B.M.
Culbertson, Plenum Press, New York, NY, USA, 1989, 601.
46. P. Anthony, S. Bandyopadhyay and S.K. De, Journal of Materials Science, 1999,
34, 2553.
47. P. Anthony, S. Bandyopadhyay and S.K. De, Polymer Engineering and Science,
39, 5, 963.
48. P. Antony and S.K. De, Journal of Applied Polymer Science, 1998, 70, 3, 483.
49. P. Anthony, A.K. Bhattacharya and S.K. De, Journal of Applied Polymer Science,
1999, 71, 8, 1257.
50. A. Murakami, H. Kanatani and T. Yoshiki, Kobunshi Ronbunshu, 1981, 38, 8, 507.
51. A. Valenza, G. Carianni and L. Mascia, Polymer Engineering and Science, 1998,
38, 3, 452.
197
Handbook of Polymer Blends and Composites
53. T. Kurian, S. Datta, D. Khastgir, P.P. De, D.K. Tripathy, S.K. De and D.G. Peiffer,
Polymer, 1996, 37, 21, 4787.
54. Y. Pan, Y. Huang, B. Liao, M. Chen, G. Cong and L.M. Leung, Journal of
Applied Polymer Science, 1997, 65, 2, 341.
55. W.Y. Hsu, T.D. Gierke and C.J. Molnar, Macromolecules, 1983, 16, 12, 1945.
57. R.D. Deanin, ACS Polymer Materials, Science and Engineering, 1987, 56, 35.
59. R.D. Deanin in Handbook of Polyolefins, Ed., C. Vasile, Dekker, New York, NY,
USA, 2000, 624.
60. R.D. Deanin and W. Chu, ACS Polymeric Materials Science and Engineering, San
Francisco, CA, USA, 1997, 76, 296.
61. R.D. Deanin, S.A. Orroth and H. Bhagat, ACS Polymeric Materials Science and
Engineering, Dallas, TX, USA, Spring Meeting, 1998, 78, 156.
62. P. Degee, R. Vankan, P. Teyssie and R. Jerome, Polymer, 1997, 38, 15, 3861.
63. B. Fisa, A. Bouti, B.D. Favis and F. Lalande, Proceedings of ANTEC 91,
Montreal, Canada, 1991, 37, 1135.
64. M. Hara, D. Wollmann and A. Eisenberg, ACS Polymer Preprints, 1984, 25, 1,
280.
67. H. Potente and X. Gao, Proceedings of ANTEC 94, San Francisco, CA, USA,
1994, 40, 3050.
68. L.A. Utracki, Polymer Alloys and Blends, Carl Hanser Verlag, Munich, Germany,
1989, 13 and 207.
198
Ionomer Polyblends
69. H.Q. Xie, J.G. Xu and S.B. Zhou, Proceedings of Polymer Blends, 1990,
Cambridge, UK, Paper No. E28/1.
70. R.D. Deanin and R.S. Ciulla, ACS Polymer Preprints, 1996, 37, 1, 384.
71. D.R. Paul, J.W. Barlow and H. Keskkula in Encyclopedia of Polymer Science and
Engineering, Eds., H.F. Mark, N. Bikales, C.G. Overberger, G. Menges and J.J.
Kroschwitz, Wiley, New York, NY, USA, 1988, Volume 12, 399.
73. L. Mascia, A. Moggi and F. Bellahdeb, Journal of Materials Science Letters, 1992,
11, 21, 1441.
74. L. Mascia and F. Bellahdeb, Advances in Polymer Technology, 1994, 13, 2, 99.
75. S. Pan, Z. Yang, Y. Xiong, W. Song and H. Xie, China Synthetic Rubber Industry,
1992, 15, 3, 169.
76. H-Q. Xie, Z-P. Ao and J-S. Guo, Journal of Macromolecular Science B, 1995, 34,
3, 249.
78. Y. Zhao, G. Yuan and P. Roche, Polymer, 1999, 40, 11, 3025.
82. A. Wiesemann, R. Zentel and G. Lieser, Acta Polymerica, 1995, 46, 1, 25.
83. R.A. Weiss and X. Lu, ACS Polymeric Materials Science and Engineering, Spring
Meeting, Dallas, TX, USA, 1998, 78, 145.
199
Polyamide-Based Blends
6
Polyamide-Based Blends
6.1 Introduction
According to the type of structural units of the polymer backbone PA are divided into
homopolyamides (with only one type of structural unit) and copolyamides (two or more
types of structural unit).
Referring only to homopolyamides, they can be subdivided into two groups, by chemical
structure [2]:
where A indicates the amino groups and B the carboxyl groups. The number of C atoms
in the individual linear monomers serves as a code number. In the case of the AA BB
type, the number of carbons of the diamine is indicated first of all, and then those of the
dicarboxylic acid [2, 3].
201
Handbook of Polymer Blends and Composites
For copolyamides one names first the predominant component, and then, in decreasing
order, the minor components. The proportion of each component is often put between
parentheses, e.g., PA66/6 (80:20) means 1,6-hexanediene, adipic acid and -caprolactam
in an 80:80:20 mol ratio.
Aliphatic PA are linear polymers and thus thermoplastic. They contain polar groups
spaced out at regular intervals, so that the polymer crystallise with a high intermolecular
attraction. These polymer chains also have aliphatic chain segments with give a measure
of flexibility in the amorphous region. Thus the combination of high interchain attraction
in the crystalline zone and flexibility in the amorphous zone leads to polymers which are
tough above their apparent glass transition temperature (Tg).
There are a number of structural variables, which can considerably affect the properties
of the aliphatic PA. The most important among these variables is the distance between
the repeating amidic groups. It is the presence of the polar groups that cause the aliphatic
PA to differ from polyethylene, and the higher their concentration the greater the
difference. As a rule, the higher the amide group concentration, i.e., the shorter the
distance between amidic groups, the higher the:
a. density;
b. forces required to mechanically separate the polymer molecules, and hence the higher
the tensile strength, rigidity, hardness and resistance to creep;
d. resistance to hydrocarbons;
e. water absorption.
The number of methylene groups in the intermediates is the second variable. It has been
observed that polymers from intermediates with an even number of methylene groups
have a higher melting point than similar polymers with an odd number of methylene groups.
These differences are due to the difference in the crystal structure of polymers with odd
and even methylene groups, which develop in order that oxygen atoms in one molecule are
adjacent to amino groups of the second molecule. Hydrogen molecule bond with an NH O
202
Polyamide-Based Blends
distance of 2.8 are produced and are the reason for the high strength and high melting
points of PA. The crystal structures of PA differ according to the type of polymer.
Specific types of PA are frequently available in forms differing in molecular weight. The
viscosity of linear PA melts is a function of molecular weight, increasing at about 3.5 times
of M w or M n . In the Newtonian shear-rate regions (shear stress less than approximately
30 kPa), the melt viscosity lies in the range of 401,000 Pa-s.
Replacement of hydrogen atoms in the amidic group by such groups as CH3 and CH2OCH3
will cause a reduction in the interchain attraction and consequent decrease of softening
point.
Except in those rare cases where monomer segments are isomorphous, copolymerisation,
as usual leads to the less crystalline and frequently amorphous materials. As might be
expected, these materials are tough, leather-like, flexible and, when unfilled, reasonably
transparent.
At room temperature all PA are soluble in phenol, formic acid, m-cresol, chloral hydrate,
fluorinated alcohols, and mineral acids. At higher temperatures, nitroalcohols, unsaturated
alcohols, and calcium chloride/methanol mixtures are the solvents. PA are resistant to
organic solvents, hydrocarbons and refrigerants.
Electrical conductance of dry PA is very low because of its low moisture regain. The
conductivity increases by several orders of magnitude as the relative humidity increases.
Mechanical properties are self-explanatory and are also subject to thermal and water
content changes. General rules are established for all aliphatic PA: tensile strength and
modulus decrease while elongation at break and Izod impact strength increase with the
rise of temperature and moisture content.
Laboratory tests and experience during use have demonstrated that the PA have extremely
good abrasion resistance. This may be further improved by addition of lubricants and by
processing under conditions that develop a highly crystalline hard surface.
The PA are commercially produced by reacting diamines with dibasic acids, by self-
condensation of an amino-acid or by opening of a lactam ring. Whatever method is
chosen, it is important that there is an equivalence between the number of amine and
acid groups, in order to obtain polymers with high molecular weight. In the case of
203
Handbook of Polymer Blends and Composites
amino-acids and lactams this is ensured using pure monomer, but with diamines and
dibasic acids it is necessary to form a salt that enable the groups equivalence. Small
quantities of monofunctional compounds are often used to regulate molecular weight.
Thermoplastic PA, e.g., PA6, PA66, PA666, PA1010, have been widely used in the
formation of mechanical and electrical parts for which excellent mechanical characteristics,
high heat resistance, stiffness, strength, oil and solvent resistance, and good durability
are required. PA are widely used for automobile, electric and electronic parts. Although
PA have good heat resistance and high mechanical strength, they show poor impact
resistance and notch sensitivity. Furthermore, PA have high density, and they are
significantly more expensive than commodity polymers. A major deficiency of PA is
their tendency to absorb water, which may result in hydrolytic degradation and loss of
performance [1, 5].
The great number of PA synthesised up to now offer a wide range of properties [1, 7].
However, not all the characteristics requested by multiple domains of use can be satisfied
by PA as unitary products. One way to extend the properties of PA or to limit deficiencies
is by blending with other polymers. Mixing PA with other polymers (thermoplastic resins
belonging to the category of commodity, engineering and specialty polymers, elastomers
and to a lesser with thermoset) is mainly intended to:
2. increase flexibility;
This group includes the blends obtained from various types of PA with mainly polyolefins
(PO) [low-density polyethylene (LPDE), linear low-density polyethylene (LLPDE), high-
density polyethylene (HDPE), ultra-high molecular weight polyethylene (UHMWPE),
polypropylene (PP)] and with ethylene and propylene copolymers.
204
Polyamide-Based Blends
Being non-polar, polyethylenes (PE) are antagonistically immiscible with PA. However,
if one manages to finely disperse PE into submicron particles, the discontinuities may
provide an adequate amount of stress concentration for toughening and reduction of
moisture sensitivity of PA.
There is little information about PA/LLDPE blends, which are mentioned only in patent
literature or as comparison test for blends also containing compatibilisation agents [7,
15, 16].
205
Handbook of Polymer Blends and Composites
One way to overcome the immiscibility of PA/PE heterogeneous blends is the introduction
of a small number of polar groups [such as maleic anhydride, vinyl acetate, (meth)acrylic
acid, or (meth)acrylic esters] in the backbone of polyolefins [11, 20]. The maximum
concentration of these modifiers varied from about 5 to 15 wt%. It has been reported
that this method of chemical compatibilisation of PA/PE blends remarkably improve the
mechanical behaviour of the resulting materials due to the increased interaction between
the phases as shown by capillarity rheometry, differential scanning calorimetry (DSC),
Fourier Transform Infra red spectroscopy (FTIR) [11, 21].
Thus, the blends of PA (either PA6, PA66, PA610) with 550 wt% finely dispersed
copolymer selected from ethylene-acrylic acid (EAA), ethylene-carbonate (ECA), ethylene-
ethyl acrylate (EEA), ethylene-methyl methacrylate (EMMA), ethylene-vinyl acetate (EVA),
ethylene-diethylaminoethyl methacrylate copolymers or similar one, are characterised
by large improvements of impact strength, tear strength and flexibility, good stability
and high toughness [7].
A great importance acquired also PA/ionomers (copolymers of ethylene with 110 wt%
(meth)acrylic salt like Li, Na, K, Mg or Zn) [7]. These blends showed high tensile strength
and HDT.
Blends of PA with partial saponified ethylene-vinyl acetate copolymer (EVAl) have been
used in packaging applications as barriers to inhibit the passage of gases or liquids.
However, miscible PA/EVAl mixtures show poor forming characteristics, and even
successfully transformed and stretched they show poor barrier properties.
Another way to improve the compatibility of PA/PE blends is to graft polar groups on
the PE chains, because this functionalised PE is able to interact with the end groups of
PA. Typical carboxyl containing modifiers are those based on maleic anhydride. They
provide a much higher level of interaction since they are capable of double covalent
bounding with a single PA chain. Since PA6 possesses only a single end group, the alloy
is in all cases an uncrosslinked graft copolymer [2224].
The use of radiation to graft polar groups in the PE chains to improve the compatibility
of PA6/PE blends has been also suggested [25, 26].
206
Polyamide-Based Blends
Comparing the effect of the two ester groups, Valenza and co-workers [27] have observed
a more homogeneous morphology for PA6/LLDPE grafted with ethyl methacrylate than
for PA6/LLPDE grafted with isobutyl methacrylate. This result can be ascribed to the
lower steric hindrance of the ethyl methacrylate with respect to isobutyl methacrylate.
The most commonly used compatibilisers for PA/PE blends are graft copolymers of PE
or of elastomers such ethylene-propylene rubber (EPR) and styrene-ethylene-butylene-
styrene copolymer (SEBS) with maleic anhydride or carboxyl functionality [2834].
Preparation of graft copolymers usually requires two processing steps, first the
polymerisation of the matrix and than the actual grafting, typically performed as free
radical grafting [3537]. Metallocene catalyst technology has offered a new route to the
functionalised PE [3843]. On the other hand, grafting can be combined with the blending
step, so that is one step less to perform [44, 45].
The blends are used in a broad range of applications, including electrical, (e.g., for
components subjected to low temperature impact), automotive (window and door handles,
207
Handbook of Polymer Blends and Composites
door panels, sunroof frames, fan blades and shrouds, exterior mirrors, splash, shields,
gears, and accelerator, brake and clutch pedals), tools (power tool housing, garden tools),
sport equipment (ski binding accessories, ice skate blade supports, roller skates, sailboat
components), telecommunications, furniture (office furniture parts, office chair seats,
rollers, coasters, etc.), and industrial (hydraulic cylinder components, gear, pulleys,
propellers, pump parts, pipe fittings and conveyor belt parts) [7].
PA/PP blends are also incompatible, their morphology and properties being determined
by PA type [4, 51, 52]. Because of their incompatibility such mixtures have low strength
and elongation and are quite brittle [4, 5161]. Each polymer nucleated separate
crystallisation of the other [57, 62]. This restricts its use in several new emerging
technologies. Chemical modification of PP through grafting offers an effective way for
setting up some desirable properties in the polymer, without adverse effects on the nature
of polymer backbone [63].
PA/grafted PP with other monomer blends has been studied little [21].
There are many studies that refer to the use of MA-g-PP as compatibiliser for PA/PP
mixtures [7, 66, 7277]. Here, the PP chains provide good miscibility with the PP phase,
whereas the MA group reacts with terminal NH2 group of the PA macromolecules.
Gonzales-Montiel and co-workers [71] reported that the PP domain size generally
decreases as the MA content in PP/MA-g-PP blends increases, while Duvalli and co-
workers [72] showed decreased domain size as the volume fraction of the MA-g-PP
increased in the 75/25 PA6/PP blends.
208
Polyamide-Based Blends
the adhesion between the two phases. By examining the interface morphology, it was
found that the more MA content in MA-g-PP, the stronger the adhesion between PA1010
and MA-g-PP [74].
It is well known [78, 79] that functionalised styrenic block copolymers are promising
candidates as an interfacial agent, whose non-polar olefinic mid-block can penetrate the
PP domains and whose grafted anhydride groups exhibit a strong affinity for the amine-
end groups of PA. Recent works by Gonzales-Montiel and co-workers [8082], Oshinski
and co-workers [83, 84], Wong and Mai [85] and Wilkinson and co-workers [86] have
shown that by altering the level of functionality in commercial available MA-g-SEBS,
drastic changes in morphology and performances of PA/PP blends. Rsch [87] and Rsch
and Mlhaupt [88] have suggested a core-shell morphology in PA/PP blends with
continuous PP matrix, wherein PA formed a rigid core encapsulated by an elastic shell
given by the functionalised block copolymers.
Tensile strength and stiffness increased with MA-g-SEBS content in the 75/25 PA66/PP
blends, but an opposite trend occurred in the 50/50 PA66/PP blends. Phase inversion
occurred in the 50/50 systems where PP becomes the continuous phase. Migration of
functionalised styrenic block copolymers from the PP phase to PA66 phase occurred in
both systems. An intrinsic strengthening behaviour by rigid polystyrene blocks of the
functionalised SEBS copolymer was believed to be cause for stiffening in the PA-rich
blends and weakening in the PP-continuous blends. It was concluded that PA inclusions
in SEBS-modified PP matrices did not provide any reinforcement effect because of the
interfacial adhesion.
The most popular systems are PA6, PA66, PA12, PA-mXD6 (copolyamide from
caprolactam, m-xylylenediamine and adipic acid) or PARA with PP. In these reactively
compatibilised blends, PA is the matrix, but mixtures with PP as matrix are also available.
It is worth noting that incorporation of PA into PP matrix causes a relatively small
improvement of properties, thus the type of PA is of little importance. On the other
hand, when PA is the matrix, the blend properties are intermediate between those of two
polymers; low water absorption and permeability results from incorporation of PP,
209
Handbook of Polymer Blends and Composites
whereas low oxygen permeability and high stiffness originate from the presence of PA.
The blends also show synergetic behaviour regarding flow (significant reduction of shear
viscosity facilitating blend processing, and making it possible to injection mould thin-
walled components). In blends, PA crystallisation rate can also increase, shortening the
injection cycle. Briefly, the performance of these blends depends on the type of PA and its
concentration, as well as on the type and extent of compatibilisation.
1. modification of PP by grafting onto its chains, acid or anhydride moieties, e.g., maleic,
fumaric, acrylic, methacrylic,
Some grades are reinforced with mineral filler or fibres. The filler can be selected from
talc, mica, glass fibre, etc. Furthermore, flame retardants, lubricants, antioxidants, weather
resistance improver and others can also be added.
The often-cited advantages of PA/PP blends are low moisture absorption, low mould
shrinkage, stiffness, impact and heat resistance [7]. Also, the blends show good
processability, dimensional stability, low density, i.e., low cost per volume, low liquid
and vapour permeability, good resistance to alcohols, glycols and mixture of gasoline
with ethane (gashol), improved heat ageing, primer-less paintability, resistance to cracking
when exposed to metal halides.
The PA/PP blends are formulated mainly for injection moulding and extrusion (profiles,
films, sheets), but compression, transfer or blow moulding grades are also available.
Owning to the immiscibility and relative brittleness of both polymers, the blends of PA
with polystyrene (PS), are rare and of limited value. By contrast, there are a large number
of high performing PA blends, with either statistical or block styrene copolymers. Block
210
Polyamide-Based Blends
or graft styrene copolymers can be performed and added separately or formed in situ by
reaction between co-reactive functional groups of the polymer or additive during the
melt blending process. Also, incorporation of compatibilising groups does not pose major
problems [7].
Many studies have demonstrated that effective reactive compatibilisers for PA blends
systems may consist of a functional group as anhydride [23, 66, 88, 92100], carboxylic
acid [101103], sulfonic acid [104], epoxy [105, 106] or oxazoline [107, 108]. These
functional groups are reported to react with PA functional groups during melt blending,
generating interchain bonds and in situ graft copolymers which diminish the interfacial
tension and promote a finer and more uniform distribution of the dispersed phase.
Most PA/styrenic blends belong to the toughened-PA alloys class, prepared by reactive
compatibilisation with acidified styrenic copolymers. However, because of styrene
reactivity, one can easily modify styrenic components of the blend, bringing in another
functionality such as flame resistance [7].
The morphology of PA/PS blends, where PS domains have a well defined spherical
shape with the diameter between about 2 and 8 m. The domain surface and surface of
the holes left by the PS spheres appears very smooth, suggesting poor adhesion between
the two phases, and hence the mechanical properties of such blends are poor too [7].
Although, there is the information that because of the low water absorption and reduced
melt viscosity, PA6/PS blends (with 510 wt% PS) were used for extrusion of profiles
[109].
For enhancing PA/PS blends properties one resorts to compatibilisers such as: maleic
anhydride graft polystyrene [110], p-carboxylated polystyrene (CPS) [103], styrene-maleic
anhydride copolymers (SMA) or zinc sulfonated polystyrene ionomers [ZnSPS] [110,
111]. Blends are constituted from PA (either PA6, PA66 or PA1010) with PS and the
compatibilisers specified above are microheterogeneous and characterised by good
mechanical properties.
The miscibility of PA with SMA depends on the acrylonitrile (ACN) content from styrenic
copolymers and the molecular weight of blend components [99].
The studies have demonstrated that melt blending PA with SMA, amide and amino groups
of PA reactions with the anhydride groups of SMA give rise to formation of graft
copolymer (X) and have a compatibilising effect. The ternary blends PA6/X/SMA have
been studied by Dedecker and co-workers [99, 112, 113] and Paul and co-workers [114,
115]. In PA/SMA blends, the SMA copolymer does not have to diffuse towards the
interface as in the case of PA/X/SMA blends. The addition of SMA to PA leads to an
increase of the melt viscosity [116].
211
Handbook of Polymer Blends and Composites
The blend of vinyl oxazoline modified PS with amorphous PA has been studied by Briber
and co-workers [117]. Recently, Steurer and co-workers [118] have studied the blend
phase morphology and interfacial graft copolymer formation in blends PA12/SMA.
The blends created from high-impact polystyrene (HIPS) and PA1010 have been intensely
studied by Chen and co-workers [98, 100, 110]. They emphasised that PA1010/HIPS
blends are macroheterogenous because of system component incompatibility. With a
view to enhancing the properties of these blends, they were compatibilised using maleic
anhydride grafted HIPS (MA-g-HIPS). The insertion of MA units in HIPS was very effective
for controlling the morphology of PA1010/HIPS/MA-g-HIPS blends. The domain size of
dispersed phase of ternary blends decreased with the increase of MA-g-HIPS content in
the blend. The properties of PA1010/HIPS/MA-g-HIPS blends are superior to binary
PA1010/HIPS blends, owning to the interaction between MA-g-HIPS and PA1010. The
compatibilised blends of PA and HIPS have been used to manufacture the liners of
refrigerators and freezers with HCFC-1416 blowing agent [7, 100].
The styrene-butadiene (SB) (or isoprene) di-, tri-, or multiblock copolymers, SEBS and
styrene-methacrylic acid copolymers are immiscible with PA. In order to compatibilise
PA/styrene block copolymers blends one resorts to their modification by grafting with
maleic anhydride (MA-g-SB; MA-g-SBS, MA-g-SEBS). Mixtures composed from PA (either
PA6 or PA66) and maleated styrenic block copolymers showed super-toughness, and
could be readily formed using any conventional method, such as extrusion, injection
moulding, blow moulding, pressure forming, rotational moulding articles having various
kinds of shaped. The articles were used for automobile electrical or mechanical parts,
packaging or building materials [119].
212
Polyamide-Based Blends
blends with an excellent balance of stiffness and toughness [120]. There is significant
interest in developing PA/ABS blends, with the goal to combine desirable properties of
the blend constituents. However, simple blends of PA with ABS have poor mechanical
properties [120]. Moreover, these mixtures undergo delamination during processing.
These drawbacks can be avoided by PA/ABS blends compatibilisation [120].
Several approaches to the reactive compatibilisation of PA/ABS blends have been reported
in the recent literature [121131]. The preferred compatibilisation strategy has been to
incorporate a functional polymer capable of reacting with the amine end groups of the
PA that is also miscible with SAN phase of ABS.
For better heat and solvent resistance it is preferable to keep PA as continuous phase.
Hence, the key to success has been optimisation of the amount and type of compatibiliser
to generate co-continuous morphology with appropriate fineness of dispersion.
The easiest way to compatibilise PA/ABS blends is functionalising the styrenic copolymer.
Thus, blends of PA6 with ABS containing 0.515 wt% unsaturated carbonamide, e.g.,
acrylamide, are characterised by high HDT and good impact strength. These alloys are
commercially promoted for use in automobile body panels. But often, ABS functionality
is achieved using MA or unsaturated carboxylic acids [7]. PA/maleated ABS mixtures
have co-continuous morphology and are characterised by good mechanical properties
and impact strength.
213
Handbook of Polymer Blends and Composites
In 1991 DuPont de Nemours disclosed blends comprising 4060 wt% of either PVC or
PVDC and 4060 wt% of PA, processable at a temperature below 220 C, such as PA6,
PA1212 or PARA. The blends were compatibilised by addition of 120 wt% of an acrylic
copolymer [ethylene-alkyl(meth)acrylate-vinyl acetate copolymer; (meth)acrylic acid-
carbon monoxide copolymer modified with succinic anhydride]. In the resulting alloys,
chlorinated polymers were the continuous phase and PA was a finely dispersed phase.
The blends have been processed by extrusion or injection moulding into films, profiles,
and mouldings such as containers. The products showed good notched Izod impact
strength, elongation and tensile strength [7].
Polyamides (either PA6 or PA12) are immiscible with PVDF, inspite of some specific
intermolecular interactions [132, 133]. For PA12/PVDF blends compatibilisation,
polyetheramide block copolymers were used [134, 135].
The studies have demonstrated that PA6 and PVDF do not co-crystallise over the entire
composition range. The crystallinity of PVDF is not very significantly changed by the
PA6 addition, but the crystallinity of PA6 goes through a minimum when the PVDF
content is increased [132].
Although PA6 and PVDF blends are immiscible over the whole composition range, strong
interactions occur at the interface, which accounts for a very fine phase morphology and
significant improvement in the ultimate tensile strength and elongation at break compared
to predictions by the additive laws. The effect of the interfacial adhesion on the impact
strength is complex and suggests that an optimum in the interfacial adhesion is critical
for the toughening of PVDF by PA6 and synergism in this property [133].
General aspects regarding PA/acrylics blends are discussed in Chapter 9. In the following
sections only PA toughened blends are referred to, constituted from PA and acrylic
copolymers. A great diversity of these blends has been developed. The early ones were
214
Polyamide-Based Blends
based on polyalkenes grafted with (meth)acrylics (see 6.1.3), the later ones on the
multicomponent core-shell latex technology. Since compatibilisation interactions between
the copolymer and PA are very important, in most cases the acrylics have either acidic,
acid anhydride or epoxy groups.
The shell phase should preferably have a Tg 60 C. The carbonyl groups contained in
the shell may be in the acid or salt form, i.e., -COOH or an alkali metal, alkaline earth
metal or ammonium salt. The preferred average particle size of the core-shell polymer
can be recovered from the latex by spray drying then extruding [7].
The early core-shell polymers developed for improving the impact strength of PA had in
the shell polymers with carboxyl groups. Blends of these with PA6 showed good notched
Izod impact strength values, but their efficiency for improving the PA66 performance
was low and the melt viscosity was high, which resulted in poor mouldability. To
ameliorate this situation, several blends were prepared in two steps, first reactively blending
the acidified core-shell copolymer with low-molecular weight PA, then in the second step
combining the prepared compatibilised impact modifier with the main, high-molecular
weight PA.
Polyamides (either PA6, PA66, PA11, PA610, PARA) can be blended with a wide range
of acrylic copolymers, such as ethylene-acrylic acid, ethylene-2-hydroxyethyl methacrylate-
methyl methacrylate, ethylene-methacrylic acid, ethylene-ethyl acrylate, ethylene-vinyl
acetate-acrylic acid, methyl methacrylate-butyl acrylate-2-hydroxyethyl acrylate, styrene-
maleic anhydride copolymers, acrylic acid grafted EVA, graft copolymer of acrylic acid,
2-hydroxyethyl acrylate, butyl acrylate, maleic anhydride, acrylamide onto PE or EVA,
215
Handbook of Polymer Blends and Composites
Depending on necessity, such blends can incorporate stabilisers, pigments, dyes, fillers or
reinforcements, etc.
Making a general analysis of this extremely wide range of PA/acrylics blends it can be
emphasised that are characterised by various properties like: toughness, tensile strength,
excellent impact strength, high flow rate and high melt strength, high HDT, oxidation
resistance, good surface properties.
PA/acrylic copolymers blends can be processed by extrusion, injection and blow moulding,
the manufactured products founding employment in automotive, electrical, electronics,
consumer and sport industries.
The fact that most of industrial manufactured PA (PA6, PA66, PA610, PA11, PA12,
etc.), are characterised by relative reduced impact strength at low temperatures, limit
their use in specific applications [8]. The low temperature impact strength of PA has
been enhanced by incorporation of such elastomers as ethylene-propylene-carbon
monoxide copolymer (COPO), carboxy-terminated nitrile rubber (CTNBR), butyl rubber
(IIR), epichlorohydrin elastomer, acrylic elastomer, styrene-butadiene di-, tri- or multiblock
copolymers (SB, SEBS), polyurethane rubber, thermoplastic elastomers (TPE).
The elastomer is usually selected by considering the desired morphology and the method
of compatibilisation. The type of morphology, size and size distribution of the dispersed
phase, as well as the degree of bonding the disperse phase and the matrix, all are critical
factors. The particle size and distribution are of prime importance since they control the
fracture mechanics, and thus are responsible for the toughness and strain at break. The
physical and chemical bonding can led to significant changes of the impact strength,
even at higher elastomer content. The elastomer modified PA exhibit resistance to initiating
and propagating stresses.
216
Polyamide-Based Blends
Some blends (PA/SB, PA/SEBS and PA/acrylic elastomers) properties have been already
presented (see 6.1.2 and 6.1.4).
Blends of PA with EPR or EPDM are highly immiscible and incompatible, thus the mechanical
properties of such blends are relatively limited too [136]. This is because the interfacial
tension between the two polymers is extremely high on account the high polarity difference.
The morphology of immiscible PA6/EPR rubber blends indicated a two phases structure
in which the high viscosity EPR phase was dispersed as domains in the continuous low
viscosity PA matrix, up to 40 wt% of the EPR concentration. Interpenetration co-
continuous phase morphology was obtained for 50/50 and 60/40 EPR/PA6 compositions.
At higher EPR concentration (> 60 wt%) the PA phase was dispersed as domains in the
continuous EPR phase. The size of the dispersed phase was found to increase marginally
with the increasing concentration of the phase that is associated with coalescence. The
coalescence was predominant only at higher concentration of the dispersed phase [136].
The properties and morphology of PA6/EPDM blends does not differ essentially from
PA/EPR blends.
To modify the mechanical properties and morphology of PA/EPR and PA/EPDM blends,
they were compatibilised using either elastomer functionalising (by grafting with MA or
unsaturated carboxylic acids) or by adding compatibilisers (often ethylene-propylene
copolymers grafted with MA) [137139].
Paul and co-workers [83, 84, 140144], studying PA6/maleated ethylene-propylene rubbers
(MA-g-EPR) have established that increasing the content of PA6 from 5 to 30 wt%, the
average size of the dispersed PA6 particles in the matrix of MA-g-EPR increased from
0.14 to 0.23 mm, while the hardness, modulus, and tensile strength of the blend increased
too. The observed values of the modulus are in reasonable agreement with those predicted
by a theoretical model. As the content of PA6 increased from 30 to 50 wt%, the physical
properties of the blends quickly changed because of phase inversion, i.e., the PA6 becomes
the continuous phase with spherical dispersed particles of MA-g-EPR.
The blends with an MA-g-EPR continuous phase have lower strength than commercial
TPE and show stress softening which indicates that PA6 phase does not strongly reinforce
the MA-g-EPR matrix. The blends based on a PA6 with reduced molecular weight made
by an hydrolysis process show somewhat improved strength and a reduced PA particle
size. The addition of magnesium oxide to these blends causes significant improvement of
in tensile properties. This may be the result of the reduced particle size caused by the
increase in the melt viscosity or the formation of ionic cluster type crosslinks [140].
217
Handbook of Polymer Blends and Composites
The domain size of the dispersed phase in the PA6/EPR/MA-g-EPR blends decreased
with increasing concentration of compatibiliser followed by levelling off at higher
concentration (>5 wt%), which is an indication of interfacial saturation. The reduced
domain size as a function of compatibiliser content collapsed into a single master curve
for different PA6/EPR blend compositions. This was accounted for by the fact that the
area occupied by a compatibiliser molecule at the interface is the same in all the blends
irrespective of the blend composition [136]. In the reactively compatibilised blends, the
size of the dispersed EPR phase was found to decrease with the increasing molecular
weight of the PA matrix. The fracture surface morphology of the reactively compatibilised
blends indicated excellent adhesion between phases, as shown by the absence of microvoids
irrespective of the nature of the dispersed phase (EPR or PA) [136].
Blending with 540 wt% acrylonitrile-butadiene rubber (CNBR) and glass fibres improved
the impact properties of PA, such as PA6. Note that here the carboxylic groups provided
an excellent source of interaction with PA, whereas the glass fibres directly bonded to PA
segments. PA blends with CNBR and ZnO in the full range of compositions are reported
to have excellent fracture resistance.
The blends constituted from PA6, COPO, and hydrogenated SBS block copolymer showed
excellent performance, such as good balance of strength and toughness.
In an immiscible blend of PA1010 as matrix and butyl rubber as dispersed phase, the use
of ethylene-methacrylic acid-isobutyl acrylate copolymer as compatibiliser can improve
the interfacial adhesion between the two phases and the dispersion of butyl rubber domains
in the PA1010 matrix. As result, the improvement in the notched impact strength of the
blend has been achieved by the addition of compatibiliser.
218
Polyamide-Based Blends
of these blends showed that the PA6 is the continuous phase while the elastomers form
domains. For all compositions, the domain size and their distribution are not
homogeneous. This is probably caused by the poor mixing conditions and by the
simultaneous formation of the graft copolymer. The degree of grafting depends on the
blend composition and the elastomer type. Blends of ECO present larger domains size
than for the corresponding blends of PEPI. In addition, the interfacial region in the
blends of ECO is larger, which is consistent with the higher grafting degree [145].
The toughened blends have improved chemical resistance, high notched impact strength
at room and low temperature, and a HDT of 5075 C; can be obtained by mixing PA
(PA6, PA66, PA12) with block or graft copolymer formed by coupling an elastomer
(either EPDM, SB or nitrile rubber) with unsaturated (di)carboxylic acid, an anhydride
of an dicarboxylic acid, monoester, sulfonic acid and an epoxy or their salts). In these
blends PA are the continuous phase and the grafted elastomer the dispersed one. The
latter had particle diameters 3.50.35 m, with microparticles of PA imbedded in it [7].
It was reported that blend properties could be further enhanced by addition of such
inorganic fillers as talc or glass fibre.
The high impact strength PA/elastomer blends can be extruded (for manufacturing films,
oriented fibres, tubing, wire coatings, etc.), and injection moulded without pre-drying.
The melt temperatures range from 260290 C. These blends are used for many
applications, namely for ski bindings, ski and roller boots, surfboard mast holders, bicycle
wheels, curtain rails, office equipment housing, handles, electrical components, gears,
safety helmets, seals, sleeves, flexible couplings, cable sheathing, wheel trim and front
spoilers for cars, cattle troughs, bicycle saddles, swivel chair basis, radiator hoses, etc. [7].
219
Handbook of Polymer Blends and Composites
TPU 1 was mixed with PA12, the miscibility was improved owing to the close size of the
Kuhns segments in the polyurethane soft block and PA.
Therefore, although thermodynamic preconditions are less favourable with TPU 1 than
with TPU 2 [solubility parameter of TPU1 soft block and PA12 differ by 4.7 (MJ/m3)0.5],
a partial compatibility in TPU 1/PA12 blends can be achieved at wider variations in the
component contents than in TPU 2/PA12 blends. Partial compatibility of PA12 with
TPU soft block is followed by the latter plasticisation that shifts by 15 C the Tg of the
soft block to the lower temperature region. PA12 was crystallised in both tested blends
without much variation of the PA crystal lattice parameters. It may be assumed that only
the amorphous portion of PA or interphase layers with irregular arrangement participle
in interphase interaction with TPU. Intensive interphase interactions lead to slower
crystallisation of PA12 in the blends where it is the dispersion phase, and, in contrast, to
faster crystallisation if PA is the polymer matrix.
Addition of PA12 to TPU provides wide variation in the polyurethane mechanical strength
and hardness. At all concentrations of PA12 (up to 20 30 wt%), mechanical strength
values for the blends were higher than additive ones. Blends containing TPU of 30
50 wt% belong to the impact resistant group: TPU 1/PA12 compositions showed higher
impact resistance at low temperature (40 C) testing, than TPU 2/PA12; this can be
explained by lower Tg of TPU 1 soft block as compared with TPU 2 [146].
6.1.7 PA/Santoprene
Some PA/elastomers blends possess characteristics of TPE. The choice of TPE over
conventional elastomers is due to the following advantages [147, 148]:
4. damage tolerance;
6. thermal stability;
220
Polyamide-Based Blends
PA/TPE blends also turned out to be very interesting, from which, PA66/Santoprene
blends are most studied [149, 150].
The research of blending of PA66 with Santoprene has revealed that both components
are immiscible, but good interaction occurs during blending, based on the results of
thermal and mechanical properties. The blends showed double melting temperatures,
corresponding to those of the PP component in Santoprene and PA66 irrespective of
blend compositions. The increase in crystallisation temperature of PA66 was due to the
EPDM component in Santoprene, which acts as a nucleating agent. A significant effect
of the blend composition on mechanical properties occurred at around 50/50 composition.
The SEM micrographs of the PA66/Santoprene blends revealed two-phase structure
confirming phase segregation of the two components especially for PA66-rich
compositions. However, it may be concluded that the addition of as little as 5% SAN
compatibilise the blend, and the interfacial adhesion of the components is improved
above 75 wt% composition as was revealed by SEM micrographs [150].
There is a large body of literature concerning PA (1)/PA (2) blends, where the two polymers
are chemically different [7, 152155]. The wide range of such blends can be divided in
two groups: semicrystalline PA (PA6, PA66, PA610, PA11 or PA12/semicrystalline PA),
and semicrystalline PA/amorphous aromatic PA.
From the first group of blends, the most studied are those with PA66 as base polymer.
From blends of PA66 with PA6, PA610, PA612, PA11 or PA12 materials were obtained
with improved processability over that of the component polymers, good solvent
221
Handbook of Polymer Blends and Composites
resistance, elongation, low temperature impact and tensile strength, as well as enhanced
barrier properties against permeation of gases and liquids, printability, clarity and
dimensional stability. Addition of a compatibiliser (sulfonated EPDM, methyl
methacrylate-styrene-maleic anhydride or styrene-methacrylic acid copolymers), a mineral
filler or aminosilane-treated glass fibres assure a supplementary improvement of the
mentioned properties and the increase of the melt viscosity, making it more suitable for
vacuum forming. These blends are commercially available for injection moulding and
extrusion. They are used in the manufacture of safety helmets, housing for electrical and
electronic parts of chassis, pulleys, fans, lock parts, levers, clamps, etc.
The blends constituted from semicrystalline PA (PA6, PA66, PA610, PA11 or PA12)
and amorphous aromatic PA [polycondensates of terephthalic, isophthalic, adipic or
azelaic acid with hexamethylene diamine, trimethyl hexamethylene diamine, m-xylylen
diamine or bis(p-aminocyclohexyl)methane] were reported to have good resistance to
heat, chemicals, oil, impact strength and toughness, good stress corrosion as well as
rigidity up to a temperature of 100 C. These blends are available for the production
of tough, high temperature, food packaging films with good barrier properties against
permeation of oxygen and different parts for automotive, electrical and electronic
industry [7].
In this case too, the polyamide-based blends can be associated with compatibilisers
and fillers, giving raise to an improvement of processability characteristics and
mechanical properties.
Although, in many cases, both PA and PEST exhibit ductile behaviour in both tensile
strength and impact properties, non-oriented PA/PEST blends reveal brittle performance
under the same test conditions [164166]. Pillon and co-workers [162] have suggested
two possible mechanisms for the decreased mechanical behaviour of these blends: (1)
poor compatibility leading to weak interphase actions and (2) enhanced crystallinity
of the PEST because of increased nucleation by PA crystalline domains.
222
Polyamide-Based Blends
The presence of polar functional groups on PA and PEST offers the possibility of both
strong specific interactions and chemical reactions between the dissimilar polymers.
Therefore, reactive compatibilisation methods to increase the interfacial bonding between
components have been of recent interest [161163]. For example, compatibilisation of PA/
PEST blends has been accomplished by in situ copolymer formation via amide-ester
interchange reactions (trans reactions) during melt processing [157163]. However, to
obtain a mechanically compatibilised blend, long processing times and/or added catalyst
are generally required for sufficient copolymer formation. As an alternative to the formation
of amide-ester copolymers, the use of specific interactions, i.e., ionic or
H-bonding, may be used to compatibilise immiscible PA/PEST blends [167]. Depending on
the functional groups present, several types of specific interactions can be promoted: acid-
base [168, 169], hydrogen-bonding [170172], charge transfer complexes [169], ion-dipole
[173175], ion-ion [173, 176, 177], and metal coordination [178].
Among PA/PEST blends, the most studied are PA/polyethylene terephthalate (PET), PA/
polybutylene terephthalate (PBT) and PA/polyarylates (PAr) blends.
As a result of the incompatibility between PA6, PA66 and PET, the blends yielded are
heterogeneous and characterised by low strength and elongation at break and brittle
behaviour. The morphology of such blends depends on the components ratio, phase inversion
taking place at a ratio of about 1/1 PA6/PET. Turned into films, stretched and then quenched
in liquid nitrogen, PA6/PET blends acquire fibrillar structure with PA6 dominant amorphous
matrix, reinforced by the preserved PET microfibrils [179]. The transreactions between
PA6 and PET lead to an improvement of the adhesion between matrix and microfibrils
resulting in high microhardness values.
Adding small proportions of PA66 (0.0510 wt%) in PET results in better crystallisation
rate and overall crystallinity in PEST, as well as decrease of spherulite size. These blends
were reported to have improved mouldability, rigidity, strength and fire resistance.
223
Handbook of Polymer Blends and Composites
attributed, for the most part, to the presence of specific interactions between the metal
counter ion of the ionomer and the amide groups [180].
In the early 1980s, reactive extrusion was introduced for compatibilisation of PA/PEST
blends. Two approaches have been used compatibilisation was achieved either by acid-
catalysed amide-ester exchange reaction or by direct coupling. For the first method were
used as compatibilisation agents ,-dimethylol propionic acid, acetobenzoic acid,
acetoaminobenzoic acid or p-toluensulfonic acid.
Although PA (PA6 or PA66) are incompatible with PBT, PA/PBT blends obtained in
molten state are highly crystalline materials with a finely dispersed minor phase. Using
this method, the components were likely to undergo mechanical degradation and
recombination. PA6/PBT block copolymers were found to form by segmental exchange,
whereby compatibility was enhanced in the system [157161, 166, 182184].
PA/PBT blends give raise to polymeric materials with good mouldability, rigidity, strength,
and fire resistance.
The properties of PA/PBT blends can be varied by adding compatibilisers to the system.
Thus, PA66/poly(m-xylenyl adipamide)/PBT blends were reported to have good resistance
to heat, chemicals, oil, high impact strength and toughness. The PA6/PBT blends
compatibilised with styrene-glycidyl methacrylate (SGMA), styrene-maleic anhydride
copolymers or epoxy compounds showed high impact strength.
Polyarylates blended with 50 wt% PA6 and 2.5 wt% PET give mouldings with particularly
good processability, high glass, chemical, mechanical, electrical and HDT [7].
224
Polyamide-Based Blends
Polycarbonate (PC) is an amorphous polymer with a high Tg (148 C), sensitive to moisture
and durable to various weathering conditions [185].
PA/PC blends are incompatible and immiscible in the whole range of composition and
temperature, their melt viscosity is low; they are susceptible to thermal and hydrolytic
degradation [185, 186]. Moulding resulted in brittle points, having poor mechanical
properties because of poor adhesion in the interface [184].
The PC of bisphenol A has been most frequently used in blends with PA, but blends of
PA with polyestercarbonates (PEC) have been realised too.
The first PA/PC blends date from 1973, when have been realised PC blends with 10
20 wt% PA12. These materials, with pear-like glass, had good high voltage arc and anti-
tracking properties, making them suitable as electric insulators.
In order to enhance and diversify PA/PC blend properties various compatibilisers were
used such as maleated polyalkene (maleated PP or EPR), polyesteramide [187], ABS and
maleated ABS (MA-g-ABS), butadiene-styrene-acrylonitrile-acrylate copolymer, SMA,
225
Handbook of Polymer Blends and Composites
For the PA6/PC blends, Horiuchi and co-workers [188191] concluded that the dramatic
improvement of the impact strength can be achieved in 15/5 PA6/PC blends, when the
total amount of added MA-g-SEBS and SEBS is held constant at 20 phr and the ratios of
MA-g-SEBS or SEBS are varied. These results were attributed to the encapsulation by
MA-g-SEBS around PC domains in the PA matrix.
Depending on the nature of compatibilisation agents, PA/PC blends show a wide range
of properties such as good processability, good stability of morphology during processing,
increased toughness, good resistance to oils and boiling water, heat stability and flexibility,
solvent resistance, dimensional stability, high modulus, elongation and impact strength [7].
The PA/PC blends have been developed for use in electric insulation (1020 wt% PA in
PC), for easy to spin PA fibres (515 wt% PC in PA), for high mouldability (PA:PC =
3:2 to 1:100), for good melt-adhesive properties (540 wt% of PC in PA).
The PA/POM blends are immiscible. In these blends, interactions between carbonyl and
NH groups of the pure PA compete with intermolecular interactions between NH
groups of the POM. For this reason, the situation in PA/POM blends is more complicated
compared with other blends [192194].
In the PA/POM blends with component ratios of either 19:1 or 1:19, for good mechanical
performance, the size of dispersed second polymer must be smaller than 10 m. This
condition is hard to realise without adding a compatibilisation agent to the system. To
improve modulus, up to 40 wt% of an inorganic powder can be added. The reinforcing
effect caused by addition of the solid particles has been small, POM does not easily wet,
thus bonding between the resin and filler particles is notoriously poor. Furthermore,
addition of filler reduced the natural lubricity of the POM and limited usefulness of the
reinforced blends.
The POM/PA (PA6, PA66 or PA610) blends were reported to have improved colour and
thermal stability.
226
Polyamide-Based Blends
The stability of these blends can be improved by mixing with 1.1 wt% polyethylene
glycol. The blends of either PA6 or PA66 with POM and with PET were reported to have
good frictional resistance.
The blends of POM with PA have a low melting point, 230 C or less (i.e., PA10, PA12,
PA12, PA6 or PA610), and filler were characterised as being excellent in high temperature,
stiffness and lubricity, with good colour tone [7].
Polyamide blends with liquid crystalline polymer (LCP) are discussed later in this book,
together with other blends (see Chapter 16).
Polysulfones (PSU) are a class of polymers which show high stiffness, hardness, and thermal
stability at temperatures from 100 to +150 C. They are transparent and have good
resistance to radiation, as well as high flame resistance. On the other hand, they require
high processing temperatures, and even at these temperatures they have high melt viscosity,
thus are difficult to process. PSU also show susceptibility to stress cracking with certain
solvents [7, 195]. Blending with PA aims to amend these reminded characteristics.
Although any type of PSU is incompatible with PA, PA/PSU blends have interesting
properties [7]. Thus, PA11/PSU were found to be easy to process into films or fibres and
to have resistance towards folding. Furthermore, injection moulded parts showed good
impact properties. The PA46/polyethersulfone (PESU) showed good balance of mechanical
and thermal properties. The blends comprising PARA and PESU have high HDT, high
strength, and a good set of mechanical properties [7].
Weber and co-workers [196200] recently carried out reactive blending of PA6 with
PSU and found that one can control PSU particle size in a wide range (140 nm), using
various functional groups attached to the PSU chain such as epoxide, carboxylic acid,
MA and phthalic anhydride. Compared with non-reactive systems, the reactive one can
generate PA-PSU graft or block copolymers, which would act as an emulsifier to prevent
particle coalescence. Significant improvement of toughness was found in PA6/phthalic
anhydride-terminated PSU (PSU-PhAS) system with finest morphology. Moderate
improvement was achieved in other reactive systems. High toughness in PSU-PhAS blend
seems to originate from the enhanced interfacial adhesion, which provide adequate stress
transfer to induce the cold drawing of the brittle PSU particles.
227
Handbook of Polymer Blends and Composites
Including glass fibres in PA/PSU blends gives rise to supplementary improvement of the
properties of these blends [200].
The linear polyphenylene sulfides (PPS) are highly crystalline with Tm in the range 285-
295 C and quoted values for Tg range from 85 C to 150 C. PPS have high strength,
hardness, stiffness, HDT and good dimensional stability, as well as excellent chemical,
solvent and flame resistance. The polymers may also be vulcanised by heat and/or sulfur.
The PA/PPS blends combine the properties of the two classes of polymers. Thus, PA
(PA6, PA66) or polyphthalamide/PPS blends are reported to have good mechanical
properties, HDT, dimensional stability and chemical resistance.
Adding bronze powder, glass and/or carbon fibres or magnesium hydroxide to PA/PPS
blends gives polymeric materials with high degradation temperature, chemical resistance,
excellent tensile strength, as well as arc and tracking resistance [7].
The most important commercial PA-based blends are presented in Table 6.1.
228
Commercial polyamide blends
Polyamide-Based Blends
229
Commercial polyamide blends
Handbook of Polymer Blends and Composites
230
Commercial polyamide blends Polyamide-Based Blends
231
Commercial polyamide blends
Handbook of Polymer Blends and Composites
232
Commercial polyamide blends Polyamide-Based Blends
233
Handbook of Polymer Blends and Composites
References
5. Nylon Grades and Properties, Brochure, Ube Industrie, Ltd., Japan, 1985.
7. L.A. Utracki, Commercial Polymer Blends, Chapman and Hall, London, UK,
1998.
10. Handbook of Polyolefins, 2nd Edition, Ed., C. Vasile, Marcel Dekker, Inc., New
York, USA, 2000.
11. Y. Jois and J. Harrison, Journal of Macromolecular Science C, 1996, 36, 3, 433.
12. B. Kyn, S. Yun and S. Jin, European Polymer Journal, 1991, 27, 349.
13. H. Raval, S. Devi, Y. Singh and M. Metha, Polymer, 1992, 36, 3, 433.
14. G. Fairley and R. Prudhomme, Polymer Engineering and Science, 1987, 27,
1495.
15. R.A. Kudva, H. Keskkula and D.R. Paul, Polymer, 1999, 40, 6003.
234
Polyamide-Based Blends
18. J. Pan and W.J.D. Show, Macrostructure Science, 1993, 20, 351.
19. J. Pan and W.J.D. Show, Macrostructure Science, 1994, 21, 95.
21. L. Lamas, G.A. Mendez, A.J. Mller and M. Parcella, European Polymer Journal,
1998, 34, 2, 1865.
22. C.C. Chen, E. Fontan, K. Min and J.L. White, Polymer Engineering and Science,
1988, 28, 1, 69.
23. G. Serpe, J. Jarrin and F. Davas, Polymer Engineering and Science, 1990, 30,
553.
24. H. Raval, S. Devi, Y.P. Sing and M.H. Metha, Polymer, 1991, 31, 493.
27. A. Valenza, G. Geuskens and G. Spadaro, European Polymer Journal, 1997, 33,
6, 957.
29. D.R. Paul in Polymer Blends, Eds., D.R. Paul and S. Newman, Academic Press,
New York, NY, USA, 1978, 2.
31. R. Fayt, R. Jerome and P. Teyssie, Journal of Polymer Science, Part C: Polymer
Letters, 1981, 19, 79.
32. R. Fayt, R. Jerome and P. Teyssie, Journal of Polymer Science, Part C: Polymer
Letters, 1986, 24, 25.
235
Handbook of Polymer Blends and Composites
33. S. Wu, Polymer Interface and Adhesion, Marcel Dekker, Inc., New York, NY,
USA, 1982.
35. S. Datta and D.J. Lohse, Polymeric Compatibilisers, Hanser Publishers, Munich,
Germany, 1996.
37. U. Antilla, C. Vocke and J. Seppl, Journal of Applied Polymer Science, 1999,
72, 877.
40. K. Hakala, B. Lfgren and T. Helaja, European Polymer Journal, 1998, 34, 1093.
41. C.E. Wilen and J.H. Nsman, Macromolecules, 1994, 27, 4051.
42. M.J. Schneider, R. Schfer and R. Mlhaupt, Polymer, 1997, 38, 2455.
45. Y.J. Sun, G.H. Hu, M. Lambla and H.K. Kotlar, Polymer, 1996, 37, 4199.
46. C.C. Chen and J.L. White, Proceedings of SPE ANTEC 91, Montreal, Canada,
1991, 49, 969.
47. B.K. Kim, S.Y. Park and S.J. Park, European Polymer Journal, 1991, 27, 349.
48. A.R. Padwa, Polymer Engineering and Science, 1992, 32, 1703.
236
Polyamide-Based Blends
51. M.I. Kohan, M. Hewel, H. Dalla Torre, H.A. Scheetz, H. Keskkula, D.R. Paul,
F.J. Rietmaijer, G. Stppelmann and P.N. Richardson in Nylon Plastics
Handbook, Ed., M.I. Kohan, Hanser Publishers, Munich, Germany, 1995, 301.
52. R.D. Deanin, M.A. Manion, C.H. Chung and K.N. Tejeswi in Handbook of
Polyolefins, 2nd Edition, Ed., C. Vasile, Marcel Dekker, Inc., New York, NY,
USA, 2000, 613.
53. M.K.S. Moon, B. Ryoo and J. Park, Journal of Polymer Science, Part B: Polymer
Physics, 1994, 32, 1427.
54. T. Tang and B. Huang, Journal of Applied Polymer Science, 1994, 53, 355.
56. T. Tang, Z. Lei, X. Zhang, H. Chen and B. Huang, Polymer, 1995, 36, 5061.
57. B. Lotz and J.C. Wittmann, Journal of Polymer Science, Part B: Polymer Physics,
1986, 24, 1559.
60. I. Grof, O. Durcova and M. Jambrich, Colloid and Polymer Science, 1992, 270, 22.
62. S.H. Jafari and A.K. Gaupta, Journal of Applied Polymer Science, 1999, 71,
1154.
63. D.L. Xie, D. Chen, B. Jiang and C.Z. Yang, Polymer, 2000, 41, 3599.
65. M.K. Naqvi and M.S. Choudhary, Journal of Macromolecular Science C, 1996,
36, 3, 601.
66. F. Ide and A. Hasegawa, Journal of Applied Polymer Science, 1974, 18, 963.
67. R. Rengarajan, V.P. Parameswaran, S. Lee and P.L. Rinaldi, Polymer, 1990, 31, 1703.
237
Handbook of Polymer Blends and Composites
70. H.C. Chung and C.D. Han, Polymer Blends and Composites in Multiphase
Systems, American Chemical Society, Washington, DC, USA, 1984.
73. H. Li, T. Chiba, N. Higashida, Y. Yang and T. Inoue, Polymer, 1997, 38, 15, 3981.
74. X. Zang, G. Li, D. Wang, Z. Yin, J. Yin and J. Li, Polymer, 1998, 39, 1, 15.
75. J.D. Lee and S.M. Yang, Polymer Engineering and Science, 1995, 35, 23, 184.
76. S.N. Sathe, S. Devi, G.S.S. Rao and K.V. Rao, Journal of Applied Polymer
Science, 1996, 61, 1, 97.
77. R.M. Holsti-Miettinen, K.P. Perttila, J.V. Seppl and M.T. Heino, Journal of
Applied Polymer Science, 1995, 58, 9, 1551.
78. M.J. Modic, Proceedings of SPE ANTEC 93, New Orleans, LA, 1993, USA, 51,
205.
79. M.J. Modic and L.A. Pottick, Polymer Engineering and Science, 1993, 33, 819.
80. A. Gonzales-Montiel, H. Keskkula and D.R. Paul, Polymer, 1995, 36, 4587.
81. A. Gonzales-Montiel, H. Keskkula and D.R. Paul, Polymer, 1995, 36, 4605.
82. A. Gonzales-Montiel, H. Keskkula and D.R. Paul, Polymer, 1995, 36, 4621.
83. A.J. Oshinski, H. Keskkula and D.R. Paul, Polymer, 1992, 33, 268.
84. A.J. Oshinski, H. Keskkula and D.R. Paul, Polymer, 1992, 33, 284.
85. S.C. Wong and Y.M. Mai, Polymer, 1999, 40, 1553.
86. A.N. Wilkinson, L. Langel, M.L. Clemens, V.M. Harding and M. Marin,
Polymer, 1999, 40, 4971.
238
Polyamide-Based Blends
89. A. Valenza, F.P. LaMantia, G.B. Demma, V. Romano and D. Aciero, Polymer
Networks and Blends, 1991, 1, 71.
90. Q. Liu, J.M. Bronk, A. Verma and J.S. Riffle, Polymer Materials Science and
Engineering, 1991, 64, 158.
91. S. Nosoda, K. Kojima and A. Ayoagi, Polymer Networks and Blends, 1991, 1,
51.
92. J. Angola, Y. Fujita, T. Sakai and T. Inoue, Journal of Polymer Science, Part B:
Polymer Physics, 1988, 26, 807.
95. C. Scott and C. Macosko, Journal of Polymer Science, Part B: Polymer Physics,
1994, 32, 205.
96. Y. Kayano, H. Keskkula and D.R. Paul, Polymer, 1997, 38, 1885.
97. S.C. Tjong and Y. Meng, Polymer International, 1997, 42, 209.
98. G. Chen, J. Yang and J.J. Liu, Journal of Applied Polymer Science, 1999, 71,
2017.
100. G. Chen and J.J. Liu, Journal of Materials Science, 1999, 34, 4375.
101. W.J. Macknight, R.W. Lenz, P.V. Masto and R.J. Somani, Polymer Engineering
and Science, 1985, 25, 1124.
102. Z. Liang and H.L. Williams, Journal of Applied Polymer Science, 1992, 44, 699.
104. H. Li and Z. Li, Journal of Applied Polymer Science, 1998, 67, 61.
239
Handbook of Polymer Blends and Composites
105. H.C. Chin, K.C. Chiou and F.C. Chang, Journal of Applied Polymer Science,
1996, 60, 2503.
106. C.C. Huang and F.C. Chang, Polymer, 1997, 38, 3185.
107. N.C. Tan Beck, E. Fortan, K.E. Min and J.L. White, Polymer Engineering and
Science, 1988, 28, 69.
110. G. Chen and J. Liu, Journal of Applied Polymer Science, 1999, 74, 857.
114. V.J. Triacca, S. Zialeel, J.W. Barlow, H. Keskkula and D.R. Paul, Polymer, 1991,
32, 1401.
115. Y. Takeda and D.R. Paul, Journal of Polymer Science, Part B: Polymer Physics,
1992, 30, 1273.
116. R. Vankan, P. Dege, R. Jerome and P. Teyssie, Polymer Bulletin, 1994, 33, 221.
117. N.C.B. Tan, S.K. Tai and R.M. Briber, Polymer, 1996, 37, 3509.
118. A. Steurer and G.P. Hellman, Polymers for Advanced Technologies, 1998, 9, 297.
120. R.A. Kudva, H. Kekkula and D.R. Paul, Polymer, 2000, 41, 1, 225.
121. Y. Aoki and M. Watanabe, Polymer Engineering and Science, 1992, 32, 878.
122. C. Carrot, J. Guillet and M.F. Mary, Plastics and Rubber and Composites
Processing and Applications, 1991, 16, 61.
123. A. Misra, G. Sawhney and R.A. Kumar, Journal of Applied Polymer Science,
1993, 50, 1179.
240
Polyamide-Based Blends
124. V.J. Triacca, S. Ziaee, J.W. Barlow, H. Keskkula and D.R. Paul, Polymer, 1991,
32, 1401.
125. B. Majumdar, H. Keskkula and D.R. Paul, Polymer, 1994, 35, 3164.
126. B. Majumdar, H. Keskkula and D.R. Paul, Polymer, 1994, 35, 5453.
127. B. Majumdar, H. Keskkula and D.R. Paul, Polymer, 1994, 35, 5468.
128. R.A. Kim, Y.M. Lee and H.M. Jeong, Polymer, 1993, 34, 2075.
129. R.A. Kudva, H. Keskkula and D.R. Paul, Polymer, 1998, 39, 2247.
130. R.A. Kudva, H. Keskkula and D.R. Paul, Polymer, 2000, 41, 1, 239.
131. R.A. Kudva, H. Keskkula and D.R. Paul, Polymer, 2000, 41, 1, 335.
132. Z.H. Liu, P. Marchal and R. Jrme, Polymer, 1997, 38, 20, 5149.
133. Z.H. Liu, P. Marchal and R. Jrme, Polymer, 1998, 39, 10, 1779.
134. G. Deleens, P. Foy and P. Marchal, European Polymer Journal, 1997, 13, 343.
137. R.J.M. Borggreve, R.J. Gaymans, J. Schuijer and J.F.I. Housz, Polymer, 1987, 28,
1489.
138. R.J. Gaymans in Rubber Toughened Engineering Plastics, Ed., A.A. Collyer,
Chapman and Hall, London, UK, 1994, 210.
139. H. Keskkula and D.R. Paul in Rubber Toughened Engineering Plastics, Ed., A.A.
Collyer, Chapman and Hall, London, UK, 1994, 136.
140. O. Okada, H. Keskkula and D.R. Paul, Polymer, 1999, 40, 2699.
141. A.J. Oshinski, H. Keskkula and D.R. Paul, Journal of Applied Polymer Science,
1996, 61, 623.
142. A.J. Oshinski, H. Keskkula and D.R. Paul, Polymer, 1996, 37, 4891.
143. A.J. Oshinski, H. Keskkula and D.R. Paul, Polymer, 1996, 37, 4909.
241
Handbook of Polymer Blends and Composites
144. A.J. Oshinski, H. Keskkula and D.R. Paul, Polymer, 1996, 37, 4919.
145. S.C.G. Da Costa, M.D.C. Goncalves and M.I. Felisberti, Journal of Applied
Polymer Science, 1999, 72, 1827.
146. S.S. Pesetskii, V. Fedorov, B. Jurkowski and N.D. Polosmak, Journal of Applied
Polymer Science, 1999, 74, 1054.
147. C.P. Rader and B.M. Walker in Handbook of Thermoplastic Elastomers, Eds.,
B.M. Walker and C.P. Rader, Van Nostrand Reinhold, New York, NY, USA,
1987, 1.
150. G.O. Shonaike in Polymer Blends and Alloys, Eds., G.O. Shonaike and G.P.
Simon, Marcel Dekker, Inc., New York, NY, USA, 1999.
155. T.S. Ellis, Polymer Engineering and Science, 1990, 30, 571.
242
Polyamide-Based Blends
159. S. Fakirov, M. Evstatiev and J.M. Schultz, Polymer, 1993, 34, 4669.
160. V.M. Nadkani, V.L. Shingankuli and J.P. Jog, Polymer Engineering and Science,
1998, 28, 1326.
161. L.Z. Pillon and L.A. Utracki, Polymer Engineering and Science, 1984, 24, 1300.
162. L.Z. Pillon, J. Lara and D.W. Pillon, Polymer Engineering and Science, 1987, 27,
989.
163. L.Z. Pillon, L.A. Utracki and D.W. Pillon, Polymer Engineering and Science,
1987, 27, 562.
164. M.R. Kamal, M.A. Sahto and L.A. Utracki, Polymer Engineering and Science,
1982, 22, 1127.
165. M.R. Kamal, M.A. Sahto and L.A. Utracki, Polymer Engineering and Science,
1983, 23, 637.
166. A. Szymczyk, E.G. Koulouri, J.K. Kallitsis and Z. Roslaniec, Polymery, 1999, 44,
1, 30.
167. T.L. Boykin and R.B. Moore, Polymer Engineering and Science, 1998, 38, 10,
1658.
169. J.N. Clark and J.S. Higgins, Polymer Engineering and Science, 1992, 32, 1, 49.
170. M.M. Coleman, J.F. Graf and P.C. Painter, Specific Interactions and the
Miscibility of Polymer Blends, Technomic Publishing, Lancaster, PA, USA, 1991.
171. C.J.T Landry, D.M. Teegarden and B.K. Coltrain, Polymer Materials Science and
Engineering, 1994, 71, 1, 83.
172. R.N. Franch, J.M. Walsh and J.M. Machado, Polymer Engineering and Science,
1994, 34, 1, 42.
174. R.A. Register and T.R. Bell, Polymer Materials Science and Engineering, 1991,
65, 238.
243
Handbook of Polymer Blends and Composites
176. C.D. Eisenbach, J. Hofmann and W.J. MacKnight, Macromolecules, 1994, 27,
3162.
177. P. Smith and A. Eisenberg, Journal of Polymer Science, Part C: Polymer Letters,
1983, 21, 223.
178. P.K. Agarwal, I. Durderani, D.G. Pfeiffer and R.D. Lundberg, Journal of Polymer
Science, Part B: Polymer Physics, 1987, 25, 839.
179. M. Krumov, S. Fakirov, F.J. Balta Calleja and M. Evstatev, Journal of Materials
Science, 1998, 33, 2858.
180. T.L. Boykin and R.B. Moore, Polymer Engineering and Science, 1998, 38, 10,
1658.
181. L.Z. Pillon and L.A. Utracki, Polymer Processing Engineering, 1998, 4, 375.
182. C.C. Huang and F.C. Chang, Polymer, 1997, 38, 2137.
183. C.C. Huang and F.C. Chang, Polymer, 1997, 38, 4285.
184. F.J. Balta Calleja, L. Giri, T.A. Ezquerra, S. Fakirov and Z. Roslaniec, Journal of
Macromolecular Science B, 1997, 17, 655.
188. S. Horiuchi, N. Matchariyakul, K. Yase, T. Kitano, H.K. Choi and Y.M. Lee,
Polymer, 1997, 38, 3065.
189. S. Horiuchi, N. Matchariyakul, K. Yase, T. Kitano, H.K. Choi and Y.M. Lee,
Polymer, 1997, 38, 59.
190. S. Horiuchi, N. Matchariyakul, K. Yase, T. Kitano, H.K. Choi and Y.M. Lee,
Polymer, 1997, 38, 6317.
191. C.H. Lee, Y.M. Lee, H.K. Choi, S. Horiuchi and T. Kitano, Polymer, 1990, 40,
6321.
244
Polyamide-Based Blends
192. A. Etxeberria, S. Guezala, J.J. Iruin, J.G. de la Campa and J. de Abajo, Polymer,
1998, 39, 5, 1035.
193. J. Hu, P.C. Painter, M.M. Coleman and T.D. Krizan, Journal of Polymer Science,
Part B: Polymer Physics, 1990, 28, 149.
194. D.E. Bhagwagar, P.C. Painter, M.M. Coleman and T.D. Krizan, Journal of
Polymer Science, Part B: Polymer Physics, 1991, 29, 1547.
195. J.A. Brydson, Plastics Materials, 5th Edition, Butterworths, London, UK, 1982.
201. N. Bulakh and J.P. Jog, Journal of Macromolecular Science B, 1999, 38, 3, 287.
202. L.A. Utracki, Polymer Alloys and Blends: Thermodynamics and Rheology,
Hanser Publishers, Munich, Germany, 1989.
245
Polyester-Based Blends
7
Polyester-Based Blends
7.1 Introduction
These polymers can be produced by a variety of techniques of which the following are
technically important:
From the extremely wide range of PEST synthesised up to now and utilised in various
fields, the aromatic-aliphatic polyesters, known as thermoplastic PEST or linear PEST
and their blends will be referred to here.
The thermoplastic PEST are formed by the polycondensation of at least one aromatic
dicarboxylic acid (selected from terephthalic acid, isophthalic acid, 2,6-
naphthalendicarboxylic acid and 2,5-naphthalendicarboxylic acid), with at least one diol
(selected from a group consisting of alkylene glycols, such as ethylene glycol, propylene
glycol, butylene glycol and hexylene glycol, or polyalkylene glycols such as polyethylene
247
Handbook of Polymer Blends and Composites
PEN is known to present a higher Tg (about 50 C higher than that of PET) and superior
oxygen and carbon dioxide barrier properties compared to PET. The oxygen permeability
of PEN is approximately one-quarter to one-fifth of that of PET. PEN can be a good
candidate for hot-fill and high barrier packaging applications, but the high cost of PEN
has limited its direct application as a common packaging material.
248
Polyester-Based Blends
A wide range of polymers can be blended with thermoplastic PEST, including: polyolefins
(PO), styrenic, acrylic and vinylic compounds, polyamides (PA), polycarbonates (PC),
polyoxymethylene (POM), polysulfone (PSU), polyphenylene ether (PPE), polyimide,
liquid crystalline polymers (LCP), etc.
Blends of PEST (either PET or PBT) and PO [low density polyethylene (LDPE), linear
low density polyethylene (LLDPE), medium density polyethylene (MDPE), high density
polyethylene (HDPE), polypropylene (PP) and ethylene copolymers] have been the subject
of considerable research activity because large amounts of these compounds are available
through recycling technology. In the area of polymer blends, the challenge is to recycle
mixed plastics while keeping at an acceptable level their properties and production cost.
However, PEST and PO (either LDPE, LLDPE, MDPE, HDPE and PP) are incompatible
and when they are mixed, the interfacial adhesion is weak, resulting in lower mechanical
properties and poor dispersion of the components [4]. An improvement of the mechanical
properties of PEST/PO blends can be achieved using chemical modification of one or
both components (via grafting and copolymerisation) or by introducing a third component
in the blend as compatibiliser [4-45].
Although the components are incompatible, PEST/PO blends exhibit some interesting
properties. Thus, PET/PE blends (0.5-50 wt%), PET/poly-4-methylene-1-pentene [2] and
PET/ethylene-propylene rubber (EPR) or ethylene-propylene terpolymer (EPDM) [2] have
good impact strength; PET/PE or PP blends show high heat and moisture resistance [2,
14, 17-21, 34, 43]; PBT/PE or PP blends present improved electrical insulation properties
[2, 30, 33].
The improvement of the compatibility between PEST/PO blend components using chemical
modification can be accomplished using PO grafting (functionalisation) with (meth)acrylic
acid (AA), succinic acid, glycidyl methacrylate (GMA), maleic anhydride (MA), etc., or
partial oxidation or epoxidation of PO [2].
Several studies have demonstrated that chemical groups grafted onto PO, to react with
end groups of PEST (OH or COOH), ensure the compatibilisation of the blends. A
comparative study has shown that GMA is a more reactive monomer for the end groups
of PEST. As indicated in these studies, the epoxide group offers a high reactivity in relation
to the end groups of PET. This reactivity is superior to the one found with the polymer
bearing MA [6, 10, 41, 46] or AA group [23, 39] because MA and AA only react with
the hydroxyl groups.
249
Handbook of Polymer Blends and Composites
The PEST/modified PO blends show good notched impact strength and oxygen barrier
properties. These alloys were found to be suitable for moulding electrical, or automobile
parts with improved balance of heat resistance, chemical resistance, impact strength,
weld strength, stiffness, reduced mould shrinkage factor and appearance [2].
To expand the utilisation area for PEST (either PET or PBT), blending with ethylene
copolymers was also considered. The copolymers used were: ethylene-vinyl acetate
copolymers (EVA) (for blends with improved processability, mechanical and impact
strength, hardness, abrasion and solvent resistance); ethylene-methyl methacrylate
copolymer (EMMA) (for high toughness and low mould shrinkage); ethylene-vinyl acetate-
methacrylic acid copolymer (EVA-MAA) (to give crystalline material for thick-walled
mouldings with good impact strength) [47]; ethylene-hydroxyethyl methacrylate copolymer
(EHEM) (alloys were found useful for moulding and fibre spinning); ethylene-methacrylate
acid sodium salt copolymer (the alloys have good processability, impact strength properties
and high crystallisation rates [27]); ethylene-methacrylic acid partially Zn2+ neutralised
copolymer; ethylene-glycidyl methacrylate copolymer (EGMA) (the alloys showed good
mouldability, toughness, resistance to moisture and heat); ethylene-butyl acrylate (EBA)-
GMA (the alloys showed good toughness at room and low temperature); EBA; acrylic
acid grafted EVA (AA-g-EVA) (the alloys have improved low temperature impact strength)
and GMA grafted EPDM (GMA-g-EPDM) (the blends showed good weld strength) [2].
Block or grafted copolymers are generally prepared separately and introduced in low
concentrations in the melt-mixing system. An alternative way of compatibilising immiscible
PEST/PO blends is to create in situ these block or graft copolymers, by adding
functionalised polymer that can react with blend components due to their inherent
chemical reactivity based upon end-capping or grafting reactions [28]. However, block
copolymers or reactive functionalised polymers usually require a separate preparation
step, and certain copolymers are difficult to obtain. The price of the blend is, hence,
much higher than the sum of the parent polymers due to high cost of the compatibiliser.
Therefore, it would be a great advantage simply to use small molecule compounds as
compatibilisers [28].
The range of compatibilisers used for PEST/PO blends is wide and includes statistical,
block and graft copolymers such as: ethylene-GMA copolymer [5-8, 11]; EMMA
copolymer [13]; ethylene-ethyl acrylate-GMA copolymer (EEA-GMA) [6, 9, 11]; ethylene-
acrylic acid copolymer [13, 15]; EVA copolymer [13, 17, 29, 30-33]; styrene-ethylene-
butylene-styrene block copolymer (SEBS) [4] and hydrogenated SEBS [4, 5, 34, 41, 42],
maleic anhydride grafted PE (MA-g-PE) [5, 12, 14, 19-21, 24] or PP (MA-g-PP) [10, 43];
acrylic acid grafted PE (AA-g-PE) [16] or PP (AA-g-PP) [10, 22, 23]; GMA grafted PP
(GMA-g-PP) [10] or EPDM; MA grafted EVA, EPR (MA-g-EPR) [5, 30, 32] or ethylene-
250
Polyester-Based Blends
The path of action of the compatibiliser is different according to its chemical structure,
its method of synthesis and the way of including it in the blend, as well as the chemical
structure of PEST and PO [12, 15, 27, 29, 34].
For polymer and functionalised copolymer type of compatibiliser, many authors consider
that compatibilisation between PEST and PO is achieved due to the reactivity of the
compatibiliser with the reactive groups of PEST.
However, each chain of PEST has only one end hydroxyl group and one end carboxyl
group. Therefore, the most probably interactions that PEST can form with compatibiliser
(polymer functionalisation) are those of hydrogen bonding [21, 26, 28, 34, 49-51].
As in the case of PEST/ethylene copolymer blends, it was established that the epoxy
function offers a high reactivity in relation to the end groups of PEST [9-11].
In the case of the reactive compatibilisation of PEST/PO blends, the amount of functional
groups is the most important factor determining the morphology and mechanical
properties. When the amount of functional groups is insufficient, a few graft or block
copolymers is produced at the interface, and they cannot provide efficient
compatibilisation. At the other extreme, when the functional groups are more numerous
than necessary, the copolymer formed in situ may have a large molecular weight;
consequently this copolymer is located in the bulk phase rather than at the interface,
resulting in a poorer dispersion of the dispersed phase [15].
PEST/PO/compatibiliser blends are characterised by good melt flow, heat resistance and
impact strength [1, 18, 20, 23, 34-40, 52], low permeability to gasoline fuels, adhesion
to metals [2].
They can be processed by extrusion and injection, being suitable for batteries and fuel
cells, moulding interior and exterior parts of automobiles, parts for electric applications
and for many other applications.
251
Handbook of Polymer Blends and Composites
Thermoplastic PEST are immiscible with polystyrene (PS). Furthermore, since both resins
require toughening, so do their blends [2]. For these reasons PEST have not been blended
with PS, but with styrenic copolymers, namely high-impact polystyrene (HIPS), styrene-
acrylonitrile (SAN), styrene-maleic anhydride (SMA), acrylonitrile-butadiene-styrene
(ABS) or methyl methacrylate-butadiene-styrene (MBS) copolymers, SEBS or hydrogenated
SEBS, styrene-butadiene rubber (SBR), etc. These blends have been developed mainly to
reduce cost and to improve the impact strength of engineering resins.
Both PEST (either PET or PBT) and PS are low-cost commodity polymers that present
unique properties individually. The blends of PEST and PS are expected to be highly
desirable due to their great commercial potential. However, the research on PEST/PS
mixtures has been sparse because of their incompatibility. The studies have emphasised
that PEST/PS blends are macroheterogenous and are characterised by poor mechanical
properties [53]. In order to ameliorate these properties, resorting to compatibilisers has
been considered. Thus, McKay [54] used the polystyrene-polycaprolactone diblock
copolymer (PS-b-PCL) as a nonreactive compatibiliser for the PET/PS blends, but obtained
reduced mechanical and thermal properties. Not even the use of SMA as compatibiliser
can give a substantial improvement of the mechanical properties of PET/PS blends. This
is because the PET carboxyl terminal groups do not react with the anhydride of SMA,
while the reaction between PET hydroxyl groups and anhydride is insignificant without
the presence of a catalyst. Addition of small amount of epoxy [tetraglycidyl ether of
diphenyl aminomethane (TGDDM)] can act as a coupling agent to produce PET-co-
TGDDM-co-SMA copolymer at the interface.
These interfacially formed copolymers tend to anchor along the interfaces and act as
effective emulsifiers. Consequently, the compatibilised blend has a finer phase domain
size of the dispersed phase and greater interfacial adhesion. Tensile and impact properties
of the PET/PS blends have been substantially improved by employing the dual
compatibilisers, SMA and TGDDM [55].
On the other hand, Moa and Chang [56] used styrene-glycidyl methacrylate copolymer
(SGMA) as an in situ reactive compatibiliser in PET/PS blends, which resulted in substantial
improvement of mechanical properties and processability. These results prove that a
reactive compatibiliser is more effective at compatibilising PET/PS blends than a
nonreactive one.
252
Polyester-Based Blends
groups, thus it is considered an impact modifier. In the second case, the PS domains have
to be toughened by incorporation of elastomeric particles [57].
HIPS is a low price, but tough thermoplastic with relatively poor solvent resistance. A
combination of PBT and HIPS would be ideal to produce property balanced blend
products. In practice, however, it is too difficult to obtain good performance because
PBT and HIPS are immiscible [2]. Mechanical performances of PBT/HIPS blends can be
improved partially by replacing HIPS from these compositions with MA grafted HIPS
[MA-g-HIPS]. The blends containing 5 wt% MA-g-HIPS component exhibit outstanding
toughness. This behaviour can be attributed to the homogeneous morphological dispersion
of HIPS and the improved interfacial adhesion between the matrix and dispersed phase
owing to the chemical reaction between PEST and MA-g-HIPS. PET/HIPS blends
compatibilised with SGMA are described as having good miscibility and high flexural
strength [2].
PBT/SAN blends are macroheterogenous, where PBT is the matrix and SAN the disperse
phase. The SAN particle size of binary PBT/SAN blends can be controlled by changing
the shear rate during processing or the PBT to SAN viscosity ratio [58-61]. In general, a
higher matrix (PBT) viscosity should produce a finer dispersion of particles (SAN) during
melt processing. In the absence of a compatibiliser, the morphology generated may be
unstable and coarsening can occur under certain conditions [61]. Adding a compatibiliser
(methyl methacrylate (MMA)-GMA-ethyl acrylate terpolymer) (MGE)) in PBT/SAN
blends greatly improves their properties by producing a finer dispersion of SAN; this
leads to a reduction in ductile-brittle transition temperature far beyond that of PBT/SAN
blends, and also to good tensile, flexural and impact strength.
The impact properties of PEST (mainly PET and PBT) can be improved by simple melt-
blending with appropriate ABS [62]. These materials, however, can only be produced
within a limited processing range and have an instable phase morphology: the ABS
domains are coalescent during certain low shear conditions (particularly at high
temperature when the PEST melt viscosity is low), the resulting melt having reduced
mechanical properties.
Apart from the processability conditions and the morphology, the properties of PEST/ABS
blends are also determined by a great number of factors, such as: the type of polyesteric
component, the type and composition of ABS, molecular weight of the components,
blending ratio, etc.
253
Handbook of Polymer Blends and Composites
MA-g-ABS [63], AA-g-ABS, SEBS, MA-g-SEBS, and MGE can be used as compatibilisers
for PEST/ABS blends [61, 62, 63, 64-66].
If one considers using only MGE as compatibiliser, studies have shown that MGE-g-PBT
graft copolymer is formed at the PBT/ABS interface by reactions between the carbonyl
end-groups of PBT and the epoxide groups along the MGE chain; the copolymer successfully
reduces the scale of dispersion, provides morphology stability, gives a broader processing
window, and improves low temperature impact toughness of PBT/ABS blends. However,
undesirable crosslinking reactions also occur involving the epoxy functionality of MGE,
catalysed by residual acid impurities in emulsion-made ABS materials, which result in a
reduction of impact strength in toughened PBT blends [64]. One practical way to solve this
problem is by changing the order of mixing the blend components, e.g., using a two-pass
extrusion method where PBT and MGE are melt-mixed together prior to extrusion with
ABS. These alloys are used first to obtain the desired graft copolymer and then for limiting
the competition for available epoxide units for crosslinking reactions [61].
In the patent literature many types of PEST/MBS blends are cited, with miscellaneous
chemical compositions for MBS [2]. With or without compatibiliser, these blends exhibit
good processability, heat resistance, ductility, HDT, high modulus, impact, tensile and
flexural strength over a wide temperature range, good adhesion, solvent, chemical and
UV resistance [2].
The uncompatibilised blends of PEST with SEBS and hydrogenated SEBS (1:1) alloy
were reported to have co-continuous morphology, good dimensional stability and good
mechanical behaviour, varying from rubber-like to flexible [2].
254
Polyester-Based Blends
A closer control over the properties of PEST/SEBS or hydrogenated SEBS blends may be
achieved through functionalisation of the elastomeric component by grafting with MMA
[50, 51, 67, 68]. Improved properties with these blends of PET that incorporated
functionalised SEBS were consistent with in situ formation of a graft copolymer by reaction
with PET end groups. It appeared that the graft copolymer acted as an emulsifier to
decrease the interfacial tension and promote adhesion [9, 63, 69, 70]. The amount of
graft copolymer formed by reaction of PET and functionalised SEBS affects both particle
size and interfacial strength. Increasing the anhydride concentration in the SEBS did not
strongly affect blend rheology [50, 51]. In contrast, particle size was strongly dependent
on elastomer functionality: the higher the functionality, the smaller the particle size and
the narrower the particle size distribution. These trends were attributed to an increase in
the degree of grafting. Notched impact strength is used to differentiate between the blends
in terms of their toughness. The most effective was SEBS with only 1 wt% anhydride.
The decrease in toughness with increasing functionality was attributed to decreasing
particle size. Variables other than the amount of grafting are also expected to influence
the effectiveness of functionalised SEBS for toughening the components and modifying
the hydroxyl to carbonyl end group ratio of the PET [71].
As for PEST/PS blends, systems comprising PEST and the main acrylic resin, polymethyl
methacrylate (PMMA), are of no commercial interest. The reason is similar, immiscibility
and brittleness of both resins, augmented by the lack of interfacial adhesion between the
phases in solid state. By contrast, the acrylic copolymers constitute an important class of
compatibilising and impact modifying components in numerous PEST blends.
255
Handbook of Polymer Blends and Composites
Thus, the PBT was blended with MMA-MA-butyl acrylate-allyl methacrylate copolymer.
The blends show good processability and high impact resistance.
PBT was blended with not more than 5 wt% of MMA-methyl acrylate copolymer. The
alloys present good mouldability, bending strength, and have a pearl-like lustre.
The main aspects regarding the compatibility and properties of PEST/vinyl blends have
been discussed in Chapter 3. Only few additional details will be given next.
256
Polyester-Based Blends
More recently, miscible blends of PET with polyvinyl phenol were announced. These blends
could be extruded into chemically resistant materials for transparent packaging films [2].
Special mention should be made of the thermoplastic elastomers (TPE) of PEST (PET or
PBT)/commodity rubber blends obtained by dynamic curing [2, 76-78]. Cai and Isayev
[76] prepared such blends by using different methods (in one or two-step, the latter
including the production of masterbatch) or techniques (twin-screw extruder, internal
mixer or open mill), and they found that the performance of thermoplastic dynamic
vulcanisates produced by twin-screw extrusion is superior to all other alternatives.
Some data regarding morphology, properties, processability and application areas of the
PEST blends with ethylene and styrene copolymers (many with elastomeric characteristics)
have already been presented (see Sections 7.2 and 7.3).
There are a peculiar group of PEST/elastomers blends in which the elastomeric component is
an acrylic rubber. These blends exhibit good processability, excellent electrical properties,
257
Handbook of Polymer Blends and Composites
high stiffness and strength, high HDT, low water absorption, good dimensional stability, no
tendency for environmental stress cracking, good low and high temperature impact strength,
high weatherability, water and chemical resistance, and resistance to heat ageing [2].
These blends are formulated for injection moulding, or extrusion, but can be compressed,
blow or transfer moulded as well.
The commercial PEST/acrylic elastomer blends have been used for engineering parts,
sport gear, automotive parts (such as door handles, windscreen wiper arms, motor housing,
printed circuit boards, housings for electronic ignition system, fuel level sensors, bumpers,
protecting strips, spoilers, head lamp housings, distributor caps, spark connectors, etc.).
Putting aside the basic chemical reactivity of functional groups and the associated
dependence on structure and temperature, the feature that affects the ability of
transreactivity to influence phase behaviour in immiscible polymer blends is the relative
frequency that reactive species come into contact with each other. This will depend upon
processing conditions, e.g., temperature and the interaction parameter. The latter is
perhaps the most important, since it can affect directly the width of the interfacial region
and hence the contact between the reactive species.
258
Polyester-Based Blends
The latter thermal treatment also masks any underlying phase transition, e.g., lower
critical solution temperature (LCST) that may be present below the melting point of the
respective components. When all of these concerns are taken into account, it is not
surprising that the literature is replete with contradictory observations and conclusions
concerning phase behaviour of PEST(1)/PEST(2) blends [79, 84-95].
In the following paragraphs, the properties and utilisation areas of the most important
of PEST(1)/PEST(2) blends, i.e., PET/PET, PBT/PBT, PET or PBT/PEN, PET or PBT/
other PEST are discussed. The two types of PET or PBT (with different molecular weights)
are mixed, with the aim of obtaining good mouldability, improved impact strength and
resistance to repeated bending fatigue.
The materials produced comprised 60-99 wt% PBT (having 40<Mn<60 kg/mol) blended
with 1-40 wt% PBT (with Mn between 21 and 27 kg/mol). These homologous blends
were found to be suitable for moulding bodies with hinges. Blends of PBT/PBT-modified
polyurethane (25 wt% PBT-TPU block copolymer) exhibit high impact strength down to
70 C.
Commercial PET/PBT blends have good processability, high HDT, good impact strength
and dimensional stability, good crystalline properties without the need to control their
molecular weights, or for nucleating agents.
259
Handbook of Polymer Blends and Composites
The alloys with PET/PBT in ratio 99.9:0.1 to 91:9 were used for production of films,
containers and fibres. The stretch-moulded parts showed low initial residual stress,
excellent shape, good surface properties, heat resistance, transparency and mechanical
strength. The PET/PBT blends, used in electrical condensers, showed a stabile level of
crystallinity and better arc resistance.
Although PET/PBT blends have a wide range of good properties, in order to intensify
some of those, there were introduced in the system other components like: IIR, Acryloid
KM 330, polyester-polyether rubber, butyl acrylate-GMA-MMA copolymer, butyl
acrylate-MMA-styrene-triallyl isocyanate copolymer, acrylonitrile-allyl methacrylate-butyl
acrylate-ethyl acrylate-MMA copolymers, ethylene-GMA copolymer (for blends with
improved processability and high impact strength copolymer); ethylene-butyl acrylate-
GMA copolymer (the alloys showed a good level of toughness, at both room and low
temperatures). With the same objectives, graft copolymers of polybutadiene and
(meth)acrylate esters, an acrylate rubber or ABS (the rubber particles should have an
average diameter of ~0.4 m, to assure high impact strength and good balance of
mechanical properties); MMA grafted polybutylacrylate (the alloys containing glass fibres
and talc had good mechanical properties as well as improved crystallisation rate and
surface appearance) have also been used [2].
260
Polyester-Based Blends
Stewart and co-workers [87] have observed that the degree of transesterification was
primarily controlled by the blending time and temperature rather than composition. The
higher the temperature was, the faster the transesterification reaction becomes. The
transesterification was also dependant on the blending time: for example, the
transesterification in a PET/PEN (50/50) blend increased from 5.8% for 1.5 minutes to
21.3% for 4.5 minutes at 295 C. They also reported that about 10% of transesterification
was required to produce transparent samples and the formation of a single amorphous
phase. The same result was reported by Aoki and co-workers [105].
There are many other types of PEST(1)/PEST(2) blends, including: PET/PHT with or
without inorganic filler (the materials obtained were reported to have good mouldability
and high impact strength); PET/PCT (the blend used to thermoform such articles as
frozen food trays, that exhibit high impact strength at low temperature); PET/copolyester-
ether prepared from butanediol, dimethyl terephthalate and polytetramethylene glycol
(the alloy showed good impact and tensile strength); PET or PBT/polyethylene
isophthalate (the alloys had good oxygen barrier) [2, 107]; PET, PBT or PEN/polyether
inside [108-110].
In the same group of PEST(1)/PEST(2) blends one can also find the systems containing
PET, PBT, PEN and other aromatic aliphatic PEST/liquid crystalline, PEST or aliphatic
PEST. Details regarding morphology, properties, and application areas are found in
Chapter 16.
The PEST (either PET or PBT)/PAr blends obtained by solution-blending are immiscible,
exhibit two Tg, and slightly positive binary thermodynamic interaction parameters [111].
The same mixtures obtained by melt blending are estimated as miscible [112-114]. The
miscibility between PET or PBT and PAr is somewhat different because of the difference
in the PAr backbone structure [115].
261
Handbook of Polymer Blends and Composites
Because of the presence of polar groups, ester exchange and possibility of graft-
modification, PEST are relative easy to blend with PAr.
Commercial PEST (either PET or PBT)/PAr blends are known for their excellent processability,
toughness and high impact strength, high HDT and good transparency. Many of these
properties can be enhanced by introducing a compatibiliser in PEST/PAr blends [2].
Thus, blends of PBT with 10-90 wt% PAr and either EPR or EPDM grafted with
glycidyloxybenzyl acrylamide showed very good processability and high impact strength.
Mixing PET, PAr and a polymer selected from thermoplastic polyurethane (TPU), PVC
or phenoxy (polyhydroxy ether of bisphenol A), gives blends suitable for injection
moulding, and the moulded parts have a good balance of mechanical properties and
stress crack resistance. Incorporation of 50 wt% polyamide 6 (PA6) in PET/PAr blends
was reported to give mouldings with excellent processability, glass, chemical, mechanical,
electrical properties and HDT [2]. Blends of PEST (PET or PBT)/PAr/ethylene-acrylate
or ethylene-glycidyl methacrylate copolymers present good balance of hydrolytic stability
and mechanical properties.
Polycarbonates are thermoplastic polymers with interesting practical characteristics, such as:
1. rigidity up to 140 C;
2. toughness up to 140 C;
3. transparency;
6. physiological inertia.
PEST/PC blends are of both scientific and commercial interest. The PC of bisphenol A
has been frequently used in blends with PEST (either PET, PBT, PEN), but some other
blends of PEST with polyester carbonates (PEC) have also been synthesised.
262
Polyester-Based Blends
increase compatibility of PET/PC [120, 121], PET/PEC [122] and PBT/PC [122-126]
blends. In this case, the induced reactions that occur during melt-blending modify the
chemical structure of the two immiscible (in principle) polymers, providing a final system
with a controlled degree of compatibility, as a result of generation of copolymers of the
two starting polymers [127, 128].
Because of their significant commercial potential, PET/PC blends have been investigated
rather intensively, particularly for their miscibility. Porter and co-workers [82] reported
a larger range of miscibility: from complete miscibility for all compositions to only partial
miscibility of the blends. For instance, the blends prepared by extruder-blending (with a
die temperature of 290 C) present two amorphous phases over the whole composition
range, suggesting that neither miscibility nor transreactions take place [129]. On the
other hand, the exchange reactions in the molten PET/PC blends (in the range from
240 to 274 C and in the presence of transesterification catalyst) give rise to four-
component copolycondensates [130].
More recently, it was reported [131, 132] that the same system is entirely miscible only
after transesterification, as per melt-mixing at high temperature (300 C) or for long
processing time. It is therefore not surprising that single extrusion-blending generates
immiscible blends, due to the insufficient reaction time [133]. Without the
transesterification reaction, this pair of polymers is immiscible.
Very recently, the sequential reordering in copolymers appearing in PET/PC blends was
also investigated in detail [134]. In agreement with previous studies [82, 135], it was
found that annealing at 280 C, for four hours (without an extra added catalyst), results
in a complete disappearance of the blends crystallisation. Shifting to this completely
amorphous state is attributed to the formation of random copolymers.
Blends of PBT and PC represent a perfect example of the complex interplay between
phase separation and crystallisation. Several authors have reported that PBT/PC blends
obtained by melt-mixing, during which transesterification occurs, are partially miscible
[116, 136-138]. Sing [139] observed a large decrease of melt flow of PC, when only 5-
10 wt% of PBT was added. This indicates some solubility of PBT in PC. Hobbs and co-
workers [137] and Delimoy and co-workers [140] observed the presence of PC in the
interlamelar zones of PBT. From this, they concluded that in the molten state PBT and
PC are miscible, but on cooling, phase separation and crystallisation of PBT take place.
The dispersed blends can exhibit either a PBT-rich continuous phase or a PC-rich
continuous phase.
The solubility of PBT in solid state PC is rather low, and in principle does not depend on
the blend composition. The estimated decreasing Tg of PC with increasing PBT content
indicates instability of the blends [116, 136, 141].
263
Handbook of Polymer Blends and Composites
Berlisson and co-workers [142, 143] observed that the phase separation in PBT/PC blends
increased by annealing at temperatures close to or above the Tg of PC. Delimoy and co-
workers [140] reported that during annealing at 220 C, the PBT lamellae were growing
from the interface, inside the PC-rich phase, depleting the latter from dissolved PBT [144].
The range of polymers added to PEST/PC blends to improve some of their properties is
wide and includes: PE (LDPE or LLDPE) and ethylene copolymers (EVA, EPR, EPDM,
etc.); polyvinyl acetate (PVAc); BR; IIR; PEC; PCL; ABS; MBS; SEBS; carboxylated,
maleated, epoxidated or aminated polymers and copolymers, e.g., GMA-g-PE, GMA-g-
PP, GMA-g-EPR, MA-g-EPR, MA-g-EPDM, aminated EPR, EPDM or NBR, etc.); styrene-
acrylonitrile copolymer grafted EPR or EPDM; acrylic elastomers (phenoxy, acryloid,
core-shell acrylic elastomers), polybutadiene grafted with styrene and MMA [145].
In the following paragraphs, only some of these blends will be presented, emphasising
their properties and application area [2].
Other blends of practical interest contain 20-60 wt% PEST (PET, PBT or PCT), 40-
60 wt% PC, 10-20 wt% aminated EPR, EPDM or NBR, a copolymer of PEST and the
selected elastomer, and a graft coupling agent such as poly(2,4,6-triisopropyl-1,3-
phenylcarbodiimide).These alloys have been useful for moulding parts, e.g., for garden
and powder tools, sport equipment and snowmobiles, and have found applications in
the automotive, electric and electronics industries.
Blends consisting of PET, PC, MBS and methyl acrylate-ethyl acrylate-butyl acrylate
copolymer exhibit improved melt-flow and increased the sag time (time in which concave
formation occurs), thus facilitating blow moulding, thermoforming and extrusion processing.
264
Polyester-Based Blends
strength, humidity resistance and dimensional stability. The blend comprising PET, PC
and a core-shell copolymer made of 99-50 wt% of MMA-butyl acrylate-MA copolymer
and 1-50 wt% of monomer units with UV-absorbing groups, such as phenyl methacrylate
or 2-ethylhexyl-thioglycolate were used as moulding materials for injection moulding
parts, for housing and bumpers.
Compounds of PBT, PC and SAN grafted EPDM show good mouldability, mechanical
properties and weld line strength.
PBT/PC/ABS blends were reported to have good mouldability, high impact strength at
low temperatures and gasoline resistance.
PBT/PC/SEBS/mineral filler blends exhibit good impact strength, low warpage and
shrinkage.
Blends of PCT with PC are well-known for high toughness, HDT and chemical resistance.
Owing to their miscibility, they show a high degree of transparency.
The PEST/PC blends, with or without compatibilisers or impact modifiers, are usually
formulated for injection moulding, extrusion and blow moulding, but they can also be
processed by transfer or compression moulding, formed in the solid state or by
thermoforming. These blends show good processability, heat resistance, ductility, HDT,
high modulus, impact, tensile and flexural strength over a wide temperature range and
good adhesion, solvent, chemical and UV resistance. They can be printed, hot stamped,
metallised and plated. The blends are mainly used in the automotive industry for
automotive body panels that require toughness at low temperature, chemical resistance,
heat resistance and dimensional stability, such as dashboard, pillars, roof claddings, heat
lamp housing and luggage racks. They are also used in outdoor power equipment (tractor
or lawn mower shrouds, consoles, grilles, trim and equipment parts), recreational
applications, for manufacture of consumer goods (vacuum cleaner housing and nozzles,
utility locks, protective face guards, beverage containers, dishes, food transport bins,
dinnerware, mixing bows and swimming pool pump housing), telecommunications (radio
housing, speakers and instrument housing), and many others [2].
265
Handbook of Polymer Blends and Composites
Both semiaromatic PEST (either PET or PBT) and aliphatic PA (PA6, PA66, PA666,
PA610) are semicrystalline, brittle resins. Since their antagonistically immiscible blends
have large interfacial tension coefficients, the dispersion is poor, and the mechanical
behaviour shows fractures and atrocious impact strength [1, 146-155]. One could expect
that, during the compounding and processing step, a degree of ester-amide exchange
reaction could take place, improving the interaction between the two phases. Furthermore,
the extent of ester-amide exchange reaction could be controlled, providing alloys with
good adhesion of the dispersed phase to the matrix. However, the blends show a brittle
behaviour [156].
Blends of PEST with polyamide 3 are extensively discussed in Chapter 6. Some aspects
of the properties and application areas of these blends will be discussed next [2].
Blends constituted of PET and PA (PA6 or PA66) have good frictional resistance. PBT
blends with 10-90 wt% PA6 to show high crystallinity and a fine dispersion of the minor
phase, which results in a good set of mechanical properties. PA blends with 15-45 wt%
PET and 0.05-10 wt% phosforyl azide reactive coupler were found to have improved
processability.
Blends of PBT (69.6 wt%) with polyesteramide show rapid crystallisation of PBT and
significant improvement of the impact strength.
PEST (PET or PBT)/PA blends compatibilised with 0.5-1.5 wt% of phenoxy exhibit
increased elongation, tensile strength, Izod impact strength and uniformity of material
after injection.
Mixing PET with polyester-polyamide copolymer yield alloys useful for manufacturing
containers with good barrier properties.
Blends containing PEST, PA, acrylonitrile copolymers, styrene copolymers and fillers
showed enhanced melt tension and improved blow mouldability, without causing parison
draw-down. Products have excellent mechanical properties and thermal resistance,
266
Polyester-Based Blends
uniform thickness and an excellent appearance, even when repeatedly recycled. These
blends are used to produce hollow mouldings to be used under severe conditions, such as
the intake manifold of an automobile, suction and exhaust parts around the engine,
vessels for high temperature liquids, chemical and solvent, pipes, floats and tubes [2].
Polyesters and PPE are nearly ideal for blending since the properties of one polymer
compensate the deficiencies of the other one. For example, in PEST/PPE blends the latter
resin provides good HDT and rigidity, whereas the former contributes toward
processability, paintability and solvent resistance.
However, the two resins are antagonistically immiscible and must be compatibilised.
Since both resins are brittle, their compatibilised blends must be toughened. Thus, the
evolution of PEST/PPE blends is the evolution of compatibilisation and impact
modification. Because of the miscibility of PPE with styrenics, toughening of the PPE
phase is relatively easy. The main difficulty is to find a method for compatibilisation of
PEST and impact modification of its phase.
The family of the compatibilisers and/or impact modifiers for PEST/PPE blends comprises:
PS, HIPS, SBS, SEBS, GMA-g-SEBS, ABS, EPR, MA-g-EPR, EPDM, PP grafted with
styrene and GMA, ethylene-GMA copolymer and EVA, GMA-MA copolymers, acrylate
elastomers, PC, PC-PBT block copolymer, PPE block copolymer, and other polymers [2].
In the following sections only some of the PEST/PPE blends will be described with emphasis
on their properties and areas of application.
267
Handbook of Polymer Blends and Composites
Blends of PEST, PPE, HIPS and styrene polymers having reactive 2-oxazoline groups
exhibit improved strength which is the highest advantage attained with a favourable
combination of properties, such as mechanical strength, resistance to chemicals and
solvents, and flow properties in the melt.
By reactive mixing of PET with PPE and GMA-g-SEBS yielded alloys owning excellent
solvent resistance, mouldability, impact and mechanical properties.
The alloys comprising PEST (PET or PBT), PPE, PC-PBT copolymer (with 3-20 wt%
PC) and SEBS have excellent rigidity and impact strength, good solvent resistance, excellent
surface finish and low coefficient of thermal expansion.
Blends with excellent processability, heat resistance and mechanical properties, are
obtained by alloying 35-98 wt% PEST (PET or PBT), 1-65 wt% PPE, 0.1-50 wt% of a
compatibiliser, and reinforcement with talc and/or glass fibres. As compatibiliser, MA-g-
EPR, a copolymer of ethylene, vinyl acetate and GMA, a monomeric MA or monomer
bis (4-phenyl isocyanate) were used. The reactive compatibilisation, carried out in high-
speed two-screw extruder, results in PPE particle size having diameter of 0.01-10 m.
The alloys can be either extruded or moulded into a variety of parts for the automotive,
electrical and electronic industries [2].
The literature provides relatively little information about PEST blends (mainly PET and
PBT) with POM, polysulfide, polysulfone, polyimide and polyester imide. Many pieces
of information on this topic may be found only in patent literature [2].
The most important commercial PEST-based blends are presented in Table 7.1.
268
Commercial PEST blends
Polyester-Based Blends
269
Handbook of Polymer Blends and Composites Commercial PEST blends
270
Commercial PEST blends
Polyester-Based Blends
271
Commercial PEST blends
Handbook of Polymer Blends and Composites
272
Commercial PEST blends
Polyester-Based Blends
PTFE: polytetrafluoroethylene
273
Handbook of Polymer Blends and Composites
References
2. L.A. Utracki, Commercial Polymer Blends, Chapman and Hall, London, UK, 1998.
3. J.O. Iroh in Polymer Data Handbook, Ed., J.E. Mark, Oxford University Press,
Oxford, UK, 1999, 349.
4. T.D. Traugott, J.W. Barlow and D.R. Paul, Journal of Applied Polymer Science,
1983, 28, 2947.
5. S.S. Dagli and K.M. Kamdar, Polymer Engineering and Science, 1994, 34, 23,
1709.
10. Y.J. Sun, G.M. Hu, M. Lambla and H.K. Kotlar, Polymer, 1996, 37, 4119.
12. J. Curry and A. Kiani in Proceedings of SPE ANTEC 90, Dallas, TX, USA, 1990,
1452.
13. G.E. Wissler in Proceedings of SPE ANTEC 90, Dallas, TX, USA, 1990, 1434.
14. P. Sumbru and S.A. Jabrin, Polymer Engineering and Science, 1993, 33, 827.
15. S. Kim, C.E. Park, J.H. An, D. Lee, E. Grulke and J. Kim, Polymer Journal, 1997,
29, 3, 274.
16. C. Pattanakul, S. Selke, C. Lai and J. Miltz in Proceedings of SPE ANTEC 88,
Atlanta, GA, USA, 1988, 1802.
274
Polyester-Based Blends
17. A. Khelifi and F. Lai in Proceedings of SPE ANTEC 88, Atlanta, GA, USA, 1988,
1824.
18. T.L. Cart and A. Moet, Journal of Applied Polymer Science, 1993, 48, 611.
19. N.K. Kalfoglou, D.S. Skafidas and D.D. Sotiropoulou, Polymer, 1994, 35, 3624.
20. N.K. Kalfoglou, D.S. Skafidas, J.J. Kallitsis, J.C. Lambert and L. Van der
Stappen, Polymer, 1995, 36, 4453.
21. L. Marquez, D.E. Gambus, M.P. Romero-Rato and M.A. Sabino in Proceedings
of SPE ANTEC 99, New York, NY, USA, 1999, Volume 3, 3903.
22. M. Xanthos and S.S. Dagli, Polymer Engineering and Science, 1991, 31, 929.
23. P. Bataille, S. Boisse and H.P. Schreiber, Polymer Engineering and Science, 1987,
27, 622.
24. L. Marquez, M.A. Sabino and I.A. Reviro, Polymer Bulletin, 1988, 41, 191.
25. L. Yao and C. Beatty in Proceedings of SPE ANTEC 97, Toronto, Canada, 1997,
Volume 2, 2577.
27. P.M. Subramanian, Polymer Engineering and Science, 1987, 27, 20, 1574.
28. H.X. Zhang and D.J. Hourston, Journal of Applied Polymer Science, 1999, 71,
2049.
32. C.J. Parikh, A.A. Donatelli and D.R. Deanin, Polymer Preprints, 1993, 34, 2.
33. T- K. Kang, Y. Kim, W-K. Lee, H-D. Park, W-J. Cho and C-S. Ha, Journal of
Applied Polymer Science, 1999, 72, 989.
275
Handbook of Polymer Blends and Composites
34. J.C. Lepers, B.D. Favis and C. Lacroix, Journal of Polymer Science: Polymer
Physics Edition, 1999, 37, 939.
35. J.W. Barlow and D.R. Paul, Polymer Engineering and Science, 1984, 24, 525.
36. I-M. Chen and C-M. Shiah, Proceedings of SPE ANTEC 89, New York, NY,
USA, 1989, 1802.
37. I-M. Chen and C-M. Shiah, Plastics Engineering, 1989, 45, 33.
38. M. Xanthos, M.W. Young, and J.A. Biesenberger in Proceedings of SPE ANTEC
89, New York, NY, USA, 1989, 1835.
39. M.W. Young, M. Xantos, and J.A Beisenberger, Polymer Engineering and
Science, 1990, 30, 355.
40. J-C. Lepers, B.D. Favis and S.L. Kent in Proceedings of SPE ANTEC 97,
Toronto, Canada, 1997, Volume 2, 1788.
42. J-C. Lepers, B.D. Favis and R.J. Tabar, Journal of Polymer Science: Polymer
Physics Edition, 1997, 35, 2271.
43. Y.X. Pang, D.M. Jia, H.J. Hu, D.J. Hourston and M. Song, Polymer, 2000, 41,
357.
48. S. Datta and D.J. Lohse, Polymer Compatibilisers, Hanser Publishers, Munich,
Germany, 1996.
276
Polyester-Based Blends
50. V. Tanrattanakul, A. Hiltner, A. Baer, W.G. Perkins, F.L. Massey and A. Moet,
Polymer, 1997, 38, 2191.
51. V. Tanrattanakul, A. Hiltner, A. Baer, W.G. Perkins, F.L. Massey and A. Moet,
Polymer, 1997, 38, 4117.
52. L. Marquez, M.A Sabino and I.A. Rivero, Polymer Bulletin, 1998, 41, 191.
54. I.D. McKay, Journal of Applied Polymer Science, 1991, 32, 1401.
55. M.Y. Ju and F.C. Chang, Journal of Applied Polymer Science, 1999, 73, 2029.
56. C.T. Moa and F.C. Chang, Journal of Applied Polymer Science, 1993, 49, 913.
57. D.Y. Chang, W.F. Kuo and F.C. Chang, Polymer Networks and Blends, 1994, 4,
157.
59. G. Serpe, J. Jarrin and F. Dawans, Polymer Engineering and Science, 1990, 30,
553.
60. B.D. Favis and J.P. Chalifoux, Polymer Engineering and Science, 27, 1591.
61. W. Hale, J-H. Lee, H. Keskkula and D.R. Paul, Polymer, 1999, 40, 3621.
62. E. Hage, W. Hale, W. Keskkula and D.R. Paul, Polymer, 1997, 38, 3237.
63. N.K. Kalfoglou, D.S. Skafidas and J.K. Kallitsis, Polymer, 1997, 37, 3387.
64. W. Hale, H. Keskkula and D.R. Paul, Polymer, 1999, 40, 3353.
65. W. Hale, L.A. Pessan, H. Keskkula and D.R. Paul, Polymer, 1999, 40, 4237.
66. W. Hale, H. Keskkula and D.R. Paul, Polymer, 1999, 40, 3676.
68. V. Tanrattanakul, W.G. Perkins, F.L. Massey, A. Moet, A. Hiltner and E. Baer,
Journal of Materials Science, 1997, 32, 4749.
277
Handbook of Polymer Blends and Composites
69. B. Boutevin, J.M. Lusinchi, Y. Pietrasanta and J.J. Robin, Polymer Engineering
and Science, 1966, 36, 879.
70. H-C. Chin, K-C. Chiou and F-C. Chang, Journal of Applied Polymer Science,
1996, 60, 2503.
71. A. Pawlak, W.G. Perkins, F.L. Massey, A. Hiltner and E. Baer, Journal of Applied
Polymer Science, 1999, 73, 1, 203.
74. J.R. Campbell, F.F. Khouri, S.Y. Hobbs and J.J. Shea, Polymer Preprints, 1993,
34, 846.
76. F. Cai and A.I. Isayev, Journal of Elastomers and Plastics, 1993, 25, 74.
77. A.J. Moffet and M.E.J. Dekker, Polymer Engineering, 1992, 32, 1, 1.
79. T.S. Ellis in Polymer Blends and Alloys, Eds., G.O. Shonaike and G. P. Simon,
Marcel Dekker, New York, NY, USA, 1999, 209.
80. A.M. Kotliar, Journal of Polymer Science: Macromolecular Reviews, 1981, 16,
367.
81. R.S. Porter, J.M. Jonza, M. Kimura, C.R. Desper and E.R. George, Polymer
Engineering and Science, 1989, 29, 1, 1989.
83. J. Devaux, P. Godard and J. P. Mercier, Polymer Engineering and Science, 1982,
22, 229.
84. J. Anderson and H.G. Zachmann, Colloid and Polymer Science, 1994, 272,
1352.
85. J.S. Kollodge and R.S. Porter, Polymer, 1993, 34, 4990.
278
Polyester-Based Blends
86. K.G. Gravalos, J.K. Kallitsis and N.K. Kalfoglou, Polymer, 1995, 36, 1393.
87. M.E. Stewart, A.J. Cox and D.M. Naylor, Polymer, 1993, 34, 4060.
88. S.C.E. Backson, A.M. Kenwright and R.W. Richard, Polymer, 1995, 36, 1991.
89. J.S. Kollodge and R.S. Porter, Macromolecules, 1995, 28, 4089.
90. J.S. Kollodge and R.S. Porter, Macromolecules, 1995, 28, 4098.
91. J.S. Kollodge and R.S. Porter, Macromolecules, 1995, 28, 4106.
92. M.M. Guo and H.G. Zachmann, Polymer Preprints, 1996, 37, 1, 825.
93. M.M. Guo and H.G. Zachmann, Polymer, 1993, 34, 2503.
94. M.M. Guo and H.G. Zachmann, Macromolecules, 1997, 30, 2746.
95. M.M. Guo and W.J. Brittain, Macromolecules, 1998, 31, 7166.
98. Y. Yu and K.J. Choi, Polymer Engineering and Science, 1997, 31, 1, 91.
99. M. Bosma, G. ten Brinke and T.S. Ellis, Macromolecules, 1988, 21, 1465.
100. A. Escala and R.S. Stein, Advances in Chemistry Series, 1979, 176, 455.
102. D.W. Ihm, S.Y. Park, C.G. Chang, Y.S. Kim and H.K. Lee, Journal of Polymer
Science: Polymer Chemistry Edition, 1996, 34, 2841.
104. S.C. Lee, K.H. Yoon, I.M. Park, H.C. Kim and T.W. Soon, Polymer, 1997, 38,
4831.
279
Handbook of Polymer Blends and Composites
107. A. Ophir, S. Kenig, A. Shai and Y. Barkaai in Proceedings of SPE ANTEC 99,
New York, NY, USA, 1999, 1010.
109. H-L. Chen, J.C. Hwang, C.C. Chen, R-C. Wang, D.M. Fang and M. J. Tsai,
Polymer, 1997, 38, 2742.
110. H-L. Chen, J.C. Hwang and R-C. Wang, Polymer, 1998, 39, 24, 6067.
111. J.C. Chung and A.K. Akkapeddi, Polymer Preprints, 1993, 34, 1, 614.
112. J. Runt, C.A. Barron, X. Zhang and S.K. Kumar, Macromolecules, 1991, 24, 3466.
113. J. Runt, D.M. Milley, X. Zhang, K.P. Gallagher, K. McFeaters and J. Fishburm,
Macromolecules, 1992, 25, 1929.
114. J. Runt, D.M. Milley, X. Zand and K.P. Gallagher, Macromolecules, 1992, 25,
3902.
115. W.B. Lian, A.S. Liu and W.Y. Chiu, Journal of Polymer Research, 1999, 6, 1, 27.
116. D.C. Wahrmundt, D.R. Paul and J.W. Barlow, Journal of Polymer Science, 1978,
22, 2155.
117. S.R. Murff, J.W. Barlow and D.R. Paul, Journal of Applied Polymer Science,
1984, 23, 3231.
118. B.D. Hanrahan, S.R. Angeli and J. Runt, Polymer Bulletin, 1985, 14, 399.
119. M.E.J. Dekkers, S.Y. Hobbs, I. Bruker and W.H. Watkins, Polymer Engineering
and Science, 1990, 30, 1628.
120. H. Yoon, Y. Feng, Y. Quiu and C.C. Han, Journal of Polymer Science: Polymer
Physics Edition, 1994, 32, 1485.
121. F.J. Balta Calleja, L. Girri, T.A. Ezquerra, S. Fakirov and Z. Roslaniec, Journal of
Macromolecular Science, Part B: Physics, 1997, 36, 5, 655.
122. M. Cortazar, J.L. Equiazabal and J.J. Iruin, European Polymer Journal, 1997, 30,
901.
280
Polyester-Based Blends
123. V.A. Danilov, I.I. Perepechko, V.V. Nizhegorodov, Y.K. Godovsky, E.V.
Konyukhova and S.I. Belousov, Vysokomolekularnye Soedinenya, 1994, 36, 1, 57.
124. B.D. Hanrahan, R.S. Angeli and I. Runt, Polymer Bulletin, 1986, 15, 455.
125. P. Sanchez, P.M. Remiro and J. Nazabal, Journal of Applied Polymer Science,
1993, 50, 995.
126. D.G. Hamilton and R.R. Gallucci, Journal of Applied Polymer Science, 1993, 48,
2249.
129. X.Y. Chen and A.W. Birley, British Polymer Journal, 1985, 17, 326.
131. L.H. Wang, Z. Huang, T. Hong and R.S. Porter, Journal of Macromolecular
Science, Part B: Physics, 1990, 29, 171.
132. L.H. Wang, M. Lu, X. Yang and R.S. Porter, Journal of Macromolecular Science,
Part B: Physics, 1990, 29, 155.
135. W. Zheng, Z. Wang and Z. Qi, Polymer International, 1994, 34, 301.
136. W.N. Kim and C.M. Burns, Journal of Applied Polymer Science, 1990, 41, 1575.
137. S.Y. Hobbs, V.L. Groshans, M.E.J. Dekkers and A.R. Schultz, Polymer Bulletin,
1987, 17, 335.
138. R.S. Halder, M. Joshi and M. Misra, Journal of Applied Polymer Science, 1990,
39, 1251.
281
Handbook of Polymer Blends and Composites
140. D. Delimoy, C. Bailly, J. Devaux and L. Legras, Polymer Engineering and Science,
1988, 28, 1, 104.
142. H. Berlisson, B. Franzn and J. Kubat, Plastics and Rubber Processing and
Application, 1988, 10, 1445.
144. A.C.M. van Bennekom, D. van der Berg, J. Bussink and R.G. Gaymans, Polymer,
1997, 38, 20, 5041.
145. N.C. Lin and H. Luang in Reactive Polymer Blending, Eds., W.E. Baker, C.E.
Scott and G.H. Hu, Hanser Publishers, Munich, Germany, 2001, 14.
146. L.Z. Pillon, J. Lara and D.W. Pillon, Polymer Engineering and Science, 1987, 27,
984.
147. L.Z. Pillon, L.A. Utracki and D. W. Pillon, Polymer Engineering and Science,
1987, 27, 562.
148. M.R. Kamal, M.A. Sahto and L.A. Utrachi, Polymer Engineering and Science,
1983, 22, 1127.
149. M.R. Kamal, M.A. Sahto and L.A. Utrachi, Polymer Engineering and Science,
1983, 23, 637.
151. T. Serhatkulu, B. Erman, I. Bahar and S. Fakirov, Polymer, 1995, 36, 2371.
154. T.L. Boykin and R.B. Moore, Polymer Engineering and Science, 1998, 38, 10,
1658.
282
Polyester-Based Blends
155. M. Krumova, S. Fakirov and F.I. Balta Colleja, Journal of Materials Science,
1998, 33, 2857.
156. L.Z. Pillon and L.A. Utracki, Polymer Processing and Engineering, 1986, 24,
375.
158. S.Y. Hobbs, T. Stanley and O. Phansteil, Polymer Preprints, 1992, 33, 2, 614.
159. L.A. Utracki, Polymer Alloys and Blends: Thermodynamic and Rheology, Hanser
Publishers, Munich, Germany, 1989.
283
Blends Based on Poly(Vinyl Alcohol) and the Products Based on This Polymer
8
Blends Based on Poly(Vinyl Alcohol) and the
Products Based on This Polymer
Silvia Patachia-Bodonea
8.1 Introduction
Polyvinyl alcohol (PVA), synthesised in 1924 has much attention focused on it because it
can be used as fibres, protection colloid, sizing agent for textiles and paper, adhesive,
moulded sheets that can replace leather, rubber and other plastic materials, such as suture
fibres in surgery, as a temporary protective film coating, quenching agent for steel, agent
for metal finishing by electrolytic method, synthetic sponges, special hoses and rollers,
cosmetics, soil conditioners and many others.
The most usual formula used today for PVA is written as a 1,3-glycol structure (see
Figure 8.1).
But some structural variables, that result from the PVA manufacturing process and that
strongly influence the PVA properties must be taken into account.
PVA can be obtained only indirectly, by the hydrolysis of different poly (vinyl esters).
The most common polyester used for obtaining PVA is poly (vinyl acetate) (PVAc).
The characteristics of PVAc determined by the polymerisation process represent the first
variable in the production of PVA. So, the degree of polymerisation, the long or short
chain branching, the structural irregularities of PVAc are transmitted to the PVA obtained.
285
Handbook of Polymer Blends and Composites
The second variable in producing PVA is the incomplete removal of the acid residue
from the molecule. Partly hydrolysed PVA can be obtained. PVA grades with different
degrees of hydrolysis are in fact copolymers, e.g., vinyl alcohol-vinyl acetate.
The distribution of hydroxyl groups along the main chain, determined by conditions of
the PVAc hydrolysis process, leads to PVA with block or statistical structure.
Because of the hydroxyl groups, that can promote inter- or intra-molecular hydrogen
bonds, PVA molecules can generate a regular structure, which means crystallinity. PVA is
a typical crystalline polymer. It is highly crystallisable even in the atactic form. The
relationship between the crystallinity and tacticity of PVA is given by the order of
crystallisation: high syndiotactic > atactic > isotactic [1].
Some factors, such as: presence of 1,2- glycol structures, residual acetate groups in the
main chain and molecular branching, tend to disrupt the crystalline structure. So, PVA
exists in two phases: amorphous and crystalline. The degree of crystallinity of PVA
influences most of the PVA properties.
The melt transition temperature (Tm) depends strongly on the tacticity of the polymer
and its purity as measure of the amount of residual acetate and 1,2-glycol content in the
polymer chain. The Tm of PVA generally follows the order: syndiotactic (230-256 C)
> atactic (228-240 C) > isotactic (212-235 C) [1]. In general, the Tm of partly hydrolysed
PVA decreases with increasing residual acetate content and depends also on the method
of preparation of copolymer. The acetate groups in the random copolymers are very
effective in lowering the melting point of PVA, although those in the block of saponified
copolymers have very little effect. The blocky copolymers prepared by alcoholysis lower
the melting point at an intermediate level [1].
The glass transition temperature (Tg) of the copolymer increases monotonically with
increasing vinyl alcohol content.
The water resistance of PVA is higher for grades completely hydrolysed and can be
improved by thermal treatment, which promotes the crystallisation process. The degree
of crystallisation increases almost linearly with the heat-treatment temperature.
The degree of swelling decreases rapidly with increasing polymer crystallinity. The
solubility of PVA depends not only on the degree of crystallisation but also on the structure
286
Blends Based on Poly(Vinyl Alcohol) and the Products Based on This Polymer
of the amorphous region of the polymer chain. Thus, PVA swelling by water is governed
by the average chain length in the amorphous region. This effect is significant at low
degrees of crystallinity. The presence of residual acetate groups decreases the crystallinity
thus increasing the swelling in water.
PVA films also have a low gas permeability coefficient see Table 8.1.
The gas permeability increases with the increase in relative humidity, with the decrease
in the degree of hydrolysis of PVA, with an increase temperature and tends to decrease
sharply as the degree of crystallinity increases. The decrease in crystallinity and the decrease
in Tg would be expected to increase the gas permeability.
Because of its low gas permeability, PVA has excellent flavour-retaining properties.
PVA with its many hydrophilic groups exhibits a high affinity for organic substances
containing hydroxyl, amino and amide groups but it is highly resistant to animal, vegetable
287
Handbook of Polymer Blends and Composites
and mineral oils, and many organic solvents such as aliphatic and aromatic hydrocarbons,
ethers, esters and ketones. PVA films do not swell in these oils and solvents but shrink
and show a slight decrease in weight, possibly because of the dehydration of the film
during immersion, as the films were not completely anhydrous. PVA is an excellent oil
resistant material.
PVA film is hard and brittle at low humidity, but soft and tough at high humidity. Tensile
strength and elongation increase with degree of polymerisation while the strength of
partly hydrolysed grades is less than that of completely hydrolysed grades.
The tensile elastic moduli of films of partly hydrolysed grades are lower over the entire
humidity range than those of completely hydrolysed grades because of their higher
equilibrium moisture content. In general, the dependence of mechanical properties on
relative humidity is greater with partly hydrolysed grades compared to completely
hydrolysed grades.
The thermal stability of PVA is dependent on the method of synthesis, on the degree of
hydrolysis and on the distribution type of hydroxyl groups along the main chain of PVA
[1, 2, 3]. Generally, acid-hydrolysed products of poly (vinyl esters) begin to eliminate
water below 200 C, but similar polymers prepared by alkaline hydrolysis decompose at
temperatures above 180 C [1].
The end-uses of PVA are based on its exceptional properties: water solubility, good thermal
stability, good film forming, high tensile strength, a lack of static charge, excellent
resistance to organic solvents and oils, excellent adhesion to organic and inorganic
materials, good abrasion resistance, non-toxicity and non-carcinogenity and
biodegradability.
In recent years, the number of applications of PVA have been increased by blending PVA
with other polymers, which has led to new materials with superior properties.
The interest in obtaining new materials based on PVA has shifted to special domains
such as biomaterials, gas and organic solvent barrier materials, membranes for separation
of azeotropic mixtures, close-boiling point compounds and mixtures consisting of heat
sensitive compounds by pervaporation and evapomeation, controlled release matrices
288
Blends Based on Poly(Vinyl Alcohol) and the Products Based on This Polymer
for drugs, sensors for pH, temperature, humidity, glucose, hydrogen peroxide, conducting
polymeric materials, catalysis and others.
Blending of PVA with other polymers is also a way to design low-cost materials which
show high properties, which have to be easy to process at low cost and have to be recyclable.
PVA has many possibilities of blending with other polymers because of its structure. The hydroxyl
groups of PVA can interact with other functional groups by physical or chemical routes.
It is very difficult to realise a very precise delimitation between the physical and chemical
blended polymers because during the blending process the reactive hydroxyl groups can
participate in many different chemical reactions such as etherification, esterification,
acetalysation, addition and so on.
So, the in situ formation of covalent bonds between blend components during melt mixing
or reactive extrusion is now well established as a technique for preparing novel materials
(see chapter on Reactive Blending in Volume 3 of this series). Thus, PVA can interact
with hydrocarbon polymers that contain conjugated double bonds, with polyelectrolytes
such as polyacids, polybases, polysalts and with synthetic and natural polymers containing
nonionisable polar groups as hydroxyl, amine, amide and ester groups. These interactions
can be of physical (hydrogen bonds) or of chemical nature.
PVA can also be compatibilised with polyolefins such as polyethylene (PE) and
polypropylene (PP) by using special compatibilisers. The latter ones contain molecules
that can interact physically or chemically with hydroxyl groups and moieties that contain
hydrocarbon chains that show a good compatibility with polyolefins (PO).
Maleic anhydride (Ma) can react with hydroxyl groups determining the ester bond
formation and the hydrocarbon chain causes good adhesion with the PO.
The hydroxyl groups of PVA can react with dialdehydes, cyclic anhydrides or alkyl halides
such as 1,2 dibromoethane that lead to a semi-interpenetrating network (S-IPN).
If a polymer blended with PVA is also crosslinked with different curing agents, a fully
interpenetrating network (IPN) is obtained.
289
Handbook of Polymer Blends and Composites
PVA also has the capacity of self-curing, thus generating ether bonds among the
macromolecular chains. The network obtained can entrap other polymer chains, leading
to a so-called snake-cage hydrogel.
An important category of polymers that can be blended with PVA is that of natural
polymers such as: starch, chitin and its derivatives, silk fibroin, polysaccharide-chitosan,
cyclodextrin, cellulose, soluble collagen, gelatin, poly(allyl biguanido-co-allylamine). These
blends are especially important from the point of view of new materials design and the
effective utilisation of natural polymers that constitute a very large biomass.
Generally, these blends are used as biocompatible materials, such as entrapping materials
for immobilisation of enzymes in biosensors, as permoselective membranes, such as
hydrogels that can control drug release, as gas barrier materials, as food packaging or as
a substrate in room-temperature phosphorimetry.
There are many ways for PVA and its copolymers to blend with other polymers: by
casting the solution polymers and solvent evaporation, by repetitive freezing and thawing
cycles, by polymer bulk melt mixing, by extrusion of the melt mixture, by conventional
or shear controlled oriented injection moulding, by dispersing the polymers in a PVA
matrix, by polymerisation in the presence of PVA solution or in a PVA swollen film.
The different forms of blends that can be obtained are film, foam, hydrogel or thick
sheets.
The final characteristics exhibited by blend products are dependent on the properties of
the individual components as well as on the phase morphology.
The phase morphology can be manipulated to obtain the desired product properties, by
optimisation of blend composition and processing conditions. For example, a
heterogeneous structure, containing oriented fibrils and lamellae has a lower gas
permeability than that of the finer dispersed particles, but the latter determines the
improvement of the mechanical properties.
The mechanical properties of blends are also influenced by the matrix type, by the
composition of the minor phase, by the presence, type and concentration of the
compatibiliser. It was generally observed that the compatibiliser determines the increase
of the tensile strength and simultaneously the increase of the elongation to break. The
tensile strength increases with the increasing of the curing degree of the polymers and
decreases with increasing moisture and plasticiser content of the blend.
290
Blends Based on Poly(Vinyl Alcohol) and the Products Based on This Polymer
The PVA/polyelectrolytes blend hydrogels show active mechanical behaviour. The blends
can generate a mechanical tension as a response to environmental changes such as: solvent,
pH or temperature.
Blends based on PVA and its copolymers show good swelling in water. The blend swelling
decreases with the increasing of the degree of crosslinking of the blend and increases
with increasing temperature.
The conductivity of blends, generally decreases with increasing PVA content and increases
with the increasing of moisture content.
The drug release rate is higher for S-IPN than that of full IPN.
The blends properties could be also changed by changing the PVA characteristics such
as molecular weight, degree of hydrolysis or tacticity.
Details on some new blends of PVA and its copolymers with other polymers are presented
in Section 8.2. Recent research has shown that PVA is still a rich source of new materials.
PVA/PPV blends were obtained by mixing aqueous solutions of PVA and a PPV precursor [4].
These blends show photo- and electro-luminescence and can be applied to light-emitting
diodes [4]. The luminescence characteristics of the blends are dependent on their composition
and conditions of treatment. The light emitted by the blends shifted from yellow-green
(550 nm) to blue (485 nm) as the PVA content increased as shown in Figure 8.2.
The addition of PVA into PPV improves the relative photoluminescence intensity per
PPV unit of PPV/PVA blend, by more than two orders of magnitude. The light emission
efficiencies of these PPV/PVA light-emitting diodes were higher than that of a PPV light-
emitting diode (about three orders of magnitude).
291
Handbook of Polymer Blends and Composites
Figure 8.2 Photoluminescence spectra of (a) PPV; (b) PPV-PVA1; (c) PPV/PVA2; (d)
PPV/PVA3 polymers [4]
Reproduced with permission from W.P. Chang and W.T. Whang, Polymer, 1996, 37, 3493,
Figure 4. Copyright 1996, Elsevier Science
292
Blends Based on Poly(Vinyl Alcohol) and the Products Based on This Polymer
Strong physical and chemical interactions like hydrogen bonds and the formation of
ester bonds can be realised between PVA and PAA functional groups. The high reactivity
of OH and COOH groups could determine the ability of the polymers to crosslink in
the presence of different curing agents such as dialdehydes, diols, acryl amide, etc., to
obtain IPN or S-IPN.
The blends obtained can be used to produce membranes for liquid mixture separation by
pervaporation or evapomeation (a membrane separation technique which avoids the
membrane swelling as the membrane is not in direct contact with fee solution only the
feed vapour), adsorbent materials for waste water treatment, ion exchange membranes,
controlled drug release systems, artificial muscles, biomaterials, sensors of humidity,
pH, temperature, etc.
The main characteristics of PVA and PAA as blend components are: high molecular weight
(from 50,000 to 186,000), and high degree of hydrolysis (80%-100%) for PVA (except
when the PVA is used as a protective colloid in the polymerisation of the acrylic acid which
has a weight average molecular weight (Mw) of 9,000-10,000 and 80% degree of hydrolysis),
and very high or very low molecular weight (250,000-450,000 or 2,000) for PAA [4-21].
1. Mixing of PVA and PAA aqueous solutions, solvent evaporation and thermal treatment
of the resulting film.
2. Mixing of PVA and PAA aqueous solutions with curing agent solutions, solvent casting
and thermal treatment of the resulting film.
3. Repetitive cycles of freezing and thawing of the aqueous solutions of polymer mixture,
in the presence or the absence of the curing initiators.
5. Acrylic acid polymerisation in the matrix of PVA, in the presence of curing agents and
initiators.
293
Handbook of Polymer Blends and Composites
7. Swelling of PAA dried hydrogels with an aqueous solution of PVA and application of
freezing and thawing cycles.
Using small amounts of acids that are catalysts for esterification reactions.
Changing the crosslinking degree of the polymers by the thermal treatment of polymer
blends, increasing the number of freezing-thawing cycles, changing of the type of
crosslinking agent that determines whether IPN or S-IPN are obtained, changing of
the amount of crosslinker, the films treated with crosslinkers in the gaseous phase, or
the film treated with salt solutions, such as CuSO4 or Na2S.
PVA and PAA are compatible polymers through the whole range of composition [5-10].
The blends obtained are homogeneous and the films are transparent showing good clarity
[5, 11] or semitransparent [12]. However an heterogeneous IPN [13], was obtained by
acrylic acid polymerisation in a PVA matrix.
The blends obtained could exhibit a continuous or a microporous [14] structure. The
latter was obtained by using the freezing-thawing technique applied to a mixture of
PVA, PAA and poly(allyl amide) hydrochloride. The characteristics of the porous structure
depend on the hydrogels cooling rate: rapidly frozen gels (3.8 C/s) had a more uniform
and fine porous structure than slowly frozen gels.
The increasing of the number of freezing-thawing cycles determines the increase of the
pore dimensions [14].
The blends are water insoluble [15], but they can swell in different solvents: water, acetone
[14], aqueous solutions of acids or alkalis [15]. The swelling equilibrium was attained in
few hours at 25 C.
The method of obtaining the blends influences their swelling ratios by inducing different
degrees of crosslinking. So, if the number of freezing-thawing cycles increases, the swelling
ratios decrease significantly [18].
In general, the swelling ratios increase with the increase the temperature up to 40 C
[13]. The swelling ratio depends on temperature. This relationship is different, according
to the blend composition [16, 17].
294
Blends Based on Poly(Vinyl Alcohol) and the Products Based on This Polymer
The pH values strongly influence the swelling behaviour of the blends. For example, the
difference between the swelling ratio of IPN 4:6 immersed in a solution of pH = 4 and
the swelling ratio of IPN 4:6 immersed in a solution of pH = 7 is 2.0 [16].
The blends show a single Tg value, like all compatible systems. A positive deviation of Tg
from the ideal mixing rule was observed [10]. The Tg gradually decreases from 129 C to
84 C with an increase of the PVA content from 0% to 50%. Above this composition the
Tg value remains constant [5].
The mechanical properties of the blends are dependent on the blend preparation procedure
and on the composition of the blend. For IPN membranes, the tensile strength and the
elongation at break decrease with the increasing of PAA content, because of crosslinking
of the latter by ethylene glycol (EG) molecules. For S-IPN membranes, a different behaviour
was observed: the tensile strength increases and the elongation at break decreases with
the increasing of PAA content [11].
Both types of hydrogels, normal and microporous, are elastic [14, 16].
The mechanical properties depend on the number of freezing and thawing cycles [14,
18] and on the nature of the swelling agent [14].
Blends have a lower initial isothermal storage modulus (at 160 C) and in time, this
increases one order of magnitude compared to that of the pure polymers [10].
A remarkable property of PVA/PAA blends is the active mechanical behaviour. The blend
could generate a mechanical tension, corresponding to different changes in environment,
such as: solvent [14], pH [15] and temperature [13] changes.
The permeation rate and separation factor are strongly dependent on membrane
composition and structure, on feed mixture composition and on temperature.
The drug release could be controlled by the degree of swelling/deswelling of IPN hydrogels
which was dependent upon pH and temperature [21]. For example, the release of
indomethacin incorporated into these hydrogels showed pulsatile patterns in response to
both pH and temperature.
295
296
Membrane composition
80/20 95.6/4.4* 9 60 1400-0 -
30 75 5800
90/10 27 60 9000
60 75 2800
80/20 65 60 1500
125 75
50/50 120 75 260
95/5 550 60 150
Membrane composition
Blends Based on Poly(Vinyl Alcohol) and the Products Based on This Polymer
Table 8.2 Continued
No. Composition of the Composition of liquid mixture Permeation Temp. C Separation Activation energy Refs.
membrane based on rate, g/m2.h factor of permeation, Ea,
PVA/PAA blend kJ/mol
3 90/10 Fusel oil** Alcohol 5000 60 10 Ea,water= 49.4-41.7 [20]
mixture with Ea,ethanol= 60.8-55.7
10%-30% of
water
4 75/25 Acetic acid- 90/10 5.6 30 795 - [17]
water
5 10/90 Methanol/ 10/90 120 30 460 - [5]
toluene
30/70 265 30 50 -
6 10/90 Ethanol/ 10-90% 25-480 30 300-80 - [5]
toluene ethanol
7 20/80 S-IPN Ethanol/ 10/90; 90/10 30; 560 50 110; 3.5 19.2 [11]
benzene
30/70 S-IPN 10/90; 90/10 12; 460 50 650; 1.9 27.6
40/60 S-IPN 10/90; 90/10 6; 360 50 1100; 53 31.4
8 80/20 Phenol/water 80/20 50 30 3580 - [6]
*azeotropic point
**fusel oil is a mixture of five alcohols ranging from C2 and C5 and water
297
Handbook of Polymer Blends and Composites
PVA/PAA blends are good adsorbent materials for different organic compounds like
Chrysoidin (104-106 mg/g), Bismark Brown Y (72 mg/g) [12].
The PVA/PAA blends films show size exclusion properties. The partition coefficients for
theophylline (K = 0.36 0.04) and for fluorescein isothiocyanate (FITC)-dextran (K = 1.0)
were determined [18].
The PVA/PAA blends could act as an ion exchange membrane if they are treated with
1,2-dibromoethane in the gas phase [9].
1. By dissolving the two polymers in a mixture of 30% dimethyl sulfoxide (DMSO) and 70%
water and then applying the freezing-thawing technique (from 50 C to 15 C) [22], or
By the first technique, an opaque hydrogel, with a heterogeneous structure was obtained,
because of the partial miscibility of PVA and PSAc [23]. The freezing-thawing cycles
influenced the hydrogel structure [22].
The microscopic changes of both kinds of polymer segments correlate with the change of
macroscopic swelling ratio as shown in Figure 8.3 [22]. The gel swelling ratio decreases
with the increasing number of the freezing-thawing cycles.
PVA/PSAc gels are hot-water-soluble. PVA/PSAc gels are a rubberlike, tough material,
showing a high elasticity and a melting point of about 60 C [22]. In an electric field,
PVA/PSAc gels, suspended in an electrolyte solution, show a bending deformation. Their
capability for direct conversion of electrical into mechanical energy and good mechanical
properties indicate the feasibility of designing a biomimetic machinery system [22].
By the second technique of gel preparation, a composite hydrogel was obtained. The
dispersed PSAc particles in the PVA matrix, produce an extremely high volume expansion
when the gel medium was changed from methanol to water. The value of the volume
expansion ratio was about 20 [23]. The swelling and dehydration kinetics of these
composite hydrogels are different from conventional composite gels because of the high
298
Blends Based on Poly(Vinyl Alcohol) and the Products Based on This Polymer
Figure 8.3 Relationship between the freezing-thawing cycles number and t2 for
different types of polymer segments and the swelling ratio of the PVA/PSAc gels [22]
Reproduced with permission from T. Shiga, K. Fukumori, Y. Hirose, A. Okada and T.
Kurachi, Journal of Polymer Science, Part B - Polymer Physics, 1994, 32, 85. Copyright
1994, John Wiley & Sons, Inc
degree of dimensional change during the release of PSAc. During the swelling and
dehydration processes a distortion of the hydrogel shape was observed (Figure 8.4 and
Figure 8.5) [23].
In spite of the high amount of water absorbed, the mechanical strength and structural
integrity are maintained although they are not chemically crosslinked. No cracks were
observed in the hydrogel during the large expansion and contraction cycles [23]. This
hydrogel could act like an osmotic pump.
299
Handbook of Polymer Blends and Composites
Figure 8.4 Observed hydrogel diameter, d, and the hydrogel shape changes during
swelling in water. (A) small cylinder, d(t0) = 1.15 cm and L(t0)= 0.26 cm; (B) large
cylinder, d(t0) = 1.15 cm and L(t0)= 1.60 cm; An elliptic hyperboid shape was observed
when large cylinders were swollen [23]
Reproduced with permission from D. Graiver, S. H. Hyon and Y. Ikada, Journal of
Applied Polymer Science, 1995, 57, 1299. Copyright 1995, John Wiley & Sons, Inc
Hydroxyl groups of PVA and sulfonic groups from another macromolecular chain could
interact through hydrogen bonding producing an increased miscibility of the two polymers.
A PVA/poly(styrene sulfonic acid) (PSSA) blend was prepared by mixing their aqueous
solutions.
300
Blends Based on Poly(Vinyl Alcohol) and the Products Based on This Polymer
The miscibility of these polymers increases with increasing molecular weight of PVA,
with decreasing degree of hydrolysis of PVA (from 99% to 88%) and with the increase of
the total concentration of the polymer solutions [24].
The PVA/PSSNa blend was obtained by casting an aqueous solution of polymer mixtures
(PVA with Mw of 124,000-186,000 and degree of hydrolysis of 99% and PSSNa with Mw
of 70,000). The weight fraction of PSSNa in membrane was between 0.35 and 0.65. The
resulting films were crosslinked with 1,2-dibromethane in the gaseous phase. A S-IPN in
301
Handbook of Polymer Blends and Composites
which polyelectrolyte (PSSNa) chains are trapped inside a PVA-based network was
obtained [9].
Although the reduced viscosity variation of the blend solutions, depending on the PSSNa
concentration, shows a partial miscibility of PVA and PSSNa within the PSSNa weight
ratio values ranging from 0 to 0.6 [24], the study of the crosslinked blends films shows a
total miscibility of these polymers [9].
Blends with very good film clarity and high mechanical resistance [9] were obtained. The
ion exchange capacity increases with the time of crosslinking (tc) [9].
Blends were obtained by dissolving TBES, in a 10% PVA aqueous solution (TBES/PVAL
<30/70 w/w) or in an aqueous potassium hydroxide solution (TBES/PVA ratio is 50/50
w/w) [25].
302
Blends Based on Poly(Vinyl Alcohol) and the Products Based on This Polymer
The blends obtained are homogeneous in a limited weight ratio (up to a TBES/PVA ratio
50/50 w/w).
TBES and TBES/PVA blends could react with metal ions such as Co2+ and Ni2+, forming
complexes that contain two chelated rings. The metal complexes of these blends form S-
IPN by chelation of TBES with metal ions (TBES-K).
The blend film complexed with Co2+ could reversibly change its colour from blue to pink
as a function of the atmospheric moisture, and could be used as moisture sensor.
The blend properties were influenced by blend composition and by complexation as can
be seen in Table 8.4.
303
Handbook of Polymer Blends and Composites
These blends were obtained by mixing aqueous solutions of PVA, SPAN or PAPSAH
with PVA [26].
The strong interaction between PVA and these self-acid-doped conducting polyanilines
leads to their partial miscibility.
The conductivity of the blends decreases slightly with the increasing PVA content in the
SPAN or the PAPSAH matrix.
The blends with a 1:1 mole ratio consist of two phases: a P3TESH phase and a P3TESH/
PVA complex phase where the two polymers are intimately mixed.
The complex phase showed a Tg and a side chain relaxation temperature higher than
that of the pure components.
The blending of P3TESH with PVA determines a significant undoping and a decrease of
the material conductivity (from 10-2 S/cm for the pure component to 10-6 S/cm for the
blend), while the P3TESNa conductivity (10-7 S/cm) drops by one order of magnitude
after blending with PVA. The doping of the latter blend by a protonic acid leads to a
higher conductivity of 10-3 S/cm [27].
The P3TESH segments that are not self-acid-doped (or containing SO3H group) exhibit
a red shift of the UV-VIS absorption maximum by 44 nm, while the P3TESNa/PVA
blend shows a higher shift of absorption maximum by 76 nm, accompanied by a colour
change from pale orange-yellow to bright orange-red [27].
304
Blends Based on Poly(Vinyl Alcohol) and the Products Based on This Polymer
PDMeDMPCl with a Mw of 85,000) [9]. The weight ratio of PDMeDMPCl in the blend
ranges from 0.25 to 0.75. Then, films were crosslinked with 1,2-dibromethane in the
gaseous phase.
The blend is totally miscible but a lower film clarity compared to PVA/PSSNa or PVA/
PAA films was observed [9].
The PVA/PDMeDMPCl blend membrane showed an ion exchange capacity that increased
with the tc [9].
The blend membranes swell in water and the swelling ratio (g) is shown as a decreasing
function of tc, and an increasing function of the weight fraction of the polyelectrolyte as
is shown in Figures 8.7 and 8.8.
305
Handbook of Polymer Blends and Composites
Figure 8.8 Swelling ratio (g) of PVA/PDMeDMPCl S-IPN membranes versus weight
fraction of polyelectrolyte [9]
Reproduced with permission from M. Metayer and C.O. MBareck, Reactive & Functional
Polymers, 1997, 33, 311. Copyright 1997, Elsevier Science
PVA/SAlg blends in hydrogel form, were obtained by mixing aqueous solutions of PVA
(Mw of 22,000 [PVA1]; 50,000 [PVA2]; 93,000 [PVA3]; degree of hydrolysis is 99.9%)
and sodium alginate (mannuronate/glucuronate = 1.78, 1.78, 0.43, respectively, and Mw
is 380,000 [SAlg 1]; 550,000 [SAlg 2]; 650,000 [SAlg 3], respectively), with two cycles
of freezing (16 h at 20 C ) and thawing (8 h at 4 C) [28].
PVA has an auto-curing capacity and other polymer chains, like SAlg, could be trapped
in its network, obtaining a snake-cage hydrogel.
Both PVA and SAlg polymers are bio-compatible and nontoxic so, its could be applied to
the drug controlled delivery systems obtaining. The mechanical properties of the PVA/
SAlg hydrogels are dependent on SAlg type and content and on PVA molecular weight.
Figure 8.9 shows the mechanical properties changes versus the SAlg content and the PVA
characteristics.
306
Blends Based on Poly(Vinyl Alcohol) and the Products Based on This Polymer
Figure 8.9 Modulus and elasticity of the PVA/SAlg blends dependence on SAlg content
and PVA characteristics [28]
Reproduced with permission from C. Desmarquet, E. Fabregue, G. Cassanas and M. Vert,
Macromolecular Symposia, 1997, 114, 285. Copyright 1997, Wiley VCH
PVA/PSA blends were obtained by the solvent-cast method from homogeneous aqueous
solutions of PVA (number average molecular weight (Mn) of 88,000; degree of hydrolysis
is 99%-100%) and PSA (Mw = 1.4 x 104; mole ratio linkage/ linkage = 27:73) [29].
The homogeneity and transparency of the blend films are proof of the high compatibility
of these polymers in a wide composition range [29].
PSA is a fully biodegradable polymer, desirable from ecological point of view, and is a
polycarboxylate polymer which makes it attractive for applications utilising
polycarboxylate functionality.
The Tg of PVA in the blend, determined by differential scanning calorimetry (DSC), shifted
towards higher temperatures with increasing the PSA content up to 50 wt % (from 70 C
to 83 C). If the PSA content increases to 70 wt %, the blend did not show any Tg [29].
307
Handbook of Polymer Blends and Composites
The shift of the bands to lower wave numbers observed in spectra is explained by the
formation of hydrogen bonds between the hydroxyl groups in PVA, as an electron acceptor,
and the carbonyl oxygen in PSA, as an electron donor.
Energy dispersive X-ray spectroscopy (EXD) measurements showed that the surface
composition of the blends is the same as the bulk composition. This is very important for
their use as biomedical and biodegradable materials because the homogeneity of the
biomedical materials gives better mechanical properties [29].
Tough and rubber-like, physically crosslinked hydrogels, which have high values for Youngs
modulus even with a relatively high water content, were obtained [30]. These moduli were
twice that of PVA hydrogel prepared by repeated freezing and thawing, and increased with
increasing PSLG content. This is explained by the physical crosslinks formed among the
PVA and the PSLG polymer chains in a polymer-polymer complex [30]. In this case, the
-OH groups of PVA form a complex with the peptide groups of PSLG, by hydrogen bonds.
A PVA/PPB polymer blend was prepared by mixing a methanol solution of PPB with a
solution of PVA in DMSO to form polymer complexes between the hydroxyl groups of PVA
and phenyl boronic acid. The resulting gel was deposed on the surface of a platinum electrode,
dried and then immersed into pyridine and then into triisocyanate solution in pyridine for
crosslinking the hydroxyl groups that did not react with the phenyl boronic acid [31].
The membranes prepared from this blend, immersed in a glucose solution, showed
reversible swelling changes with time in a glucose concentration-dependent manner.
Therefore, this property was used for the design of novel electrodes to sense glucose
concentration changes eliminating the need to use glucose oxidase. A high selectivity,
sensitivity, reproducible output and a minimal protein adsorption were found at
physiological pH conditions [31].
308
Blends Based on Poly(Vinyl Alcohol) and the Products Based on This Polymer
Blends of PVA and PQ, which are polymers with both hydrophilic quaternary ammonium
groups and hydrophobic ester groups in their structure, were studied because of their
superior properties. Considerable interest has been focused on their diverse applications
in electrically conductive polymers, polymer catalysts and antibacterial polymers.
PVA/PQ blend films were prepared by casting from aqueous solutions [32]. The PQ/PVA
mixing ratios are between 10/90 and 70/30.
Formulae of PQ and its copolyesters blended with PVA are shown in Figure 8.10.
309
310
Table 8.5 The thermal and electrical properties of PVA/PQ blends with different
compositions and different moisture contents [32]
Code Blend Thermal properties Electrical
composition properties
In dried state In presence of moisture
Tg (C) Tm (C) Tdec (C) Water content Tg (C) Tc (S/cm)
(%)
PVA 71.6 229.7 274.4 4.2 68.3 -
0.5 72 -
PQ/PVA 10/90 50.8 220.6 288.6 0.5 82.5 1.9 x 10-11
30/70 70.3 215.7 255.0 0.5 74.8 3.8 x 10-10
40/60 41.3; 93.1 205.0 253.2 1.9 66.9 4.1 x 10-10
4.6 42.3 1.1 x 10-5
Handbook of Polymer Blends and Composites
The thermal and electrical properties of blends with different compositions and different
Blends Based on Poly(Vinyl Alcohol) and the Products Based on This Polymer
Three types of PVA were used in these blends: syndiotactic (sPVA), isotactic (iPVA) and
atactic (aPVA), in order to obtain information about the influence of the PVA structure
on the miscibility of the blends.
These blends could be used as surgical sutures or long-term carriers for drugs and moulded
plastics.
1. Dispersion of PMMA microgels (200 nm diameter) covered with thin layer (prepared
by soap-free polymerisation) of poly(hydroxyethyl methacrylate) in a PVA (degree of
polymerisation is 2000) solution, under vigorous magnetic stirring for several hours
and then treated by ultrasonic irradiation for 10 hours and the film obtained by
casting the mixture.
2. The soaking of mixed film in a solution of monomers with the same composition as
the microgel and carrying out seeded polymerisation to realise the connection of the
microgel particles. A polymer blend with a bicontinuous structure was obtained [35].
The PVA could be extracted with hot water and a porous film remains.
This method is recommended for obtaining materials with controlled morphology [35].
311
Handbook of Polymer Blends and Composites
Sequential IPN technique: PVA (9% aqueous solution; Mw is 124,000), AAM, BisAAM,
K2S2O8, were heated at 70 C, for 8 hours, cooled at room temperature, than aqueous
glutaraldehyde/HCl solution was added [37].
This blend had good mechanical properties. The presence of PVA in the blend improves
the mechanical properties.
The membranes can preferentially swell in water and other different aqueous solutions
like water/ethanol and water/acetic acid mixtures [37].
This behaviour of the membranes mentioned previously, may be used in the separation
processes by pervaporation.
The ability to swell in water and the good mechanical properties suggests that the PVA/
PAAM hydrogels could be used like controlled-release systems for drugs. These polymer
networks were evaluated as drug delivery devices using Crystal Violet (CI:42555:1) and
Bromothymol Blue as model drugs [36] (Figure 8.11).
Specific conductance values of the composite solutions are higher than that of PAAM [38].
A PVA/PEEA blend was obtained by melt blending of PVA (DP = 1400; Mn = 67,000;
degree of hydrolysis is 88% and 99%) and PEEA (with 18% w/w acrylic ethyl ester)
312
Blends Based on Poly(Vinyl Alcohol) and the Products Based on This Polymer
Figure 8.11 Drug release versus time curves for Crystal Violet-loaded capsules. C-SIPN
hydrogel; D-Full IPN hydrogel [36]
Reproduced with permission from B. Ramaraj and G. Radhakrishnan, Journal of Applied
Polymer Science, 1994, 51, 6, 979, Figure 2. Copyright 1994, John Wiley & Sons, Inc
[39]. The ultimate tensile strength decreases with the increase in PEEA content [39]. The
elongation at break passes through a maximum for the blend with 10% of PEEA [36].
PEEA produces good adhesion of the PVA dispersion in the PE matrix. This blend could
be used as packaging material [39].
This blend was obtained by precipitating the intermolecular complex between PVA and
P(AN-AM-AcAc) from an aqueous solution of both polymers [40].
313
Handbook of Polymer Blends and Composites
This complex could be formed through hydrogen bonding between the hydrogen of acrylamide
group and carboxylic group from P(AN-AM-AcAc) and the oxygen of the hydroxyl group
of PVA. The optimal conditions for complex formation are: acrylic acid content of 41%-
46%, P(AN-AM-AcrAc)/PVA ratio: 1/20-1/10, pH of the solution: 5.0-5.5 [40].
This complex is superior in thermal stability and had a lower melting point, compared to
the pure components.
The mechanical properties of the complex are enhanced compared to the initial
components [41].
Better mechanical properties were obtained for a P(AN-AM-AcAc)/PVA ratio of 1/10 [41].
This material could be used to prepare new microcomposite materials with very good
mechanical properties. It could be also a way to use the polyacrylonitrile (PAN) waste
fibres, that could be transformed in P(AN-AM-AcAc) by alkaline hydrolysis followed by
acidifcation.
Two types of modified PVA were studied in blends with PVA: oxidised PVA (PVA(B))
and acetalised PVA (PVA(D)) [42].
PVA/oxidised PVA blend was obtained by mixing aqueous solutions of PVA (degree of hydrolysis
is 71.0 mole%, degree of polymerisation is 700) and oxidised PVA(B) (obtained by thermo-
oxidative treatment of aqueous solution of PVA with degree of polymerisation = 931 and
degree of hydrolysis = 92.0 mole%, at 200 C, in the presence of oxygen, for 2 hours) [42].
This blend could be used as dispersant in suspension polymerisation of vinyl chloride [42].
PVA/acetalised PVA blend was obtained by mixing aqueous solutions of PVA (DH =
71.0 mole%, DP = 700) and acetalised PVA(D) (17.7 mole % degree of acetalisation,
prepared by the reaction of PVA Solviol R with formaldehyde at 60 C) [42].
314
Blends Based on Poly(Vinyl Alcohol) and the Products Based on This Polymer
PVA/PVAc was obtained by casting solutions of the mixture of two homopolymers (PVA
with Mw = 124,000-186,000 and 87%-89% hydrolysed; PVAc with Mw = 237,000) in
ethanol/acetone mixture.
PVA/PVAc is a partially compatible blend exhibiting phase separation in the solid state
[43, 44].
PVA/PEG blend was obtained by mixing the two aqueous solution of polymers (PVA
with Mw of 24,000 and degree of hydrolysis is 98.5 mole% and PEG with Mw of
20,000) [45].
The blend of PVA with PEG-modified glucose oxidase could be used as glucose sensor
characterised by the linearity of the calibration curve in the range of concentration by
5-500 x 10-5 mol glucose/l [46].
PVA/PEG membranes crosslinked by aldehydes and sodium salts were used in water
desalination by pervaporation [47].
PVA/PEO blends were obtained by mixing aqueous solutions of PVA (Mw = 1.0 x 105)
and PEO (Mw = 1.3 x 105) [48].
PVA and PEO are incompatible polymers; a multiphase system was shown at all
temperatures and concentrations [48].
All the blend compositions determine the increase of the ultrasonic velocity with increasing
temperature and solution concentration [48]. The nonlinear behaviour indicates that the
present blend is a multiphase one (Figure 8.12).
For all compositions, the apparent viscosity decreases with increase in shear rate that
shows the pseudoplastic nature of the blend, Figure 8.13 [48].
The viscosity decreases nonlinearly with increase in the percentage composition of PEO
(Figure 8.14) [48]. This blend could be used to reduce the turbulent drag of the fluid [48].
315
Handbook of Polymer Blends and Composites
Figure 8.12 Ultrasonic velocity versus weight percentage of PEO in PEO/PVA blends
of various concentrations at 30 C [48]
Reproduced with permission from R. Paladhi and R.P. Singh, Journal of Applied Polymer
Science, 1994, 51, 1559. Copyright 1994, John Wiley& Sons, Inc
316
Blends Based on Poly(Vinyl Alcohol) and the Products Based on This Polymer
Figure 8.13 Shear rate dependency of viscosity for 7% polymer blend PEO/PVA
solutions with varying proportions of PEO at 30 C [48]
Reproduced with permission from R. Paladhi and R.P. Singh, Journal of Applied Polymer
Science, 1994, 51, 1559. Copyright 1994, John Wiley & Sons, Inc
317
Handbook of Polymer Blends and Composites
Figure 8.14 Apparent viscosity versus weight percentage of PEO in PEO/PVA blends of
various concentrations at 30 C [48]
Reproduced with permission from R. Paladhi and R.P. Singh, Journal of Applied Polymer
Science, 1994, 51, 1559. Copyright 1994, John Wiley & Sons, Inc
This composite material could be used in electrochromic displays and secondary batteries
obtaining.
318
Blends Based on Poly(Vinyl Alcohol) and the Products Based on This Polymer
Figure 8.15 Visible absorption spectra of PANI/PVA composite film at various applied
potentials [49]
Reproduced with permission from M. Morita, Journal of Polymer Science, Part B-
Polymer Physics, 1994, 32, 231. Copyright 1994, John Wiley & Sons, Inc
Films containing PANI.HCl particles dispersed in a PVA matrix had good conductivity even
at very low PANI.HCl loading (for 0.007 wt% load, the conductivity is 1.12 x 10-5 S/cm);
The conductivity increases with the increasing PANI concentration (Figure 8.16).
Because of their low percolation threshold these blends could be used to improve the
processability of PANI.HCl, hard soluble and infusible electroconducting polymer, and
these blends are cheaper than PANI.HCl [51].
319
Handbook of Polymer Blends and Composites
PVA/PMO blend films were prepared by a solvent casting method from an aqueous
solution of PVA (Mn = 88,000) and poly(2-methyl-2-oxazoline) (Mn = 10,200) [52].
PVA and PMO are miscible throughout the whole range of compositions.
320
Blends Based on Poly(Vinyl Alcohol) and the Products Based on This Polymer
A single Tg for each blend, shifted towards a lower temperature region with the poly(2-
methyl-2-oxazoline) content increasing was found (Figure 8.17) [52].
PMO could be used as a compatibiliser agent in blends of PVA with other polymers.
321
Handbook of Polymer Blends and Composites
PVA and PA6 blends were studied in binary and ternary systems. The role of the third
polymer, i.e., PEEA, introduced in the initial PVA/PA-6 blend is to produce a good adhesion
between the blend phases, to improve the mechanical properties and to ensure the
compatibility between PVA and non-polar polymers as PE.
These blends were obtained by melt mixing of PVA (with Mw = 67,000 and degree of
hydrolysis of 88% or 99%) and PA6 (Mn = 20,600) at 230 C for 7 minutes.
Blends films were made by compression moulding between Teflon sheets at 250 C and
70 kg.cm-2 [39]. Partially miscible blends were obtained.
These showed good mechanical properties in the whole range of composition: the ultimate
strength (), except for the blend with PVA/PA6 (1:2), was above 50 MPa; the strain
values (b) of PVA were improved with the addition of PA6 and remained at high values
for all the blends (550%-600%).
The strain of the blends was strongly reduced by using PVA with a 99% degree of
hydrolysis instead of PVA with 88% DH (from 510% to 24%) [39].
Except for the blends with PVA/PA6 ratios of 3/1 and 2/1 which had two Tg (57-72 C
and 53-69 C, respectively), most of the blend compositions showed a single Tg value
(from 53 C to 57 C) situated between the Tg of the pure components (73 C and 27 C,
respectively), [39].
There was a Tm depression from 218 C to 209 C for pure PA6 to a PVA/PA6 blend
(3:1), respectively, which was a function of the composition of the blend [39].
PA6 can be considered as a suitable polar polymer to facilitate the mixing between PVA-
polar polymer and PE-nonpolar polymer. PA6 can be reactively blended with different
polymers that can act as compatibilisers between PA6 and PE. These blends could be use
as packaging materials [39].
PVA/PPy blend was obtained by polymerisation of pyrrole within the PVA matrix in
presence FeCl3 as oxidant that was molecularly dispersed through the formation of chelate-
322
Blends Based on Poly(Vinyl Alcohol) and the Products Based on This Polymer
type complexes with the polymeric matrix [53]. This composite could be used as
conducting polymeric material.
Another application of this blend is for a new glucose sensor. PVA/PPy blend including
glucose oxidase (GOX) was obtained by the electro-polymerisation of Py on an electrode
spin coated with PVA/GOX. The activity of GOX is independent on film thickness, but
the current based on the reaction of GOX with glucose rapidly decreased with the increase
of the film thickness on the electrode. Kojima and co-workers [54] consider that the
electrons arising from the enzyme reaction were easily trapped in the PPy/PVA/GOX
film due to an increase of the film thickness.
PVA/PVP blends were obtained by mixing the aqueous solutions of the two polymers.
The blends exhibit a good miscibility through the entire range of composition, due to
strong interactions between the proton accepting carbonyl groups from pyrrolidone rings
and the hydroxyl groups of PVA [55].
The interaction parameters (12) were negative for all blends, independent of the molecular
weight of PVP.
The enthalpy of fusion in the blend as well as the melting point decreased with the
increasing PVP concentration.
The higher degree of hydrolysis of PVA and the lower molecular weight of PVP promoted
higher crystallinity of the blends.
cp at the Tg gave a similar behaviour for the blends, independent of the molecular
weight of PVP or degree of hydrolysis of PVA.
In spite of this evidence for PVA and PVP miscibility through the entire range of
compositions, Narayana and co-workers [56], determining the electrical properties of
the blends, found that the maximum dielectric loss corresponds to the ratio PVA/PVP of
25/75 and assigned this fact to a maximum heterogeneity in this blend, which gives rise
to maximum trapping sites [56] and X-ray diffraction (XRD) studies indicated that this
sample had a more amorphous nature.
323
Handbook of Polymer Blends and Composites
S-IPN were obtained by mixing aqueous solutions of PVA (degree of hydrolysis = 88%)
and a functionalised polyurethane (PU) with a small molecular weight (containing
chloromethyl and amino groups that could react with each other to generate polymer
crosslinking) [57].
The mechanical properties of the blend are influenced by the blend composition, heat
curing temperature and the heating time. (Figure 8.18).
S-IPN of PVA/APU show good swelling in water. The degree of swelling decreases with
the increase of the curing temperature, the increase of the heating time and with the
increase of APU content.
These blends could be used for producing tough, hydrophilic polymers which are
important for separation membranes and adsorption [57].
A PVA/PCU blend was obtained by extrusion of the polymer mixture and extraction
with the hexane/ethanol azeotropic mixture, followed by surface modification with a
polymer of 4-isocyanato butanoic acid methyl ester (as a spacer molecule), saponification
of methyl ester groups and coupling of 4-amino-2,2,6,6-tetramethylpiperidine-1-oxyl
(TEMPO) [58].
Chelate membranes were obtained from Schiff base membranes after treating a poly(allyl
amine) (PAAm)/PVA blend with salicylaldehyde and then including cobalt ions [59].
The permeability of oxygen and nitrogen decreased and selectivity increased with the
increasing PVA content in the Schiff base membrane. As cobalt ions were incorporated
324
Blends Based on Poly(Vinyl Alcohol) and the Products Based on This Polymer
Figure 8.18 The tensile strength and elongation at break of S-IPN with 25% and 50%
APU versus heating curing temperature when holding the heat curing time at 20 min. [57]
Reproduced with permission from Q-Y. Zhou, J. Zuo, M-D. Song, B-H. Zhang and B-L. He,
Journal of Applied Polymer Science, 1998, 67, 473. Copyright 1998, John Wiley & Sons, Inc
325
Handbook of Polymer Blends and Composites
into the membrane, nitrogen permeability was not affected, while oxygen permeability
increased with increasing cobalt content as can be seen in Figures 8.19 and 8.20.
Both the permeability and the selectivity of the chelate membranes containing cobalt
increased with the cobalt content. Their maximum increases are about 10 and 4 times
greater than those of Schiff membranes [59].
326
Blends Based on Poly(Vinyl Alcohol) and the Products Based on This Polymer
Figure 8.20 Effect of PVA content on gas permeability of chelate membranes of PVA/
PSAAm blend [59]
Reproduced with permission from M.J. Choi, C.K. Park and Y.M. Lee, Journal of Applied
Polymer Science, 1995, 58, 2373. Copyright 1995, John Wiley & Sons, Inc
PVA/PVC blends were obtained by casting from solutions of the two polymers in
DMSO [60].
PVC with Mw = 57,000 and two types of PVA with degrees of polymerisation of 1600
and 500, were used [60].
327
Handbook of Polymer Blends and Composites
Generally, the mixtures show multiple Tg smaller than those of the pure components
[60], which indicate that the blend should be easier to process.
The activation of SRF and PVA-F could be made by using p-benzoquinone, cynuric chloride
and p-tolylsulfonyl chloride spacers, whereas that of glycosilamine-chitosan, was made
using glutaraldehyde (GAH). The immobilised enzyme showed very good storage stability
at room temperature, an improved thermal stability and durability and a better solvent
stability than does the free -galactosidase [61].
-Cyclodextrin (
8.5.2 PVA/ -CD) Blends
PVA and -CD showed a good compatibility. The blend films are transparent [63].
328
Blends Based on Poly(Vinyl Alcohol) and the Products Based on This Polymer
The blend membranes are permoselective for different organic isomers such as n-propanol
from a mixture of n-propanol (n-PrOH) and i-propanol (iPrOH) [63] and the separation
of p-xylene from a p-xylene and o-xylene mixture [64]. It was shown that in both cases,
the separation was better by applying the evapomeation technique than that of the
pervaporation. Figure 8.21 emphasises the differences between the two techniques of
separation for this mixture.
The same situation was seen for xylene isomer separation. The evapomeation is more
efficient than that of pervaporation [64].
Figure 8.21 Effect of feed composition on the n-PrOH concentration in the permeate
passed through CD/PVA membrane (the amount of CD in the membrane was 40 wt%)
by (O) pervaporation and () evapomeation [63]
Reproduced with permission from T. Miyata, T. Iwamoto and T. Uragami, Journal of
Applied Polymer Science, 1994, 51, 2007. Copyright 1994, John Wiley & Sons, Inc
329
Handbook of Polymer Blends and Composites
A comparison between the behaviour of the PVA membrane and the CD/PVA membrane
in the separation process by evapomeation is presented in Figure 8.22.
The CD content in the membrane and the n-PrOH and p-xylene content in the feed
mixture, influence the separation sorption, and diffusion selectivity factors of the CD/
PVA membranes for the n-PrOH/iPrOH and for the p-xylene and o-xylene mixtures
prepared by evapomeation (see Table 8.6).
It could be seen that a very high separation factor of organic liquid isomers through
PVA/CD blend membranes was obtained for PrOH isomers [63, 64].
Figure 8.22 Comparison between the effect of feed composition on the n-PrOH
concentration in the permeate passed through the CD/PVA (CD content: 40 wt%)(O)
and PVA () membrane by evapomeation [63]
Reproduced with permission from T. Miyata, T. Iwamoto and T. Uragami, Journal of
Applied Polymer Science, 1994, 51, 2007. Copyright 1994, John Wiley & Sons, Inc
330
Blends Based on Poly(Vinyl Alcohol) and the Products Based on This Polymer
Table 8.6 Separation factors and sorption and diffusion selectivities of the
CD/PVA and PVA membrane for the n-PrOH/iPrOH mixture and p-xylene
and o-xylene mixture by evapomeation function on CD content in the
membrane and on feed composition [63, 64]
CD Separation of n-PrOH/iPrOH Separation of p-xylene and o-xylene
content
wt.% Content of sep sorp diff Content of sep sorp diff
n-PrOH in p-xylene in
feed (wt.%) feed (wt.%)
0 50 2.01 1.89 1.06 10 1.72 0.53 3.27
20 50 - - - 10 1.19 0.94 1.27
30 50 - - - 10 2.93 1.08 2.74
40 50 2.61 2.07 1.26 10 3.93 0.78 5.04
10 15.2 3.68 4.14 30 3.26 1.86 1.75
- - - - 50 1.86 1.06 1.75
- - - - 70 1.99 1.48 1.34
- - - - 90 0.63 1.03 0.61
The higher crosslink density in the PVA/CD hydrogels may be an additional factor in
retarding the migration of the drugs in the presence of CD.
Based on the ability of -CD to form inclusion complexes with different water soluble
substances, i.e., salicylic acid, the PVA/CD hydrogels could be used for prolonged release
of salicylic acid or other drugs [62].
The drug release from the PVA/-CD gel is nearly proportional to time [62]. A comparison
between the release profile of salicylic acid from PVA gel and PVA/-CD gel is presented
in Figure 8.23.
The release of the drug, particularly water-soluble drugs, from hydrogels could be
controlled effectively by the addition of -CD [62].
331
Handbook of Polymer Blends and Composites
Figure 8.23 Release profile of salicylic acid from (A) PVA gel and (B) PVA/-CD
gel [62]
Reproduced with permission from K. Streenivasan, Journal of Applied Polymer Science,
1997, 65, 1829. Copyright 1997, John Wiley & Sons, Inc
A PVA/CELL blend was obtained by coagulation with ethanol from mixed polymer (CELL
with a polymerisation degree of 935; PVA with Mw = 78,000; saponification value =
99.7%) solutions in N,N-dimethyl-acetamide/lithium chloride (4.5%) [65].
A strong depression of the melting temperature of the PVA component in the blends with
the increasing of CELL content was observed.
332
Blends Based on Poly(Vinyl Alcohol) and the Products Based on This Polymer
The thermodynamic interaction parameter assumes a large negative value (12 = 0.985 at
240 C) which means that PVA and CELL are thermodynamically miscible.
The blend films showed a lower gas sorption and transport coefficient than was predicted
from a simple additive relationship [65].
The thermal and volumetric properties of PVA/CELL blends are presented in Table 8.7.
The PVA/CELL blend could be used for obtaining the membranes with a low permeability
for carbon dioxide [65].
Another type of PVA/CELL blend was obtained by pressing a mixture of PVA granules
and 5% CELL powder under a pressure of 280 kg/cm2, to form a disk of 1 mm thickness
that could be used as a solid substrate for room-temperature phosphorimetry (RTP), a
method used in drug detection [66].
Another type of PVA/CELL blend with a composite structure was obtained by spin coating
of the PVA (Mw = 50,000; 99% degree of hydrolysis)/GAH solution (corresponding to
0.005 or 0.01 moles of GAH/mole of PVA repeat unit) mixture onto the regenerated
cellulose membranes (Mw = 10,000) [67].
This hydrogel coating may penetrate the larger pores of the cellulose membrane and can
exclude protein from entering them (100% protein retention). So, the hydrogel coating
reduces the irreversible fouling of the cellulosic surface. This blend is recommended as
thin-gel composite membranes for bovine serum albumin ultrafiltration [67].
333
Handbook of Polymer Blends and Composites
Foamed blends prepared by baking a mixture of granular starch (potato starch and
corn starch) and aqueous PVA (four types: Mw = 124-186,000 and DH = 98.3%; Mw
= 85-146,000 and DH = 98.3%; Mw = 124-186,000 and DH = 88.6%; Mw = 85-
146,000 and DH = 88.2%) solution inside a hot mould [69].
Depending on the temperature, two main changes of Youngs modulus occur, similar to
pure PVA. This means that between PVA and starch no interaction occurs, so these
blends revealed a limited miscibility [68]. The Tg of the blends is not dependent on blend
composition. The Tm of PVA in blends increases with the increasing starch content,
becoming higher than that of pure PVA.
Above the Tg of PVA, the blend modulus value increases with increasing starch
content [68].
The foamed form of the blend showed an improved strength, flexibility and water
resistance by addition of 10%-30% PVA to starch [69].
These properties of the foamed blends are dependent on relative humidity (RH), on
blend composition, on the type of starch used, and on molecular weight and degree of
hydrolysis of PVA and on the amount of crosslinking agents added.
The water absorption for blends containing PVA with 99% degree of hydrolysis is lower
(83%) than that of the blends with PVA 88% degree of hydrolysis (184%) [69].
PVA/starch chemically bonded composite has a higher viscosity (for example: 195 mPa-s)
than that of physically mixed paste (115 mPa-s), given the same composition of the
sample [70]. The apparent viscosity of this starch composite is dependent on the initial
molecular weight of both PVA and hydrolysed starch used [70].
334
Blends Based on Poly(Vinyl Alcohol) and the Products Based on This Polymer
could be used as disposable plates, cups, clamshells (device for transporting powdery
materials) and other biodegradable packaging materials [69].
This blend exhibits a typical biphasic behaviour. SC/PVA blends, both untreated and
treated with glutaraldehyde vapour or by dehydrothermal treatment (DHT), can be
regarded as mechanical blends in which strong interactions (hydrogen bonds) occur
between the two polymers. This blend is a bioartificial material [68].
A PVA/gelatin blend was obtained by mixing aqueous solutions of polymers. This blend
is a biphasic system showing good mechanical properties over a wide range of temperature.
The gelatin induces a stiffening effect [68]. This blend is a bioartificial material.
PVA/(Regenerated) Silk Fibroin [(R)SF] blends were obtained by casting aqueous solutions
of the two polymers [71, 72, 73]. Heterogeneous films were obtained.
The phase separation and the phase structure are dependent on the film thickness. This
is illustrated in Figure 8.24.
In the blend membrane, (R)SF, PVA and enzymes maintain their own intrinsic
properties [71].
PVA/(R)SF blends can be used as biomaterials and materials for entrapping the enzymes
(peroxidase or glucose oxidase) in biosensors (for H2O2 or glucose, respectively), [71,
72]. The enzyme immobilisation is based on a conformational transition of (R)SF from a
water-soluble silk I structure to a water insoluble silk II structure. There are two types of
silk conformational isomers that emphasise different properties like different water
solubilities. The structural differences are shown by the shifting of the IR bands.
335
Handbook of Polymer Blends and Composites
Figure 8.24 Phase diagram obtained from microscopic observation of blend cast films
SF/PVA [73]
Reproduced with permission from T. Tanaka, M. Suzuki, N. Kuranuki, T. Tanigami and K.
Yamaura, Polymer International, 1997, 42, 107. Copyright 1997, John Wiley & Sons Ltd
PVA/-chitin blend was obtained by casting the mixed polymer (PVA: DP = 2,500;
deacetylation degree = 99% and -chitin) solutions in formic acid (99%) on a glass plate
and drying in an oven at 40 C for 24 hours [74].
Transparent, optically clear and homogeneous blend films were obtained. The films are
somewhat brittle and the brittleness was reduced as PVA content increased.
The blends have good mechanical properties, even in dry and wet states and the mechanical
properties are improved compared to those of the homopolymers. The tensile strength of
the blends is greater than that of PVA but the more the strength decreases, the more the
elongation at break increases, depending on the PVA content of the blends.
Each blend has two Tg corresponding to the Tg of PVA and that of -chitin. These values
shift to inside the range determined by the two Tg values of the two pure polymers as is
shown in Table 8.8.
336
Blends Based on Poly(Vinyl Alcohol) and the Products Based on This Polymer
-Chitin - 182 -
CP82 81 181 10 0
CP73 85 165 80
CP64 78 155 77
CP55 79 153 74
CP46 80 143 63
CP37 71 145 74
PVA 69 - -
The melting point depression of PVA component in blends shows the presence of
interaction of the mixed polymers while transmission electron microscope (TEM)
measurements evidence a fine dispersion of two phases.
The dimension of microseparation was small enough (0.5 m) to infer that the two
polymers are compatible each other on the entire range of concentration [74]. This blend
is a biocompatible material and could be used for artificial skin, as sutures, as coating
materials or as a drug carrier [74].
This blend was obtained by mixing the polymer solutions (PVA and DC) in aqueous
acetic acid, followed by solvent evaporation or in dimethylacetamide-lithium chloride
with coagulation with ethanol [75].
337
Handbook of Polymer Blends and Composites
It was pointed out that the miscibility of DC with PVA is dependent on the deacetylation
degree of chitin.
A good miscibility in amorphous regions was observed for PVA and restricted chitin of
40%-60% degree of deacetylation.
Blends containing DC with a degree of deacetylation in the range of 0%-40% and 60%-
100% showed a less prominent shift in the Tg as can be seen in Figure 8.25.
338
PMOC/PVA blend films melting point
PMOC/PVA blend films glass transition temperature
Blends Based on Poly(Vinyl Alcohol) and the Products Based on This Polymer
Films of PMOC/PVA blends were prepared by solvent casting from an aqueous solution
of PVA [Mn = 88,000] and a chitin derivative having a poly (2-methyl-2-oxazoline) side
chain (Mn = 3.9 x 105; unit ratio of the main chain is 0.52:0.48 [N-acetyl-D-glucosamine
units: D-glucosamine units with poly (2-methyl-2-oxazoline)]; the repeating structural
units are randomly placed; Mw/Mn = 1.1) [53, 76].
The two polymers are miscible through the entire range of composition. The films obtained
are homogeneous and transparent.
A single Tg for each blend, which shifted towards a lower temperature region with
increasing chitin derivative content, was observed. The Gordon-Taylor equation can be
applied with k = 0.21 [53, 76].
Tm of the blends are lower than the Tm for PVA (see Table 8.9).
Table 8.9 Glass transition temperature (Tg) and melting point (Tm) of
PMOC/PVA blend films [76]
Blend sample Content of PMOC Tg (C) Tm (C)
(wt.%)
PVA/PMOC with a degree of 0 72 21 2
polymerisation of side chain = 19.6 10 10 20 8
30 68 20 6
50 67 19 7
70 67 -
1 00 66 -
PVA/PMOC with a degree of 0 69 20 7
polymerisation of side chain = 32.8 10 65 20 5
30 64 19 7
50 62 19 1
70 62 -
90 61 -
1 00 61 -
339
Handbook of Polymer Blends and Composites
A PVA/PAB blend was obtained by mixing aqueous solutions of PAB hydrochloride and
PVA and then by applying the freezing-thawing technique [77].
Hydrogels which were sensitive to the pH change of medium and salt content were
obtained. In a salt free aqueous medium, a dramatic reduction in hydrogel size was
observed above pH = 9 and below pH = 3, while in the range of pH 3-9, the hydrogel size
was little affected.
In a medium with a constant ionic strength (0.1), up to pH 10, the hydrogel size was
moderately affected but a significant size reduction was seen above pH 10. For a fixed
pH value, the hydrogel size change was gradual as the medium ionic strength was modified.
So, at pH 6, the hydrogel shrank by about 30% as the NaCl concentration was raised
from 10-6 to 2 mole/L [77].
Phase separated blends with a complex morphology were obtained [78, 79].
Mechanical properties of the blend films are strongly dependent on starch, moisture and
plasticiser, i.e., glycerol and/or water, content as well as processing history [80, 81].
The tensile strength and the modulus decrease rapidly with increase of the moisture and
the starch content for all the blend compositions, as shown in Figure 8.26 .
Irradiation of the blends by an electron beam causes the bonds scission and disruption of
intermolecular structures only for the natural polymer (starch) while the synthetic polymer
was not affected [82].
340
Blends Based on Poly(Vinyl Alcohol) and the Products Based on This Polymer
The mechanical properties of the blends are improved by irradiation and affected by
processing [82, 83].
By using SCORIM technique it was possible to induce anisotropy into the mouldings
and to obtain simultaneously enhanced values of stiffness and ductility [83].
The reduction in both Tm and DHm with the increase of the starch content gives interactions
between the two components that interfere with the crystallisation of EVOH [81].
The Tg of the EVOH in blends decreases with increasing starch and moisture content
(Figure 8.27).
341
Handbook of Polymer Blends and Composites
The blends used as biomaterials could only be sterilised for one cycle with ethylene
oxide. More cycles degraded the polymer structure and properties [83].
EVOH/starch blends are biodegradable plastics that can be used as materials for temporary
hard and soft tissue replacement.
EVOH/starch/HA blend was obtained by using two types of materials extrusion (in a
rotating drum or by twin-screw extrusion (TSE)). The compounds obtained were injection
moulded using both conventional moulding and a SCORIM technique [84]. A composite
material was obtained. The matrix was formed by two different grades of starch and
EVOH copolymer that differ mainly in their rheological behaviour (SEVOH-B and
342
Blends Based on Poly(Vinyl Alcohol) and the Products Based on This Polymer
SEVOH-C with the melt flow index values 2.64 and 0.76 g/600 s, respectively, determined
at 170 C and a pressure of 49 N). Sintered and non-sintered HA was used as
reinforcement.
Composite materials with very good mechanical properties were obtained (Table 8.10
and Table 8.11). SEVOH-C has higher strength and stiffness. The ductility is similar for
both materials.
The stiffness of these two blends is enhanced by the addition of bioactive bone-like
ceramics. This effect is enhanced for the case of sintered HA-reinforced composites due
to the much higher modulus. The ductility and absorption energy of both polymers are
strongly decreased with the HA addition [84]. The selection of processing conditions
could improve the mechanical properties of the composites obtained. So, it is possible to
increase simultaneously the strength, the stiffness and the ductility (SCORIM S2) or to
increase further the secant modulus at 1% strain (E1%) and ultimate tensile strength ()
with a decrease of the ductility (SCORIM S3) [84]
These composite materials are biodegradable and could be used for temporary tissue
replacement [84].
343
EVOH/starch/HA blends tensile test results
SEVOH-C conventional 35.6 3.2 1.81 0.16 22.4 5.8 52.5 12.7
SEVOH-C S-2 38.9 3.1 2.02 0.31 35.9 6.1 116.7 22.4
SEVOH-C S-3 41.8 0.8 2.97 0.23 7.1 2.0 15.8 5.2
SEVOH-C + 20%HA S-TSE-S4 29.3 0.5 5.64 0.41 13.2 2.2 10.2 0.2
The EVOH/SMA blend was obtained by reactive extrusion of the polymers [EVOH 67
mole% vinyl alcohol, SMA with 14 mole% Ma and a low molecular weight SMA
copolymer containing 20 mole% Ma and polystyrene (PS) precursor (as diluents)].
For all blends (except those containing 50%-60% EVOH) the phase dispersion was very
fine, the minor phase existed as dispersed droplets (diameter <2 mm). The interpenetration
of the phases in blends containing 50%-60% EVOH is emphasised.
The blends with 50%-60% EVOH exhibit different properties compared to that of the
blend with other compositions, as shown in Figures 8.28, 8.29, 8.30.
Figure 8.28 Effect of the blend composition on viscosity at a constant stress [85]
Reproduced with permission from P.S. Hope, J.G. Bonner and J.E. Curry, Pure & Applied
Chemistry, 1996, 68, 1665. Copyright 1996, IUPAC
344
Blends Based on Poly(Vinyl Alcohol) and the Products Based on This Polymer
Figure 8.29 Effect of the blend composition on the storage modulus (220 C) [85]
Reproduced with permission from P.S. Hope, J.G. Bonner and J.E. Curry, Pure &
Applied Chemistry, 1996, 68, 1665. Copyright 1996, IUPAC
Figure 8.30 Melt flow index of the blends as a function of the composition: EVOH/
SMA-first series (); EVOH/SMA-Series 2 (); EVOH/PS () [85]
Reproduced with permission from P.S. Hope, J.G. Bonner and J.E. Curry, Pure & Applied
Chemistry, 1996, 68, 1665. Copyright 1996, IUPAC
345
Handbook of Polymer Blends and Composites
It could also be observed that the melt flow index of the blends have lower values compared
to the pure components.
These findings suggest that the availability of the reactive groups is restricted by the
phase morphology of the blends during production [85].
EVOH has been used extensively in the packaging industry because of its good
processability, good mechanical properties, good oxygen barrier properties and its
biodegradability caused by the short blocks of PVA which are biodegradable and by
oligomers of PE which are biodegradable at a molecular weight lower than 500 [81].
All these blends are incompatible [88]. To attain a satisfactory level of mechanical integrity,
a satisfactory level of adhesion between the main components must be ensured by using
compatibilisers. The degree of compatibilisation and processing conditions should be
controlled so that the dispersive mixing is avoided and EVOH retains its laminar
distribution which determines its low gas permeability as film. The type and the amount
of compatibiliser influence all the properties of the blends.
These compatibilisers have two components: one that can interact with the OH groups
of EVOH, such as Ma, which can provide ester linkages or as Zn2+ which will interact by
complexation, and a component that interacts with the polyolefins.
These blends were obtained by reactive extrusion of polymers and by drawing or batch
mixing and hot pressing.
346
Blends Based on Poly(Vinyl Alcohol) and the Products Based on This Polymer
The influence of the processing type and blend composition on the blend morphology is
summarised in Table 8.12, from the experimental data obtained by Faisant and co-
workers [86].
Oxygen permeability of the blend films is influenced by the film composition and
morphology. Generally, a heterogeneous structure, containing orientated fibrils and
lamellae of EVOH exhibits a lower oxygen permeability than that of a more homogeneous
morphology with finer dispersed particles as shown in Table 8.13.
347
348
Mechanical properties of the blend are influenced by many factors such as: the matrix
type, the composition of the minor phase, the presence, type and the concentration of
the compatibiliser, the blend morphology and the processing type of the blend.
Because of their good mechanical properties and barrier properties, these blends could
be used for plastic containers with high barrier properties [84], food packaging [86],
bottle and container caps, injection moulded containers, injection blow moulded
containers, electronic components [89], and so on.
These blends combine the qualities of EVOH (good oxygen barrier properties, excellent
transparency, high oil resistance) with those of polyamides (good barrier properties against
hydrocarbon solvents, an outstanding level of toughness and good adhesion with different
substrates), minimising the drawbacks of EVOH as the relative brittleness of its films, its
relative high moisture sensitivity and its high price.
EVOH/PA blends can be obtained by melt mixing of the two polymers [90] or by casting
from a solution of the polymer in formic acid [91].
A great diversity of EVOH/PA blends can be realised by blending different types of EVOH
having different ethylene contents with different types of PA, i.e., PA6,12 (polyamide 6-
co-12) or PA4,6.
All blends of PA6,12 with any other EVOH (except that containing 59 mole% ethylene
units) were found to be immiscible at the measurable temperature (under 25 C) and on
the entire range of composition [90]. The phase behaviour is dependent on temperature.
So, for EVOH with higher ethylene units content, (i.e., 72), PA6,12 becomes attractive,
with the increasing of the temperature, when the interfacial tension tends to decrease.
These blends exhibit an upper critical solution temperature (UCST) phase behaviour. On
the contrary, for EVOH with lower ethylene units content, (i.e., 59, 44, 32), in
PA6,12 blends, the interfacial tension increases with an increase of temperature, that
means that the last systems are dewetting at elevating temperature [90] and exhibit a low
critical solution temperature (LCST) phase behaviour [90].
349
Handbook of Polymer Blends and Composites
A single Tg value was obtained by DMA and the composition dependence of Tg was
fitted by the Gordon-Taylor equation as shown in Figure 8.31.
It was concluded that these blends are miscible above the Tm [92].
De Petris and co-workers [93] studied EVOH with 29% ethylene units, PA6 blends with
different composition and concluded that biphasic blends were obtained by polymer
extrusion [93].
350
Blends Based on Poly(Vinyl Alcohol) and the Products Based on This Polymer
These results are in accordance with those obtained by Lee and Kim [94].
The effect of the composition on loss peaks temperatures (T, T) and Tg of PA6/EVOH
blends is presented in Table 8.14.
The shifting of the -peak is an indication of the interactions between the amide groups
of PA6 and the hydroxyl groups of EVOH.
351
loss peaks temperatures PA6/EVOH blends
Tg PA6/EVOH blend
Handbook of Polymer Blends and Composites
De Petris and co-workers [93] reported that these blends had a serious melt instability
shown by gel formation after few minutes of extrusion. The gelation phenomena occurred
through the etherification process between the EVOH hydroxyl groups.
The mechanical properties of these blends were affected by the presence of gels in the
films. So, the ultimate tensile strength, elongation to break and the modulus decreased
comparatively to PA6 [93]. To improve the mechanical properties by elimination of gel
formation, the authors added to the PA6/EVOH blend, carboxyl-modified EVOH (EVOH-
COOH), containing 2 wt.% of grafted acid groups, obtained by esterification of the
OH groups of EVOH with succinic anhydride, which acts as a compatibiliser. The
interactions between EVOH-COOH and end-amino groups of PA6 were shown by two
additional absorption bands in IR, at 2336 and 2042 cm-1, that are characteristic of
ammonium salts of a primary aliphatic amine. Strong intermolecular ionic linkages occur
in the ternary blend.
Contrary to these experiments, Lee and Kim [94] did not observe gel formation during
the blending time of the EVOH/PA6. Consequently, the mechanical properties of the
blends were improved by EVOH addition as shown in Figure 8.33.
The oxygen and toluene permeability of the blend film decreases as a nonlinear function
with the increase of the EVOH content as it could be seen in Table 8.15.
The use of the EVOH-COOH compatibiliser determines the increase of the permeability
of the blend film to oxygen, that become two orders of magnitude higher than that of a
co-extruded film with the same percentage of EVOH [93]. This indicates that EVOH-
COOH acts that an interfacial agent.
352
Blends Based on Poly(Vinyl Alcohol) and the Products Based on This Polymer
Figure 8.33 Tensile properties of melt blended EVOH/LPA6 (L= low viscosity): (a)
tensile strength, (b) tensile modulus, versus blend composition [94]
Reproduced with permission from S.Y. Lee and S.C. Kim, Journal of Applied Polymer
Science, 1998, 67, 2001. Copyright 2001, John Wiley & Sons, Inc
353
Handbook of Polymer Blends and Composites
EVOH/PA6 blends were used to obtain new hybrid blends by dispersing the blend of
EVOH and PA6 in a LDPE matrix.
Table 8.16 summarises the different methods of component blending to obtain hybrid
materials.
354
Blends Based on Poly(Vinyl Alcohol) and the Products Based on This Polymer
So, for blends that contain a minor phase realised by EVOH and PA6 with low viscosity
(LPA), mixed in solid state (E + LPA), all the properties are linearly dependent on the
relative portion of the components in the hybrid dispersed phases.
Oxygen barrier properties showed a linear dependence on EVOH composition in the blend
while toluene barrier properties showed negative deviation under the same morphology.
At the same time, blends with a minor phase obtained by EVOH and LPA melts mixing
(E/LPA), showed a positive deviation in the oxygen, toluene barrier properties and in
tensile strength of the blends from the additive rule (Figure 8.34).
Ha and co-workers [91] studied the miscibility of PA4,6 and EVOH in blends, obtained
by casting from 10 wt% (PA4,6 with a mole ratio of PA4 to PA6 by 69/31; EVOH with
27 mole% ethylene and 13 mole% vinyl acetate) solutions in formic acid.
Films with EVOH content between 0-50 wt% are translucent and those with 65-100 wt%
EVOH are optically clear. Only blends with more than 65 wt% EVOH are miscible.
An alteration of the crystalline structure of PA4,6 was observed when more than 65 wt %
EVOH was added.
Tensile strength of the blends decreases and the elongation at break increases with the
increasing of the EVOH content (from 331 to 4 kg.cm-2 and from 213% to 2684%,
respectively) [91].
All these blends could be used as packaging material for foods, drugs and cosmetics.
All the blends studied have lower oxygen permeability than that of pure oriented PET.
The oxygen permeability increased with the increase of the draw temperature. These
data indicate that the thinner pellets are more effective in reducing the permeability than
thicker particles. The greatest reduction in oxygen permeability was obtained for the
355
Handbook of Polymer Blends and Composites
356
Blends Based on Poly(Vinyl Alcohol) and the Products Based on This Polymer
drawn, uncompatibilised, PET/EVOH with ethylene content = 32%; MFI = 3.1 g/10 min
(at 210 C and 2.16 kg); oxygen permeability value = 1.0096 x 106 cm3.mm/m2.day.Pa)
(EVAL-F) blend stretched at 85 C, by a factor of 0.3 [95].
Parada and co-workers [96] made a very suggestive representation of phase behaviour in
EVOH/PEOX blends, function on molar percent of hydroxyl groups in EVOH
(Figure 8.35).
This representation is based on Tg, Tm, DHm and miscibility parameter - a, experimental
values and their correlation with the system compatibility.
357
Handbook of Polymer Blends and Composites
The blends containing EVOH with 56%-78% hydroxyl groups exhibits a single Tg value
over the entire range of blend composition, a strong decrease of the Tm and Hm with the
increasing PEOX content and positive values for the miscibility parameters that means a
high degree of compatibility of these polymers.
The blend with 25% hydroxyl group content in EVOH gave an opaque film, two values
for Tg, relatively constant values for Tm and negative values for miscibility parameter
that means incompatibility in this system. The blend of EVOH 56/PEOX gave negative
values for miscibility parameter above 40% PEOX, that can be interpreted as a limited
miscibility of this EVOH with PEOX on a definite range of composition [96].
Blends of PVA, its copolymers and its derivatives respond to the actual demands for new
materials: high properties, non-toxicity, biodegradability, easily processed and recycled
and to have a low cost. These blends increase the number of intelligent materials that
respond to the environmental changes. The future trends in these blends developing
seems to be biomaterials, sensors, separation membranes, conducting polymers and
packaging materials.
References
1. C.A. Finch, Polyvinyl Alcohol, John Wiley & Sons, Bristol, UK, 1973.
4. W-P. Chang and W-T. Whang, Polymer, 1996, 37, 16, 3493.
5. H.C. Park, R.M. Meertens, M.H.V. Mulder and C.A. Smoulders, Journal of
Membrane Science, 1994, 90, 3, 265.
6. J-W. Rhim, M-Y. Sohn and K-H. Lee, Journal of Applied Polymer Science, 1994,
52, 9, 1217.
7. J-W. Rhim, H-K. Kim and K-H. Lee, Journal of Applied Polymer Science, 1996,
61, 10, 1767.
358
Blends Based on Poly(Vinyl Alcohol) and the Products Based on This Polymer
8. K-H. Lee, K-K. Kim and J-W. Rhim, Journal of Applied Polymer Science, 1995,
58, 10, 1707.
9. M. Metayer and C.O. MBareck, Reactive & Functional Polymers, 1997, 33, 2-3,
311.
11. E. Ruckenstein and Y. Sun, Journal of Applied Polymer Science, 1996, 61, 11,
1949.
12. E. Ruckenstein and L. Liang, Journal of Applied Polymer Science, 1996, 62, 7,
973.
13. J.G. Byun, Y.M. Lee and C-S. Cho, Journal of Applied Polymer Science, 1996,
61, 4, 697.
16. Y.M. Lee, S.H. Kim and S.S. Cho, Journal of Applied Polymer Science, 1996, 62,
2, 301.
17. J-W. Rhim, S-W. Yoon, S-W. Kim and K-H. Lee, Journal of Applied Polymer
Science, 1997, 63, 4, 521.
18. A.S. Hickey and N.A. Peppas, Polymer, 1997, 38, 24, 5931.
19. A.V. Volkov, I.V. Karachevtsev, M.A. Moskvina, A.V. Rebrov, A.L. Volinskii
and N.F. Bakeev, Journal of Inorganic and Organometallic Polymers, 1995, 5,
3, 295.
20. C. Vauclair, H. Tarjus and P. Schtzel, Journal of Membrane Science, 1997, 125,
2, 293.
21. H.S. Shin, S.Y. Kim and Y.M. Lee, Journal of Applied Polymer Science, 1997, 65,
4, 685.
359
Handbook of Polymer Blends and Composites
23. D. Graiver, S-H. Hyon and Y. Ikada, Journal of Applied Polymer Science, 1995,
57, 11, 1299.
25. C.H. Wang, T. Takayama and S. Nakamura, Journal of Polymer Science: Polymer
Chemistry Edition, 1997, 35, 16, 3561.
26. S-A. Chen and G-W. Hwang, Polymer, 1997, 38, 13, 3333.
27. S-A. Chen and M-Y. Hua, Macromolecules, 1996 29, 14, 4919.
30. M. Nagura, H. Saitoh, Y. Gotoh and Y. Ohkoshi, Polymer, 1996, 37, 25, 5649.
32. C.H. Wang and S. Nakamura, Journal of Polymer Science: Polymer Chemistry
Edition, 1994, 32, 7, 1255.
37. L. Liang and E. Ruckenstein, Journal of Membrane Science, 1995, 106, 1-2, 167.
38. I. Kaur, S. Maheshwari, A. Gupta and B.N. Misra, Journal of Applied Polymer
Science, 1995, 58, 4, 835.
360
Blends Based on Poly(Vinyl Alcohol) and the Products Based on This Polymer
39. E.G. Koulouri and J.K. Kallitsis, Polymer, 1998, 39, 11, 2373.
40. Q. Wang, L.Y. He and J.G. Huang, Journal of Applied Polymer Science, 1997,
64, 11, 2089.
41. Q. Wang and L.Y. He, Polymer, 1997, 38, 15, 3931.
45. I. Inamura, Y. Jinbo, Y. Akiyama and Y. Kubo, Bulletin of the Chemical Society
of Japan, 1995, 68, 7, 2021.
47. P. Peng, X.L. Zhang and X.D. Li, Chemical Journal of Chinese Universities-
Chinese, 1995, 16, 7, 1122.
48. R. Paladhi and R.P. Singh, Journal of Applied Polymer Science, 1994, 51, 9,
1559.
49. M. Morita, Journal of Polymer Science: Polymer Physics Edition, 1994, 32, 2,
231.
50. P. Banerjee and B.M. Mandal, Macromolecules, 1995, 28, 11, 3940.
51. P. Banerjee and B.M. Mandal, Synthetic Metals, 1995, 74, 3, 257.
361
Handbook of Polymer Blends and Composites
55. S.N. Cassu and M.I. Felisberti, Polymer, 1997, 38, 15, 3907.
56. K.L. Narayana, Y. Dasaradhudu and V.V.R.N. Rao, Polymer International, 1994,
35, 4, 315.
57. Q-Y. Zhou, J. Zuo, M-D. Song, B-H. Zhang and B-L. He, Journal of Applied
Polymer Science, 1998, 67, 473.
59. M.J. Choi, C.K. Park and Y.M. Lee, Journal of Applied Polymer Science, 1995,
58, 13, 2373.
60. S. Patachia, C. Vasile and E. Mavru, Polymer Bulletin, 1985, 13, 301.
61. S. Rejikumar and S. Devi, Journal of Applied Polymer Science, 1995, 55, 6, 871.
65. K. Patel and R.S.J. Manley, Macromolecules, 1995, 28, 17, 5793.
67. R.H. Li and T.A. Barbari, Journal of Membrane Science, 1995, 105, 1-2, 71.
69. R.L. Shogren, J.W. Lawton, K.F. Tiefenbacher and L. Chen, Journal of Applied
Polymer Science, 1998, 68, 2129.
362
Blends Based on Poly(Vinyl Alcohol) and the Products Based on This Polymer
72. Y.C. Liu, H.Y. Liu, J.H. Qian, J.Q. Deng and T.Y. Yu, Journal of Macromecular
Science - Pure and Applied Chemistry, 1996, A33, 2, 209.
74. Y.M. Lee, S.H. Kim and S.J. Kim, Polymer, 1996, 37, 26, 5897.
76. K. Aoi, A. Takasu and M. Okada, Macromolecules, 1997, 30, 20, 6134.
77. K. Iio, N. Minoura and M. Nagura, Polymer, 1995, 36, 13, 2579.
78. S. Simmons and E.L. Thomas, Journal of Applied Polymer Science, 1995, 58, 12,
2259.
79. S. Simmons and E.L. Thomas, Polymer, 1998, 39, 23, 5587.
80. M.A. Villar, E.L. Thomas and R.C. Armstrong, Polymer, 1995, 36, 9, 1869.
81. P.J. Stenhouse, J.A. Ratto and N.S. Schneider, Journal of Applied Polymer
Science, 1997, 64, 13, 2613.
82. A.D. Sagar, M.A. Villar, E.L. Thomas, R.C. Armstrong and E.W. Merrill, Journal
of Applied Polymer Science, 1996, 61, 1, 139.
83. R.L. Reis, S.C. Mendes, A.M. Cunha and M.J. Bevis, Polymer International,
1997, 43, 4, 347.
84. R.L. Reis, A.M. Cunha, P.S. Allan and M.J. Bevis, Advances in Polymer
Technology, 1997, 16, 4, 263.
85. P.S. Hope, J.G. Bonner and J.E. Curry, Pure & Applied Chemistry, 1996, 68, 8,
1665.
86. J.B. Faisant, A. Ait-Kadi, M. Bousmina and L. Deschenes, Polymer, 1998, 39, 3,
533.
87. N.K. Kalfoglou, C.K. Samios and C.P. Papadopoulou, Journal of Applied
Polymer Science, 1998, 68, 589.
363
Handbook of Polymer Blends and Composites
88. S.Y. Lee and S.C. Kim, Journal of Applied Polymer Science, 1998, 68, 1245.
89. N. Walling and M.R. Kamal, Advances in Polymer Technology, 1996, 15, 4, 269.
91. C-S. Ha, M-G. Ko and W-J. Cho, Polymer, 1997, 38, 5, 1243.
92. N. Yamaguchi, S. Akiyama and M. Tokoh, Kobunshi Ronbunshu, 1996, 53, 12,
842.
94. S.Y. Lee and S.C. Kim, Journal of Applied Polymer Science, 1998, 67, 12, 2001.
95. R. Gopalakrishnan, J.M. Schultz and R.M. Gohil, Journal of Applied Polymer
Science, 1995, 56, 13, 1749.
364
Polyacrylic-Based Polymer Blends
9
Polyacrylic-Based Polymer Blends
Aurica P. Chiriac
9.1 Introduction
Because new polymers have been synthesised and chemical modifications made to
conventional polymers, the importance of polymer blends has increased tremendously
to satisfy the growing needs of new materials with specific properties.
Polymer blends are typically formulated to achieve a property balance. Some properties,
most notably toughness near the brittle - ductile transition, are highly dependent upon
blend composition, and slight fluctuations can effect large property differences. The mixing
process can also have a profound influence on blend performance.
One of the most numerous classes of monomers is the acrylics, namely esters of acrylic
and methacrylic acid with general formula:
365
Handbook of Polymer Blends and Composites
polymers, copolymers or often polymer blends, which have different and/or more suitable
physical properties than the homopolymers alone, are the solution.
Blends with methacrylate resins in their composition have improved solvent craze
resistance. Methacrylate resins are widely used in producing sheet, moulded parts and
articles having known, desirable acrylic properties.
Acrylic products are modified to give special properties best for particular end-uses. For
example acrylic fibres are unique among synthetic fibres because they have an uneven
surface, even when extruded from a round-hole spinneret. Among the characteristics of
the acrylics are: superior resistance to sunlight degradation, resistance to moths, oil, and
chemicals, flexibility, acrylic fibres can be dyed to bright shades with excellent fastness and
they retain their shape, etc. Acrylic fibres are the generic name of man-made fibres derived
from acrylic resins (minimum of 85% acrylonitrile (ACN)). The benefits of acrylic blends
are: superior moisture management or wickability, quick drying time (75% faster than
cotton), easy care, shape retention, excellent light fastness, takes colour easily, bright vibrant
colours, odour and mildew resistant. Thus, a 0.9 denier acrylic micro fibre is MicroSupreme
(Sterling Chemicals) available in staple and tow, natural and fluoropolycarbonate (PCF)
for yarn spinning, sliver knits, industrial and technical applications.
The Japan Chemical Fibers Association mentions in their report on composition by fibre
(Figure 9.1) the extent of acrylic fibres, their constant evolution and development [1].
The flexibility of many acrylic polymers eliminates the need for adding plasticisers. Since
the polymers can be produced by any of the common manufacturing processes, they are
available in bulk, solution, or emulsion grades, from which can be selected the type most
convenient for an application.
Acrylic elastomers are among the synthetic rubbers, which not only exhibit oil resistance
but also lack residual unsaturation in the polymer chain. Between these types of acrylic
elastomers can be specified Nipol Nitrile Elastomers with their range of products such as
Nipol Polyblends and Pre-Plasticised Nitrile Elastomers or Zetpol 1020.
There are well known industrial applications of solutions and/or dispersions of acrylic
polymers:
366
Polyacrylic-Based Polymer Blends
as coatings or impregnants,
in leather finishing,
Practically there is no field that acrylic polymers do not have applications. Polyacrylics,
including polymethacrylics, as constituent of a blend, can be divided in two categories,
Composition by fibre
Other synthetics 10%
Cellulosic 15% 10%
41% 15%
Nylon 10%
Acrylic 24%
10%
24%
Polyester 41%
Staple
Acrylic 43%
Cellolosic 17%
29% 43%
Nylon 1% 1%
17% 10%
Polyester 29%
Figure 9.1 The extent and development of acrylic fibres according to the Japan
Chemical Fibres Association
367
Handbook of Polymer Blends and Composites
acrylics that necessitate the improvement of their characteristics, e.g., increase of adhesivity
by blending with polyurethane or rubber, improvement of the ability of the resulting
blend to absorb dyes, and so on, and acrylics which bring with their presence specific
properties, e.g., compatibility, hardness, adhesivity, wetability, and so on.
Blending technologies offer many possibilities as according to the phase structures and
morphologies developed, different systems may be realised being a cheaper alternative
way to avoid the high cost synthesis of new polymers and also to optimise the properties
of polymer components and in some cases to realise new materials with unexpected
properties. Normally blending is used to combine the properties of two or more polymers.
For a successful blend the following conditions should be fulfilled:
The polymer mixture must be stable under the normal conditions for its use and no
de-mixing should occur.
The dispersed phase must have strong bonding to the surrounding polymer.
Most of the work in the past was necessarily empirical, and only in recent years has
considerable effort has been devoted to the development of a basic knowledge in the
field. In this area, research should be focused mostly on the following lines, covering
fundamental as well as technological aspects of polymer blends:
Mechanisms and kinetics of phase separation with special care to processes of phase
transitions characterised by an interplay of spinodal decomposition with other kinds
of transitions (such as crystallisation, glass transition and microphase transition of
block copolymers).
368
Polyacrylic-Based Polymer Blends
The final objective is to find new blend technologies aimed at realistic systems, acrylic
polymer-based included, characterised by a tailored, stable and reproducible overall phase
structure giving constant in time end use performances.
The most common technique adopted for preparing polymer blends are melting, mixing,
casting from common solvents, freeze-drying, and mixing via reaction.
9.2.1 Casting Method and Specific Interactions in Binary and Ternary Blends
Containing Acrylic Polymers
The blend miscibility is directly connected with the method of preparation. For example,
the miscibility of the PMMA with polyvinyl acetate (PVAc), when obtained by casting, is
known to depend on the solvent. Homogeneous solutions and transparent cast films
were obtained when chloroform was used as a solvent and opalescent films were always
obtained by casting from homogeneous solutions of N,N-dimethyl formamide (DMF) or
tetrahydrofuran (THF).
As described by Sun [5], the parameter, calculated using the equation: = KBLEND K1,
can be utilised to evaluate the presence of interactions. Thus, for example:
[ ]PMMA WPMMA + K PVAc [ ]PVAc WPVAc + 2 K PMMA K PVAc WPMMA [ ]PMMA WPVAc [ ]PVAc
2 2 2 2
K PMMA
K1 =
( WPMMA [ ]PMMA + WPVAc [ ]PVAc )
2
369
Handbook of Polymer Blends and Composites
where Ki is the Huggins constant and []i is the intrinsic viscosity of the ith pure polymeric
component (generally KBLEND is the Huggins constant of the blend and K1 is calculated
from viscometric parameters of solutions of each polymeric component).
Differential scanning calorimetry (DSC) and viscosimetric experiments revealed that the
miscibility of the PMMA/PVAc blends depends on the solvent and the solvent temperature
[6]. The lower critical solution temperature (LCST) curve was determined by
measurements of the intensity of transmitted light against temperature, at several
compositions of PMMA/PVAc. Thus, Crispim and co-authors observed that blends were
miscibile in chloroform at 30 C and 50 C, whereas in DMF at the same temperatures,
the blends were found to be immiscible. Also in toluene the miscibility depends on the
temperature: the blends are miscible at 30 C and immiscible at 50 C. Results from
viscometry match very well those from DSC. The film of PMMA/PVAc (50/50) cast from
chloroform presented only one thermogravimetric (TG) analysis (Tg). If cast from DMF,
the blend presented two Tg values, close to those for pure polymers. The thermogram of
the opalescent film, cast from DMF, was similar to the one obtained after a transparent
film, cast from chloroform, was heated above the LCST curve.
Light scattering and phase-contrast optical microscopy has also been used to investigate
the phase behaviour of blends of a liquid-crystal polymer (LCP) and PMMA as well as
the phase behaviour of the blends of PMMA and PVAc [7]. The data show a similarity
between the phase behaviour of blends of liquid crystalline and of flexible amorphous
polymers. The distinction consists of the absence of a linear regime of decomposition for
LCP-PMMA blends.
The poor miscibility of PAN is rather surprising because it possesses a-hydrogens and
cyano groups, both of which are capable of interacting with other functional groups [8].
Even so, two rare miscible PAN blend systems was reported [9]. The examples are PAN/
cellulose and PAN/polyvinylpyrrolidone (PVP) blends [10]. The miscibility was considered
to arise from dipole-dipole interactions between hydroxyl and cyano groups for the
PAN/cellulose blends and from hydrogen-bonding interactions between -hydrogens and
carbonyl groups for the PAN/PVP blends. However, no spectroscopic evidence was
provided to substantiate the suggestions.
Blends of PAN and poly(p-vinylphenol) (PVPh) [9] were prepared by solution casting
from DMF. The miscibility of PAN with PVPh was shown by thermal and spectroscopic
studies. A single Tg was found for all blend composition. FTIR studies showed that the
hydroxyl band of PVPh and the cyano band of PAN shifted to lower frequencies upon
370
Polyacrylic-Based Polymer Blends
blending, showing the existence of specific interactions between the two polymers. The
involvement of cyano groups in specific interactions was further shown by the development
of a high binding energy N1s peak (network interpenetrating network, first peak) in
each blend from X-ray photoelectron spectroscopic studies. The miscibility behaviour of
blends of PVPh has been extensively studied. The hydroxyl groups of PVPh are capable
of undergoing intermolecular interactions with proton-accepting functional groups. As
a result of such interactions, PVPh is miscible with a large variety of polymers including
polyacrylates and polymethacrylates.
Solution blends of cellulose acetate (CA) with PAN and aminated PAN (with ethylene
diamine (2DA), hexamethylene diamine (6DA) or dodecane diamine (12DA)) prepared
in DMF were tested for miscibility as a function of the chain length of diamine [11]. The
films were obtained by casting these blend solutions. It was observed that the morphology
of PAN/CA blends does not vary significantly with composition, the dominant morphology
being a particle-in-matrix structure. A significant reduction in domain size was proved
with amination, and the effect became more pronounced with increasing chain length of
diamine as well the improved miscibility. The FTIR measurements indicated that the
interactions between CA and aminated PAN come from the hydrogen bond between
ester groups of CA and the terminal amino groups of aminated PAN, which are facilitated
by the longer aliphatic segments.
Some investigations were carried out on blends of collagen (C) with PAA (C/PAA), in
which PAA was used as a model for studying the interactions between collagen and
carboxylic functional groups in aqueous solution and in the solid state [12]. These studies
showed it was possible to obtain a high degree of compatibility between these two polymers
thanks to the strong ionic interactions that occur between them. Thus, in the C/PAA
system, ionic interactions occur between the two oppositely charged macromolecules
and lead to the formation of a polycomplex that precipitates as an insoluble aggregate
because the charged groups responsible for solubility (carboxylic groups of PAA in the
form, COO and aminic groups of collagen in the form of NH3+) are involved in the
complex. The interactions between collagen and PAA show a complex of the two polymers
formed in aqueous solution, within a narrow pH range, through electrostatic interactions
involving the NH3+ groups of collagen and the COO groups of PAA.
An homogeneous solution of the monomer polymer solution is obtained and then the
bulk polymerisation is performed and this is called self-propagating frontal polymerisation
(propagating polymerisation fronts driven by the exothermicity of the polymerisation
reaction and the transport of heat from the polymerised product to the monomer).
371
Handbook of Polymer Blends and Composites
Chechilo and Enikolopyan first introduced frontal polymerisation in 1972 [13]. They
studied travelling fronts in MMA polymerisation using reactors at high pressure. Pojman
and colleagues investigated the movement of the polymerisation front in a variety of
systems, including different monomers at ambient pressure, using both liquid and solid
monomers and solution systems with high boiling point solvents [14]. The general
mechanism of formation and propagation of the polymerisation front is based on the
interaction between the production of heat by chemical reaction and its dispersion by
thermal conduction.
Using a similar concept to the one described for polymer blending, frontal polymerisation
has also been used to produce composite materials, where an inorganic powder is dispersed
in the polymer matrix to provide special conductivity [15] or thermochromic [16, 17]
properties. Better results were obtained using this method because the rapid polymerisation
in the travelling front prevents phase separation.
Acrylics and methacrylic polymers because of their available functional groups and because
of the characteristic properties conferred are efficient partners in obtaining the blends by
the IPN technique.
Fully IPN based on bisphenol A diglycidyl ether (DGEBA) and ethylene dimethacrylate
(EDMA) and semi-IPN based on DGEBA and 2-hydroxyethyl methacrylate (HEMA)
were prepared using isophronediamine and benzoyl peroxide (BPO) as curing agent [18].
It was shown that the methacrylic component acts as a plasticiser. Thus for EDMA/
DGEBA with a ratio of 18:82, the determined Tg was about 66 C while at a ratio of
25:75 the Tg value was diminished at 54 C.
372
Polyacrylic-Based Polymer Blends
The phase structure of the poly(hydroxy ether of bisphenol A)-based blends containing
SAN, prepared through in situ polymerisation, i.e., the melt polymerisation between the
DGEBA and bisphenol A in the presence of SAN, was characterised through DSC, dynamic
mechanical analysis (DMA) and scanning electron microscopy (SEM) [19]. All the blends
exhibited well-distributed phase-separated morphology. Mechanical tests effected by the
authors showed that the blends containing 5-15 wt% SAN displayed a substantial
improvement of tensile properties and Izod impact strength, which were in marked contrast
to those of the materials prepared via conventional methods.
Acrylonitrile butadiene rubber (NBR)/phenolic resin (PH) blends are known to have
excellent mechanical properties. Poly(alkyl methacrylates) are known to have a very
high wetting behaviour compared to many other polymers. As the phenolics contain
hard segments, their introduction into soft NBR may lead to formation of a constrained
layer in the matrix. A constrained layer may cause high wetting (over a narrow range of
frequency or temperature) and, at the same time, may retain high mechanical strength
over a wide and useful temperature range. Thus, IPN synthesised from several NBR-PH
blends using poly(alkyl methacrylates) as the third component are of higher strength
compared to the corresponding NBR-PH blends. The strength of the PMMA-based IPN
was found to be maximum. Thus, for NBR/PH/PMMA (51:3:46) the tensile strength
was 13.42 MPa compared to 9.97 MPa for NBR/PH/PEMA (53:3:44) or 4.42 MPa for
NBR/PH/(polybutylmethacrylate) PBuMA (46:3:51) [20].
Polymers with specific functionalities can be obtained by changing some groups on existing
polymers or copolymers with suitable reactants. It has been determined that the polymers
containing amidoxime groups have a tendency to form a complex with metal ions. Since
there are no easily available monomers with pendant amidoxime groups, these types of
polymers were synthesised by polymer-polymer conversion reactions where acrylonitrile
type polymers were reacted with hydroxylamine hydrochloride [22].
IPN of PAN and PVP were prepared using amidoximated PVP networks [23]. The blend
was synthesised by irradiation of dissolved PVP in acrylonitrile solution with a 60Co-
source. Gamma irradiation of acrylonitrile solutions of PVP yields easily IPN of PVP and
PAN. The conversion of nitrile groups in the PAN backbone can be achieved at rather
mild reaction conditions with an almost complete change into amidoxime groups.
373
Handbook of Polymer Blends and Composites
Amidoximated PVP based IPN show good swelling behaviour in aqueous media due to
the presence of PVP chains and the functional amidoxime groups are effective in the
adsorption of various metal ions.
374
Polyacrylic-Based Polymer Blends
Production of crew-cut aggregates of block copolymers has received much attention [32-
35]. The aggregates are formed through self-assembly of highly asymmetric amphiphilic
block copolymers in which the insoluble core-forming blocks are much longer than the
soluble blocks formed by corona treatment. Due to the fact that the insoluble blocks
comprise a large fraction of the total chain, the aggregates are usually prepared by indirect
methods. The procedures involve first dissolving the copolymers, such as PS-b-PAA, in a
common solvent, such as DMF, and then adding water (water being a nonsolvent for
polystyrene blocks), to a polymer solution to induce aggregation of the polystyrene
segments. One of the noteworthy phenomena associated with the crew-cut aggregate
systems is the existence of multiple morphologies, these include spheres, rods, vesicles,
lamellae, large compound vesicles, a hexagonally packed hollow hoop structure, and
several others [36, 37]. The most important morphogenic factors are the copolymer
concentration in solution, the added water content, the nature of the common solvent,
the type and concentration of added ions (salt, acid, or base), and the temperature. The
formation of crew-cut aggregates is mainly controlled by minimisation of the interfacial
energy which, in turn, is balanced by both the increase of the stretching of the core-
forming blocks in the core and the repulsion among the chains. The morphogenic effects
of the before mentioned parameters can, in general, be ascribed to their influence on the
force balance during the formation of the aggregates. The effect of the water content in
solution is related to the change of the interfacial energy. The effect of changing the
common solvent on the morphology can be ascribed to the change in the dimensions of
375
Handbook of Polymer Blends and Composites
the core-forming blocks in the core due to the changes of both the solvent content in the
core and the chain interactions.
The results of studies on the blends of phenoxy polymer and SAN prepared by bulk step-
growth polymerisation between bisphenol A and the DGEBA in the presence of SAN
involving in situ phenoxy formation, has been published [39]. Previously, the miscibility
and the phase behaviour of binary blends of poly(hydroxy ether of bisphenol A) (phenoxy)
and SAN has been reported by Vanneste and Groeninckx [40]. On the basis of their
investigation of solution casting, the phenoxy/SANx blends (composition range = 5-
34 wt %) obtained were recognised to be partially miscible. Through mechanical mixing,
Okada and co-workers [41] recently prepared the multiphase phenoxy/SAN blends. These
materials did not exhibit improvement in mechanical properties, especially in Izod impact
properties.
376
Polyacrylic-Based Polymer Blends
Ternary blends are gaining importance in the field of polymers. Paul and co-workers [43]
reported the first example of ternary blends based on polyepichlorohydrin (PECH), PMMA,
and poly(ethylene oxide)(PEO) where all three binary pairs are miscible. In 1990, two
other examples of completely miscible ternary blends were reported, one containing PMMA,
PECH/PMMA/PVAc [44, 45]. Recently, other examples of completely miscible ternary
blends were further reported by Grannock and Paul [46]. Completely miscible ternary
blends were composed of poly(vinylidene fluoride)(PVDF), PMMA and PVAc [47].
Kwei reported the first systematic study on ternary blends [48]. In his study, the addition
of PVDF to the immiscible pair PMMA/poly(ethyl methacrylate) (PEMA) was studied
and found to be miscible. In nearly all these blends, a third component either a
homopolymer or copolymer is added to homogenise an immiscible pair. Miscibility is
often achieved in cases where this third component is miscible with other polymers [49].
At the same time PVDF and PMMA blends are examples of blends with miscibility
between a semi-crystalline polymer and an amorphous one. These blends remain
homogeneous, and thus amorphous, even at low temperature provided that the PVDF
content is smaller than approximately 50 wt.%. Otherwise PVDF crystallises from the
melt with formation of two or three phases. Some discrepancy in the experimental values
of Tg for the PMMA/PVDF blends has been noted, and more likely reflects the effect of
the technique used for the sample preparation on the intimate phase morphology.
377
Handbook of Polymer Blends and Composites
Three other polymer blend systems have been made with isotactic, atactic, and syndiotactic
PMMA that were mixed with poly(styrene-co-p-hydroxystyrene) (PHS) containing
15 mol% of hydroxystyrene in 2-butanone [51]. PMMA has been preferred to show the
effect of tacticity on miscibility. DSC and FTIR spectroscopy were used to study the
miscibility of these blends. The three polymer blends were found to be miscible. All the
prepared films were transparent and there was a single Tg for each composition of the
blends. Tg elevation (above the additivity rule) was observed in all the three blends mainly
because of hydrogen bonding. It has also been shown that iPMMA and aPMMA have a
higher degree of hydrogen bonding with PHS than sPMMA.
An effective way to gain control of the morphology and to strengthen the interfacial
zone is to form block or graft copolymers in situ during blend preparation via interfacial
reaction of added functionalised polymeric components. In situ compatibilised polymer
blends based on copolymers containing glycidyl methacrylate (GMA) monomer have
attracted much attention because of their potentially broad applications. Chung and co-
workers patented a polymer based on polycarbonate (PC), polyethylene terephthalate
(PET), ABS, and styrene-acrylonitrile (SAN)-glycidyl methacrylate copolymer [52]. The
product presented excellent low-temperature impact properties. Akkapeddi and co-
workers [53] reported that polyethylene (PE)-g-GMA acted as a good compatibiliser of
blends of PC with PET and various polyolefins. Yin and co-workers [12] reported a
series of reactive compatibilisations of many blending systems based on GMA-containing
copolymers. PS-co-GMA was used as a compatibiliser for ABS/Nylon 1010 (PA1010)
blends [54-56]. The rheological, thermal, and morphological properties of ABS/Nylon
1010 blends with or without compatibiliser were investigated. ABS and PA1010 are
immiscible and incompatible with poor interfacial adhesion and large domains. This
incompatible blend has been converted into a compatible one by incorporating a PS-co-
GMA copolymer, which itself does not function as a compatibiliser but will become a
compatibiliser after reacting with PA1010 end groups during melt blending. The addition
of PS-co-GMA does not change the rheological behaviour of the blending system too
much but the viscosity is increasing compared to the noncompatibilised system. The
crystallinity of the compatibilised system is a little lower than that of the noncompatibilised
system due to the reactions between glycidyl groups with PA1010 end groups. The
morphological studies show that PS-co-GMA copolymer behaves as an effective interfacial
agent by improving the adhesion between ABS and polyamide (PA) and decreasing the
dimensions of dispersed phase.
378
Polyacrylic-Based Polymer Blends
blend components were mixed together in powdered form before extrusion. The effects
of specimen and notch geometry, ABS content and reactive compatibilisation has been
examined for understanding their influence on the fructure properties of PBT/ABS blends.
Magnetic effects can be of technological interest because they offer a new way to perform
radical processes. This is because through very weak perturbations of the magnetic field,
one can control chemical kinetics and thus the course and the rate of reactions that
normally require much higher chemical energies.
Beside the modifications brought to the evolution of the polymerisation process, the
magnetic field influences the properties of the resulting polymers obtained through
unconventional methods. The changes in the properties of the polymers synthesised in
the magnetic field are attributed to the catalytic effect of the field on the molecules that
can be re-shaped through growing of distance interactions and modification of angles
between bonds.
In grafting studies, the emphasis is generally laid on improving the textile, chemical,
hydrophobic, hydrophilic or thermoplastic properties without significantly altering the
others. Numerous methods have been developed for the grafting of acrylic monomers
onto the lignocellulosic materials, to make the cellulose-acrylic polymer mixture
compatible, with the aim of obtaining composites with good thermal stability and UV
radiation resistance. The difficulties that appear during the grafting reactions are due to
the monomer sensitivity, which leads to homopolymerisation and irreproducibility.
An increased efficiency of the grafting process from the viewpoint of minimising the
concurrent homopolymer formation can be achieved through the development of the
reactions in a continuous magnetic field. Thus a series of grafting processes in a continuous
external electromagnetic field have been realised to improve the course of the syntheses
and also to ameliorate the characteristics of the products obtained in the field [57-63].
Cellulose powder obtained from oak furfural lignocellulose by a nitration reaction was
grafted with acrylamide using a radical initiation [58]. The grafting reactions were carried
out and compared both in presence and in absence of a continuous external magnetic
field with the intensity extended between 0.15-0.35 T. The effect of monomer and initiator
concentration, time and temperature of reaction and magnetic field intensity onto the
grafting process were examined. The most important increase of conversion has been
registered for the grafting reaction in a magnetic field, namely about 25% compared to
15% for the growth process in the absence of the field, see Table 9.1. That corresponds
379
380
Graft products
Reaction temperature, (C) ([AAm] = 1.4 mol/l, [K2S2O8] = 0.154 mol/L, B = 0.25 T, cellulose = 10 g/l)
Classic 50 60 70 22.7 23.2 33.1 4.4 (4.48) 4.5 (4.57) 6.45 (6.53)
Magnetic field 50 60 70 23.2 24.1 37.9 4.4 (4.57) 4.6 (4.75) 7.45 (7.47)
to the variant having 30:100 weight ratio initiator/cellulose, compared to 18:100 weight
ratio of initiator/cellulose.
The cellulose microfibrils respond to the action of an external continuous magnetic field
because of their structure. At the same time, the field influence upon the radicals can be
conceived as a growth of the macroradicals lifetime.
The emulsion polymerisation under the magnetic field corresponds to the graft
copolymerisation of MMA onto PVA in the presence of the magnetic field and this explains
the phenomenon revealed. The cage in this case is formed by PVA macromolecules and it
contains the monomer and initiator radicals. The excited benzophenone triplet, obtained
under UV-irradiation, abstracts a hydrogen atom from the tertiary carbon of PVA and
produces a triplet radical pair consisting of ketyl and PVA. Radical pairs in the triplet
state due to the magnetic field presence and caused by the Zeeman effect, determine the
higher graft ratio of the copolymer. At the same time, the grafted copolymers obtained
under magnetic field showed improved moisture and heat resistance properties. This
was attributed to the increased graft ratio and greater stereoregularity of the copolymers
obtained under the field.
381
Handbook of Polymer Blends and Composites
Figure 9.2 Conversion of the graft reaction of acrylamide onto lignocellulosic material
initiated by: (, ) H2O2, (O, ) K2S2O8, (, ) K2S2O8/(FeSO4x7H20) in (, , )
or out (, O, )the magnetic field presence [58]
Reproduced with permission from A.P. Chiriac, I. Neamtu and C.I. Simionescu, Die
Angewandte Makromolekulare Chemie, 1999, 273, 75. Copyright 1999, John Wiley & Sons
382
Polyacrylic-Based Polymer Blends
blend system vary monotonically with composition, following Gordon - Taylor and Kwei
equations [48]. Solid-state cross-polarisation nuclear magnetic resonance (NMR)
demonstrated the monophase behaviour of these blends. The systems also exhibit single
component rotating frame spin-lattice relaxation times intermediate in value as compared
to the pure components. The formation of such a homogeneous monophase blend was
attributed to a favourable heat of mixing arising from some specific interactions between
the two component polymers. The FTIR study conducted on ACN/methyl acrylate/
butadiene terpolymer/poly(ethylene-alt-maleic anhydride) blends showed a shift in the
nitrile absorption peak of Barex 210 and in carbonyl absorption peak of PEMA. This
indicates the involvement of these groups in nonbonding interactions.
The deformation and relaxation behaviour of compatible blends of chain molecules can
be used as method to elucidate aspects of the fundamental and applied science of blends.
The deformation of polymer melts gives rise to an elongation of the coiled molecules
which in turn leads to an orientation of their segments. Optical birefringence
measurements, X-ray investigations have been used among other methods to characterise
the induced orientational order. The chain deformation, on the other hand, gives rises to
a restoring force, which is predominantly of entropic nature, so the deformation tends to
relax as the force is remove. The binary blend of PMMA/polyvinyl chloride (PVC) has
been characterised from this point of view on the basis of the friction coefficient [68].
The friction coefficient does not only depend on the respective polymer chains in blends,
but also on the coupling to their surroundings, the strength of which is controlled by the
interaction parameter. Consequently the central point in this interpretation is that a
negative interaction parameter gives rise to an enhanced friction coefficient as compared
to the pure polymer systems. The friction coefficient, in turn, sets the timescale of the
orientational relaxation process.
The high PMMA friction coefficient is attributed to the bulky methyl groups [69] which
are strongly engaged in specific interactions with PVC. Due to the strong interactions
one expects an influence of blending on the orientational behaviour of PMMA. The
PMMA/PVC system was also chosen for testing the theoretical concept on the segmental
orientational process in polymer blends owing to small difference in the Tg. That small
difference makes the ambiguity between the individual Tg and the Tg of the blend relatively
unimportant, but it cannot mask the large difference between the friction coefficients of
the blend components. The results obtained for the orientation of the PMMA in the
blend are consistent with the concept of an enhanced friction coefficient caused by a
negative interaction parameter and thus by a stronger coupling to the surrounding.
383
Handbook of Polymer Blends and Composites
The study of the properties of the binary polymer mixture containing PMMA and PVC
or PVAc has been the subject of numerous reports. It was shown that a limited
384
Polyacrylic-Based Polymer Blends
compatibility that depends on the method of obtaining the blend and also on the
characteristics of the constituents. The behaviour of these systems by TG and IR methods
was investigated too [74]. The PVC/PMMA system was characterised by
pseudocompatibility due to the interaction between functional groups, having two
maximum compatibility degrees, at PVC/PMMA (10:90) and PVC/PMMA (80:20). For
1:1 composition the system was found to be incompatible. The PVAc/PMMA system
was also described as being chain pseudocompatible for the composition range 40-90 wt
% PVAc, with a maximum degree of compatibility for PVAc/PMMA (80:20), at which
interpenetration of the two component chains takes place.
Techniques such as dynamic mechanical, impact resistance, and SEM have also been
used to study the compatibility of ethylene-propylene terpolymer (EPDM) PVC and
MMA grafted EPDM rubber (MMA-g-EPDM) PVC (graft contents of 4, 13, 21, and
32%) blends [75]. The authors found that all the regions of viscoelasticity showed two
T g for blends, and the Tg increased with increasing graft content, indicating the
incompatibility of these blends. The tan curve showed three dispersion regions for all
blends arising from the , , and transitions of the molecules. The sharp transition
peak shifted to higher temperatures with increasing concentration of the graft copolymer
in the blends. EPDM showed less improvement while six-fold increase in impact strength
was noticed with the grafted EPDM. The authors concluded that blends of EPDM PVC
and MMA-g-EPDM PVC were incompatible. Also, it was found the maximum of
impact strength of PVC was at 20% EPDM and 13% grafting. The impact strength of
PVC was increased six-fold when blended with MMA-g-EPDM rubber with 13% grafting
as compared to EPDM rubber, which improves the impact strength by two-fold.
Blends of PC with SAN, are used as a model for commercially important blends of PC
with ABS materials. The effects of PC molecular weight, ACN content of the SAN,
concentration of the dispersed SAN phase, and processing procedure on the blend
morphology was examined [76]. The effect of reactive compatibilisation using a SAN
amine on the morphology was investigated in relation to the other blend variables.
Compatibilisation by block or graft copolymers improves morphological stability
primarily by introducing a steric hindrance to coalescence. Unlike the case for polyamides
or polyesters, schemes for formation of block and graft copolymer by in situ reactions
during melt processing of PC blends have not been available until recently.
385
Handbook of Polymer Blends and Composites
(AEP) in a reactive processing scheme. The graft copolymer formed was capable of
reducing the SAN dispersed phase particle size and stabilising the blend morphology.
Dispersed phase particle size was found to increase significantly with SAN concentration
in the uncompatibilised blend due to coalescence during mixing, whereas the particle
size was relatively independent of blend composition for compatibilised blends which
were SAN-amine terminated.
PS and PMMA form an important and unique pair of polymers in that they are chemically
different; however, they display compatibility over different composition ranges at room
temperature. Thus the Tg of a series of polymer blends containing various ratios of PMMA
and PS indicated a single Tg over the compositions studied. The effect of gamma radiation
on Tg behaviour of polymer blends composed of a radiation-sensitive polymer (PMMA)
and a radiation-resistant polymer (PS) has been studied [80]. The authors found that
upon exposure of the PS/PMMA blends a further depression in Tg was observed,
particularly with polymer blends rich in PMMA ratios. The drop in Tg was attributed to
the degradation of the PMMA component. The relative rise in Tg of blends was attributed
to grafting between the blend components. This behaviour was attributed to the
restrictions of the molecular motion of the polymer chains, hence, increasing the Tg. The
calculation of the activation energy and order of reaction for PS/PMMA blends exposed
386
Polyacrylic-Based Polymer Blends
SEM has been used to investigate the composite droplet morphology development in a
polymer blends, particular a high density polyethylene (HDPE)/PS/PMMA ternary blend
model, which allows the PS and the PMMA dispersed phase to be dissolved away [82].
The authors found that the composite droplet morphology occurs within the first two
minutes of mixing and remains stable in size and shape thereafter. They demonstrated
also quantitatively that the presence of sub-inclusions generates a swelling effect, i.e., an
increase of PS droplet size.
Blends that are not miscible at high molecular weights may become compatible by lowering
the molecular weight of one or both components below critical values. Knowledge of
these critical molecular weights can be valuable for evaluating miscibility of the
components [83]. When miscible blends are obtained, the interaction energy can be
quantified if the mixture shows phase separation upon heating (LCST) or cooling upper
critical solution temperature (UCST). Low molecular weight polysulfone (PSU) was
synthesised for this purpose and the low molecular weight oligomers were tested. Modified
PSU were blended with oligomers of PS, poly(-methyl styrene) (PMS) and PMMA as
well as with styrenic copolymers to determine their interaction energies quantitatively.
The study is based on the Flory-Huggins theory and the Sanchez-Lacombe equation of
state theory. Modifications of PSU consisted into methyl substitution on the phenyl rings
and fluoro substitution at the isopropylidene unit, and combinations of these. It was
387
Handbook of Polymer Blends and Composites
observed that the modifications of PSU that favour miscibility with PS and P-MS are
generally unfavourable for miscibility with PMMA and vice versa. Increasing the polarity
of the phenyl rings of PSU while maintaining a positive outer surface charge on the
isopropylidene connector group is favourable for miscibility with PS. Substitutions at
the isopropylidene connector group of PSU decrease the polarity of the phenyl rings and
facilitate miscibility with PMMA. Blends of tetramethyl polysulfone (TMPSU) and SAN
display miscibility windows for which size depends on the molecular weight of TMPSU.
TMPSU and poly(-methyl styrene-co-acrylonitrile) (-MSAN) copolymer blends showed
similar miscibility boundaries to those for TMPSU/SAN. The minimum in the interaction
energy with ACN content coincides with the solubility parameter determinations by
Shaw. End groups have been found to significantly affect the interaction energies and
were taken into account. For molecular weights up to 12,000, the effect of the end group
(which is determined by multiple polar end groups per chain, that can interact strongly
and significantly affect the interaction energies) such as the hydroxyl groups on the PSU
oligomer, is significant. For one nonpolar end group per chain, such as the butyl group
on the styrene oligomer, the effect appears to be insignificant at molecular weights above
3,000. As a conclusion the change in interaction energy density is considered to depend
on the type, size, and concentration of the end group on each chain rather than of the
change in molecular weight itself.
The effects of miscibility and blend ratio on uniaxial elongational viscosity of polymer
blends were studied by preparing miscible and immiscible samples of the same composition
by using PMMA and SAN [84]. The study established a method to make miscible and
immiscible samples of the same composition using different combinations of PMMA
and SAN, for elongational viscosity. Miscible polymer blend samples, for the elongational
viscosity measurement, were prepared by using three steps: solvent blends, cast films,
and hot press. Immiscible blend samples were prepared by maintaining the prepared
miscible samples at 200 C, which was higher than the cloud points on the basis of a
LCST phase diagram. The miscibility was determined by the observation of whether the
fibre was transparent or opaque. It was found that strain-hardening property of miscible
blends in the elongational viscosity was only slightly influenced by the blend ratio, and it
was the case with immiscible blends.
The phase behaviour of blends involving several polymethacrylates (PMA) and a series
of poly(styrene-co-maleic anhydride) (SMA) copolymers has been investigated by DSC
and density measurements by Feng and co-workers [85]. The results of Paul and co-
workers [74] indicated miscible blends of PMMA with SMA containing MA in the range
of 6%-8% to 33%-47% by weight. It was shown that the cloud point behaviour of SMA
and PMA blends is similar to that of SAN and PMA blends. The cloud point of SMA/
PMA blends decrease in the order PMMA > PEMA > polybutyl methacrylate (PBMA).
Blends of PMMA with SMA [50 wt.% maleic anhydride (MA)] are miscible on a molecular
388
Polyacrylic-Based Polymer Blends
level. Feng concludes that the MA content in SMA needed for forming miscible blends
with PMMA is higher than that given by Paul and co-workers. The difference was
attributed to the molecular weight of SMA used in that study being an order lower than
that used by Paul and co-workers. It has been indicated that strong intermolecular
interactions between phenyl groups in SMA and carbonyl groups in PMMA. This
interaction instead of the intramolecular repulsion force within the SMA copolymer that
makes SMA/PMMA blends miscible. The strength of the intermolecular interactions
depends on the composition of the blends.
Exothermic interactions such as hydrogen bonding, ionic and charge transfer, etc., and
copolymer effect are commonly used to induce miscibility in immiscible blends. The
efficacy of these methods in promoting miscibility in poly(benzyl methacrylate) (PBzMA)
PS immiscible blends has been studied by suitably modifying the structure of the
component polymers [87]. It has been shown that these results can be extended to the
blends of poly(acrylates) or poly(methacrylates) with PS. It has been found that the
hydrogen-bonding approach is most advantageous among these approaches as it involves
the need for minimum interacting sites. To promote miscibility through ionic interactions,
sulfonic acid groups in PS chains and pyridine basic groups in PBzMA chains were
introduced. Cherrak and co-authors found that miscibility in PbzMA/PS blends have
been achieved through a copolymer effect when a minimum of 14 mol % of ACN units
were incorporated in PS. Through ionic interaction required only 4 mol % of the
interacting sites in both the components for miscibility, both their structures ought to
being modified. Besides, at higher concentrations of the interacting sites, insoluble
crosslinked gels were obtained. Miscibility through hydrogen bonding mechanism required
389
Handbook of Polymer Blends and Composites
only 5 mol % of the phenolic units to be incorporated in PS. Their results on PbzMA/PS
blends has been extended for other poly(acrylate)/PS or poly(methacrylate)/PS blends as
these systems broadly behave in the same fashion. Thus, PEMA has been shown to be
miscible with PS when the vinyl pyridine units and the sulfonic acid groups of 5 mol%
concentrations were incorporated in the respective polymers. Similarly, it has been shown
that the presence of 1 mol% of phenolic groups in PS was sufficient to cause miscibility
with PMMA and its ethyl analogue, whereas for poly(butyl methacrylate) 3 mol% of
phenolic groups was needed. PMMA required nearly 17.5 mol% of ACN units to be
incorporated in PS chains to induce miscibility through the copolymer effect mechanism.
As a conclusion, polyacrylates are not miscible with SAN copolymers, and the comparison
of the results edifies that hydrogen-bonding approach is the most convenient and potential
applicable way to induce miscibility.
One of the most sensitive techniques to prove the compatibility is to detect the separate,
smaller domain [88]. The assessment of the level of homogeneity depends on the technique
used in the investigation. Thus, it is fundamental to use suitable and coupled techniques
in studies of miscibilities of polymer blends. Significant importance has been put on the
behaviour of polymer blends under the effect of a flow field. It is well known that the
flow can change the thermodynamic state of the system and perturb the phase diagram
of polymer mixtures. Changes in the phase behaviour of a polymer blend that occur as a
consequence of the stress associated with melt flow are likely to influence significantly
the ultimate morphology of the processed and solidified materials. In this context, the
effect of simple shear flow on the phase behaviour of a PMMA/SAN blend was studied
[89]. It was observed that there was a shear-induced miscibility for all of the measured
samples, and the shear effect was found to be composition and molecular weight
dependent. The morphology of the blend under different shear rate values has also been
studied to understand the previous results for the PMMA/SAN blend and to obtain
deeper insights into the shear rate morphology relationship. The magnitude of the
elevation of the cloud points under shear was found to be composition dependent. Thus,
the cloud points of the polymer blends of PMMA/SAN indicated a high sensitivity to
shift under shear rate, higher than the sensitivity of both polymer solutions and low
molecular weight mixtures by about two and five orders of magnitude, respectively.
390
Polyacrylic-Based Polymer Blends
compositions (50% and 70% iPMMA) of the iPMMA/MPS-5 blends show three
transitions, indicative of multiple phases present in the blend.
It has been found that oligomers of bisA-PSU can be used to manipulate the phase
behaviour of polymer components and it is a useful method to obtain interaction energies
from components that are extremely incompatible, i.e., TMPSU ACN. Thus, blends
that have been considered immiscible in the past may actually be miscible at reduced
molecular weights that are still high enough to retain desired properties: impact strength,
yield stress, etc.
9.3.5 Acrylates/Other
391
Handbook of Polymer Blends and Composites
miscible with PECH and the interaction between PVMK and PECH is weak. Miscibility
of PVMK with PECH was ascribed to the hydrogen-bonding interactions between the
components.
392
Polyacrylic-Based Polymer Blends
SIN obtained. Thus, depending on the concentration of the catalyst for the PU, for the
same overall SIN composition, different morphologies were obtained, ranging from very
fine phase dispersion to more or less individualised domains. The results were confirmed
by dynamic mechanical spectroscopy (DMS) showing either a single broad transition or
two well-defined peaks in tan . DSC was used to demonstrate that samples realised by
IPN - technique changed slowly from a single phase to a two-phase morphology over a
period of eight months. The initially miscible state was termed a metastable state, which
reflects the high degree of entanglement of the linear PS chains with the poly(carbonate-
urethane) (PCU) network. That was attributed to high degree of entanglement of the
linear PS chains with the PCU network. The net effect was a limited molecular mobility
and greatly retarded phase separation in the otherwise immiscible PCU/PS system. By
contrast, the PCU/PMMA SIN were found to be stable, one-phased materials for up to
2.5 years.
The time stability of binary and ternary copolymers, as well as of mixture of binary
copolymers of ACN with vinyl acetate and/or -MS has been studied to obtain information
regarding the compatibility of such polymers [97]. The binary copolymers and
incompatible mixture are stable with time, while ternary copolymers and
pseudocompatible mixtures evolve towards an equilibrium limiting state of phase
separation, much faster in solutions than in solid state.
Cimmino and co-workers [98] have found by small-angle X-ray scattering (SAXS) and
DSC that the system PEO/iPMMA with a relatively low molecular weight of PEO (20,000)
is incompatible. Other papers show incompatibility of PEO with aPMMA. No changes
of crystallinity, melting temperature (Tm) and Tg were mentioned in mixtures of PEO
393
Handbook of Polymer Blends and Composites
(Mn ~ 2,000 and 125,000) with aPMMA (Mn ~ 500,000). The increase in Tg depression
was observed for blends with increasing molecular weight of low-molecular-weight
fractions of PEO (<7,000) and with increasing molecular weight of PMMA. Using low-
molecular-weight PEO (8,500) with aPMMA (46,800), segregation has been found at
certain compositions by inverse gas chromatography studies. A strong influence of
molecular weight on compatibility for higher molecular weights (>105) of aPMMA and
PEO in blends was observed. Recent investigations report on a local demixing in the
disordered phase of solid PEO/aPMMA blends [99].
One of the important issues is the location of amorphous diluent in the supramolecular
structure of solid mixtures. As shown by SAXS, the amorphous component can reside
between crystalline lamellae (sPMMA, aPMMA), or is excluded from interlamellar
amorphous regions. It was found that segregation of the amorphous diluent (PMMA)
strongly depends on Tg and molecular weight. The high Tg diluent resides exclusively in
the interlamellar regions, whereas the low Tg diluent is excluded at least partially from
these regions. SAXS and wide-angle X-ray scattering (WAXS) measurements also showed
that with increasing weight average molecular weight (Mw) of PMMA, the content of
PMMA in the interlamellar regions decreased [100]. The studies led to the idea that
there exists an interface at the surface of PEO lamellae from which the PMMA is excluded.
The changes of compatibility strongly influence the structural behaviour of the system. It
may be concluded that despite all research efforts, there are conflicting reports regarding
the level of mixing for PEO/PMMA, in particular concerning molecular weights of
components and the tacticity of PMMA. It has been shown that chains of low-molecular-
weight fractions (< 105) of PEO form crystalline lamellae. This has been supported by
quantified change in lamellar thickness under varying crystallisation conditions. The
most stable crystals are formed at low supercoolings without any fold. Therefore extended
chain crystals are formed.
The fracture properties of blends of PBT with ABS materials, compatibilised by a MMA
GMAEA terpolymer, MGE, have been characterised by Izod impact and single-edge
notch, three-point bend (SEN3PB) type tests [100, 101]. Thus, moderate amounts of
GMA functionality in the compatibiliser (>5%) and small amounts of compatibiliser in
the blend (<5%) significantly improve low temperature impact properties and ABS
dispersion. Higher amounts of GMA in the blend increase the room temperature impact
strength with little effect on the ductile-brittle transition temperature and increase blend
viscosity. Hale and co-workers remark that crosslinking reactions occur in the PBT/ABS/
MGE system in addition to the desired graft copolymer formation. They consider that
residual acid components in certain emulsion-made ABS materials catalyse ring-opening
394
Polyacrylic-Based Polymer Blends
Dlubek and co-workers studied the role of the free-volume holes in polymer blends and
their changes as a function of composition using positronium lifetime spectroscopy [72].
The results were discussed in light of data from dynamic mechanical analysis, DSC,
WAXS, polarised-light microscopy and mechanical testing performed by others co-authors
on specimens from polyamide 6 (PA6) and ABS, knowing the blends as a combination of
partially crystalline and amorphous components and are of practical interest because of
the expected improvement in their mechanical properties.
The fracture of thin (3.18 mm) and thick (6.35 mm) specimens of PC/ABS blends with
both standard and sharp notches was examined by standard Izod and SEN3PB
instrumented Dynatup tests [102]. Significant morphology coarsening was seen in some
PC/ABS (70/30) blends for thick (6.35 mm) injection moulded parts. Blends compatibilised
with 1% of an SAN-amine polymer exhibited well-dispersed, stable morphologies.
Studies have shown the benefits of analysis of fracture data using a two-parameter model,
which analyses the fracture energy per unit area as a function of the ligament size [103,
104]. Two mathematical approaches for analysing fracture behaviour as a function of
ligament size were introduced. Each is based on the idea first introduced by Broberg that
the region at the tip of a crack consists of an end region where actual fracture occurs and
an outer region where energy is plastically absorbed during crack propagation [105].
DSC and FTIR was used to study the miscibility of poly(vinyl methyl ether) PVME and
PBzMA [106]. It was shown that solvents and casting temperatures strongly affect the
kinetics of the formation of blends and lead to different phase structures of the blends.
The thermal transition behaviour of the PVME/PBzMA blend suggests a marginal
miscibility. Various scales of molecular aggregation might exist in the blends. FTIR results
395
Handbook of Polymer Blends and Composites
show that intermolecular interactions are weak and not specific, which is in agreement
with the Tg behaviour studies of the blends. The study showed that PVME is immiscible
with any other polymethacrylate, such as PMMA, PEMA, poly(propyl methacrylate),
poly(phenyl methacrylate), poly(cyclohexyl methacrylate), etc. The thermal transition
behaviour of the PVME/PBzMA blends indicated that miscibility exists according to the
conventional criterion of glass transition.
Miscibility of PBzMA and (poly) -caprolactone (PCL) was studied practically [107],
with acrylic polymers that possess an aromatic pendant group as polyphenyl methacrylate
(PPhMA) [108, 109]. The authors found that intermolecular interactions between the
pairs were likely nonspecific, but sufficient to lead to a thermodynamic miscibility between
the components. FTIR results do not indicate any specific interaction between C=O of
PPhMA and C=O of PCL, or through other pairs of functional groups. The miscibility
was attributed to the polarity and non-specific intermolecular interactions.
The effects of chain flexibility, concentration and molecular mass of the polymer, type of
phase separation, and stress and shear rate on phase transitions in polymer solutions
induced by the application of mechanical fields on phase diagram were studied on PE-
PVA and PE-PMMA blends using the methods of cloud point, viscometry, polarisation
microscopy, and DSC [110, 111]. It was confirmed that the application of shear stress
determines weakens of the interaction between the components, resulted by an increase
of the temperature of crystalline phase separation and for PE-PVA system, the shapes of
the boundary curves change.
The compatibilisers which can be block or graft copolymers may improve the properties
of blend materials such as polyolefins and polar polymers that exhibit poor mechanical
properties because of poor interfacial adhesion. Random copolymers with a strong specific
interaction with each individual polymer can be used as efficient compatibilisers. Ethylene-
methacrylic acid random copolymer (EMA) contains both ethylene segments and
methacrylic acid segments, and so may be used as a compatibilising agent. The role of
EMA in compatibilisation was investigated in binary blends of EMA and polyolefins,
and EMA and polar polymers [112]. Crystallisation of the linear low density polyethylene
(LLDPE) and EMA random copolymer blends was studied by DSC, WAXS and excimer
fluorescence. Then Zhao and co-workers found that ratio of the excimer emission intensity
to the emission intensity of the isolated monomer decreases upon addition of EMA
copolymer, indicating that PE segments of EMA copolymer interpenetrate mainly into
the amorphous phase of LLDPE but also in crystalline phase. As a result was suggested
the use of EMA copolymer as a compatibilising agent in recycling of polymer waste.
396
Polyacrylic-Based Polymer Blends
Polypropylene (PP) is limited in several applications due to its low surface energy, lack
of reactive sites, difficulty to dye, extremely poor hygroscopicity, poor barrier to oxygen
and many organic solvents, low melting and sticking temperatures, low impact strength,
sensitivity to oxidation, and poor compatibility with polar polymers. Additional
properties can be obtained by chemical modification, through grafting unsaturated polar
groups onto the backbone of non-polar resins. Among polar monomers that have been
reported in the literature for grafting have been acrylic (AA) and methacrylic acid (MAA)
and their esters [113]. AA and its esters were also grafted on the surface of other
polyolefins to improve the wettability and adhesion. The PP/PP-g-AA blends have been
used as an effective moisture barrier external layer. The effect of PAA has been as an
oxygen resistant central layer imparting at the same time an adequate interfacial adhesion.
Contact angle measurement is an adequate method for characterising surface polarity
of polymer blends. The method was used coupled with FTIR to characterise smooth
surfaces of PP/PP-g-AA blends, so confirming that AA was grafted onto molecular chains
of PP. Surface polarity of PP/PP-g-AA blends was augmented by an increase of PP-g-AA
content, which promotes a better wetting of the ink on the surface due to incorporation
of carbonyl polar groups.
Sun and co-workers [114] has shown that PP/PBT blends can be compatibilised in situ
by reactive extrusion. This is achieved by functionalising PP with three different monomers
AA, MA, and GMA. Therefore, the chemical functions grafted onto PP upon reaction
with the end groups of PBT (OH and COOH) ensure the compatibilisation of the blends.
A comparative study showed that GMA is the more reactive monomer towards the end
groups of PBT. The addition of GMA to PP/PBT (70/30 by weight) increases the elongation
at break from 29% to 110% and the impact strength (at 0 C) from 2 J to 19 J.
Stutz and co-workers [115] examined the reactive compatibilisation of PU/PP with PE-
co-AA at various levels of acid. They reported compatibilisation with the ethylenic
copolymer although no reaction of the carbonyl groups with the NCO functionality
was detected. In an extension of this work, Wallhainke and co-workers [116] examined
the morphology evolution of PU/PP blends compatibilised with ethylenic copolymers
containing varying contents of AA and butyl acrylate. No reaction between PU and the
compatibiliser was detected. The property improvement in the ternary blends was
attributed to dispersed-phase diminution and morphology stabilisation in the presence
of compatibiliser. To overcome the incompatibility, PE was functionalised with groups
able to interact with the terminal groups of polyamide. In this context the effect of
methacrylic acid derivatives grafted on PE chains on the PA6/PE blends structure has
been studied [117]. Ethyl-, isobutyl- and hydroxyethyl-methacrylate were grafted into
PE, so the compatibilising effect of ester and alcohol group was investigated. The functional
groups grafted in the PE chains induced compatibilisation phenomena in the blends
made with the non-functionalised PE.
397
Handbook of Polymer Blends and Composites
9.3.11 PMMA/Others
Compatible blends can be obtained from mixing diblock copolymers of THF and MMA
(PTHF--PMMA) with a homopolymer of THF (PTHF) [119, 120]. The authors presented
the PTHF--PMMA-PTHF blend as a very good system for the basic studies of microphase
separation, morphology, and crystallisation, including co-crystallisation, in different
confined regions. It was found that these blends could have an alternating PTHF and
PMMA lamellar morphology, a PTHF matrix microphase with dispersed PMMA
cylindrical or spherical microdomains, depending on the blend composition. For the
blends with PMMA spheres, it was found that the long period of PTHF microphase
increased with increasing PMMA weight fraction. This was attributed to the presence of
PMMA microdomains that could slow down the PTHF crystallisation process, resulting
in a larger long period. The morphology study of the blends showed that the crystalline
morphology is strongly dependent on blend composition, copolymer composition, and
PTHF block length, as well as the crystallisation temperature. In the blends where PMMA
spherical or cylindrical microphases were formed, the crystalline morphology changed
with the change in the PTHF microdomain size and PMMA interdomain distance. A
study of the solution-crystallised morphology of the blends at different temperatures
showed that the morphology was also dependent on the isothermal crystallisation
temperature. This behaviour suggested that the PMMA microdomains could have different
effects on the morphology formation when the blend was crystallised at different rates.
398
Polyacrylic-Based Polymer Blends
PHB phase [121]. The thermal and dielectric properties of blends of PMMA with atactic
PHB (aPHB) synthesised via anionic polymerisation have been reported. Unlike the natural
polyester, which very easily crystallises due to its complete (100%) stereoregularity, the
synthetic PHB, being atactic, is unable to develop a crystalline phase. In a aPHB/PMMA
blends, were both components are amorphous, the phase behaviour is expected to be
much simpler than when one of the components can crystallise and may phase separate
even in miscible blends above the Tg. The blends of PMMA with aPHB have been found
to be miscible while those using isotactic natural PHB showed limited miscibility with
PMMA. The difference of miscibility behaviour observed has been attributed to the large
molecular weight difference of the polyesters blended with PMMA, for example a high molecular
weight natural PHB (Mn = 350,000), and a low molecular weight aPHB (Mn = 3,000). The
authors concluded that blend miscibility increases when the molecular weight of the
component polymers decreases.
Because immiscibility usually leads to low degrees of phase dispersion and poor mechanical
properties, the control of the morphology of immiscible polymer blends has been realised
by using polymeric emulsifiers, in the form of block and graft copolymers. Modification
of blends by adding properly selected polymeric emulsifiers, often referred to as
compatibilisers, generally provides better dispersion and increased interfacial adhesion;
consequently the mechanical properties are improved. Block or graft copolymers having
in composition acrylic and/or methacrylic monomers have been used as compatibilising
agents to improve the interfacial characteristics between the different polymer phases.
However, graft copolymers are less widely used as compatibilising agents than block
copolymers because the molecular structures of these graft copolymers, which are
commonly prepared by chemical grafting reactions, are difficult to characterise and the
experimental techniques for controlling these grafting reactions are limited as well. As a
consequence the possibility of preparing copolymers with uniform grafts using the
macromonomer technique was intensively studied and handled.
The inherent brittle of the phenolic resins due to their higher crosslink density can be
moderate by incorporating the elastomeric and/or thermoplastic polymers into phenolics.
Matsumoto and co-workers [122] reported the modification of phenolic resin by a
p-hydroxyphenylmaleimide/acrylic ester copolymer to improve both flexural strength
and heat resistance. In the same context, the effect of the crosslinking density of phenolic
resins and the molecular weight of PMMA on the miscibility and structure development
of a phenolic derivative/PMMA mixture during cure was investigated [119]. Also the
mixture of phenolic derivatives with three poly(methyl methacrylate-co-ethyl acrylate)
(PMMA-co-EA) copolymers evidenced the effect of various MMA/EA compositions of
PMMA-co-EA and the fraction of multifunctional phenol in total phenols on the structure
development in phenolic/PMMA-co-EA mixtures [123, 124]. DSC measurements revealed
that the higher the fraction of multifunctional phenols in the total phenols, the faster the
399
Handbook of Polymer Blends and Composites
PMMA has been considered as a potential compatibiliser for PC/PVDF polyblends, since
PMMA is known to be miscible with PVDF and compatible with PC. PMMA might
behave as a common solvent for PC and PVDF in the melt. PC has been toughened by
core-shell rubber particles with a PMMA shell. It was suggested that the compatibility of
this PMMA shell with PC provided an interfacial adhesion strong enough for the rubber
particles to cavitate rather than to be debonded. PC envelops PMMA domains in a ternary
PC/PMMA/PBT blend. Thus, the interfacial tension between PC and PVDF as well as
PMMA and a series of PVDF/PMMA blends have been measured by the imbedded fibre
retraction method. The high interfacial tension between PC and PVDF rapidly decreased
by premixing approximately 35 wt.% PMMA with PVDF. In parallel, the PC/PVDF
interfacial adhesion increased as rapidly as the interfacial tension decreased in relation
to PMMA content in PVDF/PMMA blends.
Blends of bisphenol-A PC and PMMA have been widely studied in the literature. The
interactions between PC and PMMA are very weak as deduced from the difficulty of
preparing homogeneous mixture and the very low temperature at which these mixtures
phase separate on heating [125]. The improvement of the PC/PMMA blends with great
use for the manufacture of optical data storage media mention the enhancement of the
compatibility with acrylics or by changing the specific interactions between PC and PMMA
[64]. Some researchers intended to enhance specific interactions between the two polymers
by either changing the molecular structure of PC or by copolymerising the MMA units
with other monomer [65-67]. The improvement of compatibility between PC and PMMA
has been realised with imidised PMMA (polyglutarimide) when the imide concentration
was increased. Thus, it was shown that a slight increase of compatibility between PC
and acrylic polymers when the concentration of imide in acrylic increases. This was
400
Polyacrylic-Based Polymer Blends
attributed to the specific interactions between the highly polar imide groups and PC or
by internal repulsions in the copolymer between MMA and the glutarimide units as has
been depicted in acrylic/SAN blends [126].
Optical Polymer Research, Inc., provides fluorescent fibre in a rainbow of colours and
diameters, as Round Acrylic Core Optical Fibre. The fibre is made with a core of PS
surrounded by a clear acrylic cladding. Special fluorescent dyes are added to the core
where they absorb UV light and emit visible light. Light that is within the fibres capture
angle is guided down the fibre to the end for viewing.
Blends of ion-containing polymers have been extensively studied with the purpose to
increase miscibility as result of ion-dipole, ion pair-ion pair or acid-base interpolymer
interactions. The ionomerisation of PMMA by potassium carboxylate groups and its
effect on miscibility with PVDF, it was studied [127]. Miscibility was increased when
PMMA was modified by small amounts of ionic groups (approximately 1-2 mol%).
One of the most commonly used ionomer to compatibilise certain incompatible blends is
poly(ethylene-methacrylic acid-isobutyl acrylate) terpolymer. This ionomer has a small
number of ionic groups (less than 15% mol where the acid groups are partially neutralised
by metal ions) along nonionic backbone chains [128]. Thus it has been mentioned the
addition of the ionomer into an immiscible blend to provide a lowering of the interfacial
energy and an improvement of the interfacial adhesion between the two phases. The
final morphological effect was a reduction in the particle size of the dispersed phase in
the blend.
Because blends of PA1010 and butyl rubber (IIR) are considered to be immiscible due to
their different polarities it has been chosen the poly(ethylene-methacrylic acid-isobutyl
acrylate) terpolymer ionomer as compatibiliser for this binary system owing to the
miscibility and interaction of the ionomer segments with the two blends components.
The authors considered the fact that the ionomer hydrocarbon backbone being a common
feature with IIR, molecular intermixing occurs, to some extent, in the amorphous molten
state between the IIR and ionomer. The reactive groups of PA1010 have affinity to link
with (COO) groups present in the side chains of the ionomer. Thus the ionomer act as a
bridge holding the PA1010 and IIR components of the blend and give a compatibilised
blend. The presence of poly(ethylene-methacrylic acid-isobutyl acrylate) terpolymer as
compatibiliser improves the interfacial adhesion between the two phases and the dispersion
of IIR domains in the PA1010 matrix. SEM observations shown a great decrease in the
IIR domains size on addition ionomer. A ductile type of fracture was observed on the
401
Handbook of Polymer Blends and Composites
fracture surface of the PA1010/IIR blend compatibilised by this terpolymer. The system
presents the improvement in the notched impact strength of the blend after the addition
of the ionomer.
The literature in the field mentions the use of acrylic ionomer poly(ethylene-methacrylic
acid-isobutyl acrylate) terpolymer as a compatibiliser for incompatible PU/HDPE blends
[129]. The authors found that Zn2+ ionomer added at moderate concentrations (10-
15 wt.%) is an effective compatibiliser for melt-mixed and quenched PU/HDPE blends.
At the same time, the ionomer strongly solvates the PU component. It was considered
that compatibilisation is caused by the strong adhesion of the PU/ionomer component
due to specific forces and the mechanical compatibility of the ionomer polyolefin
components.
A random copolymer of MMA and AA (6 mol. %) has been synthesised and neutralised
to different extents by Zn cations. The effect of this PMMA modification on miscibility
with PVDF has also been studied [109]. The interaction parameter has been estimating
from melting point depression data, whereas Tg has been measured by dynamic mechanical
analysis. Modification of PMMA by 5.7 mol.% AA (random copolymer) has a deleterious
effect on miscibility with PVDF. But neutralisation of half the AA comonomer units by
Zn cations changes completely the situation as 12 decreases significantly becoming even
smaller than 12 for the original PVDF/PMMA pair. However, the complete neutralisation
is very unfavourable to miscibility, which was explained by the tendency of the Zn
carboxylated groups to self-associate into multiples and/or cluters (which is the typical
behaviour of ionomers) rather than to contribute to specific interactions with PVDF.
Blends present interesting theoretical as well as technological and practical interest. Acrylic
blends with their multiple applications in many fields such as, fibres, adhesives and binders,
paints and finishes, synthetic rubbers. As a result the interest into developing these kinds
of products is entirely justified.
References
2. E. Corradini, A.F. Rubira and E.C. Muniz, European Polymer Journal, 1997, 33,
1651.
3. E. Yilmaz, O. Yilmaz and H. Caner, European Polymer Journal, 1996, 32, 927.
402
Polyacrylic-Based Polymer Blends
5. Z. Sun, W. Wang and Z. Feng, European Polymer Journal, 1992, 28, 1259.
6. E.G. Crispim, A.F. Rubira and E.C. Muniz, Polymer, 1999, 40, 5129.
7. A.E. Nesterov, Y.S. Lipatov and V.V. Horichko, Polymer International, 1999, 48,
117.
11. Y.S. Oh and B.K. Kim, Journal of Macromolecular Science B, 1997, 36, 5, 667.
12. Z.H. Yin, X.M. Zhang and J.H. Yin, Journal of Applied Polymer Science, 1997,
63, 1857.
13. N.M. Chechilo and N.S. Enikolopyan, Doklady Physical Chemistry, 1975, 221,
392.
14. J.A. Pojman, G. Curtis and V.M. Ilyashenko, Journal of the American Chemical
Society, 1996, 118, 3783.
15. J. Szalay, I.P. Nagy, L. Barkai and M. Zsuga, Die Angewandte Makromolekulare
Chemie, 1996, 236, 97.
16. I.P. Nagy, L. Sike and J.A. Pojman, Advanced Materials, 1995, 7, 1038.
18. L.H. Sperling and V. Mishra, Proceedings of the 25th Anniversary Symposium of
the Polymer Institute, University of Detroit-Mercy, 1994, Technomic Press,
Lancaster, PA, USA, 1994.
19. E.V. Rusinova, S.A. Vshivkov and I.V. Zarudko, Polymer Science Series A, 1999,
41, 676.
20. L.H. Sperling in Interpenetrating Polymer Networks and Related Materials, Ed.,
L.H. Sperling, Plenum Press, New York, NY, USA, 1981.
21. L.A. De Graaf, J. Beyer and M. Moeller, Journal of Polymer Science: Polymer
Physics Edition, 1995, 33, 1073.
403
Handbook of Polymer Blends and Composites
22. H.S. Shin, S.Y. Kim and Y.M. Lee, Journal of Applied Polymer Science, 1997, 65,
685.
24. Y.L. Guan, L. Shao and K.D. Yao, Journal of Applied Polymer Science, 1996, 61,
393.
26. M. Krupers, P.J. Slangen and M. Moeller, Macromolecules, 1998, 31, 2552.
27. S.C. Kim, D. Klempner, K.C. Frisch and H.L. Frisch, Macromolecules, 1976, 9, 2,
263.
28. A. Morin, H. Djomo and G.C. Meyer, Polymer Engineering and Science, 1983,
23, 7, 394.
29. T. Hur, J.A. Manson, R.W. Hertzberg and L.H. Sperling, Journal of Applied
Polymer Science, 1990, 39, 1933.
31. M. Krupers, P.J. Slangen and M. Moeller, Macromolecules, 1998, 31, 2552.
38. S. Kim and J.H. An, Journal of Applied Polymer Science, 1995, 58, 3, 491.
404
Polyacrylic-Based Polymer Blends
42. E.I. Fedorchenko, V.I. Pavlov, S.S. Ishchenko and E.V. Lebedev, International
Polymer Science and Technology, 1999, 26, T/77.
43. D.R. Paul, J.W. Barlow, C.A. Cruz, R.N. Mohn, T.R. Nassar and D.C.
Wahrmund, Coatings and Plastics Preprints, 1977, 37, 1, 130.
46. G.R. Brannock and D.R. Paul, Macromolecules, 1990, 23, 5240.
48. T.K. Kwei, Journal of Polymer Science: Polymer Letters Edition, 1984, 22, 307.
51. W.P. Hsu and C.H. Yeh, Journal of Applied Polymer Science, 1999, 73, 431.
52. J.Y.J. Chung, R.P. Carter and D. Neuray, inventors; Mobay Chemical
Corporation, assignee; US Patent 4,554,315, 1985.
53. M.K. Akkapeddi, C.D. Mason and B. Van Burskirk, Polymer Preprints, 1993, 34,
848.
55. E.I. Fedorchenko, V.I. Pavlov, S.S. Ishchenko and E.V. Lebedev, International
Polymer Science and Technology, 1999, 26, T/77.
57. A.P. Chiriac and C.I. Simionescu, Progress in Polymer Science, 2000, 25, 2, 219.
58. A.P. Chiriac, I. Neamtu and C.I. Simionescu, Die Angewandte Makromolekulare
Chemie, 1999, 273, 75.
59. A.P. Chiriac, I. Neamtu and C.I. Simionescu, Synthetic Polymer Journal, 1998, 5,
43.
405
Handbook of Polymer Blends and Composites
60. A.P. Chiriac, C.I. Simionescu, I. Neamtu and G. Cazacu, Cellular Chemistry and
Technology, 2000, 34, 3.
61. A.P. Chiriac, C.I. Simionescu and I. Neamtu, Revue des Composites et des
Materiaux Avances, 1999, 9, 1, 133.
62. A.P. Chiriac, C.I. Simionescu, I. Neamtu and M. Popa, Cellular Chemistry and
Technology, 1998, 32, 5-6, 425.
63. A.P. Chiriac, I. Neamtu, G. Cazacu, C.I. Simionescu and G. Rozmarin, Die
Angewandte Makromolekulare Chemie, 1997, 246, 1.
64. C.K. Kim and D.R. Paul, Macromolecules, 1992, 25, 3097.
65. G.C. Eastmond and L.W. Harvey, British Polymer Journal, 1985, 17, 275.
66. C.C. Hung, W.G. Carson and S.P. Bohan, Journal of Polymer Science: Polymer
Physics Edition, 1994, 32, 141.
67. M.E. Fowler and D.R. Paul, Journal of Applied Polymer Science, 1989, 37, 513.
69. J.M. Willis, V. Caldas and B.D. Favis, Journal of Materials Science, 1991, 26,
4742.
70. M.T. Ahmed and T. Fahmy, Polymer Testing, 1999, 18, 589.
72. G. Dlubek, M.A. Alam, M. Stolp and H.J. Radusch, Journal of Polymer Science:
Polymer Physics Edition, 1999, 37, 1749.
73. M. Stolp, H.J. Radusch and A.K. Bledzki, Kunststoffe, 1993, 83, 7.
74. G.R. Brannock, J.W. Barlow and D.R. Paul, Journal of Polymer Science: Polymer
Physics Edition, 1991, 29, 413.
75. D. Singh, V.P. Malhotra and J.L. Vats, Journal of Applied Polymer Science, 1999,
71, 1959.
76. I. Fortelny, J. Kovar and M. Stephan, Journal of Elastomers and Plastics, 1996,
28, 106.
406
Polyacrylic-Based Polymer Blends
77. G. Wildes, S.H. Keskkula and D.R. Paul, Journal of Polymer Science: Polymer
Physics Edition, 1999, 37, 71.
78. K. Takakuwa, S. Gupta and D.R. Paul, Journal of Polymer Science: Polymer
Physics Edition, 1994, 32, 1719.
81. I. Soutar, L. Swanson, J.M.G. Cowie, I.C. Barker, N.J. Flint and M.J. Conroy,
Macromolecular Symposia, 1999, 141, 69.
83. T.A. Callaghan and D.R. Paul, Journal of Polymer Science: Polymer Physics
Edition, 1994, 32, 1847.
85. H. Feng, L. Shen and Z. Feng, European Polymer Journal, 1995, 31, 243.
87. D.E. Cherrak and S. Djadoun, Journal of Applied Polymer Science, Applied
Polymer Symposium, 1992, 51, 209.
88. S.A. Madbouly, M. Ohmomo, T. Ougizawa and T. Inoue, Polymer, 1999, 40,
1465.
90. L. Jong, E.M. Pearce, T.K. Kwei and L.C. Dickinson, Macromolecules, 1990, 23,
5071.
91. C.P. Papadoupolou and N.K. Kalfoglou, Polymer, 1999, 40, 905.
92. Q. Guo, J. Huang and X. Li, European Polymer Journal, 1996, 32, 321.
93. A. Mugica, M. Barral, J.A. Pomposo and M. Cortazar, Acta Polymerica, 1999,
50, 304.
407
Handbook of Polymer Blends and Composites
94. L.C. Cesteros, E. Meaurio and I. Katime, Macromolecules, 1993, 26, 2323.
95. C.A. Barron, S.K. Kumar, J.P. Runt and J. Fitzgerald, Polymer Materials Science
and Engineering, 1991, 65, 333.
96. V. Mishra, F.E. Du Prez, E. Gosen, E.J. Goethals and L.H. Sperling, Journal of
Applied Polymer Science, 1995, 58, 331.
97. C. Vasile, H.A. Schneider, A. Airinei and A.P. Chiriac, European Polymer
Journal, 1976, 12, 749.
100. W. Hale, J.H. Lee, H. Keskkula and D.R. Paul, Polymer, 1999, 40, 3621.
101. W. Hale, H. Keskkula and D.R. Paul, Polymer, 1999, 40, 3353.
102. G. Wildes, H. Keskkula and D.R. Paul, Polymer, 1999, 40, 7089.
103. Y. Kayano, H. Keskkula and D.R. Paul, Polymer, 1997, 38, 1885.
104. Y. Kayano, H. Keskkula and D.R. Paul, Polymer, 1998, 39, 821.
106. T.K. Mandal and E.M. Woo, Macromolecular Chemistry and Physics, 1999,
200, 1143.
108. H. Li and Z. Li, Journal of Applied Polymer Science, 1998, 67, 61.
109. H. Li, Y. Wang and Z. Li, Polymer Preprints, 1999, 40, 173.
110. S.A. Vshivkov, S.G. Kulichikhin and E.V. Rusinova, Uspekhi Khimii, 1998, 67,
261.
111. E.V. Rusinova, S.A. Vshivkov and I.V. Zarudko, Polymer Science Series A, 1999,
41, 676.
408
Polyacrylic-Based Polymer Blends
112. H. Zhao, Z. Lei, Z. Wang and B. Huang, European Polymer Journal, 1999, 35,
355.
114. Y-J. Sun, G-H. Hu, M. Lambla and H.K Kotlar, Polymer, 1996, 37, 4119.
115. H. Stutz, P. Ptschke and U. Mierau, Macromolecular Symposia, 1996, 112, 151.
119. L-Z. Liu, W. Xu, H. Li, F. Su and E. Zhou, Macromolecules, 1997, 30, 1363.
120. L-Z. Liu and B. Chu, Journal of Polymer Science: Polymer Physics Edition, 1999,
37, 779.
121. G. Ceccorulli and M. Scandola, Journal of Macromolecular Science, 1999, 36, 327.
123. B.S. Kim, G-I. Nakamura and T. Inoue, Journal of Applied Polymer Science,
1998, 11, 1829.
124. P. Rajatapiti, V.L. Dimonie and M.S. El-Aasser, Journal of Applied Polymer
Science, 1996, 6, 891.
125. C.K. Kim and D.R. Paul, Polymer, 1992, 33, 4929.
127. L.M. Leung and K.K.C. Lo, Polymer International, 1995, 36, 339.
128. E.M. Woo and T.K. Mandal, Macromolecular Rapid Communications, 1999, 20,
46.
129. C.P. Papadoupolou and N.K. Kalfoglou, Polymer, 1989, 26, 7015.
409
Rubber Toughened Epoxies/Thermosets
10
Rubber Toughened Epoxies/Thermosets
10.1 Introduction
The first synthetic plastics like cellulose trinitrate and polyvinyl chloride (PVC) were
rigid. PVC could not be processed unless it was flexibilised/toughened first. PVC and
polystyrene could become more useful because of the development of methods to toughen/
flexibilise them. The development of rubber modified thermosets and thermoplastics is
an important contribution to commercial polymer industry. A small amount of discrete
rubbery particles in a glassy plastic matrix can greatly improve the crack resistance and
impact strength of the normally brittle plastic. This must be accomplished with a minimal
loss in stiffness and thermal properties. Proper methods to toughen plastics and composites
can lead to enhanced usage of such materials in aerospace applications where weight
savings are of paramount importance.
Epoxy resin is the most widely used matrix material for aircraft structural composites. It
has many good properties like stiffness, low shrinkage, good adhesion to glass/carbon
fibre, etc. Unfortunately, the very factor contributing to its high stiffness and heat resistance
leads to its main drawback (lack of toughness), i.e., its highly crosslinked structure.
Aircraft structures are often subjected to impact loads during flight which can cause
severe damage to the structure by delamination. Toughening can help in developing
damage tolerant structural components for aircraft.
Table 10.1 shows that rubber toughened epoxy resins can match engineering
thermoplastics in fracture toughness values. Composites made out of such matrix systems
are expected to be tougher and hence more useful.
The need for toughening epoxy resins was felt in the mid 1960s. Initially, solid rubbers
like nitrile and polysulfide rubbers were found to flexibilise epoxy resins. Such resin
411
Fracture toughness
systems were found to be suitable for adhesives as these rubber modified resins could
sustain larger deformations and hence gave better peel joints than the unmodified epoxy-
based adhesive. However, as they caused unacceptable levels of deterioration of stiffness
and thermal properties for composites, better tougheners were needed.
A few scientists started working on toughening epoxy resins by using reactive liquid
rubbers synthesised by BFGoodrich Co., USA in 1965:
These rubbers had molecular weights (MW) in the range of 1000-2000. They could be
pre-reacted with the resin - they dissolve in the resin initially and precipitate during cure
to form discrete particles of diameters about 1-5 m. Often bimodal particle distributions
are encountered where two types of particles are formed. Large particles (above 5 m
diameter cause crazing while small particles (1 m or less in diameter) cause shear banding.
Shear banding is the main toughening mechanism. This is amplified by particle cavitation,
particle deformation and tearing. All these lead to strain energy release which causes
toughening. Rigid particles like treated fillers and thermoplastics contribute to toughening
by crack deflection or crack pinning mechanisms.
Most of the studies were confined to CTBN as it provided the most ideal morphology
needed for effective toughening. CTBN has two active hydrogens in its end groups so
that the resin underwent chain extension and was precipitated as spherical rubbery
particles in a glassy epoxy matrix. ATBN has four active hydrogens hence it acted as an
additional curative and provided a different morphology.
412
Rubber Toughened Epoxies/Thermosets
Until today, CTBN has been the most popular toughening additive for epoxy resins. However,
any other rubber which can precipitate as small particles during cure can also toughen epoxies
like polyacrylates, silicones (with reactive end groups) and latex based core-shell particles.
Rigid particles like alumina filler, alumina short fibre, etc., alone or in combination with
rubbers can also toughen epoxy resins effectively [2]. Rigid particle and thermoplastic
modification can toughen epoxies without decreasing the Tg, but the effectiveness on G1c
(fracture energy, see Section 10.1.2 ) is less pronounced compared with rubbers [2].
Reducing the crosslink density of the resin by chain extension with bis phenol-A, for
example, can also improve the toughness of epoxy resins. Such systems are used widely
in the adhesives field. Bertram and Burton [3] have argued that if such a chain extender
has more rigid backbones for example, tetra bromo bisphenol-A (instead of bisphenol
A), this can make the resin more ductile without decreasing the Tg of the resin. It has
been found to be so.
Presence of microvoids, and more homogenous networks, i.e., uniform crosslink densities,
can also lead to development of toughened epoxy resins [3].
For highly crosslinked resins, which were considered to be impossible to toughen, core-
shell particles, uniform crosslink distribution, thermoplastic modification and chain
extension are the approaches which are expected to lead to tougher resins and hence,
tougher composites [3].
Before going into the subject, mention may be made about how fracture toughness is
assessed. One definition is area under the stress-strain curve. More commonly, fracture
toughness is explained in terms of K1c, (stress intensity factor) and G1c (critical strain
energy release rate). These are measured by making a notch in the specimen and subjecting
it to tensile load. From the stress-strain data obtained, these parameters are calculated
using standard formulae [4].
10.2.1 CTBN
413
Handbook of Polymer Blends and Composites
For incorporating the rubber, an adduct between the epoxy and the CTBN needs to be
formed. For example, for the diglycidyl ether of bisphenol A (DGEBA) resin, 62.5 parts
of the resin may be mixed with 37.5 parts of CTBN in the presence of 0.25 parts of
triphenyl phosphine catalyst and the mixture may be heated for about half an hour at
120 C to form the adduct. The heating time may change with temperature. To incorporate
the necessary amount of rubber, this adduct may be suitably diluted with the resin.
CTBN was found to be an excellent toughening additive for epoxy resin by around
1965. Epoxy resin mostly a DGEBA CTBN hardener system has been studied
comprehensively over the years and a lot of data have been generated. The most important
thing to remember is: the efficacy of the toughening additive depends on how effective
the phase separation between the rubber and the resin, i.e., morphology of the rubber
dispersion in the resin is. This in turn depends on the type of epoxy resin, curative, cure
temperature, rubber composition, etc., and further, these factors are inter-related. So
reproducibility of the morphology is a problem and hence, consistency in the mechanical
properties of such system is also difficult to achieve [5].
If the crosslink density of the resin (Mc) is less, more ductility should result. Thus, more
plastic energy dissipation occurs during fracture. This gets magnified by the presence of
rubber particles, because this causes the formation of a large number of energy dissipating
shear zones. Thus, for an epoxy-CTBN-diamino diphenyl sulfone (DDS) hardener system,
Kinloch [7] observed the following:
For two epoxy resins, of epoxy equivalents 172 g/mol, and 475 g/mol, G1c values (in kJ/
m2) were almost equal. However the improvement in G1c caused by 15% (by wt.) CTBN
on the less densely crosslinked resin, i.e., with epoxy equivalent of 475, was up to 10 times
approximately, compared with a very small increase for the resin with an epoxy equivalent
of 172 g/mol. The improvement on a still less crosslinked resin (epoxy equivalent of
1600 g/mol), was found to be up to 50 times that of the unmodified resin.
The Tg of the resin can vary with the crosslink density and the cure schedules. Thus,
Kinloch [7] showed that for an epoxy resin-CTBN-piperidine system, for 16 hours of
cure at 120 C, crosslink density was observed to be about 610 g/mol for which Tg was
414
Rubber Toughened Epoxies/Thermosets
100 C and G1c was 2.23 kJ/ m2, for 6 hours of cure at 160 C, the crosslink density was
found to be about 7 times the first case, Tg decreased by about 10 C while the G1c
increased by more than 2.5 times that of the earlier case.
According to Levita [8], epoxy resins, when cured at higher temperatures are tougher
than those cured at lower temperatures, in spite of higher crosslink densities. This is
attributed to free volume effects at higher temperature though loss of curatives by
oxidation and volatalisation can also be cited as reasons. In epoxy - CTBN - piperidine
systems, toughening is observed only with cure temperature, above 100 C.
Levita suggested, that for formation of fine sized elastomeric particles, phase separation
should occur before gelation, when the viscosity is sufficiently high to prevent particle
coalescence, but still low enough to favour nucleation and growth. At low cure
temperatures, phase separation will be prevented.
Montarnal and co-workers [9] studied these in more detail, with DGEBA, CTBN cured
by 1,8-diamino-p-menthane (MNDA), a hindered amine hardener. The effect of cure
schedules for this system, is shown in Table 10.2.
Kim and co-workers [10] showed that if phase separation rate is faster than cure rate,
bimodal particle distribution will result, otherwise, unimodal. Thus, at higher cure
temperature, viscosities will be lower - this will also favour unimodal particle distribution.
7 h, 75 C 14 h at 190 C 2.09 32 15
415
Handbook of Polymer Blends and Composites
Levita [8] showed that CTBN particles can incorporate the resin as rigid filler and some
resin can dissolve molecularly in the rubber and form copolymers and increase strength.
The presence of bisphenol-A (a chain extender) can magnify the toughening caused by
CTBN, by causing bimodal particle distribution. This may be because, bisphenol-A can
increase the rate of the crosslinking which can make some rubber remain uncrosslinked
before cure leading to complex morphologies.
As far as the nature of the rubber is concerned, its volume fraction, particle size and its
distribution, interfacial adhesion and its Tg can all affect the extent of toughening.
Small particles (about 5 m diameter) can promote shear banding while large particles
cause crazing around large particles. Thus a combination of both, gives better results
which explains why attempts are made to get bimodal particle distributions [2].
Regarding the interfacial bonding between the rubber and the resin, this has also been
studied in detail and a few results are discussed next.
Huang and co-workers [6] reported that, compared with a difunctional liquid rubber, which
gave good interfacial strength, a nonfunctional rubber, i.e., one without reactive end groups,
gave a much poorer toughening effect, as this gave poor interfacial bonding between the resin
and the rubber particle. To compare the effect, a difunctional rubber gave a 70% improvement
in G1c against a nonfunctional rubber which provided only 20% improvement in G1c.
This may suggest that with good interfacial strength, some contribution to fracture energy
by crack bridging (by rubber particles) is possible. If there is poor interfacial strength (as
is the case with a nonfunctional rubber) this contribution will be nil because rubber
particle debonding will occur and once debonded, it cannot perform crack bridging.
Further, when we compare the interfacial bonding between two nonfunctional liquid
rubber grades, the one with higher acrylonitrile (ACN) content gives better interfacial
strength. The observed higher value of fracture energy with higher ACN content
(nonfunctional) may be due to enhanced matrix ductility caused by greater compatibility
between the rubber and the resin.
The rubber particles form copolymers with the epoxy resin but still retain their low Tg
(55 C). This means, their Poissons ratio may still approach 0.5, leading to a higher
bulk modulus. So they may not encourage large scale plastic dilatation in the matrix
unless the rubber particles undergo debonding or cavitation.
416
Rubber Toughened Epoxies/Thermosets
This has also been studied by many researchers. The ACN content in CTBN can vary.
For example in CTBN 1300 X 8 grade supplied by BFGoodrich Co., the ACN content is
8% and in 1300 X 13, it is 13%.
The change in ACN content affects the compatibility of the CTBN with the epoxy resin,
as with ACN content, polarity changes. Lee and co-workers [11] obtained the following
results for a DGEBA resin (MY 750 supplied by Ciba Geigy, cured by piperidine), modified
by CTBN:
The unmodified resin (MY 750) at 20% of CTBN (by weight), reduced the Tg by about
2 C, when the ACN content was 8%. The rubber containing 13% ACN reduced the Tg
by 7 C at any level of modification up to 30% CTBN by weight.
CTBN of 8% ACN content in this resin showed dispersed particles of 0.6-4 m diameter.
Higher amounts of CTBN (about 30%) caused phase inversion, the molecular weight of
CTBN increased by chain extension (by resin) leading to larger particles which ultimately
made the rubber, the continuous phase. In these studies, CTBN with 13% ACN content
did not show dispersed particles, i.e., no phase separation. The higher ACN content
made the rubber more polar and hence miscible with the resin.
ATBN and HTBN are also being produced as possible modifiers for epoxy resins. The
studies with ATBN have been mainly confined to adhesive applications. The amount of
ATBN that can be incorporated into epoxy resin can be much higher than that of CTBN,
because the amine group in ATBN can work as an additional hardener unlike CTBN
which can only undergo chain extension.
HTBN can be incorporated into epoxy resins with difficulty. It needs a coupling chemical
like a diisocyanate. Sankaran and co-workers [12, 13] have toughened epoxy resins with
HTBN using toluene diisocyanate as a coupler and reported an optimum performance
with HTBN at 6% by weight. In this case, toughening was accompanied by a slight
increase in modulus (and without a loss in Tg). This was explained as being due to an
increase in crosslink density caused by the addition of the diisocyanate.
417
Handbook of Polymer Blends and Composites
DGEBA cured by piperidine was the resin studied. Before mixing the resin, the unreacted
NR and MMA were removed by extraction in some cases with the appropriate solvents.
The cure temperature was 120 C for 16 hours.
Heveaplus MG, both at low (34-39 x 104) and high (8.5-9 x 105) molecular weights were
shown to toughen epoxy to a good extent. At levels of 5 phr, the low molecular weight
Heveaplus MG, increased K1c by about 2 times and the high molecular weight copolymer
(at 5 phr of the additive) to a slightly lesser extent (by about 1.8 times). The corresponding
increase in G1c are about 5 times and 2.5 times, respectively). The results for CTBN
(10 phr) were: 2-fold increase of K1c and 4-fold increase of G1c while Heveaplus MG
(both high and low molecular weight) gave a 3.5-fold increase of K1c and an 8-fold
increase of G1c. The results were also better for Heveaplus MG compared with CTBN at
low temperature tests done on toughening. This could be due to a lower Tg of NR.
Fracture surface scanning electron microscopy (SEM) showed rubber particle stretching
and tearing. This could be a major contributor to toughening, as it is well known that
NR exhibits strain induced cystallisation, prominently.
Gouri and co-workers [15] studied modification of epoxidised novolac resin by solid
CTBN, by heating in the presence of triphenylphosphine in solution. Such systems were
cured by DDS and dicyandiamide systems. Such systems reduced Tg by 19 and 10 C,
respectively. Such systems were used in adhesive formulations. The rubber was found to
be dispersed at 100-200 m particle sizes. They showed improved peel strengths.
418
Rubber Toughened Epoxies/Thermosets
10.2.3.3 Silicones
Silicones are immiscible with epoxy resins but low molecular weight silicone polymers
with reactive end groups can be made to react with epoxy groups and form block
copolymers. These can toughen epoxy resins. Yorkgitis and co-workers [16] studied poly
dimethyl siloxane (PDMS) at low molecular weights as modifiers for epoxies. They were
compatibilised with epoxy resins by using small amounts of PDMS in which some or all
methyl groups were replaced by fluoroalkyl or phenyl substituents. The silicones were
incorporated by reacting with the epoxy resin for 1 hour at 65 C.
Other advantages expected of silicone modification are: much better low temperature
performance because Tg of PDMS is 120 C compared with around 50 C for CTBN);
better moisture resistance, weather resistance, high temperature performance and lower
wear rates. The wear resistance was found to be better with silicone modification [16].
Kothandaraman [17] used amine terminated PDMS (molecular weight 10,000, supplied
by Resil Chemicals Ltd, India) in DGEBA resin (LY 556), cured by methylene dianiline
(MDA). The maximum level of silicone that could be added was about 15 phr; beyond
this lack of solubility of the hardener in the blend was suspected. Incorporation of amine
terminated resins could be done by warming DGEBA with the PDMS at 50 C for about
30 minutes. The tensile and flexural strength of the modified resins were slightly higher
than those of the corresponding CTBN modified resins. The results are shown in the
Table 10.3.
In the corresponding glass cloth composite, the modification of the resin by silicone did
not decrease the tensile and flexural strength, while in the CTBN modified composite,
the strength was found to decrease. The optimum performance was observed at the
10 phr level of amine terminated PDMS. The unnotched Izod impact test of the composite
gave similar results on those of CTBN modified composite. Further the silicone modified
composites showed lesser moisture ingress under accelerated testing conditions compared
with unmodified composites.
This unusual behaviour, i.e., toughening of a composite by a rubber without loss of strength,
could probably be explained as due to the silicone acting as an additional coupling agent
between the resin and the fibre (the results are shown in Section 10.6).
419
Handbook of Polymer Blends and Composites
Takahashi and co-workers [18] studied the effect of PDMS (MW of 2000) on epoxy-
based moulding compounds used for electronics industry (for Printed Circuit Boards
(PCB), etc). Here too, flexural strength was found to increase by addition of PDMS to
the moulding compound (they contain silica filler). The increase in strength (but not in
modulus) was explained by an improved epoxy filler interaction due to the silicone (which
could act as an additional coupling agent). Further the idea of improved epoxy filler
interaction was reinforced by the observation of a reduction in coefficient of thermal
expansion (CTE) by the added silicone, in the silica filled resin. Normally, a rubber
increases the CTE values.
A similar study due by Lin and co-workers [19] in a trifunctional epoxy-based moulding
compound for electronic application, modified by PDMS with reactive end groups also
showed similar results of increased toughness with reduced CTE and maintenance of
flexural strength by the added silicone. Another study [20] showed that PDMS with an
amine end group improved toughness with a slight increase in flexural strength. The
results were better with the rubber having a higher MW (about 5000).
Shimizu and co-workers [21] studied the effect of silicone gel on casting compounds
containing epoxy resin and silica flour as filler. Silicone gel at 15 phr caused a 7-fold
increase of K1c compared to that of the neat resin and a 2-fold increase compared to that
of the untoughened compound. For the filled compounds, the silicone modified compound
showed a small decrease in modulus by about 20%.
420
Rubber Toughened Epoxies/Thermosets
Ochi and co-workers [22] synthesised block copolymers of carboxylated PDMS with
aramid. They were used as compatibilisers for a blend of epoxy with room temperature
vulcanisable (RTV) silicone (liquid silicone rubber). Silicone particles were seen dispersed
in the toughened resin, but covered by these copolymers, providing particle matrix
compatibility. This system improved toughness (G1c) by 100%.
IPN containing dispersed silicone micro and nano phases were prepared by Buchholz
and Mulhaupt [23]. p-Hydroxyl benzoate was introduced as a terminal group into PDMS.
This was blended with epoxy resin and cured. At levels of 10%, K1c improved to 2.5-fold
compared to the unmodified resin, with a reduction in flexural modulus of 10%. The
silicone was found to be dispersed in nanometer level particle sizes.
Banthia and co-workers [24] synthesised telechelic acrylic ligomers for modifying epoxy
resins. They polymerised ethyl hexyl acrylate monomer in the presence of azobis
(4-cyano valeric acid). This system gave oligomers with COOH end groups.
This could be reacted with epoxy resin (DGEBA) in presence of 1% triphenyl phosphine
and the resultant toughened epoxy, cured by bis (4-amine cyclohexyl) methane at 100 C
for this followed by post cure at 160 C for 2.5 hours.
The MW of the rubber were varied and their effect on Tg were reported. MW in the
range of 9500 to 15000, at 10% rubber content Tg did not vary much. At a MW of
11500, change in rubber contents for 4% to 10% did not vary the Tg. Improvement in
impact strength was comparable to what could be obtained by ATBN toughening.
Kelly and Ptak [25] synthesised acrylic oligomers as latices with two different particle
sizes and mixed these with epoxy resin in presence of a solvent and subsequently the
solvent was evaporated. They also incorporated a reactive monomer (glycidyl
methacrylate) in some cases so that they could react with the epoxy resin and provide
good resin-rubber particle bonding. Large particles (reacted with the resin) showed lesser
reduction in Tg compared with smaller ones. Particles reacted with resin showed lesser
Tg reduction compared with unreacted particles. While fracture energies were the same
for large particle modified epoxies (whether reacted or not) those by non reacted small
particles was better than reacted ones. The optimum properties were seen at rubber
content of 5%-10%.
421
Handbook of Polymer Blends and Composites
Epoxy resins were also evaluated for toughening by other systems. For example,
Frischinger and Dirlikov [26] studied epoxidised soybean oil as a toughener for DGEBA
resin. At 20 phr levels, this modifier can increase K1c by about 77%-80% for MDA
cured resin and by about 50% for DDS cured resin. The effectiveness of this modifier
was considered to be on par with that of CTBN. This modifier costs much less than
CTBN and hence this provides scope for more studies. Similar results were shown by
Ratna and co-workers [27]. They also showed that such systems can improve the
performance of adhesive joints.
Bagheri and Pearson [28] showed that hollow latex particles, based on methacrylate-
butadiene-styrene are as good as liquid rubbers for toughening epoxies.
Bagheri and co-workers [29] used surface treated scrap rubber particles at 2.5 phr along
with CTBN at 7.5 phr improved K1c of the neat resin by two times. Similar observations
were also made by Boynton and Lee [30].
Toughening of a high performance epoxy resin like tetra glycidyl methylene dianiline
(TGMDA) has been a greater problem. These resins have a high Tg (190 C and upwards)
due to high crosslink density. The normal CTBN toughening has not been successful in
this case. Increase in G1c has been modest and accompanied by a drastic fall in modulus.
The effect of elastomeric additives on aluminium filled (50 phr) TGMDA (modified with
diluents) tooling resin is shown in Table 10.4.
Pearce and co-workers [32] showed a way for improving the toughness of TGMDA
resin. Instead of curing the resin, by piperidine, they used adducts of piperidine with
isocyanate or an adduct of isocyanate with methyl piperidine as curatives. Those with
CTBN (20% by weight) give much better improvement in G1c (about 24 times compared
to the unmodified resin). Piperidine seems to catalyse homopolymerisation which makes
the resin more ductile. In this condition CTBN can magnify the effect.
Preformed core-shell rubbery particles have also been studied as tougheners for epoxy
resins. Sue and co-workers [33] studied such systems as possible tougheners for high
performance epoxy resins. They are prepared by a two stage latex emulsion
422
Rubber Toughened Epoxies/Thermosets
polymerisation. The core is an elastomeric polymer which can be grafted upon. This
polymer is either crosslinkable or insoluble in the epoxy resin. The shell should have
chemically attached polymer chains and non grafted chains. The shell polymer should
have a MW in the range of 20,000-100,000. The shell should be compatible with the
epoxy resin so that the epoxy toughener additive system remains as a stable colloid until
curing is complete. For this a mixed monomer system is of help. Two types of rubbers
have been studied, (a) styrene-butadiene latex based-graft rubber concentrated (GRC)
and (b) dispersed acrylic rubbers (DAR). In the studies by Sue and co-workers [33], the
shell system studied was styrene, methyl methacrylate (MMA), ACN and glycidyl
methacrylate (GMA). The grafted shell thickness was raised by changing the core-shell
ratio from 84:16 to 75:25 or 65:35. Polarity could be raised by varying the ACN content
from 0% to 25% (of shell weight). To vary particle resin interfaced strength, GMA
content was varied from 0% to 30% of shell weight.
First, a latex of styrene butadiene rubber was prepared. This was placed under an inert
atmosphere and in presence of a radical initiator, some quantities of the monomers were
added while the latex was still warm (at about 70 C). The remaining monomers were
423
Handbook of Polymer Blends and Composites
added over a period of about 1.5 hours with a slow increase in temperature to 80 C and
followed by heating for another 2 hours. To the resultant mixture was added MEK to
form a dispersion. From this, the organic layer was separated and added to the epoxy
resin and the solvent removed by distillation. This method produces a dispersion of core-
shell particles in epoxy resin. This could be directly cured. The morphologies could be
varied by varying the GMA and ACN contents. Minor addition of GMA led to good
dispersion of the GRC particles, but, further addition led to multimodal distribution of
GRC. Some of the results of these studies are shown in Table 10.5.
The toughening mechanism was cavitation of the particle followed by shear yielding of
the matrix around the crack tip. This was irrespective of GMA content. GMA was used
for bonding chemically, the particle with the resin. Without GMA, ACN content controls
the dispersion level this is important in toughening. At lower core-shell ratios, GMA,
together with ACN, changes the dispersion from a random to a locally clustered dispersion
(probably caused by reactions involving GMA groups). This increases K1c by introducing
a new toughening mechanism called the crack deflection mechanism.
Similarly, dispersed acrylic rubber particles have been synthesised and studied by Riew
and co-workers [34]. In this case, inner cores (plastic) and middle elastomer layer and an
outer shell (copolymers of MMA and ethyl acrylate (EA) with functional groups, e.g.,
GMA) were developed and used. These show mixed results as far as effectiveness in
toughening is concerned. In some cases, they appear to toughen less effectively than
CTBN but this, unlike CTBN does not reduct Tg, stiffness or weather resistance.
Sue and co-workers [35] found that DAR are less effective than core shell rubber (CSR)
particles in toughening. DAR particles seem to remain unextended during loading and
Table 10.5 Variation of G1c with variation in core-shell ratio and GMA
contents (adapted from Tables I and II of [33]
Core-shell ratio GMA content (parts) *Change in G1c(J/m2
* with reference to the base resin - G1c of the base resin was 180 J/m2
Please note: the relative concentrations of styrene, MMA and CAN were not identified in
these cases
424
Rubber Toughened Epoxies/Thermosets
hence not able to promote crack deflection and crack bifurcation. These mechanisms
however are less effective than shear yielding for toughening. The studies show the
following results for DER 332 epoxy resin DDS, a highly crosslinked system. The
resin, hardener and core shell particles were mixed at 130 C, and cured at 180 C for
2 hours and post cured at 220 C for 2 hours.
An increase of G1c to 3.5 times that of the unmodified resin could be seen with GMA and
ACN contents in the shell being higher. The changes in Tg at all levels of CSR additions
were negligible.
However, Becu-Longuet and co-workers [36] showed that CSR particles based on butyl
acrylate showed better toughening than styrene-butadiene based CSR particles.
In studies by Jansen and co-workers [5], epoxy rubber particles were preformed by curing
droplets of alipatic epoxy resin (digycidyl ether of polypropylene oxide DGE-PPO) in
the aqueous layer and transferred into aromatic epoxy resin, and cured. For less crosslinked
epoxies like DGEBA, these particles improved toughness to a lesser extent but for tightly
cured systems (higher Tg resins: 190 C) this system gave much better toughening.
Sue and co-workers [37] studied three high Tg epoxy resin systems, i.e., DGEBA (chain
extended with varying amounts of bisphenol A), DGTMBA-diglycidyl ether of tetra methyl
bisphenol A with tetramethyl bisphenol A (TMBA), and diglycidyl ether of tetra bromo
bisphenol A (DGTBBA) with TBBA, toughened with core shell particles. They found in
the case of DGTBBA resin, more chain extender (TBBA) led to more rigid (less tough)
resins. Actually in these systems, longer chain lengths between crosslinks led to lower Tg
as usual but this did not translate to greater flexibility. This was because of the change in
curing systems or changes in epoxy rigidity when changing from DGEBA to DGTBBA
monomers.
While CTBN toughening has been effective in epoxy resins it suffers from two major
drawbacks: (1) the improvement in toughness is at the cost of thermal properties and
modulus and (2) CTBN toughening is found to be less effective in tightly crosslinked
425
Handbook of Polymer Blends and Composites
resins, e.g., TGMDA resin. In 1983, thermoplastics were studied as possible tougheners
for epoxy resins with some success [38].
The plastics studied for this purpose are: polyimides, polyether sulfones (PES), polyether
ketones (PEK), polybutylene terephthalate (PBT).
The mechanisms which could operate in these systems are (i) crack pinning, (ii) particle
bridge, (iii) crack path deflection, (iv) shear banding and (v) microcracking.
The efficiency of the toughening additive will depend on (i) size of the particles, (ii)
strength of the particle, (iii) particle matrix adhesion and (iv) particle distribution.
PBT modification of epoxy resin was studied by Kim and Robertson [39]. PBT was
dissolved in phenol, precipitated from methanol, ground to a powder and mixed, with
the resin. To find the best method of blending, the PBT and epoxy were mixed at various
temperatures, with and without mechanical mixing. The best improvement in fracture energy
was observed by heating PBT particles with epoxy resin without curing at 10 C/min to
193 C and suddenly cooling to room temperature, and then the curative was added.
Such a system showed a 4.5 fold improvement in fracture energy. The reason for the
success seemed to be formation of an interpenetrating network.
PES was studied as a possible toughener by a number of authors, e.g., in 1983, Raghava
and co-workers [2] reported a modest increase in fracture toughness due to PES.
Hendrick and co-workers [40] used two low molecular weight of hydroxyl terminated
poly sulfones in DGEBA epoxy cured with DDS (MW 5000-10000) at 15 phr loading
which showed an increase in K1c of about 0.7 MPa.m0.5. The polysulfone was found to
be precipitated as particles of about 0.2 m.
The toughening was thought to be as due to strengthening of the plastic epoxy interface.
Polyimide was tried as toughener for epoxy resins by a few researchers. For example,
Wu and co-workers [44] used a readily processible maleimide resin for toughening epoxy
resin, which produced good results.
PEK at various MW showed variation in particle sizes - between 0.5 and > 100 nm. The
improvements in K1c values obtained were +0.2 MPa.m0.5, molecular weight of about 7000,
+0.6 MPa.m0.5 at 15000 and +2 MPa.m0.5 at 18000. The better toughness at higher MW was
attributed to increased ductility of the thermoplastic at higher MW. In most of the successful
cases, the thermoplastic underwent phase separation from the epoxy resin, during cure. It
was also shown that phase separation is not sufficient, though essential, to improve toughness.
426
Rubber Toughened Epoxies/Thermosets
Above certain levels of addition of the thermoplastic, the blends of epoxies with
engineering thermoplastics undergo phase inversion where the discontinuous phase is
now the crosslinked epoxy resin. They may be thought of, as semi interpenetrating
networks (IPN) between the thermoplastic and the epoxy resin. They too have good
toughness, without decrease in Tg and high modulus (compared with the rubber modified
systems). As far as processability is concerned these are better than thermoplastics as
matrix materials for advanced composites. However, their resistance to creep, solvents,
stress cracking, etc., are expected to be low and hence such systems may pose problems
on these counts [3].
DGEBA-DDS-polysulfone + 0.7
TGMDA-anhydride-PES +0.2
DGEBA-MDA-PBT +1.1
DGEBA-DDS-PEK +1.4
The use of polyester resin in composites is more widespread than epoxy resins mainly
because of their lower costs. Polyester resin composites in various forms are used in
automotive, auto-electrical, civil engineering, and other areas. The major disadvantages
of polyester resin and its composites are poorer chemical resistance (especially alkali),
lesser mechanical strength and toughness. The toughness is so low that sometimes,
products made by sheet moulding compound (SMC) technology, break even during
demoulding or assembly in their systems. Unlike epoxy resins, a correct toughening
additive (like CTBN) is yet to be specified though many papers and patents in this area
are available.
The main difficulty in toughening polyester resin is that its very molecular architecture
prevents development of the necessary morphology. Furthermore, the presence of a reactive
427
Handbook of Polymer Blends and Composites
solvent monomer makes the resin undergo greater shrinkage during cure and causes
formation of microvoids which can counter-balance the advantages offered by a
toughening additive (Suspene and co-workers in [41]).
However, many solid rubbers have been added as a solution in styrene, to the resin in
SMC and they have shown improvements in impact strength and surface finish of the
products. The improvement in impact strength is always at the expense of stiffness. The
incorporation of such an additive is difficult as it may require the rubber to be milled in
a mixing mill before mixing with the solvent. Grossman [42] argues that addition of
solid rubbers will be of a greater advantage compared with liquid rubbers. Unlike glass-
epoxy composite, in polyester composite, failure begins at the fibre-resin interface. So, if
a rubber with lesser polarity but with polar end groups like amine or COOH is used, and
if it can cure slower than the polyester resin, its high MW which was earlier thought of
as a disadvantage will now become an advantage as it can contribute favourably to the
work of adhesion between the resin and the fibre. Thus, Grossman observed that, out of
a few solid rubbers like solid carboxylated nitrile rubber, solid amine terminated nitrile
rubber, solid carboxylated ethylene-propylene terpolymer rubber (EPDM), etc., amine
terminated nitrile rubber improved the toughness of SMC while slightly increasing the
mechanical strength.
For a hand lay up glass fibre reinforced plastic (GFRP), presence of a solution of styrene-
butadiene rubber up to 0.4% by weight in the resin was found to increase impact strength
of the composite [43].
For liquid rubbers, HTBN was found to be a fairly good toughener. Tong and co-workers
[44] showed its effectiveness can be increased drastically by prereacting it with the resin,
i.e., free OH in the HTBN, can be reacted with the free COOH groups in the resin and
form block copolymers with the resin.
Further, BFGoodrich Co., developed a butadiene ACN copolymer with vinyl terminal
groups (VTBN) which can be physically blended but can react with polyester resin during
peroxide cure and toughen the resin. However, Tong and co-workers [44] showed that
chemically reacted HTBN was superior to VTBN in toughening the resin and its SMC.
Table 10.7 shows the advantages of forming block copolymers between polyester resin
and HTBN.
Further, in the SMC composites, increase in the reacted HTBN content of up to 15% by
weight, increased flexural strength by 2.5 times, tensile strength by 5 times and impact
strength by 3 times, although modulus, hardness and heat distortion temperatures
decreased with compared to the control resins.
428
Rubber Toughened Epoxies/Thermosets
Sometimes, changing the reactive monomer from styrene (partially or fully) to one which
can give a rubbery polymer on polymerisation, for example, ethyl or butyl acrylate, can
effectively flexibilise/toughen polyester resin [45, 46].
Table 10.8 shows the effect of EA when added to a general purpose polyester resin, in
addition to the styrene already present in the resin [45].
Lin and co-workers [47] showed that a tougher polyester composite can be made by
developing an interpenetrating network between epoxy and polyester resins. They mixed
DGEBA resin (with m-xylene diamine as a curative) and a general purpose polyester
resin (curative: benzoyl peroxide) and cured the resins together, after prereacting at 60 C
for some time and then cured at 120 C for 6 hours. Epoxy-polyester ratios between
25:75 and 50:50 showed good balance in properties. This is attributed to interpenetrating
network formation. Networks of both resins get interlocked during cure leading to parts
of both resins getting partly cured. These parts act as plasticisers and cause more shear
yielding, leading to toughening.
429
Handbook of Polymer Blends and Composites
Subramaniam and McGarry [48] found another novel system for toughening polyester
resins: here too IPN were formed, but here it was a rubber toughened (ATBN) epoxy-
polyester system. Polyester resin was precured with a catalyst and then the ATBN dissolved
in styrene was added, followed by the epoxy resin and its hardener. The whole system
was heated to complete the epoxy resin cure.
The rubber content was varied. The rubber content at 6.7% provided the best set of
properties - tensile strength was maintained compared with the rubber-less composition,
while strain-to-failure increased by about 50%, K1c almost doubled, while G1c increased
to 4 times that of the rubber-less compositions. The variations of these properties with
rubber content, was not smooth. This could indicate that at rubber contents such as
6.7% and 17.7% drastic changes in morphology could occur. Beyond 10% rubber content,
toughening was observed to be accompanied by a drastic loss in stiffness.
A higher performance polyester resin called vinyl ester combines some of the advantages
of epoxy and polyester resins. Siebert and co-workers [49] studied toughening of elastomer
modified vinyl ester resin using a rubber called epoxy terminated ACN butadiene copolymer
(ETBN). This is made by reacting CTBN with bisphenol-A in a 1:2 molar ratio.
ETBN was dissolved in styrene as a 50% solution and added to an elastomer modified
vinyl ester resin at various levels. The Tg did not decrease much by the addition of the
ETBN while at 15.6 phr ETBN content, G1c increased to 4.5 times that of the unmodified
rubber modified vinyl ester resin.
430
Rubber Toughened Epoxies/Thermosets
The effect of CTBN modifier in filament wound GFRP (fibre content 72% by weight) is
shown in Table 10.9 (Epoxy resin: LY 556 cured by MDA, CTBN: Hycar CTBN 1300 X 8).
Work by Scott and Phillips [52] showed that it may not be possible to reproduce accurately,
toughness measurements in CFRP due to a few possible misalignments in fibres leading
to additional contributions from the resultant fibre pull out, debonding, etc.
Kim and co-workers [53] argued that delamination is the most prevalent life limiting
damage mode which can cause unacceptable reduction in stiffness and other mechanical
properties. Rubber modified resins give much better post impact compressive strength to
the composite than unmodified resins. Impact by drop weight along thickness direction
does lesser damage to a composite based on rubber modified resin through a little
delamination and intralaminar transverse shear cracks. For unmodified resin composites,
fibre and resin fracture are extensive delamination in all plies. These studies further
show that at small fibre volume fractions, the loss of strength caused by the rubbery
additive was severe. This may be due to formation of small air bubbles in the resin-fibre
interface (this could be aggravated by the high viscosity of the rubber additive). Application
of pressure can reduce the bubbles but this will increase the fibre content and this will
constrain the plastic deformation of the resin.
A few cases of work done on toughening of epoxy resin composites are discussed next.
McGarry [55] proposed that instead of mixing the rubber in the resin, the rubber could
be coated onto the fibre and the composite made without the rubber in the resin. The
results were encouraging for glass fibres but not as much for carbon fibres. The rubber
used was ATBN. At 1.5% of ATBN (in resin, by weight) interlaminar shear strength
increased by 40% and Izod impact strength increased by 30%. SEM studies showed
some fibre pull out and that the volume of the distorted matrix around the fibre appeared
to increase with coating thickness. Mostly cohesive failure was seen in the rubber layer.
In another study, Barcia and co-workers [56], grafted hydroxyl terminated polybutadiene
(HTPB) on the carbon fibre surface and then prepared the composite with epoxy resin.
The grafting was done by reacting the fibre with HTPB in the presence of a catalyst like
tin octoate under heat. The HTPB was also reacted with the resin and the impact strengths
were measured. Impact strength was found to improve drastically when the fibre was
grafted with HTPB and also the resin contained 10% by wt of HTPB. This was due to
improved polymer-fibre interaction.
Table 10.10 shows the effect of amine terminated PDMS (incorporated into the epoxy
resin) on hand lay-up composites with glass cloth (based on epoxy resin LY 556).
Table 10.11 shows the effect of high temperature with moisture on composites unmodified
or modified (by 10 phr of amine terminated PDMS [17]).
Low and co-workders [57] studied the effect of elastomeric additive (CTBN) on short
alumina fibre filled epoxy resins. The properties were found to be the best at 15% of
CTBN and alumina fibre. The fracture energy values were also better for a wider
temperature range for the rubber modified composite than the unmodified composite.
432
mechanical properties of Bidirectional glass cloth composite
effect of silicone on Glass-epoxy composites
Jang and co-workers [58] studied a few methods to toughen CFRP by modifying the
moulding techniques. They observed that polyamide and PBT as modifiers can improve
toughness in such composites. Also, TGMDA-DGEBA mixtures were observed to be
tougher than pure TGMDA as matrix materials for the composites.
Bascom and co-workers [59] observed that, thermoplastic modified thermoset composites
can be made to perform as well as thermoplastic composites for damage tolerance. They
were optimistic that use of thermoplastic composites will develop in time, as structures based
433
Handbook of Polymer Blends and Composites
on such systems would require costly equipment and hence toughened thermoset-based
composites will still, be favoured for aerospace use for some time to come.
10.8 Summary
1. CTBN modification is the most studied and popular approach to toughen epoxy
resin. Its main limitation is that, the necessary morphology may be achieved only by
very careful control of the mixing and curing conditions. Hence it will be difficult to
get highly reproducible results. CTBN toughening is always accompanied by a loss of
thermal and mechanical properties.
2. Rubbers less compatible with epoxy resins, but with reactive end groups, like silicone
oligomers can toughen epoxy resins but with less loss of stiffness and Tg.
3. For high performance epoxy resins like TGMDA, conventional liquid rubber
toughening is less effective but other techniques could be tried.
4. Core shell particles are easy to mix with epoxy resins and can give consistent results.
Toughening is not accompanied by a reduction in Tg nor is it dependent on processing/
curing conditions. Synthesising them is complex. They are suitable for high performance
epoxy resins.
6. For toughening high performance epoxy resins, approaches such as using a cure system
which can provide uniform crosslink density, core-shell particles, use of chain extension
with rigid side chains, etc., can be useful.
7. Other approaches like use of scrap rubber, solid rubbers, hollow spheres, combinations
of rubbers with hard particles, hard particles have also been successfully demonstrated
by many authors.
8. Polyester resin and its composites are more difficult to toughen for various reasons.
Still some practical solutions are possible.
9. The improvement in toughness in the resin is not fully transmitted into the composite
for various reasons. This can be solved in some cases by promoting resin-fibre
434
Rubber Toughened Epoxies/Thermosets
10.Toughening of epoxy resins by liquid and solid rubbers can be applied successfully
for improving the performance of adhesives.
Acknowledgements
The authors thank RAPRA Technology Ltd, for giving this assignment and for the
literature survey. Thanks are also due to Professor A.J. Kinloch of Imperial College of
Science, Technology and Medicine, London, UK for sending reprints of his papers, and
Professor S. Bandyopadhyay of University of New South Wales, Australia, for giving
information on papers published by him and his associates.
References
1. W.D. Bascom and D.L. Hunston in Rubber Toughened Plastics, Ed., C.K. Riew,
Advances in Chemistry Series No.222, ACS, Washington, DC, USA, 1989, 135.
2. R.S. Raghava in Reference Book for Composites Technology, Ed., S.M. Lee,
Technomic Publishing Company Inc., Lancaster, PA, USA, 1989, 77.
3. B.L. Burton and J.L. Bertram in Polymer Toughening, Ed., C.B. Arends, Marcel
Dekker, New York, NY, USA, 1996, 339.
4. S.J. Shaw in Rubber Toughened Engineering Plastics, Ed., A.A. Collyer, Chapman
and Hall, London, UK, 1994, 165.
5. B.J.P. Jansen, K.Y. Tamminga, H.E.H. Meijer and P.J. Lemstra, Polymer, 1999,
40, 5601.
6. Y. Huang, A.J. Kinloch, R.J. Bertsch and A.R. Siebert in Toughened Plastics-I
Science and Engineering, Eds., C.K. Riew and A.J. Kinloch, Advances in
Chemistry Series No.233, ACS, Washington, DC, USA, 1993, 189.
435
Handbook of Polymer Blends and Composites
10. S.C. Kim, M.B. Ko and W.H. Jo, Polymer, 1995, 36, 2189.
11. W.K. Lee, K.A. Hodd and W.W. Wright in Rubber Toughened Plastics, Ed., C.K.
Riew, Advances in Chemistry Series No.222, ACS, Washington, DC, USA, 1989,
263.
12. S. Sankaran and M. Chanda, Journal of Applied Polymer Science, 1990, 39, 1459.
13. S. Sankaran and M. Chanda, Journal of Applied Polymer Science, 1990, 39, 1635.
14. A.H. Rezaifard, K.A. Hodd and J.M. Barton in Toughened Plastics-I Science and
Engineering, Eds., C.K. Riew and A.J. Kinloch, Advances in Chemistry Series
No.233, ACS, Washington, DC, USA, 1993, 381.
16. E.M. Yorkgitis, C. Tran, N.S. Eiss, Jr., T.Y. Hu, G.C. Wilkes and J.E. McGrath in
Rubber-modified Thermoset Resins, Eds., C.K. Riew and J.K. Gillham, ACS
Advances in Chemistry Series No.208, ACS, Washington, DC, USA, 1984, 137.
18. T. Takahashi, N. Nakajima and N. Saito in Rubber Toughened Plastics, Ed., C.K.
Riew, Advances in Chemistry Series No.222, ACS, Washington, DC, USA, 1989, 243.
19. L-L. Lin, T.H. Ho and C.S. Wang, Polymer, 1997, 38, 8, 1997.
20. J.Y. Shieh, T.H. Ho and C.S. Wang, Angewandte Makromolekulare Chemie,
1997, 245, 125.
436
Rubber Toughened Epoxies/Thermosets
24. A.K. Banthia, P.N. Chaturvedi, V. Jha and V.N.S. Pendyala in Rubber Toughened
Plastics, Ed., C.K. Riew, Advances in Chemistry Series No.222, ACS,
Washington, DC, USA, 1989, 343.
25. F.N. Kelly and K.N. Ptak, Polymers and Composites - Recent Trends, Oxford and
IBH Publishing Co., New Delhi, India, 1989, 219.
27. D. Ratna and A.K. Banthia, Journal of Adhesion Science and Technology, 2000,
14, 1, 15.
29. R. Bagheri, M.A. Williams and R.A. Pearson, Polymer Engineering and Science,
1997, 37, 2, 245.
30. M.J. Boynton and A. Lee, Journal of Applied Polymer Science, 1997, 66, 2, 271.
32. P.J. Pearce, C.E.M. Morris and B.C. Ennis, Polymer, 1996, 37, 1137.
33. H-J. Sue, E.I. Garcia-Meitin, D.M. Pickelman and P.C. Yang, in Toughened
Plastics-I Science and Engineering, Eds., C.K. Riew and A.J. Kinloch, Advances
in Chemistry Series No.233, ACS, Washington, DC, USA, 1993, 259.
34. C.K. Riew, A.R. Siebert, R.W. Smith, M. Fernando and A.J. Kinloch in
Toughened Plastics II, Novel Approaches in Science and Engineering, Eds., C.K.
Riew and A.J. Kinloch, ACS Advances in Chemistry Series No.252, ACS,
Washington, DC, USA, 1996, 33.
35. H-J. Sue, E.I. Garcia-Meitin and D.M. Pickelman in Polymer Toughening, Ed.,
C.B. Arends, Marcel Dekker Inc., NewYork, NY, USA, 1996, 131.
437
Handbook of Polymer Blends and Composites
37. H-J. Sue, P.M. Puckett and J.L. Bertram in ACS Polymeric Materials Science and
Engineering, Fall Meeting, 1998, 79, 216.
38. R.A. Pearson in Toughened Plastics-I, Science and Engineering, Eds., C.K. Riew
and A.J. Kinloch, Advances in Chemistry Series No.233, ACS, Washington, DC,
USA, 1993, 405.
39. J. Kim and R.E. Robertson in Toughened Plastics-I, Science and Engineering,
Eds., C.K. Riew and A.J. Kinloch, Advances in Chemistry Series No.233, ACS,
Washington, DC, USA, 1993, 427.
40. J.C. Hendrick, N.M. Patel and J.E. McGrath in Toughened Plastics-I, Science and
Engineering, Eds., C.K. Riew and A.J. Kinloch, Advances in Chemistry Series
No.233, ACS, Washington, DC, USA, 1993, 293.
41. L. Suspene, Y.S. Yang and J-P. Pascault in Toughened Plastics-I, Science and
Engineering, Eds., C.K. Riew and A.J. Kinloch, Advances in Chemistry Series
No.233, ACS, Washington, DC, USA, 1993, 163.
42. R.F. Grossman in Rubber Toughened Plastics, Ed., C.K. Riew, Advances in
Chemistry Series No.222, ACS, Washington, DC, USA, 1989, 415.
43. P. Ghosh and N.R. Bose, Journal of Applied Polymer Science, 1995, 58, 2177.
44. S.N. Tong and T.K. Wu in Rubber Toughened Plastics, Ed., C.K. Riew, Advances
in Chemistry Series No.222, ACS, Washington, DC, USA, 1989, 375.
47. M-S. Lin, C-C. Liu and C-T. Lee, Journal of Applied Polymer Science, 1999, 72,
4, 585.
48. R. Subramaniam and F.J. McGarry, Journal of Advanced Materials, 1996, 27, 2, 26.
49. A.R. Siebert, C.D. Guiley, A.J. Kinloch, M. Fernando and E.P.L. Heijnsbrock in
Toughened Plastics II, Eds., C.K. Riew and A.J. Kinloch, ACS Advances in
Chemistry Series No.252, ACS, Washington, DC, USA, 1996, 151.
438
Rubber Toughened Epoxies/Thermosets
52. J.M. Scott and D.C. Phillips, Journal of Materials Science, 1975, 10, 551.
53. J. Kim, C. Baillie, J. Poh and Y.W. Mai, Composites Science and Technology,
1992, 43, 283.
55. F.J. McGarry in Rubber Toughened Plastics, Ed., C.K. Riew, Advances in
Chemistry Series No.222, ACS, Washington, DC, USA, 1989, 173.
56. F.L. Barcia, B.G. Soares, M. Gorelova and J.A. Cid, Journal of Applied Polymer
Science, 1999, 74, 6, 1424.
57. I.M. Low, Y-W. Mai and S. Bandyopadhayay, Composites Science and
Technology, 1992, 43, 3.
58. K. Jang, W-J. Cho and C-S. Ha, Composites Science and Engineering, 1999, 59,
7, 995.
59. W.D. Bascom, S-Y. Gwein and D. Grande in Toughened Plastics-I, Science and
Engineering, Eds., C.K. Riew and A.J. Kinloch, Advances in Chemistry Series
No.233, ACS, Washington, DC, USA, 1993, 519.
439
Blends Containing Thermostable Heterocyclic Polymers
11
Blends Containing Thermostable
Heterocyclic Polymers
Maria Bruma and Ion Sava
High temperature polymer blends are being increasingly used in many applications, from
the automotive to the aircraft industry. They have also attracted much interest from the
academic world. In the recent years, temperatures in excess of 200 C for periods of
hundreds of hours and longer have become the stated requirements for some materials.
Particularly severe in this regard are demands from aircraft industry for engine components
where high service temperatures for long periods of time are often normal. One approach
to meeting these material requirements is to synthesise new polymers. Another method is
to tailor the materials properties is by blending two polymers with the aim of benefiting
from the positive features of both materials while attempting to eliminate the negative
features.
This chapter will describe blends of high temperature heterocyclic polymers with
traditional ones, or blends in which both components are heterocyclic polymers, and it
will present the advantages of producing such blends. The chapter will concentrate mostly
on the progress made in this field in the last five years, since some reviews have been
reported previously [1-4].
Usually, high temperature polymers are defined as materials which have a continuous
service temperature of above 175 C for tens of thousands of hours. Similarly, high
temperature polymer blends are defined by the same use temperature. In this chapter
polymer blends means physical mixtures of polymers, but not systems which react
chemically in the mixed state, as have been described in other chapters. The physical
state of mixing in polymer blends can be miscible, partially miscible or immiscible. It is
not always possible to define the exact scientific criteria ijolved with compatible blends.
However, the term compatible is used to describe blends which possess a desirable physical
property regardless of the actual state of mixing.
441
Handbook of Polymer Blends and Composites
glass transition temperature (Tg), provided that the difference between the two Tg of the
respective partners is higher than 20 C. It is generally assumed that a polymer blend
which displays a single Tg is mixed at the molecular level. If the two components of the
blend are miscible only within a certain composition range the blends are called partially
miscible. An immiscible blend contains two or more distinct phases and it is heterogeneous
in nature.
The reason for blending two polymers is to produce a material which has a property
tailored to a certain performance level. Since the resultant physical properties are
determined by the miscibility and phase behaviour it is very important to control the
phase structure in blend. Most high temperature heterocyclic polymers possess restricted
rotation in their backbone and therefore their Tg is quite high. Based on the rigid
conformation of such polymers it is expected that the criteria for miscibility will be more
severe than for random coil polymers. However, in the last few years several high
temperature miscible polymers have been discovered. The reason for miscibility observed
in such systems is the existence of functional groups in the polymer backbone, which can
interact with the functional groups of the other polymer in the mixture. A very important
aspect with miscible polymer blends is the temperature range of their miscibility. When
such materials are to be processed in the melt state, there must be a temperature window
between the Tg and the phase separation temperature, which should be large enough for
melt processing. If this processing window is too narrow, the blends will either phase
separate or have a viscosity that is too high for melt processability. Another category of
polymer blends contains immiscible systems. Two types of immiscible blends have been
described. The first type is constituted by molecular composites, which are generally
accepted as partially miscible blends but are not truly miscible at the molecular level.
Molecular composites are based on the mixing of a rigid rod polymer and a random coil
polymer. Most of the work on molecular composites has focused on delay of the phase
separation with a desirable morphology before complete immiscibility occurs.
The second type of immiscible blend involves thermotropic liquid crystal polymers (LCP)
blended with conventional thermoplastic polymers. The advantage is that the LCP behaves
as a processing aid, lowering the overall melt viscosity. The groups of LCP molecules,
called domains, slide past each other and the overall effect is lubrication of the polymer
melt and a subsequent lowering of the melt viscosity of the blend which means that the
blend can be processed at a lower temperature. The reduction of processing temperature
leads to lower energy consumption and less degradation of polymers that are sensitive to
high temperature. Another effect in blends of LCP with thermoplastic polymers is the
utilisation of LCP as reinforcement agent for flexible polymers, which gives improved
mechanical properties. However the use of LCP as reinforcement agent has a drawback,
namely the poor adhesion to the matrix.
442
Blends Containing Thermostable Heterocyclic Polymers
Polyimides (PI) were the first of the thermally stable heterocyclic polymers introduced in
the mid 1960s. The most common synthetic method for preparing PI consists of a two-
step reaction between a tetracarboxylic dianhydride and a diamine yielding a soluble
polyamic acid (I) which can be converted to polyimide (II) by thermal or chemical loss of
water [5-7] (see Figure 11.1).
Wholly aromatic polyimides do not behave as thermoplastics even though they have a
linear structure. The polymer backbone comprised a rigid, inherently stable, aromatic
phenylene and the imide ring imparted excellent thermooxidative properties, but at the
same time made these products extremely difficult to process because of their very high
and sometimes lack of melting point. In the beginning, researchers interest was
concentrated mainly upon thermal stability with little regard to cost, ease of fabrication
and end-use temperature. Lately, emphasis has shifted towards more practical, specific
applications which made PI in these days, among the most common commercial high
performance materials. Modified PI containing various flexibilising groups such as ether,
ester, amide, sulfone, sulfide, hexafluoroisopropylidene and others, have been studied
intensively to improve their processing properties such as solubility, flexibility, Tg, melt
viscosity, cost and ease of fabrication although with a little sacrifice of their thermal
stability [8-20]. Their outstanding properties include continuous service at temperatures
443
Handbook of Polymer Blends and Composites
above 200 C, wear resistance, low friction, good strength, toughness, thermal stability,
high dielectric strength and radiation resistance, and low outgassing. Polyimides are
available as precured films and fibres, curable enamels, adhesives, and resins for composite
materials. In addition, due to their stable backbone structure, and resistance to acid
hydrolysis and bacterial attack, PI are used as membranes for various processes.
Polyimides have been blended with conventional homopolymers and copolymers including
polytetrafluoroethylene (PTFE), polyvinylidenefluoride (PVF), epoxy and phenolic resins,
polyphenylenesulfide, polysulfone, polyesters, polycarbonates, polyamides, polyethers,
polyether-ketones, polyaniline, polyurethanes and with elastomeric materials such as
nitrile rubber; most of these blends will be discussed later. Also, PI have been blended
with other heterocyclic polymers such as polybenzimidazoles, polyquinoxalines, and
polyoxadiazoles. Blends have been also made by mixing PI of various structures. The PI
are either a major or minor component of the mixture. In either case the inherent properties
of the major blend component are amended by the addition of a minor component until
the properties of the blend have attained the optimum desired.
Aromatic PI are increasingly used in microelectronic devices because of their good dielectric
properties and high thermal stability. The requirements of polymer properties for more
advanced microelectronic applications are quite severe, so that they can not be usually
met by a single component polymer. Some of these requirements include high Tg, above
300 C, and low thermal expansivities along with mechanical toughness and good
adhesion properties. While PI with rigid rod-like conformation provide the required
thermo-mechanical properties, these materials have inherently poor adhesion
characteristics. One approach to obtain such a combination of desired properties is
blending a rod-like polymer with a flexible one which could provide enhanced adhesion
and increased toughness.
Blends of PI having various structures have been prepared mostly by mixing the solutions
of the two polyamic acids in N-methylpyrolidone (NMP) or dimethylacetamide (DMAc),
followed by casting into thin films, and drying and thermal imidisation at high temperature.
The imidisation degree of PI-PI blends was evaluated with the aid of
perylentetracarboxydiimide which shows unique fluorescence when molecularly dispersed
in polyamic acid/imide mixtures. Their fluorescence is efficiently quenched in the polyamic
acid films, due to amide groups functioning as a quencher, and increases abruptly when the
imidisation process proceeds. On the other hand in mixtures of polyamic acid PI based
444
Blends Containing Thermostable Heterocyclic Polymers
However, later studies showed that mixtures of polyamic acids can lead to the formation
of segmented blockpolymers via a transamidation reaction rather than molecularly mixed
blends of the two homopolyimides [24]. To prevent the transamidation reaction and
obtain better miscibility of the blends, the ethyl or methyl esters of polyamic acids were
used [25].
445
Handbook of Polymer Blends and Composites
Figure 11.3 Chemical structure of a polyimide based on PMDA and ODA (V)
Other blends are based on polyimides prepared from ODA and various dianhydrides
such as 3,3,4,4-diphenyloxytetracarboxylic dianhydride, 3,3,4,4-
benzophenonetetracarboxylic dianhydride or 3,3,4,4-diphenyltetracarboxylic
dianhydride, and are characterised by high hardness and high softening temperature as
well as improved frictional behaviour [27].
The use of silane coupling agents with blends made from diphenylsulfone tetracarboxylic
dianhydride based polyimides provided excellent adhesives for PI films [30]. Blends prepared
446
Blends Containing Thermostable Heterocyclic Polymers
by melt mixing followed by injection moulding of such PI showed no change after treatment
with 35% H2SO4 for ten days and distortion of 7.2% at 250 C for 5 hours [31].
Because of their high inherent thermal stability and excellent electrical properties, fluoro-
containing aromatic PI have excellent potential for use in aerospace composites, aircraft
wire and cable, adhesives, and flexible wire boards [32, 33]. The PI containing
hexafluoroisopropylidene (6F) groups gave very interesting blends, fully miscible, when
they were based on the same 6F-containing dianhydride and the diamine was 4,4-6F-
dianiline or 3,3-6F-dianiline, (VIII) and (IX), respectively, (see Figure 11.5). Their
miscibility arises from the fact that their dianhydride parts have identical bond angle,
rigidity and length and their diamine segments also have 6F groups [34, 35].
A rod-like PI, prepared by the reaction of PMDA with m-phenylene diamine, was blended
with a flexible polyimide based on 6F dianhydride and 2,2-bis(p-aminophenoxy-p-
phenylene)-hexafluoropropane. The blends were made by mixing the solutions of the
corresponding polyamidic acids (X) and (XI) in NMP, one of the component being an
ethylester, followed by solvent evaporation and thermal imidisation [24] (see Figure 11.6).
Mixtures of polyamic acids were made over a large range of composition at a total
concentration of solids of 15%-20% and showed complete transparency. The films were
447
Handbook of Polymer Blends and Composites
prepared by spin-casting polyamic acid blends followed by drying and imidisation at 400 C.
The resulting films have a thickness in the range of 20-40 m. The molecular weight of the
polymer has a significant influence on the miscibility. Mixtures containing up to 15% low
molecular weight 6F polyimide remained transparent and this was an indication of polymer
miscibility. In any composition a phase-separation took place during the drying of the
polyamic acid mixtures, when the solid content reached a value of approximately 65%.
This morphology is then preserved throughout the thermal imidisation cycle at 400 C.
The particle size of the dispersed phase, approximately 1 m or smaller, is determined by
molecular weight (MW) and by the processing conditions. The mixtures containing low
molecular weight 6F polyimide (MW = 10,000) gave more homogeneous films. The mixtures
containing up to 25% low molecular weight 6F polyimide gave transparent films. The Tg
of the neat 6F polyimide was found in all the mixtures regardless of their compositions,
indicating complete demixing of rod-like and flexible polyimides. The mixture containing
rod-like polyimide as a major component exhibited excellent dimensional stability up to
500 C and low coefficient of thermal expansion comparable to the rigid-rod polyimide
itself, and a lower dielectric constant of less than 3.0. The surface properties and adhesion
strength of the mixtures are dominated by the 6F polyimide due to the segregation of this
component to the surface region resulting in excellent self-adhesion properties.
A polyimide structure (XII) that is utilised in a large variety of blends is based on regular
repeating units of ether and imide linkages and is prepared by the reaction of 2,2- bis [4-
(3,4-dicarboxyphenoxy)phenyl] propane dianhydride with aromatic diamines [36] (see
Figure 11.7).
448
Blends Containing Thermostable Heterocyclic Polymers
The dianhydride structure used in (XII) is the basis of a series of high performance
engineering thermoplastics known under the trade name of ULTEM (General Electric).
These polymers exhibit high heat resistance, excellent mechanical properties, inherent
flame resistance and outstanding electrical properties. This is why this specific
polyetherimide (PEI) structure has been utilised in blends with other polymers.
In the blends of PEI with polyimides based on traditional aromatic dianhydrides such as
PMDA, the thermal stability was dependent on the PEI content while the -relaxation
occured at a slower rate and was broader and more cooperative in some blends compared
to the neat polymers. Physical ageing of the blends in the glass state revealed changes in
the relaxation rate. The dynamic-mechanical ageing behaviour was attributed to an
increase in density on blending which decreased the free unoccupied volume and increased
the constraints of molecular mobility. The adhesive properties and bond strength were
found to vary with blend composition and test temperature [37]. Other blends based on
PEI were made with polyamic-esters precursors of PI prepared from
diphenyltetracarboxylic dianhydride and p-phenylene diamine, by dissolving them in
DMAc and subsequently coagulating them in methanol. After thermal imidisation the
blend showed a single Tg in both differential scanning calorimetry (DSC) and dynamic
mechanical analyses (DMA) and the Tg increased with increasing PI content. These Tg
values are reproducible in repeated heating cycles suggesting the true miscibility of the
blend [38]. To improve the processability in thermoplastic PI, a method named the reactive
plasticiser approach, was developed by using a small amount (5%-15% mol) of a less
activated weak nucleophilic diamine comonomer as a reactive plasticiser to obtain resins
which possess relatively low viscosity at low temperature and can be readily processed
through autoclave cycle at low pressures. During a high temperature treatment the reactive
plasticisers help the formation of high molecular weight polyimides and the desired
properties are achieved. The most effective plasticisers are aromatic diamines such as
2,6-diaminopyridine. Such a reaction took place very efficiently in mixtures of PEI with
a PI based on 1,4-bis(p-aminophenoxy) benzene [39]. Blends prepared from two different
449
Handbook of Polymer Blends and Composites
PEI showed full miscibility when the synthesis of the two polymers was performed in
situ in a melt of benzoic acid. Such blends showed a single Tg which took an intermediate
position between the corresponding Tg of each of the two polymers. On the contrary, for
the physical blend of the two PEI, the system showed a well-pronounced two-phase
character with two Tg. The observed difference in the morphology is due to the fact that
the in situ synthesis leads to the formation of some quantity of a block copolymer which
functions as a compatibilising agent for the two PEI homopolymers [40]. Blends prepared
from PEI and polyamide-imides based on trimellitic anhydride, in dimethylformamide
(DMF) solution, were used to produce membranes having high water permeability:
1.5 cm3/min-cm2 by applying 1 kg/cm2 pressure [41]. The thermoplasticity and flexibility
of electrical insulating varnishes based on PI are increased by addition of unsaturated
PEI based on maleic anhydride, trimellitic anhydride, diaminodiphenyl methane and
ethylene glycol [42].
Blends of polyimides and PEEK were prepared and they were miscible over the entire
composition range as the blends of different compositions exhibited a single Tg. The
crystallisation of PEEK was hindered by polyimide, at which the dominant morphology
was interlamellar [44]. Also, blends of PEEK with polyetherimide have been studied.
Polyetherimide is fully miscible with PEEK in the amorphous state but it is rejected from
PEEK crystals during crystallisation. For the blends with a low content of PEEK (up to
450
Blends Containing Thermostable Heterocyclic Polymers
40%) a semicrystalline structure with neatly resolved sub-spherulitic details was shown
by electron microscopy. At the finest scale, it was possible to see the stacks of lamellae as
well as singular lamellae separated by zones of almost pure PEI. When the polyimide
content is low, for example 40%, it becomes impossible to discern lamellar features in
the spherulites, which clearly indicates that a large PEEK dilution is required in the
blends to get lamellar structure, because these conditions ensure a sufficient openness of
the spherulites. The values of Tg for pure polymers and blends are: PEI 218.5 C; PEEK/
PI 70/30 161 C; PEEK/PI 60/40 167.7 C and PEEK 143 C. Upon crystallisation, the
Tg of PEEK and blends PEEK/PI was found to increase significantly [43, 45].
The investigation of PEI/PEEK blends, by both DMA and dielectric behaviour, showed
an -relaxation process which was dependent on composition and the temperature of
which increased with increasing the PEI content. The low temperature relaxation
behaviour of PEI and the dielectric relaxation time in the Tg region was found to be
affected by the addition of PEEK. The variation of loss and storage permitivities showed
that the asymmetric broadening was strongly dependent on the blend composition,
indicating a high degree of interaction between the components [46].
Blends of PEI with PEEK showed miscibility over the full range of compositions, with a
single Tg varying almost linearly with composition between the two homopolymers.
Enthalpic relaxation was a useful technique for studying the degree of compatibility of
the system. Physical ageing was observed to embrittle the blends and there was a close
correlation between the extent of enthalpy ageing and the change in mechanical and
impact behaviour. The yield stress increased and the elongation to break decreased
progressively with relaxation, in addition to a reduction in impact strength [47]. The
crystallinity and mechanical properties of blends with different amounts of semicrystalline
PEEK and amorphous PEI, prepared by melt mixing were studied by DSC to find the
miscibility between the two components and to find proper conditions for the maximum
development of crystallinity [48]. The sliding frictional behaviour of such blends follow a
similar trend to that of the neat PEEK but the overall friction is higher for blends. There
is a significant reduction in the failure potential of PEEK when PEI is incorporated. The
potential for catastrophic thermally induced contact failure in the PEEK/PEI blend is less
than that observed for unblended PEEK [49]. The dynamic-mechanical behaviour of PEEK/
PEI blends was studied over a wide temperature range incorporating the -transition and
sub-Tg transition. The segmental motions responsible for the -process were significantly
slowed by blending and were dominated by PEI. Two sub-Tg transitions in PEI and the
blends and one in PEEK were detected. The -transition in PEI which occurred at around
100 C was strongly affected by the presence of PEEK. The -transition of the blend shifts
its temperature position with composition in a linear manner between the -transition of
the PEEK (69 C at 1 Hz) and the -transition in PEI (81 C at 1 Hz) [50]. The study of
mechanical performance of an annealed PEEK/PEI blend prepared directly during injection
451
Handbook of Polymer Blends and Composites
The deformation behaviour of thin films prepared from PEEK/PEI blends was studied
over a wide temperature range by optical microscopy and tunnelling electron microscopy
which showed that all the materials had localised shear deformation at temperatures
well below Tg. In pure PEI and in blends with up to 60% PEEK content, a transition
from shear deformation to disentanglement crazing occured as the temperature was raised.
This transition was absent in PEEK which was deformed by shear over the whole
temperature range, and similar behaviour was found for blends containing 80% PEEK.
That means that at high PEEK content disentaglement crazing is supressed by strain-
induced crystallisation and some evidence for crystallisation order in the deformed region
of initially amorphous PEEK thin films was obtained by electron diffraction. The thin
film deformation of the blend was also consistent with their bulk deformation behaviour,
a high temperature ductile-brittle transition being observed at low PEEK contents in
tensile tests [57].
From blends of PEEK/PEI, high performance foams were prepared by using water as
blowing agent in the foaming process which took place at elevated temperatures in an
extruder. The properties of the foams depend of the composition of the blend mixing,
time and the temperature of the process [58].
Blends were made from PEEK with a polyesterimide prepared from N,N-hexamethylene-
bis(trimellitimide), 4,4-dihydroxybenzophenone and p-hydroxybenzoic acid. The melt
viscosity of such blends was significantly lowered which made them easily processable
452
Blends Containing Thermostable Heterocyclic Polymers
and mechanical properties were improved. When the content of polyesterimide in the
blend was above 50%, a phase separation occured at 340 C consisting of PEEK islands
in a polyesterimide continuous phase. Optical textures observed at various temperatures
support the phase separation phenomenon [59].
Sulfonated PEEK were prepared by sulfonating PEEK with 95% H2SO4 and were used in
blends with PEI, followed by solution casting into films. A single Tg was observed by DSC
in such blends which indicates phase miscibility [60]. The influence of doping with HCl
and H3PO4 on the properties of sulfonated PEEK/PEI was studied in membranes made
from such blends. Blending with PEI first results in an increase and then in a decrease in
membrane swelling at concentrations higher then 5%. The electrical conductivity of blend
membranes follows the same trend. Doping with the acids enhanced conductivity several
fold and the effect of doping with HCl is more significant. PEI forms spherical particles
dispersed in the PEEK matrix and at the same time partially dissolves the swelling of the
matrix at higher PEI concentrations. The increase in the membrane capacity to absorb
water at low PEI content is due to the formation of new water absorption sites along the
interface between the dispersed particles and the matrix [61].
453
Handbook of Polymer Blends and Composites
Blend films were cast on glass plates and heated first at 80 C to partially evaporate the
solvent, then they were stretched and again heated at 300 C. Such films showed high
elasticity, strength and good productivity [62]. Very good mechanical characteristics and
heat resistance resulted when the polyamic acid was prepared from a mixture of two
dianhydrides, such as PMDA and benzophenonetetracarboxylic dianhydride (BTDA),
with a mixture of two aromatic diamines, such as ODA and 4,4-diaminodiphenylmethane,
in m-cresol. The blend with PA showed a strong adhesion to copper foil, had a thermal
decomposition above 450 C, peel strength of 2.2 kg/cm, and no peeling in solder heat
test at 260 C for 10 seconds [63].
Blends of polyamic acid based on PMDA and ODA with aromatic polyamide Kevlar
(Monsanto), poly(p-phenylene-terephthalamide) (XVI), were moulded to give sheets with
low thermal expansion, flexural modulus of 300 kg/mm2 and Izod impact strength of
8.7 kg cm/cm [64] (see Figure 11.10).
Aliphatic polyamides, such as Nylon 6, were also blended with aromatic PI based on
dianhydrides, such as PMDA or BTDA, and diaminodiphenylether to give materials
with good chemical stability. Thus, the stability of the blend films in concentrated H2SO4
and in 40% aqueous NaOH are equivalent to that of the control film and slightly better
than commercially available Kapton PI film (DuPont de Nemours). The weight gains of
the blend films in solvents like CH2Cl2, methylethylketone and water are dependent on
the nature of the solvent and its interaction with the blend, but the BTDA based blend
films showed better chemical resistance [65].
Polyetherimides were also blended with aromatic or aliphatic PA [66, 67]. In blends with
aliphatic PA, such as Nylon 66, DSC analysis showed that crystallisation in such blends
was retarded compared to that of pure Nylon 66 by incorporation of PEI with high Tg.
The lowest growth rate of the spherulites was observed in the blends containing 10% or
15% fraction of PEI. The Tg increased with the content of PEI in the blends.
454
Blends Containing Thermostable Heterocyclic Polymers
Ternary blends of PEI, PA and sulfonated PEEK were prepared and used in the manufacture
of ionomer membranes for electrodialysis. DSC traces of the blend membranes showed
only one Tg. In the blend membranes swelling was reduced by specific interactions due
to hydrogen bonds between PA and PEI. The acid-base interactions also led to a decrease
of ionic conductivity by partial blocking of SO3 groups for cation transport. The acid-
base blends showed better ion permoselectivities, even at high electrolyte concentrations,
and thus better performance in electrodialysis. The thermal stability of the blends was
very good and even better than that of pure PEEK-SO3H. The performance of these
ionomer blend membranes is better than the performance of binary blend PEEK-PA,
especially in the electrodialysis experiments with higher NaCl concentrations. This is
mainly due to the lower swelling and thus better ion permoselectivity of the acid-base
(ionomer) blend membrane, compared with the PEEK-polyamide blend [69].
Polyetherimides have been the most used PI-type polymers in blends with polyesters.
Binary blends of PEI with PET were miscible in the melt, but showed simultaneous
crystallisation and liquid-liquid demixing below the melting point. Small-angle X-ray
scattering showed that the crystal thickness of PET was not perturbed upon blending
with PEI. A larger amorphous layer thickness in the blend was identified, showing that
some PEI was incorporated inside the interlamellar regions after crystallisation. Despite
the swelling of the amorphous layer, the amorphous layer thickness was relatively
independent of the blend composition. Significant extralamellar placement of PEI occurred
455
Handbook of Polymer Blends and Composites
when the PEI composition was higher than 20%. This morphology structure is interpreted
in terms of simultaneous occurrence of liquid-liquid demixing and crystallisation [73].
The melting behaviour of PEI/PET blends was investigated by DSC which showed the
multiple melting endotherms of PET. The study of the effects of blend composition,
heating rate and crystallisation duration on the multiple melting of the blend shows that
the extent of recrystallisation during DSC heating decreased with increasing of PEI
composition and crystallisation duration [74].
Blends of PEI with polyarylates were miscible when 90/10 and 80/20 weight ratios of the
two components were used. All other compositions were biphasic, one phase being almost
pure polyarylate and the other being PEI containing a fairly constant polyarylate component
of roughly 25%. This phase behaviour agreed with both the observed transparency and
the fracture observed by scanning electron microscopy (SEM). The mechanical properties
of the blends as a function of composition showed values close to additivity or even enhanced
with an unexpected synergism in ductility [75]. Blends of aromatic polyesters with polyimides
based on PMDA showed remarkable improved mouldability and crystallinity, and they
were free from the bleeding of mouldings and excellent in weathering and delamination
resistance. Such a test piece showed a volume change smaller than 0.1% in 75% hot water
[76]. Other PI/polyester blends were prepared by mixing the polyamic acids with a liquid-
crystalline aromatic polyester followed by injection moulding at 400 C. Such a test piece
showed tensile strength of 1349 kg/cm2, elongation of 4.2% and Notched Izod strength of
13.87 kg.cm/cm [77]. In another procedure, polyester and PI are blended in the presence of
low-molecular weight imide compounds having very low melting point for example an
imide obtained from 1,2-cyclohexane dicaboxylic anhydride and butyl amine was used as
a solvent for a blend of a PI based on PMDA and a polyester such as polyethylene-2,6-
naphthalene dicarboxylate. The blend was prepared at high temperatures (300 C) and
then the solvent was distilled off and the resulting blend composition had a Tg of 122 C
and a melting point of 267 C [78].
Polyester/PI blends with high adhesivity were prepared from PI, based on tetracarboxylic
dianhydride and diisocyanate, and polyesters based on adipic acid and hexandiol, and
they were used as adhesive layers or sealing materials for lead frames in semiconductor
devices [79]. Blend films with high mechanical strength were manufactured by mixing a
PI (XVII) having a Tg below 250 C prepared from butantetracarboxylic acid and m-
xylylendiamine with PET followed by extrusion and drawing to give biaxially oriented
films with tensile moduli of 8.8 GPa and 7.4 GPa, in the machine and transverse direction,
respectively [80] (see Figure 11.11).
Blends of polyesters with PI containing siloxane units which exhibited good antisoiling
properties, were used to cover metal plates followed by baking at 150-300 C [81].
456
Blends Containing Thermostable Heterocyclic Polymers
A blend was prepared from a PI (XVIII) based on cyclobutane dianhydride and 2,2-
bis(p-aminophenoxyphenyl) propane and polyvinylcinnamate (XIX) (see Figure 11.12).
Microscopic and thermal analysis revealed the miscibility of the two polymers in the thin
layers. From UV spectroscopic investigations the existence of photo-dimerisation and
the development of structural dichroism in the blend layers were confirmed. The photo-
dimerisation reaction of the cinnamate was not significantly influenced by addition of
PI, but the blend alignment layers provided uniform liquid-crystalline alignment, and
better thermal stability of liquid-crystalline alignment compared to the pure
polyvinylcinnamate alignment layer. The increase of the thermal imidisation temperature
enhanced the thermal stability of the liquid-crystalline alignment, which indicates that
457
Handbook of Polymer Blends and Composites
Gas transport membranes were prepared from blends based on PEI with an aromatic
polyester and CO2 permeation and sorption capacity were compared with PEI films at
different pressures and temperatures. In all cases permeability, diffusion and sorption
coefficients decreased when the amount of polyester increased. The results were consistent
with a polyester structure in the blend intermediate between an interfibrilar and a laminar
morphology, according to electron microscopy evidence [86].
Blends were also prepared from PEI with polyethylene naphthalate (PEN) [87]. PEN is a
relatively new polymer with good thermal, mechanical and barrier characteristics [88,
89]. PEN exhibits neck formation upon stretching from the amorphous state at
temperatures between Tg and cold crystallisation. To enhance the optical and mechanical
properties, blends were made with a PEI in a composition range varying from 100/0 to
70/30. PEN and PEI exhibit melt miscibility. Blends were melt casted into sheets or films,
which were further biaxially stretched in the full compositional range. The increase of
PEI fraction causes an increase of Tg while slightly decreasing the melting point. This is
a result of increased inter-chain friction between the PEN chains in the presence of bulky
PEI chains which disrupt the cooperative alignment of naphthalene groups parallel to
the surface of the films. This was found to widen the processing window where films of
458
Blends Containing Thermostable Heterocyclic Polymers
uniform thickness can be obtained. The DSC studies of the amorphous blends show an
increase of Tg with increasing PEI fraction, and a breadth of the Tg interval as well. This
suggests the presence of a certain level of microphase segregation in these blends. The
crystallisability of these blends is also influenced by composition as judged by the location
of a cold crystallisation peak. The increased fraction of stiff and bulky PEI chains shifts
the position of these peaks to higher temperatutres. This is mainly a result of dilution
effect combined with the stiffening of the environment of crystallisable PEN chains at
higher concentrations of PEI whose Tg (218 C) is substantially higher than that of PEN
(Tg = 120 C). Below 30% weight PEI concentration the crystallisability is substantially
diminished. To preserve the crystallisability of the blends the PEI concentration should
not exceed 20% weight. The biaxial stretching of blend films was preformed at 20 C
above their Tg. When PEN films were stretched to low stretch ratios, neck regions appeared
through the part. At high stretching ratios the films became uniform again after the
disappearance of the necks. In 20% PEI blend films the cold crystallisation peak occurs
immediately above the Tg and from the area below it can be seen that there is a larger
percentage of oriented uncrystallised PEN chains in the blends whose crystallisation was
hindered by the presence of PEI. In these films the melting endotherm also becomes
broader. As for optical properties, the refractive indices increase with increased proportion
of PEI and the values measured in all three directions are quite close to each other indicating
that these films are nearly optically isotropic. When the films are stretched significantly,
optical anisotropy rapidly develops. The refractive indices in the normal direction decrease
with the stretching ratio. The addition of PEI to PEN, for the same stretching ratio,
shows that the refractive indices in the normal direction increases with the addition of
PEI. This increase is partially caused by the increased proportion of higher refractive
indice in the blend, as a result of the disruption of preferential tendency of flat naphthalene
group parallel to the surface, which was also, augmented with the data of X-ray analyses.
Biaxially oriented PEN films show bimodal orientation, in the machine direction and in
the transverse direction. Addition of a little PEI (10%) to PEN converts this bimodal
orientation to in-plane isotropy in the equal biaxially stretched films. This conversion is
the result of incresed interchain friction in the presence of stiff and bulky PEI chains that
hinder the preferential orientation of the naphthalene planes paralel to one another and
to the film surface. This orientation behaviour is also reflected in the mechanical properties.
While PEN films exhibit bimodal orientation and show significantly lower moduli, the
presence of PEI eliminates this minimum and the films exhibit a plane mechanical isotropy.
A blend of 90/10 PEN/PEI showed a higher elongation to break than PEN films in all
directions; PEN films exhibited very low elongation to break in the transverse direction.
Blends of PEI with PEN were also prepared by injection moulding, having a clear
apperreance, Rockwell hardness of 104 L and flexural modulus of 2.4 GPa [90]. In another
procedure the blends were prepared by melt polycondensation of naphthalenedicarboxylate
with ethylene glycol in the presence of PEI, showing a Tg of 163 C [91].
459
Handbook of Polymer Blends and Composites
Ternary blends were prepared from PEI and two polyesters, PET and PBT, showing a
rare case of thermodynamic miscibility in the amorphous state (quenched as well as
molten state) with a single glass transition temperature observed in DSC and DMA studies
[92]. Ternary blends of thermotropic liquid-crystalline polyesters with PEI and PEEK
prepared by melt mixing exhibited a single Tg in DSC analyses. Although individual
PEEK/PEI blends and the polyester undergo catastrophic tribologic failure at their
respective Tg, the ternary blend was able to maintain sliding at temperatures in excess of
the Tg of their constituent phases. This synergistic behaviour is explained by polyester
reinforcement and PEEK cold crystallisation. PEI exhibits miscibility with both PEEK
and polyester (PEN), these two ether polymers are known to be immiscible [93].
Binary blends of PEEK/PEN are immiscible but PEI acts as a solvent for both of them
and ternary blends containing more than 40% weight PEI become miscible. The blending
of these homopolymers improved the deformation behaviour of PEN films by eliminating
necking which was observed in uni and biaxially stretched PEN films. While PEN and
PEEK form separate crystal structure upon stretching and subsequent annealing, in blends
containing more than 40% weight PEI, PEN crystallisation is hindered. Although PEEK
can be crystallised at a concentration as low as 10 wt%, PEN remains amorphous [94].
Ternary blends prepared from a polyetherimide (XX), polyarylketone and PTFE had
good impact resistance and low kinematic coefficient [99] see Figure 11.13.
460
Blends Containing Thermostable Heterocyclic Polymers
Ternary blends of a polyimide, PTFE and a polysiloxane were used to prepare antistick
films for use with metals, ceramics, glass or enamelled surfaces. The adhesion of blend
films to the substrate was very much influenced by the particle size of PTFE [100].
Polyimides prepared from PMDA and ODA were also used in ternary blends with PTFE
and liquid crystalline polyesters. The blends were injection moulded to give materials
with a coefficient of friction 0.25-0.32 [101]. Other ternary blends with good mechanical
strength, sliding, and electric insulation properties were prepared by mixing polyimide
powder, PTFE and novolak, followed by injection-moulding. Such a moulded part showed
a bending strength of 110 MPa, charpy impact strength of 3.5 kJ/m2, electric resistance
of 1.2 x 1010 , and a dynamic friction coefficient of 0.2 [102].
Ternary blends prepared from a polyamidimide, PTFE and polysiloxane are the main
components in the bearings of slidable structures such as earthquake proof footings of
buildings, where reduced friction and good heat resistance is required [103].
Blends of PEI with polysulfones in various proportions were prepared by solution mixing
followed by co-precipitation. Mechanical properties and morphology of the blends were
studied by using tensile tests and SEM. Tensile moduli and ultimate strength exhibit
positive deviations from simple additivity. The blends are not perfectly homogeneous,
with fine dispersions, but the interphase between the two phases are well-bonded. These
two polymers are partially miscible on the segment level [104]. The miscibility of
polyimide-polysulfone blends was studied by DSC, rheological and X-ray scattering.
The blends rich in polysulfone form a miscible blend when prepared by solution casting
from a common solvent. Heating to temperatures above the Tg leads to phase separation
into polysulfone rich domains and polyimide rich domains. Gas separation membranes
461
Handbook of Polymer Blends and Composites
made from such blends showed that helium permeability is controlled by the polyimide
component [105].
Blends of PEI and polyethersulfones were used in the preparation of hollow fibre
membranes. The blend was dissolved in a mixture of NMP and ethanol followed by
coagulation. The internal coagulant plays a very important role in this process and in
this particular case it includes water, methanol, ethanol, 2-propanol and mixtures of
these alcohols with water. Water is always used as the external coagulant [107].
462
Blends Containing Thermostable Heterocyclic Polymers
Blends were prepared from PEI with polyphenylene sulfide by melt mixing followed by
injection moulding to give products with improved flowability and good heat resistance
and tensile strength. The spiral flow was 92 mm at 350 C, melt index retention 70%
and tensile strength 530 kgf/cm2 [111].
Blends containing more than 80% wt PEI, less than 20% wt PC and up to 7% wt
polyacrylate have been prepared for improving the environmental stresscrack resistance
properties of polyetherimides. Such a blend was injection moulded to tensile bars which
were mounted on testing devices with induced stress of 27 MPa and strain 0.83% and
the bars showed no effect after immersion in trichloroethane for 15 minutes, compared
to rupture for specimen made from PEI alone [114].
Other blends have been prepared by mixing polyetherimide with PC and polysiloxane
and they exhibited high heat distortion temperature, improved room-temperature impact
properties and improved impact strength and ductility at low temperature [115].
463
Handbook of Polymer Blends and Composites
Polyurethanes (PU) have excellent abrasion resistance and properties of both rubber and
plastics. They are becoming more important as engineering materials. Conventional PU
exhibited small resistance to heat, for example the acceptable mechanical properties
(strength, modulus) disappear from about 80-90 C, and thermal degradation takes place
above 200 C [116]. Because of the poor heat resistance of pure PU their applications
were limited. One possible method to improve the thermal stability of PU is by blending
them with highly thermally stable heterocyclic polymers such as PI which have remarkable
heat resistance and superior mechanical and electrical properties [117, 118]. Such blends
were prepared by mixing a PU (XXIII) prepared by the reaction of a polyester polyol
with 2,4-tolylene isocyanate and end-capped with phenol with poliamidic acid, which is
a precursor of polyimide, prepared from PMDA and ODA end-capped with phthalic
anhydride (see Figure 11.15). The two polymer solutions in NMP were blended at room
temperature in various ratios and they were cast into films followed by thermal treatment
at various temperatures. With the increase of PU component, the films changed from
plastic to brittle and then to elastic. The elastomeric blend showed excellent mechanical
properties and moderate thermal stability. The elongation of film was above 300%.
TGA and DSC studies indicated that the thermal degradation was in the range of 250-
270 C which is significantly superior to PU itself. Films prepared from such blends
showed good solvent resistance. The tensile modulus of the blend increased with increase
of the imide component, but at the same time the elongation decreased greatly. At
polyimide/polyurethane (50/50) the film was brittle and the elongation became the
smallest. With further increase of polyimide/polyurethane ratio, elongation became larger
and the tensile properties of films were close to those of polyimide. The tensile properties
can be explained by the phase separation between PI and PU.
Bisimide oligomers (XXIV) obtained from phthalic anhydride and a diamine containing
polysiloxane groups were used in blends with a polyimide obtained from 1,3-bis(3-
464
Blends Containing Thermostable Heterocyclic Polymers
Conducting polymers such as polyaniline have been widely investigated for both academic
and industrial purposes over the last two decades. Their attractive electrical and optical
properties have enabled their use in many potential applications. However, their poor
mechanical properties and instability limit some industrial applications. Thus, considerable
465
Handbook of Polymer Blends and Composites
attention has been given to producing polyaniline/polyimide blends and blend films.
Sheets useful for electrodes in batteries, electromagnetic shields and others have been
prepared from a blend containing polyimide and polyaniline which had a modulus of
elasticity of 200 kg/mm2 and a volume resistivity of 107-1014 cm. The sheets were
prepared by casting the blend of polyamic acid with polyaniline, in NMP solution,
followed by heating up to 250-400 C [124, 125].
Also, blends of polyaniline and polyamic acid were processed into free standing, fully
dense films showing improved thermal stability relative to polyaniline and they were
used in gas separation membranes. These membranes showed greater gas selectivity and
an increase in permeability. They were also used for the separation of liquid mixtures.
The selectivity of such membranes for water - acetic acid mixtures was considerably
improved compared to that of neat polyaniline and was comparable to that of HCl-
doped polyaniline. This means that the polyamic acid acts as a polymeric dopant.
Experiments with membranes based on blends of polyaniline with polyimide obtained
from BTDA ODA had permanent flux for acetic acid/water that was intermediate
between polyaniline and polyimide membranes and selectivity that was intermediate
between doped and non-doped polyaniline [126, 127].
466
Blends Containing Thermostable Heterocyclic Polymers
Blend films of polyimide/polyaniline doped with camphor sulfonic acid had a good stability
of conduction at high temperature [129].
Polyimide/polyaniline (emeraldine salt) blend was used as a hole transport layer in light
emitting diodes (LED). The emissive layer was a blend containing oligophenylene vinylene,
1,4-distyryl benzene and 2-(4-biphenylyl)-5-4-(tert-butylphenyl) 1,3,4-(oxadiazole). These
layers were sandwiched between indium and indium-tin oxide (ITO) electrodes. To
increase the electron injection into emissive materials a thin magnesium-layer was inserted
between the indium and the polymer blend. The electroluminescence spectra of these
LED showed noticeable enhancement of the oscillator strength of the oligophenylene
vinylene peak at 2.76 eV. This implies improved quantum efficiency of this blue LED
resulting from the excitonic migration from oxadiazole to oligophenylene vinylene. The
electroluminescence device with polyimide/polyaniline blend as host polymer displayed
increasing device performance, lowering the turning point in intensity-voltage
characteristics, compared to that of LED without polyaniline. Under normal illumination
conditions, the devices with polyaniline showed visible blue-violet colour at room
temperature after applying a bias exceeding 8 V [130].
Blends of polyimides, for example based on PMDA and ODA, with elastomers such as
nitrile rubber, in which PI is the major component or the minor one have been also
reported in the literature [133]. A blend made from nitrile rubber as a major component
and a polyether-polyester imide was moulded into a transparent test piece with a resistivity
467
Handbook of Polymer Blends and Composites
of 2 x 1013 /cm2 initially and after washing with detergents and having good mechanical
strength [134].
This miscibility results from strong intermolecular interactions due to H-bonding between
the NH of PBI and the CO of the PI. The strength of intermolecular interaction between
PBI and various polyimides was measured by DSC, thermomechanical analysis (TMA)
and FTIR and it was found to be higher in BTDA-based blends than DSDA systems. This
difference was seen to arise from electron affinity in the carbonyl spacer of benzophenone
in BTDA and the SO2 spacer of diphenylsulfone in DSDA [135-139]. When the polar
linkages CO, SO2 were replaced by non-polar groups such as the hexafluoroisopropylidene
group, the miscibility of PBI/PI blends was significantly reduced [140].
The study of films prepared from such solution blends showed that the solvent (DMAc)
retained by the films significantly influenced their thermal stability. Vapour absorption
468
Blends Containing Thermostable Heterocyclic Polymers
and FTIR studies showed that solvents enter into strong interaction with the two polymers,
especially with PBI component and therefore the thorough removal of solvents is an
essential condition for the utilisation of films prepared from these blends [141, 142].
When PBI was blended with a polyimide containing siloxane groups the resulting blends
could be melt processed [143].
469
Handbook of Polymer Blends and Composites
to be within the PI layer. In the case of PAEBI/PAA blend, a relatively higher peel strength
was found compared to the PAEBI/PAE blend, in the adhesion joint. Therefore, higher
miscibility resulted in higher adhesion strength [144, 145].
The blends of polybenzimidazoles with PEI were widely investigated in the last decades
because the miscibility of the two polymers occurs over a wide composition range, and
thus the blends display single Tg which is intermediate. Thus, polybenzimidazole was
mixed with a polyetherimide prepared by the reaction of 2,2-bis (4-(3,4-
dicarboxyphenoxy)phenyl) propane tetracarboxylic dianhydride with m-
phenylenediamine. The latter is a commercial polyetherimide named ULTEM
1000 (General Electric). The two polymers were dissolved in DMAc at a concentration
of 2% each. The solutions were mixed and cast into films. Various mixtures were prepared
in which the composition range was from 0 to 1 mass fraction of PI. Each blend
composition gave optically clear films. FTIR studies showed that the strength of interaction
between the two polymers is influenced by the traces of moisture remaining in the blend
after washing and drying the films. There is a relatively weak interaction in the absence
of moisture which can be explained by the stiffness of the chains composed of aromatic
rings as well as self association of the component: electron-acceptor diimide groups and
the electron-donor aromatic units of the polyimide forms charge transfer complexes
resulting in aggregation of the solid polymer. Also, in polybenzimidazole the self-
association is indicated by its high Tg. The blends exhibited single intermediate Tg. The
water absorption by the blend is dependent on the composition. At high polybenzimidazole
content the water uptake is diffusion controlled while at higher PI contents the absorption
mechanism is changed and becomes a two-step process: it starts with a relatively high
diffusion rate followed by a relaxation controlled mechanism. This study shows that the
water absorption is increasing by increasing the PI content in the blend and it also showed
that the strength of interaction between the polybenzimidazole and PI is lower than that
of polybenzimidazole/water. However, in spite of the measured single intermediate Tg
and evidence of intermolecular hydrogen bonding, polybenzimidazole and PI do not
form thermodynamically homogeneous blends, in other words there is only partial
miscibility between the two polymers imhe blend [146].
470
Blends Containing Thermostable Heterocyclic Polymers
Blend films have been cast from such solutions and they showed single tan relaxation
in DMA. The intermolecular interaction due to NH in imidazole and CO group in the
imide ring played an important role in the miscibility of these two polymers. A blend
film PNI/PI having the composition 80/20% wt has the highest tensile strength and
modulus, in a series of blends having various compositions [149]. Blends were also
prepared from polynaphtimidazole, obtained by the reaction of tetraaminodiphenylether
with isophthaloyl bis(naphthalic anhydride), with polyetherketone [150].
PBI was used in blends with polyvinylacetate-vinyl alcohol copolymers. DSC, DMA,
FTIR spectroscopy, SEM and optical measurement show that as the vinyl alcohol content
of the copolymer component increases, the level of dispersion of the two components of
the blends improves. The blends are not miscible at the molecular level, but the levels of
dispersion in blends with polyvinylacetate/poly vinyl alcohol compositions of 56 and
471
Handbook of Polymer Blends and Composites
73 mol% vinyl alcohol were good enough to allow the formation of optically clear films.
Enhanced storage modulus for certain blends, which were larger than that of PBI alone,
was also obtained. Evidence for significant hydrogen bonding between the blend
component was found from FTIR measurements, but none of the blends exhibited a
single Tg. Consequently it is estimated that the phase domain sizes for the well-dispersed
blends lie in the range of 20-500 nm, which are larger than expected from single phase
miscible blends [151].
Attempts were made to prepare blends of PPQ with PEI, but a compatibility study by a
light scattering technique illustrates a significant self-agglomeration tendency of the PPQ
resulting in the incompatibility of these polymers. The same self-agglomeration
phenomenon was found in blends of PPQ with polymethylmethacrylate containing either
electron-donor or electron-acceptor groups. Optical light- and atom force-microscopy
(AFM) investigations reveal a better dispersion in these blends, but the self-agglomeration
tendency is so intense that even the use of electron-donor components does not produce
compatible blends [153].
472
Blends Containing Thermostable Heterocyclic Polymers
quenched due to the charge transfer between the two polymers. A new and faint orange-
yellow photoluminescence emission which has photonic energy consistent with the energy
difference of the band of polycarbazole and the * band of polyquinoxaline has been
observed. Light emitting electrochemical cells fabricated from these polymer blends show
strong current injection and bright electroluminescence and new emission colour, which
is due to the interpolymer radiative recombination of the electrons of the n-type polymers
and holes from the p-type polymers. Such an independent p-doping of polycarbazole
and n-doping of polyquinoxaline in the blend and interpolymer radiative recombination
provide an interesting way of generation of new emission colour in the light emitting
electrochemical cells systems [154].
Other heterocyclic conductive polymers, such as polypyrrole, have also been used in
blends with insulating host polymers such as polycarbonate, polyamide. Several
spectroscopic and thermal analyses of free standing blend films show an interaction
between the host matrices and conducting polymer [155].
473
Handbook of Polymer Blends and Composites
polyamide, a phase separation appeared and the POD was spherically dispersed in the
polyamide, as shown by SEM. Also, blends of POD with polyoxymethylenes were prepared
by mixing the two powdered components, followed by melting, cooling, extruding or
moulding, to give products with good mechanical properties [159].
Acknowledgements
References
2. M. Jaffe, P. Chen, E.W. Choe, T.S. Chung and S. Makhija, Advanced Polymer
Science, 1994, 117, 297.
4. H. Tang, L. Dong and Z. Feng, Cailiao Yanjiu Xuebao, 1996, 10, 449.
474
Blends Containing Thermostable Heterocyclic Polymers
11. M. Bruma, I. Sava, F. Mercer, I. Negulescu, W. Daly, J. Fitch and P. Cassidy, High
Performance Polymers, 1995, 7, 4, 411.
13. M. Bruma, J. Fitch and P. Cassidy in Polymeric Materials Encyclopedia, Ed., J.C.
Salamone, CRC Press, Boca Raton, FL, USA, 1996, 4, 2456.
21. M. Hasegawa, J.I. Ishii and Y. Shindo, Journal of Polymer Science, Part B:
Polymer Physics, 1998, 36, 5, 827.
475
Handbook of Polymer Blends and Composites
23. M. Hasegawa, J.I. Ishii and Y. Shindo, Macromolecules, 1999, 32, 19, 6111.
24. S. Rojstaczer, M. Ree, D.Y. Yoon and W. Volksen, Journal of Polymer Science,
Part B: Polymer Physics, 1992, 30, 2, 133.
25. D.H. Lee, S.Y. Koo and S.T. Kim, Polymer (Korea), 1998, 22, 1, 6.
26. R.F. Sutton and D.E . Coverdell, inventors; E.I. Du Pont de Nemours, assignee;
US5,939,498, 1999.
27. A.P. Krasnov, I.A. Gribova, V.N. Adericha and Ya.V. Genin, Trenie Iznos, 1999,
20, 2, 221.
30. O. Oka and T. Sato, inventors; Tomoegawa Paper Company Ltd., assignee; JP
Patent 11 100,565,1999.
33. P. Cassidy and J. Fitch in Modern Fluoropolymers, Ed., J. Scheirs, Wiley, New
York, NY, USA, 1997, 173.
34. T-S. Chung, Proceedings of Antec 97, Toronto, Canada, 1997, Volume 2, 2658.
35. T-S. Chung, P. Foley and M. Jaffe, Polymers for Advanced Technologies, 1997, 8,
9, 537.
37. J.A. Campbell, A.A. Goodwin, F.W. Mercer and V.N. Reddy, High Performance
Polymers, 1997, 9, 3, 263.
476
Blends Containing Thermostable Heterocyclic Polymers
39. Y. Kim, M. Guo, L. Zhu, D. Kim, F.W. Harris and S.Z.D. Cheng, Chinese Journal
of Polymer Science, 1999, 17, 2, 171.
40. A.A. Kuznetsov, I.D. Egorov, G.K. Semenova, V.I. Berendyav, E.I. Akhmetieva and
B.V. Kotov, Vysokomolekulyarnye Soedineniya Seriya A and B, 2000, 42, 4, 683.
41. T. Okada, inventor; Nok Corp Japan, assignee; JP Patent 08 57,277, 1996.
42. G-D. Bidulescu, L. Tarko, I. Sitaru, F. Deak and T. Voinescu, inventors; Institutul
de Cercetare Stiintifica si Inginerie Tehnologica Pentru Electrotehnica, Bucuresti,
Romania, assignee; RO Patent 105,703, 1992.
43. D.A. Ivanov, P.D.M. Lipnik and A.M. Jonas, Journal of Polymer Science, Part B:
Polymer Physics, 1997, 35, 15, 2565.
44. X. Kong, H. Tang, L. Dong, F. Teng and Z. Feng, Journal of Polymer Science,
Part B: Polymer Physics, 1998, 36, 13, 2267.
45. D.A. Ivanov, and A.M. Jonas, Journal of Polymer Science, Part B: Polymer
Physics, 1998, 36, 5, 919.
49. B.H. Stuart and B.J. Briscoe, High Performance Polymers, 1996, 8, 2, 275.
50. A.A. Goodwin and G.P. Simone, Polymer, 1997, 38, 10, 2363.
52. J.F. Bristow and D.S. Kalika, Polymer, 1997, 38, 2, 287.
53. H.L. Chen and R.S. Porter, Journal of Polymer Research, 1999, 6, 1, 21.
54. C.H. Lee, T. Okada, H. Saito and T. Inoue, Polymer, 1997, 38, 1, 31.
477
Handbook of Polymer Blends and Composites
57. R. Gensler, C.J.G. Plummer, H-H. Kausch and K. Mnstedt, Journal of Materials
Science, 1997, 32, 11, 3037.
58. D.K. Brandom, J.P. Desouza, D.G. Baird and G.L. Wilkes, Journal of Applied
Polymer Science, 1997, 66, 8, 1543.
59. Y.S.Ni, J.I.Jin, S.J. Jeon and B.W. Jo, Korean Polymer Journal, 1998, 6, 5, 349.
60. H.S. Kwon, Y.S. Chun, H.C. Jung, S.B. Kim and W.N. Kim, Pollimo, 1997, 21, 5,
786.
61. S.D. Mikhailenko, S.M.J. Zaidi and S. Kaliaguine, Journal of Polymer Science,
Part B: Polymer Physics, 2000, 38, 10, 1386.
62. S. Demura and K. Haraguchi, inventors; Dainippon Inc and Chemicals Inc.,
assignee; JP Patent 09,143,282, 1997.
63. K. Goto and T. Akke, inventors; Japan Synthetic Rubber Co., Ltd., JP Patent 09
77,972, 1997.
65. S. Niyogi, S. Maiti and B. Adhikari, Polymer Degradation and Stability, 2000,
68, 3, 459.
66. A. Etxeberria, S. Guezala, J.J. Iruin, J.G. de la Campa and J. De Abajo, Journal
of Applied Polymer Science, 1998, 68, 13, 2141.
67. J.H. Lee, S.G. Lee, K.Y. Choi and J. Liu, Polymer Journal, (Tokyo), 1998, 30, 7,
531.
68. M. Weber, H. Fisch, G. Pipper and A. Gottschalk, inventors; BASF AG, assignee;
EP737,719, 1996.
478
Blends Containing Thermostable Heterocyclic Polymers
70. J.M. Martinez, J.I. Equiazabal and J. Nazabal, Journal of Applied Polymer
Science, 1996, 62, 2, 385.
71. J. Choi and M. Cakmak, Proceedings of ANTEC 99, New York, NY, USA, 1999,
Volume 2, 1656.
72. Y. Jin, M.J. Lindway, K. Xi and R.L. Utley, inventors; General Electric Company,
assignee; WO Patent 12,604, 2000.
73. H.L. Chen and M.S. Hsiao, Macromolecules, 1998, 31, 19, 6579.
74. H.L.Chen, J.C.Hwang and C.C. Chen, Polymer, 1996, 37, 24, 5461.
75. S. Bastida, J.I. Equiazabal and J. Nazabal, Polymer, 1996, 37, 12, 2317.
77. M. Kaku, inventor; E.I. Du Pont De Nemours and Co, USA, assignee; WO Patent
98 22,533, 1998.
78. T. Matsuki and J. Sadanobu, inventors; Teijin Ltd., Japan, asssignee; JP Patent 10
195,293, 1998.
82. H.T. Kim and J.K. Park, Japanese Journal of Applied Physics Part 1, 1999, 39,
1A, 201.
83. H.S. Chen, J.C. Hwang, C.C. Chen, R.C. Wang, D.M. Fang and M.J. Tsai,
Polymer, 1997, 38, 11, 2747.
84. H.G. Kim and R.E. Robertson, Hanguk Somyu Conghakhoechi, 1999, 36, 1, 17.
479
Handbook of Polymer Blends and Composites
86. C. Uriarte, J. Alfageme and J.J. Iruin, European Polymer Journal, 1998, 34, 10,
1405.
87. J.C. Kim, M. Cakmak and X. Zhou, Polymer, 1998, 39, 18, 4225.
89. Y. Ulcer and M. Cakmak, Journal of Applied Polymer Science, 1996, 62, 10,
1661.
90. C.E. Scott, inventor; Eastman Chemical Co., assignee; US5,648,433, 1997.
91. T. Ishiwatari and J. Sadanobu, inventors; Teijin Ltd. Japan, assignee; JP Patent
11 343,334, 1999.
92. S.N. Yau and E.M. Woo, Macromolecular Rapid Communications, 1996, 17, 9,
615.
93. J. Hanehi and N.S. Eiss, Tribology Transactions, 1997, 40, 1, 102.
98. H. Li, D. Xu and H. Cheng, Gaofenzi Cailiao Kexue Yu Gongcheng, 1999, 15, 3,
81.
101. J.S. Bloom, inventor; E.I. DuPont de Nemours and Co., US5,700,863, 1997.
480
Blends Containing Thermostable Heterocyclic Polymers
103. T. Nakamaru and Y. Yamamoto, inventors; Oiles Industries Co. Ltd., Japan,
assignee; JP Patent 74,136, 2000.
104. J. Huang and Q. Guo, Journal of Macromolecular Science B, 1997, 36, 3, 423.
105. G.C. Kapantaidakis, S.P. Kaldis, G.P. Sakellaropoulos, E. Chira, B. Loppinet and
G. Floudas, Journal of Polymer Science, Part B: Polymer Physics, 1999, 37, 19,
2788.
108. O.M. Ekiner, inventor; Lair Liquide S.A., France, assignee; US5,608,014, 1997.
109. X. Peng, Y. Wu, J. Liu and J. Zheng, Gaofenzi Cailiao Kexue Yu Gongcheng,
1998, 14, 6, 79.
110. B. Shi, Y. Wu, J. Liu, Q. Kong and X. Peng, Mo Kexue Yu Jishu, 1999, 19, 6, 48.
112. Y.S. Chun, H.S. Lee and W.N. Kim, Polymer Engineering and Science, 1996, 36,
22, 2694.
113. S.M. Cooper, D. Nazareth and R.A. Greenberg, inventors; General Electric Co,
assignee; EP Patent 711,810, 1996.
114. M.J. El-Hibri, inventor; General Electric Co, assignee; US5,905,120, 1999.
115. R. Puyenbroek, inventor; General Electric Co, assignee; EP Patent 926,204, 1999.
116. B. Masiulanis and R. Zielinski, Journal of Applied Polymer Science, 1985, 30,
2731.
117. J.A. Cella in Polyimides Fundamentals and Applications, Eds., M.K. Ghosh and
K.L. Mittal, Marcel Dekker, New York, NY, USA, 1986, 343.
481
Handbook of Polymer Blends and Composites
118. M. Zuo and T. Takeichi, Journal of Polymer Science, Part A: Polymer Chemistry,
1997, 35, 17, 3745.
120. Y.B. Lee, H.B. Park, J.K. Shim and Y.M. Lee, Journal of Applied Polymer Science,
1999, 74, 4, 965.
125. Y. Uetani, M. Nakamura and M. Abe, inventors; Nitto Denko Corp Japan,
assignee; JP Patent 08,259,709, 1996.
126. T.M. Su, I.J. Ball, J.A. Conklin, S.C. Huang, R. K. Larson S.L. Nguyen, B.M.
Lew and R.B. Kaner, Synthetic Metals, 1997, 84, 1-3, 801.
127. I.J. Ball, S-C. Huang and R.B. Kaner, Proceedings of Antec 98, Atlanta, GA,
USA, 1998, Volume 2, 1301.
128. M.G. Han and S.S. Im, Journal of Applied Polymer Science, 1999, 71, 13, 2169.
129. M.G. Han and S.S. Im, Journal of Applied Polymer Science, 1998, 67, 11, 1863.
130. J.G. Lee, B.C. Park, H.S. Woo, Y. Kim, C.S. Ha, C.M. Lee, K. Jeong, J.H. Ha and
Y.R. Kim, Solid State Communications, 1997, 102, 12, 895.
131. C. Hying and E. Stande, Journal of Membrane Science, 1998, 144, 1-2, 251.
132. Yu.N. Sazanov and A.N. Gribanov, Zhurnal Prikladnoi Khimii (St. Petersburg),
1999, 72, 11, 1896.
133. J.F. McNamara, inventor; E.I. Du Pont De Nemours and Company, USA,
assignee; WO Patent 99 00,453, 1999.
482
Blends Containing Thermostable Heterocyclic Polymers
135. T.K. Ahn, M. Kim and S. Khoe, ACS Polymeric Materials Science and
Engimneering, 1996, 75, 1, 22.
136. T.K. Ahn and M. Kim and, Macromolecules, 1997, 30, 11, 3369.
137. S. Lee, S.H. Chun, J.G. Lee, H. Lee and S. Choe, Polymer Preprints, 1997, 38, 1,
284.
139. S. Lee, J.G. Lee, H. Lee and S. Choe, Macromolecules, 1999, 32, 18, 5961.
141. E. Fekete, Z. Peredy, E. Foldes, F.E. Karasz and B. Pukanszky, Polymer Bulletin,
1997, 39, 1, 93.
143. W.J. MacKnight, S.W. Kantor and H.Zhu, Proceedings of Antec 96,
Indianapolis, IN, USA, 1996, Volume 2, 1594.
144. J.Y.M. Ree, T.J. Shin, X. Wang, W. Cai, D. Zhou and K.W. Lee, Journal of
Polymer Science, Part B: Polymer Physics, 1999, 37, 19, 2806.
145. J.Y.M. Ree, T.J. Shin, X. Wang, W. Cai, D. Zhou and K.W. Lee, Polymer, 2000,
41, 1, 169.
146. E. Fldes, E. Fekete, F.E. Karasz and B. Pukansky, Polymer, 2000, 41, 4, 975.
147. T.S.Chung and Z.L. Xu, Journal of Membrane Science, 1998, 147, 1, 35.
149. Y.H. Kao and L.W. Chen, Materials Chemistry and Physics, 1997, 47, 1, 51.
150. A.L. Rusanov, A.P. Kiselev, L.D. Frolova, E.G. Bulycheva, Z.B. Shifrina and I.V.
Yaroshenko, Plasticheskie Massy, 1995, 5, 1, 38.
483
Handbook of Polymer Blends and Composites
151. G.V. Adams and J.M.G. Cowie, Polymer, 1998, 40, 8, 1993.
152. G. Yang and F. Lu, Chinese Journal of Polymer Science, 1992, 8, 1, 87.
154. Y. Yang and Q. Pei, Applied Physics Letters, 1997, 70, 15, 1926.
159. H-J. Sterzel and M. Meyer, inventors; BASF AG, assignee; EP Patent 517025A1,
1992.
484
Blends and Interpenetrating Networks Based on Polyurethanes
12
Blends and Interpenetrating Networks
Based on Polyurethanes
12.1 Introduction
It is well known that polyurethane (PU) chemistry opened the way to a new class of high
performance materials such as coatings, adhesives, elastomers, fibres, foams, etc. [1].
Segmented polyurethanes are often used in multicomponent polymeric systems, such as
polymer compounds, blends and interpenetrating polymer networks (IPN) [2-4]. These
multicomponent systems provide the possibility of combining properties of different
polymeric materials.
Incorporating a PU elastomer in a blend [7-12] or IPN [13, 14] increases the toughness
and impact resistance. The resulting polymer blend is immiscible, but compatible because
of the intermolecular hydrogen bonds [15].
485
Handbook of Polymer Blends and Composites
Blends of rubbers are fairly common in rubber compounding because of two factors: the
ability to improve the properties of one rubber by the addition of a second, and the
ability to lower the cost of the more expensive rubber by blending with a less expensive
rubber, and hopefully keeping most of the benefits of the expensive one [17].
From this point of view, urethane compounds can generally improve the properties of
polymers it is added to and when small portions of other polymers are added to it, a
lower cost compound with many of the good properties of urethane can be the result.
Optimising the cure system and polymer blend ratio, in most cases can result in compounds
that perform in the most cost effective manner [16].
PU are block and graft copolymers with many advantageous properties, classified as
thermoplastic elastomers, which have lead to their use in a wide range of applications.
The PU elastomer morphology may be described as resulting from the occurrence of
phase-segregated regions [6]: a region with low glass transition temperature (Tg), where
the dominant chemical species do not undergo considerable hydrogen-bonding; a region
with high Tg, where strong hydrogen-bonding is the rule and a region of interface,
populated with both species. The mechanical properties, which are the consequence of
this morphology, make these block copolymers behave as thermoplastic elastomers.
Their basic chemical structure is the urethane group (R-NHCOO-R) and they are
synthesised from the reactions of polyether-based or polyester-based diols and
diisocyanates, followed by the introduction of chain extenders to form the macromolecules
[6, 23]. One appropriate mechanism for elastomeric PU synthesis has been proposed by
Pappas [24]. In the first step, dibutyltin dilaurate (Bu2SnL2), a catalyst of polymerisation,
reacts with the alcohol and then with the isocyanate, which results in an intermediate (I).
In the second step an intramolecular transfer occurs, the intermediate (I) becomes very
unstable and reacts with another alcohol molecule giving the urethane bond and active
species that can reinitiate the process of polymerisation (Scheme 12.1).
486
Blends and Interpenetrating Networks Based on Polyurethanes
Due to the different polarity and chemical nature of both blocks they separate into two
phases designated as soft or hard phases (Figure 12.1).
487
Handbook of Polymer Blends and Composites
Indeed, segmented polyurethanes are typical block copolymers of the type (A-B)n, where
the block A consists of relatively long and flexible soft segment (SS) and the block B is a
highly polar hard segment (HS). Very often the SS is a polyether or polyester and the HS is
diisocyanate extended with a low molecular weight diol. Microphase separation at
mesoscopic length scales, resulting in a morphology consisting of microdomains rich in HS
(HS microdomains) and a microphase rich in SS (SS microphase), is a fundamental property
of segmented polyurethanes, arising from the incompatibility of HS and SS [25-28].
488
Blends and Interpenetrating Networks Based on Polyurethanes
The hard blocks of the molecules are connected with each other by intermolecular
hydrogen bonding, generating domains that act as junctions of physical crosslinking for
the soft blocks. Hydrogen bond distribution influences the extent of microsegregation in
PU and a set of properties typical for these materials. As consequence, the properties of
PU blended systems with other polymers would depend on how the added polymer
influences the extent of microsegregation of PU [29].
In fact, many of the versatile properties of segmented PU, which make these systems
useful for diverse technological applications, are related to their incomplete microphase
separation. Thus, it is not surprising that several studies have been devoted to the
investigation of the factors controlling microphase separation and to the development of
methodologies to quantitatively study microphase separation in PU [25-28, 30-34]. In
addition, methods have been developed to calculate the microphase composition and to
characterise the degree of microphase separation [35].
The concentration of soft and hard segments is usually expressed by Equations 12.1
and 12.2:
Some types of PU used in blends based on polyurethanes are summarised in Table 12.1.
Solid urethane rubbers are commonly used due to their excellent properties, especially
tensile and tear strength, and very high abrasion resistance. PU also have excellent
489
Polyurethane elastomers
490
Polyurethane elastomers
491
Handbook of Polymer Blends and Composites
resistance to oils and similar fluids because of their polar nature. These special properties
make urethane rubbers the ideal base polymers for compounding [53]. Thus PU are
blendable with other polar rubbers such as acrylonitrile butadiene rubber (NBR),
acrylonitrile rubber/polyvinyl chloride (NBR/PVC) and hydrogenated nitrile rubber
(HNBR), or metal ionomers, typically giving intermediate properties for the respective
blends. They may also be blended with less polar polymers such as polydimethylsiloxanes
(PDMS), butadiene rubber (BR), styrene butadiene rubber (SBR) and Kraft lignin, or
with the nonpolar ethylene propylene diene rubber (EPDM) [16, 38]. The interactions
that could appear in such blends generally include hydrogen bonding, Coulombian
attraction forces between ions and ion-dipole interactions.
The benefits of adding small portions of urethane rubber to other conventional polymers
as NBR/PVC will improve abrasion resistance, tensile and tear strength, retain high
temperature, oil and solvent resistance properties and low temperature properties. On
the other hand, adding several portions of conventional polymers to urethane improve
reversion resistance, compression set (NBR), resistance to water, glycol, methyl ethyl
ketone (MEK) and lower the cost of compounds.
Although there are many types of PU blends, those containing PVC, are among the most
important from both a scientific and a commercial point of view [7, 54, 55]. There are
contradictory results concerning miscibility and the properties of PVC/PU blends, in that
PU plays the role of permanent plasticiser for PVC [56]. However, there is no established
systematic variation of properties, since various authors have used different PU systems.
Blending PU with SAN results in many property variations due to the numerous
combinations of different PU types (hard and soft segment character, domain separation,
etc.), or with different SAN copolymers (weight% of acrylonitrile and different
distributions of styrene and acrylonitrile) that are possible [3, 57].
In recent years PDMS, due to their good performance, has received considerable interest
as a component in PU alloys and blends aiming at functional polymeric materials [10,
11, 41, 58, 59].
The main types of blend formulations based on PU are reviewed in Table 12.2.
492
Polyurethane elastomers
493
Handbook of Polymer Blends and Composites
The mixing operation is the most important step in blend preparation and it is well
known that certain properties depend on it [60, 61]. Two different reactive polymers can
chemically interact at high temperature to form interchain crosslinked polyblends. Das
and co-workers have developed such interchain, crosslinkable, elastomer-elastomer and
elastomer-plastic blends [62-65].
1. preblending technique (the two elastomers are first blended in given ratios, allowed
to equilibrate, and then curatives are incorporated into the blend at high temperature),
Recent studies also showed that PU are able to crosslink with polyethylene co-EVA [40,
45, 46, 66] at high temperature without any curatives. De and co-workers have reported
that rubbers having appropriate functional groups interact with each other when blended
and thus crosslink at high temperature in the absence of any curatives [67-72]. The
temperature profile can be adjusted based on the processing temperatures of both
individual materials and their weight ratios in the blends. Also, speed of mixing can be
adjusted to obtain different ratios to accomplish the final thickness requirements.
494
Blends and Interpenetrating Networks Based on Polyurethanes
The main techniques for obtaining highly homogenous PU blends, such as those presented
in Table 12.3 (solution or press cured mixing) and Table 12.4 (melting or mechanically
mixing), allow mixing of PU with almost all conventional plastics, rubbers, fibres, etc.
495
Handbook of Polymer Blends and Composites
Differential scanning calorimetry (DSC) was often used to study the thermal properties of
various samples. A Tg at low temperatures and a melting peak at high temperatures were
observed for every PU sample associated with the non-crystalline region of soft segments
and strong hydrogen-bonding/crystalline region of the hard segments caused them.
The morphology of the chlorinated PU-1/PVC (Table 12.2) blends, studied by SEM, is of
a uniform type, but with differing degrees of roughness [7]. So, a melting endotherm for
the PU-1 hard segments occurs in the temperature range of 230-250 C. This observation
agrees well with morphogical models proposed for different PU-1/PVC systems [82-84],
which suggest that the PVC molecules may disperse in the soft segment phase with little
influence on the microphase separation and mixing of the hard and soft segments.
The ring-like voids surround the dispersed spheres in polytetramethylene adipate (PTMA)-
PU-17/SAN/EVA ternary blend, was obtained. This interesting morphology clearly
indicated that the most PU-17 - PTMA is located between the dispersed EVA phase and
SAN matrix acting as a compatibiliser. Although PU-17 - PTMA and SAN are immiscible,
there are moderate specific interactions between them [45].
The urethane and ether groups of the PU are of particular importance, especially when
blending with polyamides [85, 86]. It is well known that PU contains polar functionality
that can also interact with the polar component of polyamide 6 (PA6). The interaction
between PU and PA6 in the corresponding blends can enhance the toughness of polyamide
due to its elastic properties. Hydrogen bonding has been observed between molecules of
either PU-PA6 and PA6-PA6, and it is sufficiently strong to cause the polymer chains to
distort rather than disrupt the hydrogen bonds. When groups of like polarity, such as
carbonyl groups, come into proximity, the polymer chains again distort from their regular
conformation because of mutual repulsion. The hydrogen bonding interaction is what
determines the compatibility of PA and PU and the interfacial adhesion in their blends [87].
Micrographs were made of the top surface of CA, PU-28, CA/PU-28 blend and the cross
sections of blend membranes [52]. The micrographs showed the existence of phase
separation and compatibility in the blend membranes (CA/PU-28). The sponge like
structures in the PU membranes was similar to those observed by Koenhen and co-workers
496
Blends and Interpenetrating Networks Based on Polyurethanes
The compatibility of a new polymer blend is often the first property studied. There are
many experimental techniques to characterise the compatibility level of the polymer
blends, but DSC and dynamic mechanical analysis (DMA) are the most widely used.
These techniques, by measuring the Tg of each phase, indicate the number of phases
present. The miscibility level, i.e., the amount of the B component miscible in the A
phase may be deduced by means of the shift of Tg of each phase compared to that of the
pure component.
Since PU can be easily modified by gradual replacement of soft segments, such as PTMA,
PTMO, PCL, PPO, etc., the relationship between chemical structure of the PU and its
miscibility with PVC can be readily investigated. Wu and co-workers [46] studied the
miscibility of the PU blended with chlorinated polyvinyl chloride (CPVC) by DSC and
FTIR spectroscopy. DSC results showed that the polyester polyurethane blends, PCL-
PU-25/CPVC and PTMA-PU-17/CPVC (Table 12.2), are miscible over the entire
composition as they exhibit only one major Tg. The Tg of the blends regularly increased
with increasing CPVC content. Increasing the molecular weight of the glycol can decrease
the miscibility due to the crystallisability of the soft segments. Two distinct Tg for all
polyether polyurethane blends (PPO-PU-26/CPVC and PTMO-PU-18/CPVC) were
obtained, which did not change with the changing of the blends composition, suggesting
497
Handbook of Polymer Blends and Composites
that the blends of are immiscible in all cases. On the other hand, the FTIR spectroscopic
results showed a strong interaction between CPVC and the polyester soft segments. In
addition, it was found that the miscibility of PU and CPVC decreased upon increasing
the hard segment content. Finally, they found that polyester polyurethane blends with
CPVC form miscible systems while polyether polyurethane-CPVC blends are immiscible.
The DSC plots of PVC/chlorinated PU-1 systems revealed a single Tg in the region of 55-
65 C, showing the blends miscibility [7]. The compatibility of these blends can be due
to hydrogen-bonding between the C=O groups of the polyester and the -hydrogen of
chlorinated polymer [89, 90] or a dipole-dipole C=O.Cl-C- interaction [91, 92].
In the ester-PU/PA12 blends, miscibility can be achieved owing to close solubility parameters
of soft segment of PU-6 and that of PA12. These differ by 1.2 (MJ/m3) 0.5, while for oligoester-
PU-5/PA12 blends, the solubility parameters differ by 4.7 (MJ/m3) 0.5 [29].
498
Blends and Interpenetrating Networks Based on Polyurethanes
12.3.2 Degradation
There is a definite correlation between the chemical structure of a polymer and its
degradation temperatures. Blends prepared through the different blending techniques,
having the same compounding formulation and same blending ratio differ in thermal
behaviour.
Pielichowski and Hamerton [7] examined, using DSC, SEM and thermogravimetric
analysis (TGA), either alone or coupled with FTIR, the thermal behaviour of a series of
solution-cast blends of PVC)/chlorinated TDI-based polyurethane (PVC/PU-1) polymers.
They found that the decomposition proceeds through a two-step route; the main, decisive
degradation stage in the 200-320 C temperature range was found to be a result of
parallel reactions of PVC and PU-1 decomposition. This was also confirmed by Ozawa-
Flynn-Wall kinetic analysis, the activation energy remained constant for degrees of
conversion greater than 0.3 [98-102]. The reasons for the better thermal stability of
some PVC/PU-1 blends can be explained by analysis of specific interactions between the
499
Handbook of Polymer Blends and Composites
C=O groups of the urethane segments and the -hydrogen of the chlorinated polymer or
a dipole-dipole C=O..Cl-C- interactions. On the other hand, the rate of diffusion of
volatile products through microphase domain structure may differ due to changes in
morphology arrangement thus considerably affecting the overall decomposition route.
There have been extensive investigations on the electronic transport properties of pseudo
semiconducting PU [51, 104-108].
Seki and co-workers [109] reported that the complexes of ether-based PU prepared from
MDI/BD/PTMO with lithium salts, exhibit ionic conductivity of approximately 10-6/S/
cm at room temperature. Also, the improvement of the ionic conductivity of the PU-27/
lithium salt complexes by blending polysiloxanes and polyether-modified polysiloxane
was studied [51].
= n q ( + + ) (12.3)
500
Blends and Interpenetrating Networks Based on Polyurethanes
where, n is the ion density, q is charge and + and - are the cationic and anionic mobility,
respectively.
= d 2 /(E t) (12.4)
Both the cationic and anionic mobility (+ and -) of both the electrolytes increased with
increasing content of PDMS-co-PMPhS and PES. In conclusion, ionic conductivity of PU-
27-based electrolyte increased by blending PDMS-co-PMPhS or PES. The improvement of
the conductivity for the PU-27/PES electrolyte is attributed to the increase of both ion
mobility and ion density. On the other hand, it is thought that the increase of ion mobility,
which compensates for the decrease of ion density, contributes to the improvement of the
conductivity for the PU-27/PES electrolyte. It is interesting that ion mobility increases by
the addition of polysiloxane even in the phase-separated PU-27/PDMS-co-PMPhS blend.
Since polyurethanes show high porosity, low weight to volume ratio, good resilience
character, abrasion resistance and oil resistance, they can be incorporated into CA to
501
Handbook of Polymer Blends and Composites
introduce balanced hydrophilicity in the resultant blend membrane and hence achieve
optimum membrane performance in terms of better rejection and flux.
Blends of PP/PU are widely used in the automobile and electronic industry as well as for
household and sport supplies, and in medical components.
Polymer network systems composed of least two kinds of components with different
physicochemical properties are expected to be promising functional polymeric materials
[111-113]. The attempts to connect some different chemical components by chemical
bonds provide a convenient route for the modification of properties to meet specific
needs and were first made for the copolymers synthesised from multi-components with
block or random sequences [10, 11]. Among these methods, those leading to IPN, which
could be considered as a subgroup of the broad class of polymer blends, represent a new,
essential approach to solve the problem of mutual incompatibility of polymers [13, 14,
114]. IPN generally possess enhanced physical and engineering properties compared to
the normal polyblends of their components because of the synergistic effect of individual
polymers.
IPN are traditionally defined as a combination of two or more chemically distinct polymers
in network form, held together ideally and solely by their permanent mutual
entanglements, in which at least one is synthesised and/or crosslinked in the immediate
presence of the other. IPN could also be called polymer alloys and have been intensively
studied [6, 14, 111, 113, 115-135].
The most commercially important PU elastomers used in IPN are phase-separated systems
consisting of a MDI or TDI based hard segment and a polyether or polyester soft segment.
Frequently synthesised and studied series of polyurethane IPN consists of crosslinking
their networks with acrylic [129-131, 133, 136-140], styrene [141-148], polysiloxanes
[149-155], unsaturated polyester [126, 156-162] and epoxy polymers [120, 163- 164].
Some representative examples are illustrated in Table 12.5.
502
Blends and Interpenetrating Networks Based on Polyurethanes
Two basic methods are currently used for making IPN: when both polymers are formed
and crosslinked simultaneously, the network is known as a simultaneous interpenetrating
network (SIN), whereas if the crosslinking of polymer II follows that of polymer I, the
material is a sequential IPN. When only one polymer is crosslinked, a semi-IPN is
produced, and when all polymers are in network form it is a full-IPN (Scheme 12.4).
There are also classes of thermoplastic IPN (involving physical crosslinks rather than
chemical crosslinks so that they could be considered as hybrids between polymer blends
and IPN), latex IPN, which often exhibit core-shell morphologies, and gradient IPN,
when the crosslink density or composition varies from one network location to another.
In fact, most IPN, including polyurethane IPN, do not interpenetrate on a molecular
503
Handbook of Polymer Blends and Composites
scale. Polymers in the system may form finely divided phases of only hundreds of ngstrm
units in size. However, these phases may be continuous on a macroscopic scale (dual
phase continuity). In some cases, true molecular interpenetration is thought to take
place at the phase boundaries [118].
Many methods of polyurethane IPN preparation have been reported in the literature
(Table 12.6).
504
Polyurethane IPN methods of
preparation
505
Handbook of Polymer Blends and Composites
It is well known that introducing filler into the reaction system at the stage of formation
of crosslinked polymer, specifically PU, essentially affects the reaction, the conditions
and the degree of microphase separation.
Like blends, graft or block copolymers, IPN are multicomponent polymer materials. However,
IPN exhibit specific characteristics: in solvents, flow and creep are suppressed, and they
swell but do not dissolve; also, the domain size can be tailored [118]. The multipolymeric
506
Blends and Interpenetrating Networks Based on Polyurethanes
systems thus obtained display a broad range of properties from toughened elastomers to
high impact plastics, based on the selected composition ratio and synthetic details.
It was found that the chemical bonds between the two networks have a great effect on
the kinetic behaviours of SIN formation as well as the morphology development and the
physical properties of such SIN.
The degree of segregation (DS) in polyurethane IPN can be calculated from the parameters
of the relaxation maxima [187]:
h l + h2 l2 + h m l m
DS = h1 + h 2 = 1 1
L
(h
0
1 + h 02 )
where h01 and h02 are the maximum values of mechanical loss for the pure components
(case of complete phase separation); h1 and h2 are the values for each component at
different degrees of segregation; hm is the maximum loss for the relaxation transition,
responsible for the interfacial layer appearance; L is the interval between the Tg of the
pure components; l1, l2 and lm are the shifts of the respective maxima in temperature scale.
Most of the materials considered are chemically crosslinked. So, the structure of
polyurethane IPN can be discussed on the basis of molecular theory of rubber elasticity
using the stress-strain data [154-155, 188-189]. The stress-strain data were interpreted
in terms of the reduced stress |f*| and the Mooney-Rivlin semi-empirical equations:
f
f = (12.5)
A( 2 )
and/or
f = 2C1 + 2C2
L
where, f is the applied force, is the elongation = , L and L1 are the stretched and
Li
unstretched lengths, respectively), A is the cross-sectional area of the unstressed sample,
C1 and C2 are constants.
The reduced stress can be plotted as a function of reciprocal elongation to determine the
crosslink density of the sample. The empirical Equation 12.5 provides a linear relation
between |f*| and 1/.
507
Handbook of Polymer Blends and Composites
If one compares reduced stress for 1 according to Equation 12.5 and Flory expression
for the modulus [189] the following equation is obtained:
32 2
2C1 + 2C2 = [(1 2 )(1 + fc fph )] 1 RTV2c = A ' RTV2c 3
M M
where, Af designates the term in square brackets, M means the molecular weight between
crosslinks, is the sample density, is the crosslink functionality, fc is the contribution
due to the constraints imposed by the junction fluctuations in the polymer coils, fph is the
contribution due to the assimilation of the polymer network with a phantom network,
V2c = V/V0 (V0 is the initial volume of the sample, and V is the volume of the sample
strained with a force f).
However, the synthesis of IPN-like materials having a thermoplastic nature can involve
physically crosslinked systems. These bonds of a physical nature may arise from ionic
portion of an ionomer, glassy portion of a block copolymer, crystalline portion of a semi-
crystalline polymer, or hydrogen bonding.
In the preparation of simultaneous IPN, different formation rates of the two networks
have been shown to give a material with a range of morphologies and mechanical
properties [190-192].
The most popular and well-studied polyurethane IPN is PU/PMMA, which is a good
system for modelling and yields to a variety of materials with useful properties [133,
138]. The highly immiscible PU/PMMA systems could reveal different morphologies as
function of polymerisation routes, more precisely as function by the events order of PU
and PMMA gelation and phase separation of these two polymers. These also explain the
differences in morphology and modulus between PU/PMMA 50/50, for that first gel
PMMA, and PU/PS 50/50 IPN, for that first gel PU.
For PU/PADC, when formation of the PU network occurs well ahead of allyl diglcol
carbonate (ADC) gelation, extensive phase separation resulted, while the simultaneous
formation of both networks or ADC first gelation give a fine continuous or very fine
dispersed morphology and better mechanical properties [135]. The dispersed domains
of PU/PS IPN (7/3) was significantly reduced by introducing crosslinks in PS domains
using DVB [139]. From the applications point of view it is desired that phase separation
occurs after at least one component gels, as is the case of PU/PMMA 50/50 IPN, where phase
separation appears after 8 minutes of reaction and gelation after 50 minutes of reaction.
Comparative studies of IPN resulting from the transesterified castor oil PU and PMMA
with the corresponding unmodified castor oil PU and PMMA IPN could reveal the effect
508
Blends and Interpenetrating Networks Based on Polyurethanes
The PU/PBMA semi IPN exhibited microphase separation that is increased by slowing
the reaction rates; each phase should be treated as an independent IPN. The maximum
compatibility of the two polymers was observed at the blend composition of 50% PU-
50% PBMA, since this proportion of the blend exhibits maximum interpenetration [123].
Lipatov and co-workers [187, 192-194] found that reaction kinetics and viscoelastic
properties of PU/PBMA semi IPN are interrelated, so that the PBMA formation rate
affect the PU formation and higher rates will give higher degrees of segregation and
higher differences between the two corresponding Tg. Kinetics and microphase separation
for PU/PBMA system could be adjusted by introducing a finely dispersed filler, either
inorganic (talc), or polymeric (TEGDM). These fillers bring a restriction into the molecular
mobility of PU and PBMA, initiator efficiency and microphase separation. At low initiator
concentrations, use of filler leads to an increase of PU and PBMA formation rate for any
filler concentration, whereas for high initiator and low filler concentrations an opposite
effect takes place. Moreover, high initiator coupled with high filler amounts (>20%)
results in decrease of both PU and PBMA network formation.
For PU/unsaturated polyester (UPE) IPN, the driving force of phase separation was the
formation of the PU phase and is due to the incompatibility between the resulting PU
polymers and the unreacted monomers of PU and UPE resin. Final morphology is strongly
dependent of curing temperature and reaction rate of PU and UPE [130].
During the formation of PU/vinyl ester simultaneous IPN it was found that the kinetics
of formation for the two networks significantly depends on the morphology development
and chemical binding between them [195-199].
The IPN formation occurs by two simultaneous processes, i.e., curing and phase separation
arising from the thermodynamic incompatibility of growing polymeric chains. This largely
determines the structure of IPN and therefore their end properties. There have been a
number of studies on IPN formation kinetics [200-202] but only few of these deal with
the connection between IPN formation kinetics and microphase separation.
Based on an analysis of the shift, form and size of the carbonyl group stretching vibration
band, an assumption was made that, with a low content of one of the polyurethane IPN
components (PEA) in the initial reaction mixture, contacts between growing chains of
heterogeneous networks are preferable, and so in the initial stages of the reaction, a
retardation of the PU formation takes place [192]. As curing proceeds, the composition
of the system and the intermolecular interactions continuously change and at a certain
509
Handbook of Polymer Blends and Composites
stage thermodynamic incompatibility appears. It has been shown that after the onset of
microphase separation, the influence of the admixture networks diminishes markedly.
As the concentration of PEA in the IPN increased, the time before the onset of microphase
separation reduced sharply. The formation of three-dimensional structures and microphase
separation proceeding simultaneously leads to a continuous alteration in phase
composition and in interaction forces. The influence of the PEA networks on the urethane
formation rate in IPN of the same composition depends strongly on the PEA curing rate,
determined in the present case by the initiator concentration. The PEA micro-regions
having been heterogeneous in the reaction initial stages are now uniformly distributed in
the PU matrix as fillers [165].
In this way, the PU IPN formation kinetics and the ratio of the constituent network
curing rates determine the rate and degree of microphase separation. This in turn
determines the boundary layer composition and structure and hence the structure of IPN
based materials.
Formation of polymers both in bulk and on various hard surfaces is often accompanied
by the appearance of internal shrinkage stress, which affects the polymeric material
properties. Therefore, it is advisable for polymeric composite production to be carried
out under conditions such that the stress relief time is comparable with the time of polymer
formation. In this respect, the production of multicomponent polymeric compositions
based on the IPN principle, where the individual component formation processes can be
separated for an appropriate time, appears to be the most promising.
All the fillers, regardless of their nature, shorten the time of appearance and increase the
internal shrinkage stress values in the filled IPN with predominantly high-modulus
component concentration as compared with unfilled systems. However, the onset of the
internal shrinkage stress and the absolute value are determined by the chemical nature of
the filler.
According to the method suggested in [203] the value of the polymer-filled interaction
g*P-F in the above IPN was estimated. The calculation was carried out based on the
analysis of the vapour sorption by an unfilled polymer, the filler and filled polymer
according to the equation:
510
Blends and Interpenetrating Networks Based on Polyurethanes
where g1 and g111 are the free energies of the filled and unfilled polymers interaction
with a large amount of solvent; g11 is the free energy of filler with a large amount
solvent.
For filled IPN the sign of the value of g*P-F depends on the ratio of components, on
introducing the filler into the composition: for small ionomer contents the value of g*P-F is
negative, i.e., the system is thermodynamically stable, at an ionomer content of 30% or
greater, the value of g*P-F is positive, i.e., the system is destabilised.
When fillers are introduced into an IPN consisting of components which are
thermodynamically unstable under certain conditions, e.g., filler content, components
ratio, affinity of only one component, negative g*P-F values are observed, i.e., in presence
of a filler, systems with higher thermodynamic stability are formed as compared with
unfilled systems. In this case the filler favours component compatibility in the IPN.
Because of their phase separated elastomeric nature, polyurethanes opened the way to a
new class of high impact IPN materials such as coatings, adhesives, elastomers, fibres
and foams. Numerous applications have been patented in the fields of electrical insulation,
coatings and encapsulants, noise and vibration damping, adhesives and membranes,
materials for optics.
Urethane alkyds are usually used to make conventional paints that cure by air oxidation
of the unsaturated groups in the presence of metallic driers to give tough coatings with
high abrasion resistance. Uralkyds have superior adhesion, hardness, abrasion resistance,
and durability and chemical resistance as compared to traditional alkyds. Moreover,
cured uralkyd resin is elastomeric in nature. This property of the uralkyd resin is used to
develop elastomers.
Polysiloxane crosslinked in presence of aromatic urethanes gave IPN with wear rates
much lower than pure polyurethanes or pure silicones. This type of IPN can be applied
as biomedical materials, synthetic fabrics, or composite foams with adhesive layers.
511
Handbook of Polymer Blends and Composites
Many plastics, such as PS, are brittle and may be toughened by incorporation of small
quantities of elastomeric PU [13].
Pernice and co-workers [204] pointed out that SIN composed of polyurethane in
combination with epoxy resin, polyacrylates, or unsaturated polyester resin could be
applied in reaction injection moulding (RIM) processes. Also, SIN consisting of different
polyurethane acrylates and vinyl ester resin in which the main backbone of epoxy is
maintained have been synthesised and studied for a RIM process to prepare toughened
epoxy materials [120, 205-207].
Concluding Remarks
Because plant design is not expensive, a broad spectrum of development projects are
envisaged in this area, which will enrich continuously the existing range of PU materials.
In the some cases PU products, which are not reproducible even with a sophisticated
technology, can be obtained by blending.
References
4. S. Dabdin, R.P. Burford and R.P. Chaplin, Polymer, 1996, 37, 785.
512
Blends and Interpenetrating Networks Based on Polyurethanes
6. Z.S. Petrovic and J. Ferguson, Progress in Polymer Science, 1991, 16, 695.
8. M. Maity, B.B. Khatua and C.K. Das, Polymer Degradation and Stability, 2000,
70, 263.
10. L.A. Utracki, Polymer Alloys and Blends: Thermodynamics and Rheology,
Hanser Publishers, Munich, 1989.
11. Polymer Blends, Eds., D.R. Paul and S. Newman, Academic Press, New York,
NY, USA, 1978.
14. G.Y. Wang, Y.L. Wang and C.P. Hu, European Polymer Journal, 2000, 36, 735.
15. M.M. Coleman, J.F. Graf and P.C. Painter, Specific Interactions and the
Miscibility of Polymer Blends. Practical Guides for Predicting and Designing
Miscible Polymer Mixtures, Technomic, Lancaster, PA, USA, 1991.
16. T.L. Jablonowski, Proceedings of the 155th ACS Rubber Division Meeting,
Chicago, IL, USA, Spring 1999, Paper No. 46.
20. D.G. Bucknall and J.S. Higgins, Polymer, 1992, 33, 4419.
21. J. Heuschen, J. Vion, R. Jerome and P. Teyssie, Polymer, 1990, 31, 1473.
23. W.Y Chiu, P.S Wang, and T.M. Don, Polymer Degradation and Stability, 1999,
66, 233.
513
Handbook of Polymer Blends and Composites
24. S.P. Pappas, Notes from Coatings Science Course, North Dakota State University,
USA, 1988.
25. L.M. Leung and J.T. Koberstein, Macromolecules, 1986, 19, 706.
26. Y.V. Savelyev, E.R. Akhranovich, A.P. Grekov, E.G. Privalko, V.V. Korskanov, V.I.
Shtompel, V.P. Privalko, P. Pissis and A. Kanapitsas, Polymer, 1998, 39, 3425.
27. J.T. Koberstein, A.F. Galambos and L.M. Leung, Macromolecules, 1992, 25, 6195.
28. B. Chu, T. Gao, Y. Li, J. Wang, C.R. Desper and C.A. Byrne, Macromolecules,
1992, 25, 5724.
29. S.S. Pesetskii, V.D. Fedorov, B. Jurkowski and N.D. Polosmak, Journal of
Applied Polymer Science, 1999, 74, 1054.
30. V.P. Privalko, E.S. Khaenko, A.P. Grekov and Y.V. Savelyev, Polymer, 1994, 35,
1730.
32. Y. Li, J. Kang, O. Stoffer and B. Chu, Macromolecules, 1994, 27, 612.
33. G. Pompe, P. Pohlers, P. Poetschke and J. Piontek, Polymer, 1998, 39, 5147.
34. L. Cuve, J.P. Pascault, G. Boiteux and G. Seytre, Polymer, 1991, 32, 343.
35. J.T. Koberstein and L.M. Leung, Macromolecules, 1992, 25, 6205.
36. S. Parnell, G. Suppiah and K. Min, Proceedings of Antec 99, New York, NY,
USA, 1999, Volume 3, 3625.
37. R. Xu, J.L. Mead, S.A. Orroth, R.G. Stacer and Q.T. Truong, Presented at the
155th ACS Rubber Division Meeting, Chicago, IL, USA, Spring 1999, Paper No.
78.
38. D. Feldman and M.A. LaCasse, Journal of Applied Polymer Science, 1994, 51,
701.
39. S.D. Seneker, N. Barksby and B.D. Lawrey, Proceedings of Polyurethanes Expo
96, Las Vegas, NV, USA, 1996, 305.
514
Blends and Interpenetrating Networks Based on Polyurethanes
40. J. Xie, H. Ye, Q. Zhang, L. Li and M. Jiang, Polymer International, 1997, 44, 35.
41. T. Bremner, D.J.T. Hill, M.I. Killeen, J.H. ODonell, P.J. Pomery, D. St. John and
A.K. Whittaker, Journal of Applied Polymer Science, 1997, 65, 939.
42. B.P. Grady, E.M. OConnell, C.Z. Yang, and S.L. Cooper, Journal of Polymer
Science: Polymer Physics Edition, 1994, 32, 2357.
43. E.M. OConnell, C.Z. Yang, T.W. Root and S.L. Cooper, Macromolecules, 1996,
29, 6002.
46. W. Wu, X.L. Luo and D.Z. Ma, European Polymer Journal, 1999, 35, 985.
47. C.P. Papadopoulou and N.K. Kalfoglou, Polymer, 1999, 40, 905.
52. Plastics Additives and Modifiers Handbook, Ed., J. Edenbaum, Chapman and
Hall, London, UK, 1996.
54. M. Sivakumar, M. Carmel, L.A. Utracki, J.P. Szabo, I.A. Keough and B.D. Favis,
Polymer Engineering and Science, 1992, 32, 1716.
55. J.T. Haponiuk and A. Balas, Journal of Thermal Analysis, 1995, 43, 215.
515
Handbook of Polymer Blends and Composites
56. A.J. Varma, S.V. Deshpande and P. Kondapalli, Polymer Degradation and
Stability, 1999, 63, 1.
58. Siloxane Polymers, Eds., S.J. Clarson and J.A. Semlyen, PTR Prentice Hall,
Englewood Cliffs, NJ, USA, 1993.
60. J.A. Manson and L.H. Sperling, Polymer Blends and Composites, Plenum Press,
New York, 1976.
61. F.P. La Mantia, A. Valenza and D. Acierno, Polymer Degradation and Stability,
1985, 13, 1.
62. S.K. Singha Roy and C.K. Das, Polymer Composites, 1995, 3, 5.
63. A.R. Tripathy and C.K. Das, Journal of Applied Polymer Science, 1994, 51, 245.
64. A.R. Tripathy and C.K. Das, Plastics & Rubber & Composites Processing &
Applications, 1994, 21, 5.
65. S.K. Singha Roy and C.K. Das, Journal of Polymer Engineering, 1995, 14, 175.
66. M. Maity and C.K. Das, International Journal of Polymeric Materials, 1998, 45,
123.
67. R. Alex, P.P. De and S.K. De, Polymer Communications, 1990, 31, 67.
68. S. Mukhopadhyay and S.K. De, Journal of Applied Polymer Science, 1992, 45, 181.
69. S. Mukhopadhyay, P.P. De and S.K. De, Journal of Applied Polymer Science,
1991, 43, 347.
70. T. Bhattacharya and S.K. De, European Polymer Journal, 1991, 27, 1065.
71. S. Mukhopadhyay, T.K. Chaki and S.K. De, Journal of Polymer Science: Polymer
Letters Edition, 1990, 28, 25.
72. S. Mukhopadhyay and S.K. De, Journal of Applied Polymer Science, 1991, 42,
2773.
516
Blends and Interpenetrating Networks Based on Polyurethanes
73. L.C. Sawyer and D.T. Grubb, Polymer Microscopy, Chapman and Hall Ltd.,
London, UK, 1987.
74. G. Binnig, C.F. Quate and C. Gerber, Physics Review Letters, 1986, 56, 930.
77. L.M. Eng, H. Fuchs, K.D. Jandt and J. Petermann, AIP Conference Proceedings,
1992, 241, 262.
78. L.M. Eng, K.D. Jandt, H. Fuchs and J. Petermann, Applied Physics A, 1994, 59,
145.
79. P. Maivald, H.J. Butt, S.A.C. Gould, C.B. Prater, B. Drake, J.A. Gurley, V.B.
Elings and P.K. Hansma, Nanotechnology, 1991, 2, 103.
81. D. Reifer, R. Windeit, R.J. Kumpf, A. Karbach and H. Fuchs, Thin Solid Films,
1995, 264, 148.
82. G.M. Estes, R.W. Seymour and S.L. Cooper, Macromolecules, 1971,4, 452.
83. C.B. Wang and S.L. Cooper, Journal of Applied Polymer Science, 1981, 26, 2989.
84. F. Xiao, D. Shen, X. Zhang, S. Hu and M. Xu, Polymer, 1987, 28, 2335.
86. T.K. Kang, Y. Kim, W.J. Cho and C.S. Ha, Polymer Testing, 1997, 16, 391.
87. A. Genovese and R.A. Shanks, Computational and Theoretical Polymer Science,
2001, 11, 57.
88. D.M. Koenhen, M.H.V. Mulder and C.A. Smolders, Journal of Applied Polymer
Science, 1977, 21, 199.
517
Handbook of Polymer Blends and Composites
91. D. Allard and R.E. Prudhomme, Journal of Applied Polymer Science, 1982, 27, 559.
93. R. Hernandez, J.J. Pena, L. Irusta and A. Santamaria, European Polymer Journal,
2000, 36, 1011.
94. T. Tang, X. Jing and B. Huang, Journal of Macromolecular Science B, 1994, 33,
287.
95. H. Stutz, P. Potschke and U. Mierau, Macromolecular Symposia, 1996, 112, 151.
97. C.P. Papadopoulou and N.K. Kalfoglou, Polymer, 1998, 26, 7015.
99. J.M. Criado, M. Gonzales, A. Ortega and C. Real, Journal of Thermal Analysis,
1988, 34, 1387.
101. T. Ozawa, Bulletin of the Chemical Society of Japan, 1965, 38, 1881.
102. J.H Flynn and L.A Wall, Polymer Letters, 1966, 4, 232.
103. D.J.T. Hill, M.I. Killeen, J.H. ODonnell, P. Pomery, D. St. John and A.K.
Whittaker, Journal of Applied Polymer Science, 1996, 61, 1757.
104. Electronic Properties of Polymers, Eds., J. Mort and G. Pfister, Wiley, New York,
NY, USA, 1982.
106. Organic Semiconducting Polymers, Ed., J.E. Katon, Marcel Dekker, New York,
NY, USA, 1968.
518
Blends and Interpenetrating Networks Based on Polyurethanes
109. M. Seki, K. Sato and R. Yosomiya, Die Makromolekulare Chemie, 1992, 193, 2971.
113. Interpenetrating Polymer Networks, Eds., D. Klempner, L.H. Sperling and L.A.
Utracki, Advances in Chemistry Series No. 239, ACS, Washington, DC, USA,
1994.
114. S.C. Tang, C.P. Hu and S.K. Ying, Polymer Journal, 1990, 22, 70.
115. J.R. Millar, Journal of the Chemical Society, 1960, 263, 1311.
116. IPNs Around the World, Science and Engineering, Eds., S.C. Kim and L.H.
Sperling, Wiley, New York, NY, USA, 1997.
117. Y.S. Lipatov and L.M. Sergeeva, Interpenetrating Polymer Networks, Naukova
Dumka, Kiev, 1979
118. R. Bischoff and S.E. Cray, Progress in Polymer Science, 1999, 24, 185.
119. V. Athawale and S. Kolekar, European Polymer Journal, 1998, 34, 1447
120. F.J. Hua and C.P. Hu, European Polymer Journal, 1999, 35, 103.
121. L.W. Barret and L.H. Sperling, Trends in Polymer Science, 1993, 1, 45.
123. S. Raut and V. Athawale, European Polymer Journal, 2000, 36, 1379.
124. E.E.C. Monteiro and J.L.C. Fonseca, Polymer Testing, 1999, 18, 281.
519
Handbook of Polymer Blends and Composites
126. S. Ajithkumar, N.K. Patel and S.S. Kansara, European Polymer Journal, 2000,
36, 2387.
127. G.C. Marti, E. Enssani and M. Shen, Journal of Applied Polymer Science, 1981,
26, 1465.
128. L.H. Fan, C.P. Hu and S.K. Ying, Polymer Engineering and Science, 1997, 37, 338.
129. L.H. Sperling and V. Mishra, Macromolecular Symposia, 1997, 118, 336.
130. Y.C. Chou and L.J. Lee, Polymer Engineering and Science, 1994, 34, 1239.
132. W.Y. Chiang and D.M. Chang, Proceedings of Antec 94, San Francisco, CA,
USA, 1994, Volume 1, 959.
133. V. Mishra and L.H. Sperling, Journal of Polymer Science, Polymer Physics
Edition, 1996, 34, 883.
134. Y.S. Lipatov, L.V. Karabanova and L.M. Sergeeva, Polymer International, 1994,
34, 7.
135. S. Dadbin, R.P. Burford and R.P. Chapin, Polymer Gels and Networks, 1995, 3,
179.
136. S.C. Kim, D. Klempner, K.C. Frisch, W. Radigan and H.L. Frisch,
Macromolecules, 1976, 9, 258.
137. H. Djomo, A. Morin, M. Damyanidu and G.C. Meyer, Polymer, 1983, 24, 65.
138. V. Mishra, F.E. Du Prez, E. Gosen, E.J. Goethals and L.H. Sperling, Journal of
Applied Polymer Science, 1995, 58, 331.
140. Y. Rharbi, A. Yekta, M.A. Winnik, R.J. DeVoe and D. Barrera, Macromolecules,
1999, 32, 3241.
141. S.C. Kim, D. Klempner, K.C. Frisch, H.L. Frisch and H. Ghirardela, Polymer
Engineering and Science, 1975, 15, 339.
520
Blends and Interpenetrating Networks Based on Polyurethanes
142. K. Kaplan and N.W. Tschoegl, Polymer Engineering and Science, 1975, 15, 343.
144. N. Devia, J.A. Manson, L.H. Sperling and A. Conde, Polymer Engineering and
Science, 1979, 19, 12.
145. W. Ku, J. Liang, K. Wei, H. Liu, C. Huang, S. Fang and W. Wu, Macromolecules,
1991, 24, 4605.
146. X. He, J. Widmaies and G.C. Meyer, Polymer International, 1993, 32, 289.
147. L.H. Sperling, Polymer Engineering and Science, 1985, 25, 517.
148. N. Devia, J.A Manson, L.H. Sperling and A. Conde, Macromolecules, 1979, 12,
360.
149. H. Xiao, Z.H. Ping, J.W. Xie and T.Y. Yu, Journal of Polymer Science, Polymer
Chemistry Edition, 1990, 28, 585.
150. J.R. Ebdon, D.J. Hourston and P.G. Klein, Polymer, 1984, 25, 1633.
151. J.R. Ebdon, D.J. Hourston and P.G. Klein, Polymer, 1988, 29, 1079.
156. H.L. Frisch, K.C. Frisch and D. Klempner, Polymer Engineering and Science,
1974, 14, 646.
157. L.T. Nguyen and N.P. Suh, Polymer Engineering and Science, 1986, 26, 799.
158. S.S. Lee and S.C. Kim, Polymer Engineering and Science, 1991, 31, 182.
159. T.J. Hsu and L.J. Lee, Journal of Applied Polymer Science, 1988, 36, 1157.
521
Handbook of Polymer Blends and Composites
160. T.J. Hsu and L.J. Lee, Polymer Engineering and Science, 1988, 28, 956.
161. Y.S. Yang and L.J. Lee, Macromolecules, 1987, 20, 1490.
162. J.H. Kim and S.C. Kim, Polymer Engineering and Science, 1987, 27, 1243.
163. F.J. Hua and C.P. Hu, European Polymer Journal, 2000, 36, 27.
164. H.L. Frisch, R. Foreman, R. Schwartz, H. Yoon, D. Klempner and K.C. Frisch,
Polymer Engineering and Science, 1979, 19, 284.
165. L.M. Sergeeva, S.I. Skiba and L.V. Karabanova, Polymer International, 1996, 39,
317.
166. G.M. Yenwo, L.H. Sperling, J. Pulido, J.A. Manson and A. Conde Polymer
Engineering and Science, 1977, 17, 251.
167. N. Devia, J.A. Manson, L.H. Sperling and A.J. Conde, Journal of Applied
Polymer Science, 1979, 24, 569.
168. V.G. Kumar, M.R. Rao, T.R. Guruprasad and K.V.C. Rao, Journal of Applied
Polymer Science, 1987, 34, 1803.
169. S. Bai, D.V. Khakhar and V.M. Nadkarni, Polymer, 1997, 38, 4319.
170. S.B. Pandit, S.S. Kulkani and V.M. Nadkarni, Macromolecules, 1994, 27, 4595.
171. M. Rutkowska and A. Eisenberg, Journal of Applied Polymer Science, 1984, 29,
755.
172. K.C. Frisch, D. Klempner, H.X. Xio, E. Cassidy and H.L. Frisch, Polymer
Engineering and Science, 1985, 25, 758.
173. K.H. Hsieh, L.M. Chou and Y.C. Chiang, Polymer Journal, 1989, 21, 1.
174. J.H. Lee and S.C. Kim, Macromolecules, 1984, 17, 2193.
175. J.H. Lee and S.C. Kim, Macromolecules, 1986, 19, 644.
176. H. Chen and J.M. Chen, Journal of Applied Polymer Science, 1993, 50, 495.
177. G.M. Yenwo, J.A. Manson, J. Pulido, L.H. Sperling, A. Conde and N. Devia,
Journal of Applied Polymer Science, 1977, 21, 531.
522
Blends and Interpenetrating Networks Based on Polyurethanes
178. M. Patel and B. Suthar, Journal of Polymer Science, Polymer Chemistry Edition,
1987, 25, 2251.
180. P.L. Nayak, S. Lenka, S.K. Panda and T. Patnaik, Journal of Applied Polymer
Science, 1993, 47, 1089.
181. P. Chakrabarty and B. Das, Journal of Applied Polymer Science, 1996, 60, 2125.
182. P. Nayak, D.K. Mishra, D. Parida, K.C. Sahoo, M. Nanda and S. Lenka, Journal
of Applied Polymer Science, 1997, 63, 671.
186. V. Athawale and S. Kolekar, European Polymer Journal, 1998, 34, 1451.
187. Y.S. Lipatov, T.T. Alekseeva, V.F. Rosovitsky and N.V. Babkina, Polymer
International, 1995, 37, 97.
188. L.R.G. Treloar, The Physics of Rubber Elasticity, 3rd Edition, Clarendon Press,
Oxford, UK, 1975.
191. D.S. Lee, J.H. An and S.H. Kim in Interpenetrating Polymer Networks, Eds., D.
Klempner, L.H. Sperling and L.A. Utracki, Advances in Chemistry Series No.
239, ACS, Washington, DC, USA, 1994.
192. Y.S. Lipatov, G.M. Semenovitich, S.I. Skiba and L.V. Karabanova, Polymer, 1992,
33, 361.
193. Y.S. Lipatov, T.T. Alekseeva, V.F. Rosovitsky and N.V. Babkina, Polymer, 1992,
33, 610.
523
Handbook of Polymer Blends and Composites
194. Y.S. Lipatov, and T.T. Alekseeva, Polymer Communications, 1991, 32, 254.
195. N.P. Chen, Y.L. Chen, D.N. Wang, C.P. Hu and S.K. Ying, Journal of Applied
Polymer Science, 1992, 46, 2075.
196. L.H. Fan, C.P. Hu, Z.P. Zhang and S.K. Ying, Journal of Applied Polymer
Science, 1996, 59, 1417.
197. L.H. Fan, C.P. Hu and S.K. Ying, Polymer, 1996, 37, 975.
198. L.H. Fan, C.P. Hu, Z.Q. Pan, Z.P. Zhang and S.K. Ying, Polymer, 1997, 38,
3609.
199. L.H. Fan, C.P. Hu and S.K. Ying, Polymer Engineering and Science, 1997, 37,
338.
200. L. Hermant and G. Meyer, European Polymer Journal, 1984, 20, 85.
202. N.J. Harrick, Internal Reflection Spectroscopy, Wiley, New York, NY, USA, 1967
204. R. Pernice, K.C. Frisch and R. Navare, Journal of Cellular Plastics, 1982, 3, 121.
205. N.P. Chen, Y.L. Chen, D.N. Wang, C.P. Hu and S.K Ying, Journal of Applied
Polymer Science, 1992, 46, 2075.
206. L.H. Fan, C.P. Hu, Z.Q. Pan, Z.P. Zhang and S.K. Ying, Polymer, 1997, 38,
3609.
207. L.H. Fan, C.P. Hu and S.K. Ying, Journal of Polymer Engineering and Science,
1997, 37, 338.
524
Blends and Networks Containing Silicon-Based Polymers
13
Blends and Networks Containing
Silicon-Based Polymers
13.1 Introduction
The positioning of the silicon atom in the third period and fourth main group of the
periodic table accounts for its chemical behaviour being similar to that of its neighbour,
carbon. However, the higher atomic value and the presence of unoccupied 3d orbitals
give marked differences between the properties of silicon atoms and silicon derivatives
as compared to their carbon analogues. The higher electropositivity and the vacant 3d
orbitals of the silicon atom diminish its tendency to self-linking and induce a strong
affinity for electronegative elements possessing unshared p electrons, especially oxygen.
Thus, the common silicon-based polymers are not the polysilanes but rather the
polysiloxanes. However, Si-Si and Si-C incatenation in polysilanes and polycarbosilanes,
respectively, is also possible.
The properties of silicon-based polymers are strongly dependent on both the structure of
chain backbone (Si-Si, Si-C or Si-O-Si) and the nature of the organic substituents attached
to the silicon atom. Polysilanes are characterised by interesting photo- and electrochemical
behaviour related to the conjugation of electrons into the Si-Si skeleton. Polycarbosilanes,
especially those containing aromatic groups, are generally thermally stable compounds.
In polysiloxanes, the double inorganic (Si-O backbone) - organic (organic radicals linked
to the silicon atom) nature of the macromolecular chains determines a quite unique
combination of properties. Polydimethylsiloxane (PDMS), the most simple and used
siloxane polymer, is characterised by high flexibility, hydrophobicity and physico-chemical
stability, low surface tension and surface energy, low solubility parameter and dielectric
constant and UV transparency. In addition, these properties show only relatively small
variation over a wide range of temperatures. Polysiloxanes also possess film forming
ability, high gas permeability, release action, and chemical and physiological inertness.
This is why Rochow wrote some years ago For decades I avoided mixtures or copolymers
of organic resins with silicone, even abhorred them, because I did not want the pristine
properties of my silicon brain children to be degraded or ruined by adulteration [1].
However, the poor mechanical properties of polysiloxanes impelled the development of
two- and multi-component polymeric materials. Several reasons encouraged the
525
Handbook of Polymer Blends and Composites
a. the wish to impart their useful properties to the newly obtained material and to diminish
the shortcoming of their poor mechanical properties,
In blend systems of silicon-based polymers the low morphological stability of the polymeric
material is a consequence of the immiscibility of blend components that could induce a
macrophase separation. Copolymer networks are composed from two different polymer
sequences chemically linked into a single network. In such systems the crosslinking
prevents the macrophase separation and only a microphase separation is possible. The
IPN contain two different polymer networks (full-IPN) or a linear polymer and a polymer
network (semi-IPN); the structure of the polymeric material is stabilised by crosslinking
and chain entanglement.
This chapter considers the recent developments in blend systems, copolymer networks
and IPN, containing as one of the partners, polysiloxanes, polycarbosiloxanes, polysilanes
or polycarbosilanes possessing the properties presented in Table 13.1.
526
components Silicon-based
polymers
Blends and Networks Containing Silicon-Based Polymers
PPhMS Tg = 85 [5]
= 9.0;
= 26.1
Permeability: O2: 24
POLYCARBOSILOXANES
PTMPS Tg = 15 [8]
Tm = 129
Crystallinity = 45
527
Handbook of Polymer Blends and Composites
POLYCARBOSILANES
PDMSiM Tg = 88 [12]
PSiMS Tg = 22 [13]
PDMSiPh Tg = 15 [8]
Tm = 100-160
Crystallinity = 45
528
Blends and Networks Containing Silicon-Based Polymers
The simplest relationship describing the miscibility of polymer blends is the well-known
Flory-Huggins theory [28, 29] giving the free energy of mixing Gm as a function of
interaction parameter, :
where R = gas constant; T = absolute temperature; V and Vr = total and reference volumes;
Vi (i = 1 or 2) = volume of a chain polymer of component i and i = volume fraction of
component i.
The first two terms in Equation 13.1 represent the combinatorial entropy of mixing. In
most cases, where both blend components are high molecular weight polymers, the entropy
of mixing is negligible and the phase behaviour of the blend are governed by the sign and
529
Handbook of Polymer Blends and Composites
p = p / p*, v = v sp / v*, T = T / T *
pv / T = v1 / 3 / v (1 / 31) 1 / vT
and the free-energy of mixing also contains free-volume contributions, which were
completely neglected by Flory-Huggins theory. For EOS theories based on the Prigogine
statement [31] it is possible to derive an expression for which a priori contains entropical
contributions by using the more simple Flory-Huggins theory to extract the parameter
from EOS theories. However, there are different ways to determine a composition
dependent parameter from these theories and the results may be different. The most
used EOS theories are Flory-Orwoll-Vrij [32-34], Patterson [35, 36] or modified cell
model of Dee and Walsh [37]. All these theories provide the expression of a total interaction
parameter tot as a sum of int (enthalpic contribution accounting for the exchange of
energy of the two blend components) and fv (free-volume contributions):
530
Table 13.2 Thermodynamically characterised blend systems containing silicon-based polymers
Silicon-based polymers
Blend Silicon-based polymer Organic or silicon polymer Blend characteristics Ref.
no.
Code Mw Tg (C) Code Mw Tg (C)
(kg/mol) (kg/mol)
1 PDMSiM 721 88 PIB 1.5-85 81 - Completely miscible for MPIB 1.5 [12]
64 - UCST and LCST to hour glass
behaviour with increasing molecular
Comparing the blend systems based on polyisobutene (PIB) or polystyrene (PS) and their
carbosilane analogues (obtained by the replacement in the backbone of a carbon atom
with a silicon atom) to those containing PDMS (blends 1, 2, 3a and 4 in Table 13.2), the
following general features can be deduced:
The phase behaviour of the polymer blends studied is a result of a balance of interactional
and free-volume contributions. The nature of the backbone rather than the nature of the
constituents controls the repulsive enthalpic interactions between the vinyl polymers
and their carbosilane analogues as int values are similar in the systems studied (blends 1
and 3a, Table 13.2). Also, in PDMSiM/PIB blends int and fv are of the same order of
532
Blends and Networks Containing Silicon-Based Polymers
A computational method based on the Flory-Huggins lattice model, the extended random
copolymer theory and the molecular modelling method were used to show a non-negligible
effect of end groups on the value and temperature of phase separation in blends of
PDMS and polymethylphenylsiloxane of low molecular weight [46].
The variation of interfacial tension with temperature and molecular weight in polymer
blends is strongly influenced by the miscibility of its components, measured by the
segregation strength [50]. Three regimes of segregation are characterised by the differences
between and its value at the critical point, c: weak segregation, c; strong segregation,
1 > >> c; very strong segregation > 1.
533
534
Siloxane blend
to pure PS
2 11.3 or PIB 1.5 or 3.3 Harmonic mean Density UCST > 200 C (increases with [38]
19.6 or 5.0 relation [54] measurements increasing MPIB);
Spinning-drop for PDMS11.3/PIB5, UCST = 262 C
and sessile-drop
measurements
3 3.0 PHxMS 11.3 Mean field theory increases with the length of the [55]
Ising-3D model corresponding tie line
4 2.5 or CR XPS, SEM [56]
150 or
625
5 2.5 or PVC 120 or 190 XPS, SEM The degree of surface segregation [41]
150 or or 260 decreases with increasing MPDMS
625
6 4-177 PEO 41 Sessile-drop [57,
measurements 58]
PHxMS = polyhexylmethylsiloxane; CR = polychloroprene; PVC = polyvinyl chloride; PEO = polyethylene oxide;
XPS = X-ray photoelectron spectroscopy; SEM = scanning electron microscopy; MPDMS: molecular weight of PDMS
Blends and Networks Containing Silicon-Based Polymers
The values and temperature dependence of interfacial tension between PDMS and organic
polymers are strongly dependent on the nature and on the molecular weight of the organic
partner (Figure 13.2). For mixtures of dimethylsiloxane and styrene or
hexylmethylsiloxane oligomers [39, 55] it was possible to measure the interfacial tension
near the critical point. The ODMS/OS blend system is characterised by a value of critical
temperature slightly higher than 100 C.
535
Handbook of Polymer Blends and Composites
The very low solubility parameter and surface energy of polysiloxanes and especially
PDMS, determine their immiscibility with almost all organic polymers. In polymer blends
they migrate at the air contact surface of the material. Since the blend system tends to
decrease its energy by enriching the surface with the component of the lower surface
tension and/or by diminishing the number of energetically disadvantageous contacts of
different components in its bulk by squeezing out one component to the surface [59], the
surface composition of siloxane blends is strongly different from the bulk composition.
XPS associated with SEM proved to be appropriate methods for the investigation of
surface profile of siloxane blends. For both PDMS/CR [56] and PDMS/PVC mixtures
[41] the solvent cast film blends contain PDMS enriched surface layers. PDMS surface
enrichment dramatically decreases with increasing PDMS molecular weight (from 100%
for PDMS2.5 - an oligomer - to less than 50% for PDMS625) and no evidence of a clear
influence of the molar mass of organic polymer upon surface segregation was observed.
The explanation of this strange dependence of surface segregation on the molecular weight
was thought to be the higher mobility of low molecular weight molecules during solvent
evaporation (the samples were prepared by solution casting method).
The migration of the siloxane sequences to the air/blend surface was also observed in
siloxane containing block or graft copolymer/PS systems [60]. A deep insight into the
surface activity of siloxane copolymers was performed by Teyssie and co-workers on the
poly(styrene-block-dimethyl siloxane (poly(S-b-DMS)/PS) system [61]. Both surface
tension (measured by wettability and contact angles) and surface composition (determined
by X-ray photoelectron spectroscopy and secondary ion mass spectrometry) investigations
showed the accumulation of PDMS blocks at the air/polymer interface when small
amounts of the block copolymer (0.2%-2.0%) were added to the PS matrix and practically
complete covering of the surface with PDMS blocks. A rather surprising observation
was the block copolymer enrichment of the interface between polymer and the high
536
Blends and Networks Containing Silicon-Based Polymers
energy substrate (glass, stainless steel or aluminium), a specific behaviour for copolymers
containing a polar sequence, for non equilibrium spinning-cast films. However, after
annealing, the copolymer migrated to the polymer/air interface.
= EF
= AB
F being close to 0.5 for all systems; B = 2.4; A and E are variable parameters that are not
universal; is the relative distance from the critical point.
Another proposed relationship offering a simple means to grasp the effect of polydispersity
correlated with the length of the corresponding tie line, ( = volume fraction):
= A M0.5B
The interfacial tension between two polymer components of a blend can be modified by
the addition in their mixture of low or high molecular weight additives such as solvents
or copolymers with various molecular structures.
The theory of Broseta and Leibler [48] was verified by the addition of a good solvent
(toluene) for both components of the PDMS/PIB blends [38]. A sharp decrease of was
observed with increasing the weight fraction of the solvent. Methyl ethyl ketone (MEK),
a good solvent for PDMS and a poor one for PIB, produces a (wMEK) curve characterised
by a minimum value for the weight fraction wMEK 0.4. A rather inverse effect was
obtained by adding a small amount of an incompatible polymer (PDMS or PS-PDMS
block copolymer) to a phase separated polymer solution of PS in cyclohexane, when an
important increase of was observed [63].
The modification of values of PDMS/PEO [57, 58] and PDMS/PPhMS [64] blends in
the presence of copolymers was also investigated. Different chemical structures of
polymeric additives were determined, i.e., di-, triblock and random copolymers having
537
Handbook of Polymer Blends and Composites
structural units of the same nature as those of the blend components or copolymers with
chemical structures of the blocks different from those of the blend components. The
addition of relatively small amounts of all these copolymers induced a substantial decrease
of . For PDMS/PEO blends decreases from 10 mN/m to 1-4 mN/m by adding 0.2%
by weight of di- or triblock PDMS-PEO copolymers with different architectures [57,
58]. Contrary to theoretical expectations, it was established that the total number of Si-O
units in additives is more decisive for their efficiency than copolymer architecture. PS-
polymethyl methacrylate (PMMA) diblock copolymers proved to be less effective additives
for the same PDMS/PEO blends (the addition of 4% by weight of PS-PMMA copolymer
determines a decrease of only three units for ). The efficiency of PDMS-
methylphenylsiloxane random copolymers was demonstrated by the decreasing of value
of PDMS/PPhMS blend from 2.2 to 1.6 mN/m [64]. Two different kinetic processes were
observed during the equilibration, (a) interdiffusion of the components in the vicinity of
the phase boundary and (b) hydrodynamic relaxation of the droplet.
Very recently Nose and co-workers [65] reported on the influence of oligodimethylsiloxane
additives upon the thermodynamic and interfacial tension behaviour of PDMS/
polytetramethyldisiloxanylethylene blend characterised by weak segregation. The addition
of ODMS produces a decrease of the critical temperature and shifts the - T curves to
lower temperatures as the concentration of the additive increases, suggesting that the
adsorption effects of additive on interfacial tension are weak. Both square gradient theory
and dynamic mean field calculations compare well with the experimental behaviour.
The surface activity of PDMS containing block and graft copolymers in bicomponent
polymer/copolymer systems at air/polymer interfaces was also determined in relatively
older reports [7, 66-73]. The surfactant behaviour was observed to be dependent on
copolymer architecture and on the nature of the polymeric partner of PDMS in the
copolymer structure. The complete surface coverage by PDMS occurs at copolymer
concentrations as low as 0.1%-2% by weight.
538
Blends and Networks Containing Silicon-Based Polymers
Very few mixtures composed of low molecular weight silicon-based oligomers were found
to be miscible with other polymers in narrow intervals of compositions. Because of the
increased similarity between the chemical structures of carbosilane polymers and organic
analogues, a higher miscibility (larger composition intervals at reasonable temperatures)
was observed for their blend systems (see Table 13.2). Moreover, blend systems of PDMS
of different molecular weights [51] or PDMS isotopic blends [40] are characterised by
small but positive values of interaction parameter, UCST behaviour and phase separation
at room temperature for a symmetric composition. As for blend systems based on polysilane
components, a negative value of parameter and a poor degree of miscibility was found
for polymethylpropylsilane/polydi(n-hexyl)silane system, while polydi(n-hexyl)silane and
polymethyloctadecylsilane yield immiscible blends [77]. Existing methods to improve blend
miscibility using specific molecular interactions such as hydrophobic interaction or hydrogen
bonding were not often used for blends containing silicon-based polymers. However, an
increased miscibility window was reported for blends of 4-vinylphenyldimethylsilanol
homopolymer or its styrene copolymers (hydrogen donors) with polybutyl methacrylate
[24, 25] or poly(N-vinylpyrrolidone) [78] (hydrogen acceptors based on ester or amide
carbonyl groups). Highly miscible blends were also obtained by solvent-casting of
polyhydroxymethylsiloxane and poly(N-vinylpyrrolidone) or PEO mixtures [6].
An increased miscibility window based on hydrogen bond formation was also reported
for other organic polymer/siloxane copolymer mixtures. For example, miscible blends of
poly(siloxane-imide) segmented copolymers and polybenzimidazole were prepared and
evaluated as potential high performance aerospace materials [27].
Rigid-rod polyimides grafted with PDMS were synthesised and proved to possess a good
compatibility with linear PDMS elastomers for PDMS side chain in the grafted copolymer
longer than 40 siloxane units [79]. Electron microscopic imaging techniques were used
to study the microphase separated mixtures of poly(styrene-phenylmethylsiloxane) block
copolymer with both styrene and siloxane homopolymers [80].
539
Handbook of Polymer Blends and Composites
As a result of the hydrogen bonding between the vinyl sequences, for an almost 1:1 MAA/
NVP molar ratio, a single Tg for a complexed specimen at a value with 40 C higher than
the Tg of PMAA blocks was observed. Copolymer complexes characterised by MAA/
NVP ratios far away from the equivalence present DSC curves with three different Tg
values in the positive temperature interval. They correspond to uncomplexed polyvinyl
sequences (PMAA and PNVP uncomplexed sequences) and to PMAA-PNVP complexed
domains, respectively.
Figure 13.3 Typical DSC curves for (1) PDMS-poly(N-vinyl pyrrolid) (PNVP)
copolymer; (2) PDMS-polymethylacrylic acid (PMAA) copolymer; (3) mixture of (1)
and (2), MAA/N-vinyl pyrrolidone (NVP) = 1/1 molar ratio; (4) mixture of (1) and
(2), MAA/NVP = 0.6/1 molar ratio (Tg = glass transition temperature; COMPL =
PNVP-PMAA complexed specimen)
540
Blends and Networks Containing Silicon-Based Polymers
13.2.4 Rheology
Rheology of polymer blends is a subject of great interest, particularly for their processing
behaviour in the two-phase state. Recent publications provide correlations of rheology
parameters and interfacial tension [82-84], descriptions of structure development during
spinodal decomposition [85] or as a result of flow stress [86-97] and the influence of the
phase state on the rheology in the vicinity of phase separation or far from the phase
separation [94-103]. Near the phase separation temperature and far from the Tg the
linear viscoelastic behaviour of a polymer blend is controlled by critical concentration
fluctuations and the rheological determination of both the binodal and spinodal diagrams
is possible by an appropriate quantitative account of these fluctuations.
In the phase separated state the blend system consists of isolated droplets of the less
viscous PPhMS-rich phase dispersed in the matrix of the more viscous PS-rich coexisting
phase and follows the scaling relation established by Onuki [105] (Equation 13.5):
541
Handbook of Polymer Blends and Composites
Based on their thermal stability, silicones were earlier tested as non-halogenated fire
retardants in blends with polyolefins but their efficiency was proved to be poor in
uncrosslinked polyolefins [110, 111].
Generally, the thermal stability of polymer blends is a result of the behaviour and relative
proportion of blend components. However, some peculiarities not directly linked to their
partners thermal stability could arise after blending. A deep insight into the effect of
blending of different vinyl polymers (low density polyethylene, polyethyl acrylate and
ethylene-ethyl acrylate copolymer) with vinyl-terminated PDMS was found by McNeill
and Mohammed [112] by using different experimental techniques (thermogravimetry,
volatilisation analysis, DSC). Both polymer components were stabilised by mixing and
the degradation of the blend systems was observed to occur at temperatures higher than
predicted, accounting for the interaction between blend partners before starting the
decomposition processes. Mass spectrometry data on the resulting volatile products led
to a stabilisation mechanism based on the trapping of the formed radicals by the siloxane
chains through their methyl or vinyl groups or on the reaction of ester/carboxylic groups
with siloxane units (Scheme 13.2).
Scheme 13.2 Reactions involved in the stabilisation of vinyl polymers by blending with
vinyl-terminated PDMS [112]
542
Blends and Networks Containing Silicon-Based Polymers
Highly oriented fibres were prepared by gel drawing of mixtures of ultrahigh molecular
weight polyethylene and poly(di-n-pentylsilane) known to form columnar liquid crystalline
phases [10, 11]. The mechanical deformation of the irradiated blends showed the
improvement of the fibre strength [11].
543
Handbook of Polymer Blends and Composites
The introduction of flexible siloxane sequences into the main chain of polymer electrolytes
was proved to provide an increase of the ionic conductivity [123-126]. An improvement
of mechanical properties of rigid electroconducting polymers without the degradation of
the electrical characteristics was also reported [127]. The same beneficial influence was
observed by blending siloxane polymers (PDMS-phenylmethylsiloxane copolymers
(PDMPhMS) of different molecular weights) or siloxane-organic copolymers
(polydimethylsiloxane-g-alkylene oxide, PDMS-AO) with poly(etherurethane) (PEU)
[128]. An increase of the ionic conductivity from 10-7 up to 10-5 S/cm at room temperature
was achieved. Blend morphology analysis (PEU/PDMPhMS blends are phase separated,
while PEU/PDMS-AO blends are compatible), dielectric constant and ion mobility
measurements suggested that the improvement of the conductivity was determined by
the increase in ion mobility that compensates for the decrease of ion density in incompatible
blends and by the augmentation of both characteristics in compatible electrolytes.
The surface resistivity of polypropylene and PS was reduced from 1017 to 1016 and
1013 ohmcm, respectively, by their blending with low proportions of polydimethylsilylene-
co-diphenylsilylene [9].
544
Blends and Networks Containing Silicon-Based Polymers
An extensive literature, including review papers [7, 142], deals with the synthesis,
morphology, properties and application of block and graft copolymers containing silicon-
based sequences. Section 13.3 is concerned only with crosslinked structures such as
copolymer networks and interpenetrating networks.
Different polymers were considered as partners for such complex structures. Thus, Shiomi
and co-workers [143] reported the synthesis and swelling behaviour of a phase separated
two-component polymer network of PS and PDMS as copolymer components starting
from telechelic PDMS (pyrrolidinium end groups) and PS (containing a low proportion
of main chain acrylic acid groups) and Hill and co-workers [144] described the dynamic
mechanical properties of crosslinked siloxane-divinylbenzene copolymers.
The combination of siloxane polymers and modified epoxy resins focused the interest of
fundamental and applied research in this area. Epoxy resins are widely used in electronic
applications, because of their properties, excellent heat, solvent, moisture and chemical
resistance, good adhesion to many substrates, superior mechanical and electrical
characteristics. To encapsulate microelectronic devices through crosslinked protective
layers, epoxy resins seem to be one of the preferred materials. To flexibilise such rigid
structures, the incorporation of siloxane moieties is an alternative. In this respect, the
545
Handbook of Polymer Blends and Composites
literature reports the development of naphthalene-type epoxy resins modified with amino-
terminated PDMS to reduce internal stress while conserving a high Tg [145, 146]. The
crosslinked material was evaluated for semiconductor encapsulation. The same effect
was observed for three functional epoxy resins modified by hydrosilation with side chain
methylhydrogensiloxanes [147], the copolymer network being proposed for electronic
applications. Zheng and co-workers used an epoxy-terminated PDMS as modifier for
bisphenol A-type epoxy resins [148]. The crosslinking was performed with aromatic
amine curing agents. It was observed that the toughness of the epoxy resin was enhanced
without loss of the thermal resistance and of substantial modulus. The crosslinked epoxy-
rich phase (matrix) presented a higher Tg when compared to pure epoxy resins cured
with aromatic amines.
Polysiloxanes possessing different organic substituents to the silicon atoms were widely
used as components of IPN being combined with many organic polymers such as vinyl
or diene polymers (PS, polymethyl methacrylate, polymethacrylic acid, polyethylene,
polypropylene, melt-processable fluorocarbons, polybutadiene), PU, PEO, polyoxazolines,
polyesters, polycarbonates, polyamides. All types of IPN were obtained, full- and semi-
IPN, by both sequential and simultaneous preparation, as well as thermoplastic, latex
and gradient IPN [157].
As a consequence of their strong incompatibility with almost all organic polymers, the
formation of siloxane-based IPN presents peculiar aspects. Special care has to be taken
to ensure the appropriate dispersion of the components all along the network formation
process, i.e., to avoid gross phase separation before the network formation. Owing the
propensity of siloxane chains to migrate to the air surface of the polymeric materials, to
stabilise the bulk structure of IPN, the siloxane component is introduced as crosslinked
polymer. The crosslinked polysiloxane is formed independently of the matrix resin thus
546
Blends and Networks Containing Silicon-Based Polymers
The hydrosilation reaction presents the advantages of a high rate, a high final conversion
and the absence of low molecular weight by-products. The condensation of silanol groups
is accompanied by the evolution of low molecular weight alcohols or hydrochloric acid.
The simultaneous presence of radicals and oxygen (implied in the crosslinking of organic
polymer) was proved to diminish the efficiency of stannous octoate catalyst [158].
The morphology and properties of IPN are depending on IPN type and processing
technique, on polymer compatibility and interfacial tension, on crosslinking density and
relative proportion of components. The compatibility between components controls the
degree of phase separation and of molecular interpenetration. The modification of the
chemical nature of the substituents to the silicon atoms was used to achieve an appropriate
degree of mixing of IPN components, in other words, to overcome their extreme
incompatibility which results in a gross phase separation or the full solubility leading to
alloys with poor properties. The degree of phase separation is also controlled by
crosslinking density. Usually, the increase of this parameter determines the decrease of
domain size and the augmentation of molecular interpenetration. The crosslinking density
of siloxane component is determined by the molecular weight of linear polymer precursor
and/or by its functionality.
Every interpenetrating system has to be treated in a quite specific manner. While for
some vinyl monomers (styrene) the swelling of a preformed siloxane network into the
monomer and the formation of the IPN by the subsequent polymerisation of the vinyl
547
Handbook of Polymer Blends and Composites
monomer is quite evident, for polar monomers one has to use a common solvent to
avoid phase separation. For IPN made from mixtures of prepolymers, the
compatibilisation of the system is achieved by linking different organofunctional groups
to the siloxane chains. Despite the method of preparation, the final idea is to obtain
materials with a high degree of the entanglement of different macromolecular chains in
order to improve the mechanical or surface properties of the final material.
13.4 Conclusion
It was the intention of this review presentation to underline the role and importance of
silicon-based macromolecular structures in building up new and performant polymeric
materials such as blend systems and copolymer/interpenetrating networks. All these
materials take advantage of the quite unique properties of the Si-O, Si-Si and Si-C skeleton
and are highly diversified through the attachment of a large variety of organic radicals to
the silicon atoms. While for polymer blends the applications are limited by the strong
incompatibility of mixture components, in copolymer and interpenetrating networks the
crosslinking and/or permanent entanglement determine a more stable macrostructure
and long-term end use properties.
548
Table 13.4 Interpenetrating polymer networks
No. Organic Siloxane polymer Properties Applications Ref.
polymer
POLYSILOXANE/POLYSILOXANE IPN
1 - Polysiloxane 1 Elastomeric properties Model IPN (static moduli [158-
Polysiloxane 2 - network structure 164]
References
2. W. Noll, Chemistry and Technology of Silicones, Academic Press, New York, NY,
USA, 1968.
3. Polymer Handbook, Eds., J. Brandrup, E.H. Immergut and E.A. Grulke, John
Wiley & Sons, Inc., New York, NY, USA, 1999.
4. I. Yilgor and J.E. McGrath in Reactive Oligomers, Eds., F.W. Harris and H.J.
Spinelli, ACS Symposium Series, 1985, 282, Washington, DC, USA, Chapter 14.
6. S. Lu, M.M. Melo, J. Zhao, E.M. Pearce and T.K. Kwei, Macromolecules, 1995,
28, 4908.
8. M.M. Werlang, M.A. De Aranjo, S.P. Nunez and I.V.P. Yoshida, Journal of
Polymer Science, Part B: Polymer Physics, 1997, 35, 2609.
10. S. Sheiko, H. Frey and M. Mller, Colloid and Polymer Science, 1992, 270, 440.
11. M. Mller, H. Frey and S. Sheiko, Colloid and Polymer Science, 1993, 271, 554.
12. R.D. Maier, M. Kopf, D. Maeder, F. Koopmann, H. Frey and J. Kressler, Acta
Polymerica, 1998, 49, 356.
14. S.H. Jacobson, D.J. Gordon, G.V. Nelson and A. Balazs, Advanced Materials,
1992, 4, 198.
15. O. Olabisi, L.M. Robenson and M.T. Shaw, Polymer - Polymer Miscibility,
Academic Press, New York, NY, USA, 1979.
16. Polymer Blends, Eds., D.R. Paul and S. Newman, Academic Press, New York, 1978.
552
Blends and Networks Containing Silicon-Based Polymers
18. R.G. Kirste and B.R. Lehnen, Die Makromolekulare Chemie, 1976, 177, 1137.
19. P. Bajaj, D.C. Gupta and S.K. Varshney, Polymer Engineering and Science, 1983,
23, 820.
22. G. Grigorescu, S. Ioan, M. Pinteala and B.C. Simionescu, Journal of the Serbian
Chemical Society, 1998, 63, 961.
24. S. Lu, E.M. Pearce and T.K. Kwei, Journal of Macromolecular Science A, 1994,
31, 1535.
25. S. Lu, E.M. Pearce and T.K. Kwei, Journal of Polymer Science, Part A: Polymer
Chemistry, 1994, 32, 2607.
27. C.A. Arnold, D.H. Chen, Y.P. Chen, R.O. Waldbauer, M.E. Rogers and J.E.
McGrath, High Performance Polymers, 1990, 2, 83.
28. M.L. Huggins, Annals of the New York Academy of Science, 1942, 43,1.
30. I.C. Sanchez in Polymer Blends, Eds., D.R. Paul and S. Newman, Academic Press,
New York, 1978.
32. P.J. Flory, R.A. Orwoll and A. Vrij, Journal of the American Chemical Society,
1964, 86, 3507.
553
Handbook of Polymer Blends and Composites
33. P.J. Flory, R.A. Orwoll and A. Vrij, Journal of the American Chemical Society,
1964, 86, 3515.
34. P.J. Flory, Journal of the American Chemical Society, 1965, 87, 1833.
37. G.T. Dee and D.J. Walsh, Macromolecules, 1988, 21, 811 and 815.
40. G. Beaucage, S. Sukumaran, S.J. Clarson, M.S. Kent and D.W. Schaefer,
Macromolecules, 1996, 29, 8349.
41. I.O. Volkov, A.I. Pertsin, L.V. Filimonova, M.M. Gorelova and E.M. Belavtseva,
Polymer Science, Series B, 1997, 39, 310.
42. C.M. Kuo and S.J. Clarson, Macromolecules, 1992, 25, 2192.
43. C.M. Kuo and S.J. Clarson, European Polymer Journal, 1993, 29, 661.
44. C.M. Kuo and S.J. Clarson, Polymer, 1994, 35, 4623.
45. C.M. Kuo and S.J. Clarson, Polymer, 2000, 41, 5993.
48. D. Broseta and L. Leibler, Journal of Chemical Physics, 1987, 87, 7248.
49. A. Vrij, Journal of Polymer Science, Part A: Polymer Chemistry, 1968, 6, 1919.
50. G.H. Fredrickson, Physics of Polymer Surfaces and Interfaces, Ed., I.C. Sanchez,
Butterworth - Heinemann, London, UK, 1992.
51. G.T. Dee and B.B. Sauer, Macromolecules, 1993, 26, 2771.
554
Blends and Networks Containing Silicon-Based Polymers
56. M.M. Gorelova, A.J. Pertsin, I.O. Volkov, L.V. Filimonova and E.S. Obolonkova,
Journal of Applied Polymer Science, 1996, 60, 363.
60. Y. Cai, D. Gardener and G.T. Caneba, Journal of Adhesion Science and
Technology, 1999, 13, 1017.
61. S. Petitjean, G. Ghitti, G. Fayt, R. Jerome and P. Teyssie, Bulletin des Societes
Chimique Belges, 1990, 99, 997.
63. A. Schneider and B.A. Wolf, Macromolecular Rapid Communications, 1997, 18,
561.
64. A. Stammer and B.A. Wolf, Macromolecular Rapid Communications, 1998, 19, 123.
66. Y. Kawakami, R.A. Murthy and Y. Yamashita, Polymer Bulletin, 1983, 10, 368.
70. N.M. Patel, D.W. Dwight, J.L. Hedrick, D.C. Webster and J.E. McGrath,
Macromolecules, 1988, 21, 2689.
555
Handbook of Polymer Blends and Composites
72. I. Yilgor, W.P. Steckle Jr., E. Yilgor, R.G. Freelin and J.S. Riffle, Journal of
Polymer Science, Polymer Chemistry Edition, 1989, 27, 3673.
73. S.D. Smith, J.M. DeSimone, H. Huang, G. York, D.W. Dwight, G.L. Wilkes and
J.E. McGrath, Macromolecules, 1992, 25, 2575.
74. M.M. Gorelova, A.J. Pertsin, V.Y. Levin, L.I. Makarova and L.V. Filimonova,
Journal of Applied Polymer Science, 1992, 45, 2075.
75. V.S. Popkov, G.G. Nikiforova, I.M. Raygorodsky and I.P. Storozhuk, Polymer
Science, 1995, 37B, 428.
76. V.S. Popkov, G.G. Nikiforova, V.G. Nikolsky, I.A. Krasotkina and E.S.
Obolonkova, Polymer, 1998, 39, 631.
78. S. Lu, E.M. Pearce and T.K. Kwei, Polymer Engineering and Science, 1995, 35,
1113.
79. M. Itoh and I. Mita, Journal of Polymer Science, Part A: Polymer Chemistry,
1994, 32, 1581.
80. A. Du Chesne, G. Lieser and G. Wegner, Colloid and Polymer Science, 1994, 272,
1329.
81. V. Harabagiu, New Initiators for the Synthesis of Polymers and Copolymers, Gh.
Asachi Technical University, Iasi, Romania, 1992. [Ph. D. Thesis]
85. J. Lauger, R. Lay, S. Maas and W. Gronski, Macromolecules, 1995, 28, 7010.
86. I. Vinckier, P. Moldenaers and J. Mewis, Journal of Rheology, 1996, 40, 613.
87. J.A. Zawada, G.G. Fuller, R.H. Colby, L.J. Fetters and J.E.L. Roovers,
Macromolecules, 1994, 27, 6851.
556
Blends and Networks Containing Silicon-Based Polymers
88. J.A. Zawada, G.G. Fuller, R.H. Colby, L.J. Fetters and J.E.L. Roovers,
Macromolecules, 1994, 27, 6861.
89. R.M. Kannan and J.A. Kornfield, Journal of Rheology, 1994, 38, 1127.
90. B.H. Arendt, R.M. Kannan, M. Zewail, J.A. Kornfield and S.D. Smith,
Rheologica Acta, 1994, 33, 322.
92. S. Mani, M.F. Malone and H.H. Winter, Macromolecules, 1992, 25, 5671.
93. R. Wu, M.T. Shaw and R.A. Weiss, Journal of Rheology, 1992, 36, 1605.
95. R. Horst and B.A. Wolf, Rheologica Acta, 1994, 33, 99.
96. E.K. Hobbie, A.I. Nakatari and C.C. Han, Modern Physics Letters B, 1994, 8, 1143.
97. M.L. Fernandez, J.S. Higgins and S.M. Richardson, Polymer, 1995, 36, 931.
103. S.K. Kumar, R.H. Colby, S.H. Anastasiadis and G. Fytas, Journal of Chemical
Physics, 1996, 105, 3777.
106. V.T. Tsakalos, P. Navard and E. Peuvrel-Disdier, Journal of Rheology, 1998, 42,
1403.
557
Handbook of Polymer Blends and Composites
110. J.L. Falender, S.E. Lindsey and J.C. Saam, Polymer Engineering and Science,
1976, 16, 54.
111. A.L. Schroll and M.R. MacLaury, Journal of Applied Polymer Science, 1984, 29,
3883.
112. I.C. McNeill and M.H. Mohammed, Polymer Degradation and Stability, 1995,
50, 285.
113. I.C. McNeill and S. Bassan, Polymer Degradation and Stability, 1993, 39, 139.
114. S. Zulfigar and S. Ahmad, Polymer Degradation and Stability, 1999, 65, 243.
115. S. Zulfigar and S. Ahmad, Polymer Degradation and Stability, 2001, 71, 299.
119. X. Chen, J.A. Gardella, T. Ho and K.J. Wynne, Macromolecules, 1995, 28, 1635.
120. R. Benrashid and G.L. Nelson, Journal of Polymer Science, Part A: Polymer
Chemistry, 1994, 32, 1847.
121. D.J.T. Hill, M.I. Killeen, J.H. ODonnell, P. Pomery, D. St. John and A.K.
Whittaker, Journal of Applied Polymer Science, 1996, 61, 1757.
122. T. Bremner, D.J.T. Hill, M.I. Killeen, J.H. ODonnell, P.J. Pomery and D. St.
John, Journal of Applied Polymer Science, 1997, 65, 939.
123. P.G. Hall, G.R. Davies, J.E. McIntyre, I.M. Ward, D.J. Bannister and K.M.F.
LeBrocq, Polymer Communications, 1986, 27, 98.
558
Blends and Networks Containing Silicon-Based Polymers
124. D. Fish, I.M. Khan, E. Wu and J. Smid, British Polymer Journal, 1988, 20, 281.
126. Z. Ogumi, Y. Uchimoto and Z. Takehara, Solid State Ionics, 1989, 35, 417.
129. P. Xie, L. Sun, D. Dai, D. Liu, Z. Li and R. Zhang, Molecular Crystals and
Liquid Crystals, 1995, 269, 75.
132. K.J. Quinn and J.M. Courtney, British Polymer Journal, 1988, 20, 25.
133. B.D. Ratner in Treatise on Clean Surface Technology, Volume 1, Ed., K.L. Mittal,
Plenum Press, New York, NY, USA, 1987, p.247.
135. M.D. Lelah and S.L. Cooper, Polyurethanes in Medicine, CRC Press, Boca Raton,
FL, USA, 1986.
138. L. Tang, M-S. Shen, T. Chu and Y.H. Huang, Biomaterials, 1999, 20, 1365.
139. R.L. Powers, Rubber and Plastics News, 1999, 28, 21, 5.
140. G.D. Friends, J.F. Kunzler and R.M. Ozark, Macromolecular Symposia, 1995,
98, 619.
559
Handbook of Polymer Blends and Composites
142. B.C. Simionescu, V. Harabagiu and C.I. Simionescu in The Polymeric Materials
Encyclopedia: Synthesis, Properties and Applications, Ed., J.C. Salamone, CRC
Press, Boca Raton, FL, USA, 1996, 10, 7751.
144. D.J.T. Hill, M.C.S. Perera, P.J. Pomery and H.K. Toh, Polymer, 2000, 41, 9131.
145. C.S. Wang and M-C. Lee, Journal of Applied Polymer Science, 1998, 70, 1907.
146. J-Y. Shieh, T-H. Ho and C-S. Wang, Die Angewandte Makromolekulare Chemie,
1997, 245, 125.
147. L-L. Lin, T-H. Ho and C-S. Wang, Polymer, 1997, 38, 1997.
157. R. Bischoff and S.E. Cray, Progress in Polymer Science, 1999, 24, 185.
560
Blends and Networks Containing Silicon-Based Polymers
158. X.W. He, J.M. Widmaier, J.E. Herz and G.C. Meyer, Polymer, 1989, 20, 364.
159. S. Wang and J.E. Mark, Journal of Polymer Science, Part B: Polymer Physics,
1992, 30, 801.
161. W. Shuhong, X. Ping and J.E. Mark, Macromolecules, 1991, 24, 6037.
162. E.E. Hamurcu and B.M. Baysal, Polymer, 1993, 24, 5163.
164. G.J. Gibbons and D. Holland, Journal of Sol Gel Science and Technology, 1997,
8, 599.
165. D.W. McCarthy, J.E. Mark and D.W. Schaefer, Journal of Polymer Science, Part
B: Polymer Physics, 1998, 36, 1167.
166. X.W. He, J.M. Widmayer, J.E. Hertz and G.C. Meyer, Polymer, 1992, 33, 866.
167. X.W. He, J.M. Widmayer, J.E. Hertz and G.C. Meyer in Advanced
Interpenetrating Polymer Networks, Eds., D. Klempner and K.C. Frisch,
Technomic Publishing Company, Lancaster, PA, 1994, 4, 321.
169. H.L. Frisch, L. Wang, W. Huang, Y.H. Hua, H.X. Xiao and K.C. Frisch, Journal
of Applied Polymer Science, 1991, 43, 475.
170. T.C. Gilmer, P.K. Hall, H. Ehrenfeld, K. Wilson, T. Bivens, D. Clay and C.
Endreszl, Journal of Polymer Science, Part A: Polymer Chemistry, 1996, 34,
1025.
171. J-R. Caille, inventor; Dow Corning Ltd., assignee; FR 2 270 289, 1997.
174. C.J.T. Landry, B.K. Coltrain, J.A. Wesson, N. Zumbulyadis and J.L. Lippert,
Polymer, 1992, 33, 1496.
561
Handbook of Polymer Blends and Composites
175. F.O. Ecshbach and S.J. Huang, Polymer Preprints, 1992, 33, 2, 526.
176. D. Knauss, T. Yamamoto and J-E. McGrath, Polymer Preprints, 1992, 33, 2, 988.
177. J.P. Kennedy and G.C. Richard, Macromolecules, 1993, 26, 567.
178. B. McGarey, A.D.W. McLenaghan and R.W. Richards, British Polymer Journal,
1989, 21, 227.
179. C. Robert, C. Bunel and J-P. Vairon, Inventors; Essilor International, assignee;
E.P. 643 083, 1997.
180. W. Yin, J. Li, J. Wu and T. Gu, Journal of Applied Polymer Science, 1997, 64,
903.
181. W. Yin, T. Gu, H. Liu and B. Yin, Polymer, 1997, 38, 5173.
182. E.E. Hamurcu and B-M. Baysal, Macromolecular Chemistry and Physics, 1995,
196, 1261.
183. M.E. Dillon and D.K. Lange, Polymer Materials Science and Engineering,1990,
62, 814.
184. M.E. Dillon and D.K. Lange, Polymer Materials Science and Engineering, 1991,
65, 84.
185. E. Pearce, T.K. Kwei and S. Lu, Polymer Preprints, 1997, 38, 2, 392.
186. T. Miiata, J.I. Higuchi, H. Okuno and T. Uragami, Journal of Applied Polymer
Science, 1996, 61, 1315.
187. D.S. Lee, W.T. Kim, H.J. An, S.C. Kim and J. Kim, Pollimo, 1994, 18, 815.
189. B. Arkles and J. Crosby, Advanced Chemistry Series, 1990, 224, 181.
190. M. Zolotnitsky, Proceedings of ANTEC 95, Boston, MA, USA, 1995, Volume 3,
3576.
191. S.A. Ali, D.J. Hourston, K. Manzour and D.F. Williams, Journal of Applied
Polymer Science, 1995, 55, 733.
562
Blends and Networks Containing Silicon-Based Polymers
192. H. Xiao, Z.H. Ping, J.W. Xie and T.Y. Yu, Journal of Polymer Science, Part A:
Polymer Chemistry, 1990, 28, 585.
193. S. Boileau, L. Bouteiller, R.B. Khalifa, Y. Liang and D. Teyssie, Polymer Preprints,
1998, 39, 1, 457.
194. S. Boileau, L. Bouteiller, R.B. Khalifa, Y. Liang and D. Teyssie in Silicones and
Silicone-Modified Materials, Eds., S.J. Clarson, J.J. Fitzgerald, M.J. Owen and
S.D. Smith, Washington, DC, USA, ACS Symposium Series, 2000, 729, 383.
196. L. Price, R. Ryntz, K. Frisch, H. Xiao, V. Gunn, R. van den Heuvel, K. Baars and
H. van den Reijen, Journal of Coatings Technology, 1996, 68, 65.
197. P-H. Sung and C-Y. Lin, European Polymer Journal, 1997, 33, 231.
198. S. Marturunkakul, J. Kumar and S.K. Tripathy, Polymer Materials Science and
Engineering, 1993, 69, 436.
200. C.J.T. Landry, B.K. Coltrain, J.A. Wesson, N. Zumbulyadis and J.L. Lippert,
Polymer, 1992, 33, 1496.
201. W. Huang and H.L. Frisch, Die Makromolekulare Chemie - Supplement, 1989,
15, 137.
202. H.L. Frisch and W. Huang, Journal of Polymer Science, Part A: Polymer
Chemistry, 1991, 29, 131.
204. A. Du Chesne, K. Wenke, G. Lieser and G. Wenz, Acta Polymerica, 1997, 48,
142.
563
Lignin-Based Blends
14
Lignin-Based Blends
14.1 Introduction
The rapid development of petrochemistry in the late five decades was dominated by the
research and production of the synthetic polymers to the detriment of compounds arising
from renewable resources. Today, several reasons justify the necessity to pay an increasing
attention to the biomass as a provider of micro- and macromolecular products, namely:
ii. the renewable character of these sources independently on geographical and geophysical
situations;
iv. the fact that solar energy ensures continuously their regeneration. However, research
and production efforts in the direction of the biomass complex upgrading are still
modest in comparison with its enormous potential, both quantitatively and
qualitatively, for the three major compounds, cellulose, lignin and hemicelluloses.
Lignin is the main aromatic component of the vegetable tissues; it represents about 20-
30 wt% of mass of higher plant tissues and it exists into cell walls and intercellular
spaces. The lignin notion does not reflect a defined structure substance, it refers to a
family of heterogenous biopolymers resulting from the oxidative polymerisation of: trans-
coniferyl, trans-sinapyl and trans-coumaryl alcohols, that contain limited number of
branches and/or crosslinks [1, 2]. About 10,000-12,000 scientific papers, patents, and
books have been published concerning this important natural polymer.
From wood, lignin may be produced in many ways using different procedures such as:
sulfite [lignosulfonate (SSL)] and alkaline (soda or Kraft) pulping and by processes based
on the involvement of mineral acid (acid hydrolysis lignin, AHL), by water and steam
treatments at various temperatures and pressures (autohydrolysis and steam explosion
565
Handbook of Polymer Blends and Composites
lignin, SEL), organic solvent mixtures (organosolv lignin, OSL) and by mechanical wood
milling (MWL). Production of lignin products is concentrated in a few companies and is
dominated by lignosulfonate derived by spent sulfite liquor. In Table 14.1 are listed the
main lignin producers.
The recent interest concerning renewable resources, and also, the concern about reducing
the pollution of our environment have encouraged many efforts into researching the
potential applications of lignins as a phenolic raw material and as a structural material.
Efforts devoted to incorporate lignin preparations into useful polymeric materials have
started in 1944, but intensive work has been done in this field, especially in the last
twenty years.
Thus, for the first time a laminated plastic product named Tomlinite that contains a
lignin adhesive was developed in 1944 by Tomlinson [3]. In the following period, studies
on the lignin had continued to increase the production of lignin-based materials.
During 1975-1980, many more studies have been dedicated to the examination of those
lignin properties such as polyfunctionality, brittleness and rigidity (glassy nature) that
appear to be useful for the development of thermosetting and thermoplastic materials
[4-9]. In 1990-2000, significant advances have been made in the study of compatibility
and structure-properties relationship of lignin-based materials [10-18] (Scheme 14.1).
In commercial applications epoxy resins are rarely used without the incorporation of
some other materials. Thus, a way to enhance their performance is by providing additional
mechanical properties or modifying the physical characteristics of the blends. Several
approaches have been chosen to incorporate non-modified or chemically modified lignin
(see Table 14.2) into epoxy resins (ER), aiming at partially replacing oil-based materials.
In 1988, Feldman and co-workers [19] obtained an adhesive system by simply blending
of the kraft lignin powder (up to 20% lignin) with the liquid epoxy prepolymer and
hardener.
DSC, DMA for 5%-20% lignin content in epoxy resin-lignin blends shown a single glass
transition temperature (Tg), which is characteristic for monophasic systems [20] but two
Tgs for blends having higher content of lignin [21, 22]. The significant improvement in
the adhesion of the lignin-epoxy resin blends (LER) to aluminium substrate could be
explained by an interaction on the one hand between lignin-polyamine hardener and on
the other hand between lignin-epoxy prepolymer (LE) [19, 22, 23].
566
Lignin producers
Lignin-Based Blends
567
Handbook of Polymer Blends and Composites
568
Table 14.2 Epoxy resin (ER) and lignin or epoxy-modified lignin-based blends
System Investigation methods General remarks Source
Lignin-Based Blends
Lignin: NH4-LS;Ca-LS; - thermo-optical analysis; - good compatibility after crosslinking 32]
Phenol; 2,2 - thermogravimetry; at high temperatures;
bis(4-hydroxyphenylpropane); - optical and electronic microscopy; - homogeneous structure;
phenolformaldehyde resin; - mechanical and dielectric - moulding mass and adhesive films;
Hardening agents: anhydride, properties measurements; - good electric and mechanical properties;
amine, imide - biodegradability tests - high biodisintegration rate of binary
569
The study on the epoxy resin-lignin blends was extended using different lignin types, epoxy
prepolymers, hardeners and other additives. The results obtained show that epoxy resin-
lignin blends containing hardwood lignin (Tomlinite (TO), Eucalin (EU), SEL) separated
by the Kraft process or isolated by steam explosion, impart a better adhesion than those
having softwood lignin (Indulin AT (AT)) [30]. The improvement can be correlated with
the structure and the molecular weight of lignin fraction. A significant improvement of the
adhesive strength and of the weatherability of epoxy was observed by the introduction of
organosilanes (2 wt% epoxyalkylsilane) into a lignin/epoxy blend (10% lignin) [35].
Tomita and co-workers [39] described a new L-EP adhesive in which lignin was modified
by ozonisation. Nieh and Glasser [31] reported the synthesis of an EP resin based on
hydroxypropyl lignin. Hofmann and Glasser [26, 27, 31] prepared the lignin-epoxy resins
from hydroxyalkyl lignin derivatives with varying degrees of alkoxylation. It was shown
that these lignin-based epoxy resins can be crosslinked with aromatic diamine (m-
phenylene diamine, m-PDA) to form strong thermosets whose properties depend on lignin
content and polyether chain length, and have Tg ranging from 0 C to 100 C. The
results of the tensile stress-strain tests reveal that the value of breaking stress and tensile
modulus are higher than those for the epoxy resin control. A trend to stiffer and stronger
materials can be observed with increasing lignin content, while the ultimate extension
simultaneously decreases [26].
Ito and Shiraishi [40] treated Kraft lignin with bisphenol A and then with epichlorohydrin
in the presence of a catalyst which gives a waterproof adhesive with the improved
glueability. The increase in degree of lignin phenolation led to increase of the Tg and the
storage modulus in the rubbery plateau region of the cured films developing a better
three-dimensional structure.
Tai and co-workers [41] synthesised lignin-based epoxy resin from Kraft lignin, Kraft
lignin derivatives (including bisguiacyl lignin) and phenolated Kraft lignin. The epoxy
content was found to be approximately 16 eq/100 g. The solubility of lignin epoxides in
organic solvents was reported to be in the order:
These lignin epoxides were tested as adhesives on aluminium and beech wood using
anhydride and diamine as curing agents, and the results were good [41].
571
Handbook of Polymer Blends and Composites
Simionescu and co-workers [29, 33, 38, 42, 43] reported the synthesis of the lignin
epoxy resins by coupling reactions with simple (phenol, -naphthol, bisphenol) and
macromolecular (novolak) phenols of the lignosulfonate (ammonium, calcium, iron and
chromium lignosulfonate) followed by reaction with epichlorohydrin in alkaline medium.
Two fractions of lignin-epoxy resin resulted: a liquid and a solid one. The physico-chemical
properties [29, 33] of the liquid fractions permit their utilisation in binary mixtures
containing epoxy resin, as new materials.
Incorporation of some epoxy and phenolic hydroxyl groups by successive reactions with
phenol and epichlorohydrin in lignosulfonate macromolecule led to improvement of their
compatibility with epoxy resin. This is reflected in the obtaining of some uniform and
homogeneous structures formed from ER/LER cured with phthalic anhydride, which
have a Tg at about 50 C and good dielectric and mechanical properties (Table 14.3).
Further phenolation had a positive effect on the dielectric properties. It was noted that
the volume resistivity increase was associated with the decrease of dielectric constant
and loss dielectric tangent angle [29].
Compatibility between the lignin-epoxy resin and epoxy resin is very poor at room
temperature, but at higher temperatures (>75 C) it is much improved (Figure 14.1) [30].
The resin mixtures containing 25-50% LER present a high uniformity, while the blends
with <20 wt% and >75% wt% LER have a heterogeneous structure. These results are in
good agreement with those obtained by thermo-optical analysis (TOA) (Figure 14.2).
The blends with various LER/ER ratios (1:3, 1:2 and 1:1) show only one Tg (Figure 14.2),
indicating the development of a single phase during the crosslinking reaction and post-
treatment, as a result of irreversible chemical linking. Incorporation of the phenolated
lignin-epoxy resin (Ph-LER) in epoxy resin matrix gave also a uniform and homogeneous
polymer network with a high glass transition temperature (52 C), indicating that the
phenolation reaction has a positive effect on degree of crosslinking [30].
572
Phenol-lignosulfonate-epoxy resin/epoxy resin blends
Table 14.3 Dielectric and mechanical properties of phenol-lignosulfonate-epoxy resin/epoxy resin
blends (Ph-LER/ER; 1:3) cured with phthalic anhydride
Sample Volume Surface Dielectric Loss in Dielectric Charpy Bonding
resistivity resistivity constant dielectric rigidity impact strength
(ohm cm) (ohm) 50 Hz tangent angle (kV/mm) strength (N/m2)
103 Hz (kJ/mm2)
ER 1.0 x 1010 1.0 x 1010 10.00 0.2000 10.00 3.00 55.00
Fe-LER/ER 1.7 x 1014 7.0 x 1014 5.60 0.0100 8.30 2.80 45.00
Ph-Fe-LER/ER 5.3 x 1014 3.2 x 1014 8.18 0.0080 9.86 2.50 37.00
Fe-Cr-LER/ER 3.4 x 1014 4.0 x 1014 5.96 0.0030 6.5 2.43 35.00
Ph-Fe-Cr-LER/ER 8.2 x 1013 3.0 x 1012 8.63 0.0100 9.3 2.50 37.00
Ca-LER/ER 2.3 x 1014 1.6 x 1012 6.10 0.0200 13.3 2.43 39.00
Ph-Ca-LER/ER 3.1 x 1014 8.0 x 1012 5.76 0.0060 10.2 2.40 35.00
NH4-LER/ER 2.8 x 1015 1.4 x 1012 7.30 0.0200 8.00 2.42 48.00
Lignin-Based Blends
Ph-NH4-LER/ER 6.2 x 1014 1.8 x 1014 8.31 0.0010 9.54 2.20 40.00
573
Handbook of Polymer Blends and Composites
(a)
(b)
(c)
ER 93 83 75 50 25 10
LER 7 17 25 50 75 90
Figure 14.1 Microstructures of LER/ER binary mixtures versus composition: (a) initial
uncured blends;(b) uncured blends after heating for 1 h at 100 C; (c) electron
micrographs of LER/ER casting crosslinked with phthalic anhydride, 20 h at 80 C [30]
In mild conditions of curing the increase in Charpy impact strength is significant, but the
samples had a poor weathering resistance being degraded in about six months of soil
burial compared with the samples cured in severe conditions [34].
The action of environmental factors, such as: soil type, microorganisms, humidity,
temperature, type of plant culture, is more pronounced on the samples containing epoxy-
modified lignin, which has a greater ability for water absorption and favours the
574
Lignin-Based Blends
Figure 14.2 Thermo-optical curves for LER/ER blends for the indicated mixing ratios:
___ film; - - - cured mixtures 20 h, 80 C
a. the plants growth has been less influenced by the presence of samples of LER/ER
blends;
The samples from the LER/ER blend crosslinked under severe conditions (25%-40% of
crosslinking agent, curing time of 45-240 minutes, 150 C) followed by post-treatment
(120-160 C and reaction time of 18-24 h), presented a better resistance than those
samples crosslinked in the mild conditions (without post-treatment), tested using the soil
burial test [34].
575
Handbook of Polymer Blends and Composites
Figure 14.3 The variation of transition temperatures for different LER/ER blends for
the indicated mixing ratios: ___ film; - - - cured mixtures 20 h, 80 C
Including some fillers (lead soap, alum earth, aluminium trioxide, talc, chalk, sand, silica,
glass fibres), plasticiser (polyester C6, dibutylphthalate), pigments (iron oxide, titanium
oxide) in lignin/epoxy blends in a proportion of 10-40 wt% gives epoxy-modified lignin/
epoxy composite materials without affecting crosslinking reactions and properties [10,
30, 44].
The shear strengths of the samples glued with adhesive from LER/ER (1:2) blend cured
with phthalic anhydride had values ranging from 18.2 to 26.5 kgf/cm2, in many cases
exceeding the corresponding value for epoxy resin adhesive, 21 kgf/cm2.
576
Lignin-Based Blends
i. lignin and phenol are pre-reacted and the lignin-phenol adduct is reacted with
formaldehyde [51, 52];
ii. lignin is reacted with phenolformaldehyde and then the resulting product is reacted
with phenol or a phenol-formaldehyde resin previously prepared by standard
procedures [8, 53];
The phenolation of lignin may take place in the presence both of catalysts (HCl, H2SO4
or BF3) [28] and alkaline hydroxide [51], at elevated temperatures, obtaining the soluble
and fusible phenol-modified lignin resins, which by treatment with formaldehyde are
transformed in thermoreactive resins. It is known that a condensation reaction occurred
between the o- or p- position of phenol and - position substituted by OH-, O-R1, =O, or
=C-R2 (R1 and R2 lignin residues) of the side-chain of the phenylpropane units of lignin [55].
Kraft lignin, acid hydrolysis, OSL, SEL and lignosulfonates (LS) as lignin component,
and phenol, cresol, resorcinol or bisphenol as phenol have been used for manufacturing
lignin-phenolic resins obtaining adhesives with satisfactory strength properties. Lignin-
polyphenolic resins are predominantly used for production of weather-resistant wood
products, such as plywood, particleboard, fibreboard, flake board and strandboards.
Generally, these board products are manufactured using lignocellulosic materials (wood
veneer, random geometric configuration wood chips or wood fibres) and an adhesive.
These blends are cured under the heat and pressure conditions and predetermined time,
see Table 14.4.
Cook and Sellers [71] reported the adhesive preparation by replacement of phenol-
formaldehyde (PF) resin with 35-40% OSL. Results indicated that the replacement up to
35%-40% of PF resin with crude OSL strength values are comparable with those obtained
with commercial PF, resin while using of pure OSL determines a decrease of strength
values.
Ono and Sudo [72] prepared an adhesive resin by treatment of SEL with phenol in the
first stage, and in the second stage with formaldehyde. Adhesive formulations consisting
of lignin-modified resins, extender (wheat flour), additive and water were used to glue of
panels and tested by TBA. Results revealed that phenolated steam explosion lignin-based
resin (LP resin) presents an intrinsic retardation in curing. The tensile strength
measurements shown that phenolated SEL adhesive provides excellent strength after
repeated boils indicating a post-cure process. The cure rate is improved by the increase
of pH of the resin. Muller and co-workers [8] have investigated the effect of phenolated
577
578
Lignin-Based Blends
temperature
- Good resistance to boiling solvents
- High crosslinking in the presence of
curing agent of lignin-novolak resins
579
Handbook of Polymer Blends and Composites
lignin/phenol-formaldehyde resin on the adhesion quality and they found that the strengths
have the following order in relation to lignin type:
phenolated-steam explosion lignin < kraft lignin < neat phenolic resin.
Klasnya [46] prepared an LPF resin adhesive by polycondensation of kraft lignin with
phenol and formaldehyde at 100 C in alkaline medium. The strength properties of
plywood bonded with this adhesive depend on temperature and the time of pressing.
Values obtained for shear strengths for LPF resin (30%-50% lignin replacement of phenol)
are 3.31-3.59 N/mm2 in comparison with 2.83-3.21 N/mm2 in the case of control-PF resin.
If a high lignin content (>50%) is used, the shear strengths decreased (1.81-2.05 N/mm2).
Oh [60] and Sellers [61] reported the substitution of phenol with lignin in the synthesis
of lignin-modified PF resin. The resin synthesis includes a first stage of reaction for a
maximum condensation of formaldehyde with lignin to introduce of hydroxymethyl
groups, and a second stage of reaction designed for copolymerisation between the
hydroxymethylated lignin and phenolic resin molecules. Oh and Sellers used the lignins
extracted with methanol, ethanol or dioxane, from various solid residues of the acid
hydrolysis process of newsprint from municipal solid wastes (MSW). The lignins presented
a satisfactory compatibility at a 30%-35% phenol substitution level in PF resins used as
wood binders. The characteristics and molecular weight of the LPF resins were comparable
with those of control PF resins, see Table 14.5.
The strandboards glued with LPF resin using four resin solids, level of concentrations
(3.5, 4, 4.5 and 5%) and two pressing times (4 and 5 minutes) were tested for physical
strength and dimensional stability properties The physical property tests included tension
shear perpendicular to the surface (IB), and static bending (MOR strength retention and
MOE) (Table 14.6).
Senyo [73] found that by replacing 20% of PF resin with methylolated-low molecular
weight OSL, the formaldehyde emissions decrease during the pressing process of the
chipboards. MOR and IB properties of the wood chipboards bonded with 20%
methylolated-lignin/80% PF resin are comparable to those glued with 100% PF resin
(MOR 17.8 and 19.9, respectively; IB 13.3% and 14.0%, respectively).
In the past, spent sulfite liquor has been used as adhesive, but because it is soluble in
water, its use is rather limited. Roffael [74, 75] and Allan [76] reported preparation of
some adhesives with good mechanical properties and dimensional stability by mixing PF
resin with the spent sulfite liquor.
580
Lignin-Based Blends
According to Stout and Ludwig [51] the sodium lignosulfonate isolated from spent sulfite
liquor and phenol are pre-reacted under alkaline conditions, and then the lignosulfonate
adduct is reacted with formaldehyde. Up to 50% of phenol was substituted with Na-LS.
Coyle [53] and Dering [54] reported the preparation of LPF-based adhesive by reacting
of a PF pre-condensed with methylolated-lignosulfonate or lignosulfonate to form a lignin-
modified PF precursor resin. The chemical interactions of lignin with PF resin are
responsible for the improvement of mechanical properties of the end-products [47].
Another adhesive for the manufacture of plywood, fibre board, particle board and similar
products have been reported by Forss and Fuhrmann [77]. This adhesive contains PF
resin (resoles) and a lignin derivative, such as lignosulfonates and alkali lignin in the
form of alkali or earth alkali salts. Forss found that the molecular weight distribution of
the lignin derivatives is a significant factor which influences the adhesive characteristics
[78]. The weight ratio of the lignin derivative to the PF resin depends on their applications,
for example in fibre board production the adhesive may be added as a dilute water
solution with a solid content of 5%, in plywood manufacture of 30% and in particle
board production of 50%. The strength properties of end-products manufactured with
lignin/PF resin adhesive (bending and tensile strength) meet the requirements of the
standards.
581
Handbook of Polymer Blends and Composites
Shen, Fung and Calv [79] have shown that the use of low molecular weight lignosulfonate
fractions offer the best conditions to glue the particle boards with a thickness of 6 and
13 mm, at low pressing time obtaining the higher mechanical properties (MOR=19.4-
23.2 MPa; MOE = 3.5-4 GPa; IB = 320-389 kPa) in comparison with the standard
requirements (MOR = 14 MPa; MOE = 2.7 GPa; IB = 280 kPa).
Sodium, calcium or sodium-ammonium lignosulfonate types were used for the 15%
phenol substitution in the resol-type liquid phenol-formaldehyde resin, involving a
582
Lignin-Based Blends
methylolated lignin condensation in the first step and then treatment with phenol-
formaldehyde resin in the second step [62]. Shear test results (wood failure and shear
strength) of plywood bonded with lignin-modified PF resins shown an influence of the
lignosulfonate nature. Values are in the following order:
Replacement of phenol with 15% lignin-based materials in PF resin adhesive for plywood
production gives cost savings of up to 30% [62].
Ysbrandy and co-workers [81] reported obtaining a synthetic mixture with interesting
properties formed from phenolated pitch/lignin resin to moulding products using certain
catalyst system (anhydrous ZnCl2, phenolsulfonic acid or a combination of (Na2S +
NaOH)). The tensile properties of mouldings were the best for 25% resin content. At a
higher resin level the phenolated pitch/lignin systems present a poor compatibility due to
the substituted nature of the phenolic pitch. The dimensional stability of the mouldings
decreased with decrease in resin content.
The lignin role in polyolefin (PO) blends depends on the lignin nature, separation
procedure and content. Due to its phenolpropanoic structure similar to that of hindered
phenols, lignin can act as stabiliser [11] or initiator of PO degradation [82-84].
Incorporation of lignin in PO material influences the useful properties of the polymer
and its biodegradation characteristics, see Table 14.7.
583
584
Lignin-Based Blends
DTG: derivative thermogravimetry
SEM: scanning electron microscopy
TG: thermogravimetry
PP: polypropylene
HDPE: high density PE
iPP: isotactic PP
585
Lignin-polyolefin blends
Kharade and Kale [85] have incorporated up to 30 wt% dry lignin powder in low-
density polyethylene (LDPE), high-density polyethylene (HDPE) and polypropylene (PP).
Table 14.8 shows that the tensile strength and elongation at break values decreased with
lignin content for all three polymer blends indicating a poor compatibility or interaction
between lignin and the matrix polymer.
The use of ethylene acrylic acid copolymer (EAA) and a titanate coupling agent lead to
the improvement of tensile properties.
586
Lignin-Based Blends
Rusu and Tudorachi [89, 90] reported results regarding the mechanical properties and
thermal stability of mixtures based on the various amounts of lignin (5-30 wt%) from
wood sulfite pulping (lignin) and LDPE.
Incorporation of lignin (up to 10 wt%) in PE leads to the decrease of melt flow index
(MFI), but addition of larger quantities of lignin didnt change the MFI values more. The
tensile strength is reduced by 25%-28%, from initial value for an increased content up
to 5 wt% lignin, but for higher content (up to 30%) is decreased by a percentage of 10-
12 wt% only. The elongation at break decreases drastically up to 5% lignin, after which
the decreases becomes insignificant. Regarding the modulus of elasticity, a characteristic
providing information on the rigidity of the polymeric materials, one can see that an
increase in the lignin content of the system induces an increasing in the values
corresponding of this property. The introduction of lignin and its increased content induced
a decrease in the Izod impact strength values. The water absorption which serves as an
indication of the ability of PE/L blends to biodegrade in the environment, increases with
increasing of the L content [90, 91].
Interpretation of the TG and DTG diagrams of the PE/L shows that in the temperature
domains of 0-220 C and 220-340 C, the weight loss was around of 3%-4% and 7%-
9%, respectively. This can be explained by the fact that lignin acts as stabilisation agent
for LDPE degradation.
The biodegradability tests achieved both in vitro (in the presence of Aspergillius niger
and Trichoderma viride fungi) and in vivo (in the presence of a compost sludge sewage
mixture) showed that the susceptibility to biodegradation increases with increasing the
lignin content in blends [91]. The burial soil test (after 24 months) shows that the LDPE/
20 wt% L blends presents a decrease in tensile strength of about 30% and in elongation
at break of 56%.
587
Handbook of Polymer Blends and Composites
SEM has revealed the different morphologies of the broken sections of PE/L-g-PE blends
where it can see either the particles that were extracted from other half of the samples,
but with enhanced roughness of the surface of spheres, or broken particles indicating a
better lignin-grafted/matrix adhesion [92].
Another way to improve of the compatibility between lignin and polyolefins is utilisation
of a compatibilising agent such as ethylene-vinyl acetate copolymer (EVA) [93]. Alexy
and co-workers [94] showed by SEM that a homogeneous structure is obtained when
EVA (28% vinyl acetate units content) is used in an LDPE/L blend. The concentration
of lignin was varied from 15 wt% to 35 wt% and for EVA from 2 wt% to 25 wt%.
Results show that tensile strength and elongation at break are significantly improved
by EVA copolymer addition. Ciemniecki and Glasser [103] have observed that the
increasing amount of vinyl acetate groups incorporated into the backbone of PE
produces the increase of interactions between lignin (hydroxypropyl lignin) and PE
matrix. This leads to improvement of compatibility of blend components. In another
work Alexy [86] observed that the addition of lignin at 10% level and conventional
additives (amide of erucic acid; phenolic antioxidant; phosphitic antioxidant) has a
similar effect on the processing stability of LDPE and PP. A different degradation
behaviour between PE/lignin and PP/lignin blends has been observed mainly at higher
concentration of lignin. In the case of PE blends, lignin acts as an initiator of the
degradation process after short UV light exposure while at long-term heat stress as a
stabiliser. For PP blends having high lignin content it has been observed that lignin acts
as an initiator of the degradation process at long UV light exposure. The mechanical
properties of PE and PP decrease with increasing lignin content.
Vasile and co-workers [95-99, 104] showed that the incorporation of low amounts of
lignosulfonates (3%-8%) or epoxy-modified lignosulfonates in PO leads to an important
increase of the physico-mechanical indices, i.e., the tensile strength is 2 times higher
and elongation at break 6 times higher to those of LDPE or iPP (Table 14.9). The
percentage of lignin can be raised up to 30 wt% by using EVA or C5 (petroleum resin
as compatibilising or dispersion agents). Also, to achieve an improved compatibility
and increased mechanical resistance, a partially cure by the addition of crosslinking
agents, such as DDM or PhA is recommended.
588
physico-mechanical indices PO/LS
blends
Lignin-Based Blends
In other experiments, Vasile and co-workers [10, 95, 96] have improved the compatibility
of PO with lignin by mixing epoxy-modified lignin with glycidylmethacrylate-grafted-
polypropylene (PP-g-GMA) as compatibilising agent and partially crosslinking. The data
obtained by DSC, SEM, contact angle measurements, X-ray diffraction have confirmed
the improvement of compatibility due to the chemical reaction between components and
also due to their good dispersion in the blends [104].
589
Handbook of Polymer Blends and Composites
Tg: glass transition temperature; cp: difference of heat capacities in the Tg interval; Tm
and Hm: melting temperature and heat of melting
Pascu and co-workers [105, 106] showed that by a combination of the chemical procedures
of component modification with physical ones of surface treatment, e.g., photooxidation,
plasma and electron beam treatment, bio/environmental disintegrable polyolefin materials
can be obtained. UV-irradiation of the PO/L blends induced the change of the surface
properties [105]. The results obtained by IR spectroscopy, DSC and X-ray diffraction
have shown that plasma and electron beam treatments on the iPP/LER/PP-g-GMA films
are very efficient imparting a high polarity and an increased hydrophylicity of blends
[105].
The surface of iPP is modified due to both the initiation of lignin degradation and
superficial modification of blends. Superficial and deep cracks appear, starting with LER-
particles and the changes in all components of the free surface energy are characteristic
for the unstable surfaces for photooxidation [11, 107]. The weight loss increased and
elongation at break decreased.
590
Lignin-Based Blends
In PP/OSL or oxidised lignin with up to 2-3 wt% lignin an increase in photostability was
shown (radical scavenger) [82] while at higher lignin content it initiates radical reactions
[84]. Other effects of lignin incorporation in PO are: increase in conductivity,
hydrophilicity, improved printability imparts biodegradability characteristics [108, 109]
Kelley [110] has presented the thermal and mechanical properties of various PU which
contain only lignin as the polyols in relation to the polyurethanes which contain both
lignin and poly(ethylene glycol) (PEG). The Tg, MOE and tensile strength increase with
increasing lignin content (25%-40% lignin), while the ultimate strain decreased. Properties
of lignin-based PU depend on the preparation method, lignin content and molecular
weight of PEG and the requirements of end product application. For the network with
high ultimate strain is recommended the preparation of chain-extended hydroxypropyl
lignin. Thus, PU derived from a toluene diisocyanate (TDI)-PEG-kraft lignin exhibited
very high values of the ultimate strain [110].
The effects of crosslinking density, the molecular weight of lignin or lignin derivatives on
the PU were investigated by Yoshida using a kraft lignin-polyether triol-poly(methylene
diphenyldiisocyanate) system [112, 113]. It was found that at low and intermediate NCO/
OH ratios, the tensile strengths and Youngs modulus increase with lignin content up to
25%-30%. For high NCO/OH ratios, the tensile strengths and Youngs modulus attained
values of 45 MPa and 1.2 GPa, respectively, when kraft lignin content is between 5%
and 10%. The ultimate strains of the PU prepared from higher molecular weight fractions
decrease monotonically with kraft lignin content, while those of the PU produced from
lower molecular weight fractions attained a maximum, then they decreased rapidly with
the increase of lignin content. It can be concluded that behaviour of mechanical properties
591
592
Eucalin
Indulin AT
Tomlinite
Lignin-polyurethane blend
[116], CGSB CAN2-190-M77) and PU matrix
- compression testing (ASTM D395 - Lignin not interacts with PU
[117]) - Lignin acts as a filler
PU/MA-g-L - IR spectroscopy - MA-g-L no interacts with PU elastomer [122]
PU formulation: - GPC but acts as a good reinforcing agent
polyisocyanate prepolymers; - DSC - MA-g-L increases swelling degree
polyether - swelling test
MA-g-L - mechanical properties (ASTM D412
[116])
EW: equivalent weight; MA-g-L: maleic anhydride grafted lignin; GPC: gel permeation chromatography
Lignin-Based Blends
of the PU did not present any uniformly systematic relationship with respect to the
molecular weight of the lignin fractions [113].
Feldman [118] investigated the viability of blending kraft lignin with PU-based sealants.
Tensile strength values of a L-PU sealant blends vary according to the amount of lignin
present in the matrix, to the type of substrate to which it is adhered and exposure conditions.
Toughness and modulus values of blended sealants increase with the addition of lignin
The effects of NW are more severe than that of the AW because moisture, pollutants and
other environmental factors act combined on the PU-L sealant blends. Loss of modulus
is 20% in AW specimens and a 30% for NW samples in comparison to control PU
sealant. The addition of small quantities of lignin reduces the loss of modulus in L-PU
blends [118].
The results of tensile tests showed that lignin acts a reinforcing agent in the two-phase
polymeric particulate system. SEM micrographs show both the uniform distribution of
lignin particles, and the different morphologies of the constituent phases [119]. The
effect of AW (thermal shock and UV radiation) on the surface of L-PU specimens was
illustrated by the presence of lignin particles entrapped into the polymeric matrix.
The L/PU blends have been characterised by several thermal transitions occurring at
approximately 52 C (attributed to the Tg of the soft segments), at 9 C (attributed to
the micro-crystalline melting of the soft segments), at 90 C (range ordered hard segments)
and at 125 C (Tg of lignin) [119].
Lignin as a filler was found to restrict the degree of swelling of PU less than other inorganic
fillers, i.e., siliceous clay/TiO2 mixture. Data based on the swelling method shows that
lignin does not interact with elastomer matrix to a great extent [120, 121]. The extent of
the restraint varies with type of lignin and the interaction potential of different lignins
decreases in the following order: Tomlinite > Indulin AT > Eucalin [35].
The addition of mineral fillers or different kinds of lignins to the elastomer formulation
increases the hardness at a given time contributing to the increase in the Youngs modulus
[120]. The modulus ratio (modulus of the modified to the nonmodified elastomer E/E0)
is dependent on the filler loading. The addition of lignins decreases both the stress and
strain at break of this elastomer in relation to the unfilled blend. The results obtained for
L-PU blends are 2.5 times greater than that unfilled elastomer (2.7%). The interactions
between the PU-based elastomer and various kraft lignin fillers were investigated by the
surface analysis establishing the contributions of the filler to the work of adhesion. The
modulus E values appears to be a function of the work adhesion at different volumetric
loadings indicating that the modulus values are proportional with the work adhesion
[123], see Figure 14.4.
593
Handbook of Polymer Blends and Composites
Figure 14.4 Dependency of the work adhesion, WA, on the tensile modulus at different
volumetric filler loadings [123]
Lignin modification by grafting with maleic anhydride was used to improve its
compatibility with PU elastomer [122]. Swelling studies indicated that MA-g-L produces
a reduction in the effective number of crosslinking and consequently an increase in
swelling, with a corresponding increase in MA-g-L content in PU blends. PU/MA-g-L
blends present an improvement in mechanical properties (ultimate strain, EB, and ultimate
stress, TB) compared to the PU control. The best results were found for PU/MA-g-L
(15% or 20%) blends. Tensile properties are in function of MA-g-L content. Ultimate
strain EB shows a maximum at 15% MA-g-L, which could be explained by the fact that
MA-g-L produces a reduction of the intermolecular bonds per unit volume of PU.
594
Lignin-Based Blends
The microscopic aspects show a non-uniform globular distribution of LER which becomes
much more uniform by mixing of components in a Brabender plastograph at 200 C and
by partially crosslinking with DDM. The different behaviour of the PET-containing blends
with respect to COP-containing ones have been shown by DSC (see Table 14.12).
In the LER-containing blends, two separated regions (Tg at 74 C and 55 C) are present
indicating incompatibility of the components. The DDM crosslinking reaction is much
more efficient in COP-containing blends, when a single Tg interval appears, the
compatibility of components being improved. The cold crystallisation temperatures of
the blends shift towards low values, while the melting temperatures increase especially
after crosslinking. The crystallisation heats increase with LER content, while those of
melting decrease. These modifications indicate the morphological change by LER
incorporation.
595
Handbook of Polymer Blends and Composites
DTG data show a slight increase of peak temperature in the blends that means a thermal
stabilisation of the polyesteric component by LER incorporation.
In another paper, Cazacu and co-workers [125] studied behaviour and UV- and non-
exposure of the COP/LER blends to the sewage sludge test, observing the important
changes in their bulk and surfaces properties after only two months of ageing. The mass
of the blend samples increase with about 0.5-1 wt%/month, while the thickness of the
sheets also increases with about 10%-12% due to the water absorption from medium.
The water absorption favours the degradability starting with the hydrophilic component
(LER) from blends. After ageing, the Tg becomes less evident and shifted to higher
temperatures. The cold crystallisation temperatures and cold crystallisation heats of
the degraded blends shifted towards lower values, while the melting temperatures increase
with about 20 C especially in the case of the UV irradiated and degraded COP/LER
samples [126].
After degradation, the surface properties of the blends were modified. Results indicated
that more susceptibility of surface to the attack of microorganisms are those with high
LER content and UV-irradiated.
596
Lignin-Based Blends
Poly(vinyl chloride) (PVC) is known for its ability to form miscible systems with other
low or high molecular weight compounds, acting as plasticisers. The selection of lignin
for obtaining the particular blends was justified by the criteria of miscibility of two
polymers, such as their close solubility parameters, the difference between their molecular
weight and their functionality [127].
Feldman reported preparation of the L/PVC blends from PVC compound formulated
with several amounts of TiO2 (10%) and a kraft lignin and an OSL [127]. Although, the
two polymers are either not miscible or partially miscible, the kraft lignin presence in a
proportion up to 7.5 phr does not affect thermal stability and processability of PVC. In
the case blending of the OSL with PVC SEM micrographs indicate their non-miscibility.
The DSC thermograms recorded for the L/PVC blends after 19 minutes of annealing
time present only one Tg located at around 85 C. This could be correlated with an
increased miscibility by thermal treatments. Mechanical properties of unweathered blends
depend on the lignin load. The maximum tensile strength is about 9% higher and the elongation
at break is reduced about 6% than the respective values of the PVC control [128].
The results of DMA analyses, (i.e., a gradual decrease of Tg and a gradual increase of
tan of the blends as function of lignin loading), suggest that the lignin presence led to a
certain degree of breaking of the intermolecular bonds existing within PVC structures.
Higher relaxation energy data of the blends in comparison with PVC control indicated a
reduction of the free volume, probably due to the polar-polar interaction between carbonyl
or hydroxyl groups of lignin and -hydrogen or the chlorine of PVC.
Feldman [129] extended this study using different lignin grades. The influence of various
lignin types on the processability of the L/PVC blends is reflected in the fusion characteristic
and equilibrium torque. The equilibrium torques are slightly lower and reflects a lower
viscosity of the melt for the blend containing partial water-soluble lignin (PWSL), OSL
and kraft lignin. The strength at yield and break and IR spectra demonstrated an
interaction occurring between OH groups of lignin and a-hydrogen of PVC. This
interactions decreases in the following order:
597
Handbook of Polymer Blends and Composites
The thermal analysis data for the LCL/PVC blends reveals a single Tg whose value is
intermediary between Tg of the individual blend component, being an indicator for their
compatibility. For concentration up to 50% LCL copolymer a plasticisation effect of
PVC is observed.
The stress-strain properties showed a decrease in modulus from 250 MPa to 10 MPa
and the increase of elongation from 200% to 600% with the rise of LCL load up to
50%. For higher LCL content than 50% an increase of modulus and a reduction in
elongation, due to partial crystallisation of PCL component, was observed.
For LCL copolymer/PVC blends aged for up to 6 months at room temperature, the
modulus data changed with ageing time and blend composition. SEM micrographs reveal
a single-phase morphology, where the fine lignin-copolymer particles were uniformly
dispersed in PVC matrix, forming a homogeneous blend from the two miscible polymers.
Transmission electron microscopy (TEM) micrographs of thin films of LCL/PVC blends
illustrated at two-phases at microstructure level with domain sizes ranging from 10-30 nm.
Results have been reported in the literature of some attempts focused on the ability of
lignin or lignin derivative to interact with another macromolecule having a linear
backbone, such as poly(vinyl alcohol) (PVA), poly(methyl methacrylate) (PMMA) and
polystyrene (PS).
In last years, preparation of lignin/PVA blends by mixing of polymers either in the form
of solution [103] or solid [132] have been reported. SEM results suggested a partially
miscibility for hydroxypropyl lignin (HPL)/PVA blends explained by a secondary
association between hydroxyl groups of HPL and partially hydrolysed PVA [103]. The
presence of polymer-polymer interaction between the two types of macromolecules is
shown by reduction in melt endothermic area of blends.
598
Lignin-Based Blends
Corradini [132] in his study emphasised the importance of the hydrolysis degree of PVA
and the blend preparation manner (with or without grinding). Determination of heat of
fusion and of apparent activation energy of the thermal decomposition reaction showed
dependence on the chemical structure of the L/PVA blend.
Another polymeric system composed from HPL and PMMA has been studied by
Ciemniecki and Glasser [103]. Observations obtained by SEM revealed that HPL/PMMA
blends are characterised by a two-phase morphology regardless of solvent, weight of
HPL fraction, molecular weight of PMMA or method of preparation [133]. DSC and
dynamic mechanical thermal analysis (DMTA) thermograms revealed the presence of
two endothermic transitions in the initial run and a single, broad transition in the second
run. Low molecular weight HPL fractions are a plasticisation effect on the material
while high-molecular weight HPL fractions have anti-plasticisation effect on the molecular
composite structure [133].
Feldman studied the L-acrylic copolymers by DSC and he observed the incompatibility
of those polymers; however the presence of lignin produce an improving of the mechanical
properties and durability of an acrylic sealant [136].
The grafting of synthetic polymers to lignin represents another way to obtain of engineering
products with controlled structure and properties. Thus, Narayan and co-workers [137]
reported synthesis of lignin-polystyrene graft copolymers (L-g-PS) by anionic
polymerisation. The L-g-PS graft copolymers can function as compatibilisers/interfacial
agents in preparing blends of kraft lignins with PS to obtaining new materials.
The strength properties of very high lignin content plastics depend on the degree of
intermolecular association between kraft lignin components. The ultimate strain extended
599
Handbook of Polymer Blends and Composites
to over 65% in the case of plastics containing the dissociated kraft lignin preparation in
comparison with the materials based on the associated lignin preparation. The tensile
strength and Youngs modulus increased linearly with the average molecular weight of
kraft lignin preparations incorporated at 85% (w/w) level in polymeric materials [142].
These results show the birth of the new generation of polymeric materials that are based
directly on the molecular structure of lignin or lignin derivatives.
The use of lignin (hydrophobic materials) as filler in starch films was studied to improve the
hydrophilic-hydrophobic balance and also to obtain a new lignin carbohydrate system [143].
Baumberger [144, 145] obtained thin films by solvent casting or by thermal moulding of
100/0 to 60/40 wt/wt starch/lignin mixtures using various lignin preparations
(lignosulfonates, kraft lignin, OSL, and Granit lignin).
The macroscopic investigation of starch/lignin blends revealed that lignin had either a
plasticising or reinforcing effect on the mechanical resistance of starch matrix. The use
of lignosulfonates lead to increase of starch plasticity without improving its water
resistance, but OSL and kraft lignins plasticised starch films reducing at the same time
the water affinity of films [146, 147]. Light microscopy showed an incompatibility between
starch and the high molecular weight fraction of kraft lignin. In the case of the use of the
low molecular weight fraction kraft lignin an improvement of compatibility and to impart
both plasticity and hydrophobicity [147]. Maximum stress and % elongation at break
increase with the increase of amylose content [144].
Another starch-based blend which containing lignin or lignin derivatives was reported by
Ghosh [142]. Ghosh blended in various ratios lignin or its ester (lignin butyrate, LB) with
a starch-caprolactone copolymer (SCL) by solvent casting or melt-processing procedure.
600
Lignin-Based Blends
process, while LB has a more pronounced effect during solvent-casting processing. These
results suggest that high Tg of nonmodified lignin play a role as a nucleating agent for
PCL, enhancing crystallisation and melting, and low-Tg of lignin esters (LB) shows a
significant compatibility with PCL reducing crystallisation and fusion [142]. Addition of
lignin or LB up to a concentration of 10%-20% by weight improves the mechanical
properties of SCL.
Rials [150] observed that the modification of hydroxy functionality of organosolv lignin
by ethylation or acetylation influences the state of miscibility and phase morphology of
blends prepared with HPC, obtaining a maximum interaction at a degree of substitution
of 23%-40% of the OH group. The higher degree of substitution (DS > 40% diminishes
the HPC/lignin interaction giving a completely incompatible system at a DS=90%.
In the case of a blend formed from ethyl cellulose (EC) and lignin, addition of lignin (up
to 40% lignin) lead to the appearance of two separate phases with creation of a liquid
crystal mesophase [151]
In order to increase the phase compatibility between lignin and cellulose (derivatives)
block copolymers were synthesised containing covalent-bonds between lignin and cellulose
ester segments [12, 17]. Lignin-thermoplastic cellulose derivatives copolymer exhibited
phase distinctions with the individual block having molecular weights as low at 103 dalton
[17, 152].
601
Handbook of Polymer Blends and Composites
DSC data revealed that lignin-cellulose propionate (CP) block copolymer was unable to
provide an improvement in lignin/cellulose (derivatives) blend compatibility [12]. Results
show that lignin favours the formation of an amorphous or liquid crystalline mesophase
structure through establishing strong interactive association between lignin and the
polysaccharide component.
The addition of lignin reduced the dynamic elastic modulus of cellulose ester solutions
but the modulus of the cellulose solutions (in dimethylacetamide (DMAc)/LiCl) raises at
all shear frequencies. This is explained by the formation of secondary interactions of
lignin with cellulose in the system DMAc/LiCl solvent system [152]. However, when
cellulose ester/cellulose acetate butyrate (CAB)/L mixtures were spun into continuous
fibres from DMAc solution, both fibre tensile strength and modulus increased significantly.
This increase was explained by a positive effect of lignin (even at small amounts of
lignin, i.e., 4%), on the molecular order of the cellulose derivatives in solution and in
solid state.
The addition of lignin enhances the liquid crystalline mesophase order in non-crystalline
cellulose derivatives and due to this order the strength properties increase. TEM of a
CAB/lignin (20%) film revealed a well-ordered cholesteric arrangement which is more
pronounced in the presence of lignin [152-154].
The degree of compatibility between a commercial cellulose ester (CAB) and lignin esters
having different ester groups substituents (acetate, LA; butyrate, LB; hexanoate, LH and
laurate, LL) has also been investigated [151]. In the case of the use of LA and LB a
compatible blends with the strongest interactions with domains sizes of 15-30 mm have
been obtained. The DSC and DMTA data confirmed this observation revealing a single
Tg. The Tg decreases with the increase of the lignin esters, showing a plasticising effect of
lignin esters on the CAB.
The LH and LL containing blends form two distinct transitions on different levels [142].
CAB blends with LA and LB are compatible for all level compositions from 10 to 50 mm
level, CAB blends with LH and LL become incompatible when lignin esters content rise
to <20%.
14.9.3 Lignin/Polyhydroxyalkanoates
602
Lignin-Based Blends
Ghosh and co-workers [142] blended in melt and solution OSL and its ester (organosolv
butyrate, LB) with polyhydroxy-alkanoate, such as poly-3-hydroxybutyrate (PHB) and
poly-3-hydroxybutyrate-co-3-hydroxyvalerate (PHBV). Presence of lignin, in proportion
of 10%-20%, inhibits or retards crystallisation of PHB at any cooling rate. The Tg values
of the PHB phase shifted towards the glass transition temperatures of L or LB indicating
a polymer-polymer interaction between blend components.
Taking into account both the accessibility of the natural vegetable resources and their
capacity to be environmentally friendly the possibilities to achieve materials based on
natural/synthetic polymers are multiple. Lignin represents one of the main component of
biomass and it could be considered as an important source in the obtaining of new
materials.
The use a huge quantities of lignin along with other natural and synthetic polymers
could be alternative to decrease their costs by substitution of the petrochemical products
and to improve the compatibility with the environment.
The last discoveries from lignin and macromolecular chemistry, specially concerning
their association at intermolecular level could allow us to create some polymeric materials
with very high lignin content having performed properties.
The problems which must be solved are connected with the development of effective
plasticisers and compatibilising agents which could lead to the obtaining of new type of
thermoplastics possessing a broad range of physico-mechanical properties.
603
Handbook of Polymer Blends and Composites
References
1. K.V. Sarkanen and C.H. Ludwig in Lignin, Occurrence, Formation, Structure and
Reaction, Eds., K.V. Sarkanen and C.H. Ludwig, Wiley-Interscience, New York,
NY, USA, 1971, 1.
3. G.H. Tomlinson II, Pulp and Paper Magazine of Canada, 1944, 45, 817.
4. O. Hsu and W.G. Glasser, Applied Polymer Symposia, 1975, 28, 297.
5. V. Saraf and W.G. Glasser, Journal of Applied Polymer Science, 1984, 29, 5, 1831.
6. V. Saraf, W.G. Glasser, G. Wilker and J.M. Grath, Journal of Applied Polymer
Science, 1985, 30, 2207.
8. D. Muller, S. Kelley and W.G. Glasser, Journal of Adhesion, 1984, 17, 3, 185.
9. J.J. Lindberg, T.H. Kuusela and K. Levon in Lignin: Properties and Materials,
Eds., W.G. Glasser and S. Sarkanen, ACS Symposium Series No. 397, American
Chemical Society, Washington, DC, USA, 1989, p.190.
10. C. Vasile, M.M. Macoveanu, V.I. Popa, M.C. Pascu and G. Cazacu, Roumanian
Chemical Quarterly Reviews, 1998, 6, 2, 85.
11. C. Vasile, M.M. Macoveanu, G. Cazacu and V.I. Popa, Proceedings of the 5th
Mediterranean School on Science and Technology of Advanced Polymer-Based
Materials, Eds., E. Martuscelli, P. Musto and G. Ragosta, 1997, Capri, Italy, p.185.
12. W.D. Oliveira and W.G. Glasser, Polymer, 1994, 35, 9, 1977.
13. W.D. Oliveira and W.G. Glasser, Journal of Wood Chemistry and Technology,
1994, 14, 1, 119.
604
Lignin-Based Blends
16. W. Oliveira and W.G. Glasser in Lignin, Properties and Materials, Eds., W.G.
Glasser and S. Sarkanen, ACS Symposium Series No.397, American Chemical
Society, Washington, DC, USA, 1989, p.414.
19. D. Feldman and D. Banu, Journal of Polymer Science: Chemistry Edition, 1988,
26, 973.
20. D. Feldman, D. Banu and M. Khoury, Journal of Applied Polymer Science, 1989,
37, 877.
21. D. Feldman and M. Khoury, Journal of Adhesion Science and Technology, 1988,
2, 2, 107.
23. J.L. McCarthy and A. Islam in Lignin: Historical, Biological and Materials
Perspectives, Eds., W.G. Glasser, R.A. Northey and T.P. Schultz, ACS Symposium
Series No.742, American Chemical Society, Washington, DC, USA, 2000, 2-99.
24. J.D. Gargulak and S.E. Lebo in Lignin: Historical, Biological and Materials
Perspectives, Eds., W.G. Glasser, R.A. Northey and T.P. Schultz, ACS Symposium
Series No.742, American Chemical Society, Washington, DC, USA, 2000, p.304.
26. K. Hofmann and W.G. Glasser, Journal of Wood Chemistry and Technology,
1993, 13, 1, 73.
27. K. Hofmann and W.G. Glasser, Journal of Adhesion, 1993, 40, 229.
28. N. Shiraishi in Lignin, Properties and Materials, Eds., W.G. Glasser and S.
Sarkanen, ACS Symposium Series No.397, American Chemical Society,
Washington, DC, USA, 1989, p.488.
29. C.I. Simionescu, C.I. Turta, S. Bobcova, M.M. Macoveanu, G. Cazacu and C.
Vasile, Izvestia Akademii Nauk Respublik Moldova, Biologiceskie i Himiceskie
Nauki, 1994, 3, 270, 62.
605
Handbook of Polymer Blends and Composites
31. W.G. Glasser, W. de Oliveira, S.S Kelley and W.L.S. Nieh, inventors; Centre for
Innovative Technology, assignee; US Patent 4,918,167, 1990.
34. C.I. Simionescu, C. Vasile, M.M. Macoveanu, G. Cazacu and A. Ioanid, Cellulose
Chemistry and Technology, 1994, 28, 517.
37. K. Hofmann and W.G. Glasser, Polymer Preprints, 1990, 31, 1, 657.
38. C.I. Simionescu, G. Cazacu and M.M. Macoveanu, Cellulose Chemistry and
Technology, 1987, 21, 5, 525.
41. S. Tai, J. Nagano and N. Migita, Journal of Japan Wood Research Society, 1967,
13, 257.
42. C.I. Simionescu, V. Rusan, C.I. Turta, S.A. Bobcova, M.M. Macoveanu, G.
Cazacu and A. Stoleriu, Cellulose Chemistry and Technology, 1993, 27, 627.
43. C.I. Simionescu, V. Rusan, G. Cazacu, M.M. Macoveanu, R. Lipsa and G. Stoica,
Cellulose Chemistry and Technology, 1990, 24, 397.
606
Lignin-Based Blends
45. H.H. Nimz in Wood Adhesives. Chemistry and Technology, Ed., A. Pizzi, Marcel
Dekker, New York, NY, USA, 1983, p.247.
46. B. Klasnya and S. Kopilovic, Holz als Roh und Werkstoff, 1992, 50, 282.
47. E. Roffael and B. Dix, Holz als Roh und Werkstoff, 1991, 49, 199.
50. N.G. Lewis and T.R. Lantzy, in Adhesives from Renewable Resources, Eds., R.W.
Hemingway, A.H. Conner and S.J. Branham, ACS Symposium Series No. 385,
American Chemical Society, Washington, DC, USA, 1989, p.13-26
51. C.H. Stout and A.W. Ludwig, inventors; Georgia-Pacific Corporation, assignee;
US Patent 3,658,638, 1972.
52. Y.Z. Lai in Chemical Modification of Lignocellulosic Materials, Ed., D.N.S. Hon,
Marcel Dekker, New York, NY, USA, 1996, p.35-95
55. R.F. Christman and R.T. Oglesby in Lignins. Occurrence, Formation, Structure
and Reaction, Eds., K.V. Sarkanen and C.H. Ludwig, Wiley-Interscience, New
York, NY, USA, 1971, p.769-793.
56. ASTM D905, Standard Test Method for Strength Properties of Adhesive Bonds in
Shear by Compression Loading, 1998
57. D.G. Naae, Proceedings of the 5th International Forum on Commercial Outlets
for New Lignins & Definition of New Projects of International Lignin Institute,
Bordeaux, France, 2000.
607
Handbook of Polymer Blends and Composites
58. JIS K 6851, Testing Methods for Tensile Shear Strength of Wood-to-Wood
Adhesive Bonds, 1999.
59. ASTM D1037, Standard Test Methods for Evaluating Properties of Wood-Base
Fiber and Particle Panel Materials, 1999
60. Y. Oh, T. Sellers, Jr., M.G. Kim and R.C. Strickland, Forest Products Journal,
1994, 44, 2, 25.
61. T. Sellers, Jr., M.G. Kim, G.D. Miller, R.A. Haupt and R.C. Strickland, Forest
Products Journal, 1994, 44, 4, 63.
63. R.H. Gobran, A. Takahashi and E.P. Reilly, inventors; American Can Company,
assignee; US Patent 4320036 1982.
64. ASTM D731-95, Standard Test Method for Molding Index of Thermosetting
Molding Powder, 1999.
65. ASTM D256, Standard Test Methods for Determining the Izod Pendulum Impact
Resistance of Plastics, 2000.
66. ASTM D570, Standard Test Method for Water Absorption of Plastics, 1998.
67. ASTM D648, Standard Test Method for Deflection Temperature of Plastics
Under Flexural Load in the Edgewise Position, 2000.
68. ASTM D790, Standard Test Methods for Flexural Properties of Unreinforced and
Reinforced Plastics and Electrical Insulating Materials, 2000.
69. ASTM D955, Standard Test Method of Measuring Shrinkage from Mold
Dimensions of Thermoplastics, 2000.
70. ASTM D543, Standard Practices for Evaluating the Resistance of Plastics to
Chemical Reagents, 1995.
71. P.M. Cook and T. Sellers in Lignin. Properties and Materials, Eds., W.G. Glasser
and S. Sarkanen, ACS Symposium Series No. 397, American Chemical Society,
Washington, DC, USA, 1989, 324-333.
608
Lignin-Based Blends
72. H.K. Ono and K. Sudo in Lignin. Properties and Materials, Eds., W.G. Glasser
and S. Sarkanen, ACS Symposium Series No. 397, American Chemical Society,
Washington, DC, USA, 1989, p.334.
73. W.C. Senyo, W.A. Creamer, C.F. Wu and J.H. Lora, Forest Products Journal,
1996, 46, 6, 73.
76. G.G. Allan, J.A. Dalan and N.C. Foster in Adhesives from Renewable Resources,
Eds., R.W. Hemingway, A.H. Conner and S.J. Branham, ACS Symposium Series
No. 385, American Chemical Society, Washington, DC, USA, 1989, p.55-67
78. K.G. Forss and A.G.M. Fuhrmann, Forest Products Journal, 1979, 29, 7, 39.
79. K-C. Shen, D.P.C. Fung and L. Calv, inventors; Canadian Patents and
Developments Ltd., assignee; US Patent 4,265,846, 1981.
80. A. Pizzi, Advanced Wood Adhesives Technology, Marcel Dekker Inc., New York,
NY, USA, 1994, p.219.
81. R.E. Ysbrandy, G.F.R. Gerischer and R.D. Sanderson, Cellulose Chemistry and
Technology, 1997, 31, 213.
82. B. Kosikova, M.J. Kacurocova and V. Damianova, Chemical Papers, 1993, 47,
132.
85. A.Y. Kharade and D.D. Kale, Journal of Applied Polymer Science, 1999, 72,
1821.
609
Handbook of Polymer Blends and Composites
87. ASTM D638, Standard Test Method for Tensile Properties of Plastics, 2000.
89. M. Rusu and N. Tudorachi, Journal of Polymer Engineering, 1999, 19, 5, 355.
94. P. Alexy, B. Kosikova, H.J. Radusch, T.L. Lupke, E. Spirk and G. Podstrnska,
Proceedings of the Polymerwerkstoffe 2000, Halle/Saale, Germany, 2000, p.97-100.
98. C. Vasile, M. Downey, B. Wong, M.M. Macoveanu, M.C. Pascu, J.H. Choi, C.
Sung and W. Baker, Cellulose Chemistry and Technology, 1998, 32, 61.
101. T.A. Bubnova, G.S. Shifris, E.Y. Kalmykova, R.S. Salakhova, E.G. Balakhonov,
V.M. Kirilets, E.Y. Plopskij, F. Tegaj and V.V. Aliulin, inventors; Nauchno-
issledovatelskij i Proektno-Konstruktorskij Institut po Problemam Razvitiya
Kansko-Achinskogo Ugolnogo Bassejna, assignee; SU 1,812,193, 1993.
610
Lignin-Based Blends
103. S.L. Ciemniecki and W.G. Glasser in Lignin, Properties and Materials, Eds., W.G.
Glasser and S. Sarkanen, ACS Symposium Series No. 397, American Chemical
Society, Washington, DC, USA, 1989, p. 452-463
105. M.C. Pascu, M.M. Macoveanu, C. Vasile, A. Ioanid and R.C. Oghina, Cellulose
Chemistry and Technology, 2000, 34, 399.
110. S.S. Kelley, W.G. Glasser and T.C. Ward in Lignin: Properties and Materials, Eds.,
W.G. Glasser and S. Sarkanen, ACS Symposium Series No. 397, American
Chemical Society, Washington, DC, USA, 1989, p.402.
611
Handbook of Polymer Blends and Composites
114. ASTM C719-91, Standard Test Method for Adhesion and Cohesion of
Elastomeric Joint Sealants Under Cyclic Movement (Hockman Cycle), 1998.
115. ASTM D2240, Standard Test Method for Rubber Property-Durometer Hardness,
2000.
116. ASTM D412-98a, Standard Test Methods for Vulcanized Rubber and
Thermoplastic Rubbers and Thermoplastic Elastomers-Tension, 1998.
117. ASTM D395, Standard Test Methods for Rubber Property-Compression Set,
1998.
118. D. Feldman, M. Lacasse and R.St.J. Manley, Journal of Applied Polymer Science,
1988, 35, 247.
119. D. Feldman and N.A. Lacasse in Electronic Packaging Materials Science IV, Eds.,
K.A. Jackson, R.C. Sundahl, R. Jaccodine and E.D. Lilley, Materials Research
Society Symposium Proceedings, MRS, Warrendale, PA, USA, 1989, 154, 265.
120. D. Feldman and M.A. Lacasse, Journal of Adhesion Science and Technology,
1994, 8, 9, 957.
121. M.A. Lacasse and D. Feldman, Journal of Adhesion Science and Technology,
1994, 8, 5, 473.
123. D. Feldman and M.A. Lacasse, Journal of Applied Polymer Science, 1994, 51,
701.
124. G.E. Agafitei, M.C. Pascu, G. Cazacu, A. Stoleriu, N. Popa, R. Hogea and C.
Vasile, Angewandte Makromolekulare Chemie, 1999, 267, 44.
125. M.C. Pascu, G. Cazacu and C. Vasile, Proceedings of the Symposium on Progress
in Biomaterials Science, Piatra Neamt, Romania, 2000, p.93-94.
126. G. Cazacu, M.C. Pascu, G.E. Agafitei, A. Ioanid and C. Vasile, Macromolecular
Materials and Engineering, 2000, 283, 93.
612
Lignin-Based Blends
129. D. Feldman and D. Banu, Journal of Applied Polymer Science, 1997, 66, 1731.
131. W.D. Oliveira and W.G. Glasser, Journal of Applied Polymer Science, 1994, 51,
563.
132. E. Corradini, E.A.G. Pineda and A.A.W. Hechenleitner, Polymer Degradation and
Stability, 1999, 66, 199.
134. W.G. Glasser and H-X. Wang in Lignin. Properties and Materials, Eds., W.G.
Glasser and S. Sarkanen, ACS Symposium Series No. 397, American Chemical
Society, Washington, DC, USA, 1989, p.515-522.
135. W.G. Glasser in Adhesives from Renewable Resources, Eds., R.W. Hemingway,
A.H. Conner and S.J. Branham, ACS Symposium Series No. 385, American
Chemical Society, Washington, DC, USA, 1989, p.13-26 and p.40-54.
137. R. Narayan, N. Stacy, M. Ratcliff and H.L. Chum in Lignin: Properties and
Materials, Eds., W.G. Glasser and S. Sarkanen, ACS Symposium Series No. 397,
Washington DC, USA, 1989, p.476-485
138. S-Y. Guan, J. Mlynr and S. Sarkanen, Phytochemistry, 1997, 45, 911.
139. S. Dutta, T.M. Garven and S. Sarkanen in Lignin. Properties and Materials, Eds.,
W.G. Glasser and S. Sarkanen, ACS Symposium Series No. 397, American
Chemical Society, Washington, DC, USA, 1989, p.155-176.
141. Y. Li, J. Mlynr and S. Sarkanen, Journal of Polymer Science B: Polymer Physics,
1997, 35, 1899.
613
Handbook of Polymer Blends and Composites
142. I. Ghosh, R.K. Jain and W.G. Glasser in Lignin: Historical, Biological and
Materials Perspective, Eds., W.G. Glasser, R.A. Northey and T.P. Schultz, ACS
Symposium Series No.742, American Chemical Society, Washington, DC, USA,
2000, p.331-350.
143. S.J. Huang and P.G. Edelman in Degradable Polymers: Principles and
Applications, Eds., G. Scott and D. Gilhead, Chapman & Hall, New York, NY,
USA, 1995, Chapter 2, p.18-28.
148. W.G. Glasser, T.G. Rials, S.S. Kelly and V. Dave in Cellulose Derivatives.
Modification, Characterisation and Nanostructure, Eds., T.J. Heinze and W.G.
Glasser, ACS Symposium, Series No.688, 1998, American Chemical Society,
Washington, DC, USA, p.265-282.
149. T.G. Rials and W.G. Glasser, Journal of Applied Polymer Science, 1989, 37,
2399.
150. T.G. Rials and W.G. Glasser, Polymer, 1990, 31, 1333.
151. I. Ghosh, R.K. Jain and W.G. Glasser, Journal of Applied Polymer Science, 1999,
74, 448.
153. P. Zugenmaier, Journal of Polymer Science: Applied Polymer Symposia, 1983, 37,
223.
154. V. Dave, W.G. Glasser and G.L. Wilkes, Polymer Bulletin, 1992, 29, 565.
614
Environmentally-Friendly Polymers and Blends
15
Environmentally-Friendly Polymers
and Blends
15.1 Introduction
Almost 100% of todays plastics are made from oil or natural gas. As concern mounts
about the potential effects of petroleum-based plastics on the environment and the
increased dependence on oil and gas imports, degradable polymers could become an
important solution to the puzzle [1].
Environmentally degradable polymers (EDP or GreenPla in Japan) and their blends are
some of the materials suitable for sustainable development. The research and market in
this field are developing and growing explosively, e.g., a 50 fold increase of production
in the last decade in the developed countries. In the 1980s the number of references and
patents was in the order of tens, in the 1990s the numbers increased to hundreds at the
beginning of decade and had reached thousands by the end [2, 3].
One option may be to make plastics directly from corn starch or to develop new classes
of polymers. EDP have the requirements for two main fields: biomedical applications (as
absorbable sutures, internal bone fixing implants, orthopaedic repair products, polymeric
drug and drug delivery systems, bioadhesives, tissue engineering, etc.), and environmental
applications [4]. EDP have been developed as biorecyclable materials to reduce the plastic
waste accumulation.
EDP and eco-friendly materials are fully biodegradable. Living organisms transform them
into carbon dioxide and water without any harmful by-products, they are completely
biodegradable within a composting cycle and are able to reduce environmental impact in
terms of energy consumption and greenhouse effect, in specific applications. Their
combustion does not yield any toxic gas or residue, they are recyclable, most of them are
thermoplastic and can be recycled several times and they are derived from renewable
resources.
615
Handbook of Polymer Blends and Composites
Cargill Dow Polymers is using sugars derived from corn as feedstock [5] for its commercial
polymers and polymer blends (Tables 15.1 and 15.2) because of its low cost and
abundance. But the company can also use other plant material such as rice, wheat, sugar
beet and even agricultural waste. They have set up a joint venture to produce PLA at a
2700 metric tonnes per year pilot facility near Minneapolis, MN, USA. With landfill
disposal costs already high and increasing almost everywhere, organic waste recycling
has become an economical option for solid waste disposal.
Many large-scale commercial generators of organic wastes, such as hotels and schools,
have been able to reduce their waste disposal costs by composting. The utilisation of
biodegradable plastics will definitely help to shift communities from landfill disposal to
the recycling of organic wastes. The collection of organic wastes in biodegradable and
compostable bags will eliminate the need for costly and inefficient debagging. This
significant cost saving, will in some part be passed on to the customer, thereby increasing
collection using biodegradable and compostable bags for organic wastes. This will be
welcomed by communities considering organic waste recycling and composting as it
avoids the substantial capital investment necessary for separate pickup and collection of
organic and non-organic solid waste.
Bags are not the only biodegradable and compostable products on the market. There are
many applications for this technology: examples include food utensils such as cutlery,
plates and cups, medical items such as syringes and gloves and packaging materials.
Imagine how much more of the solid waste stream could be diverted from landfills if all
of the plates, cups, knives, forks and spoons used daily in the United States and Canadas
fast-food restaurants were biodegradable and compostable.
Daicel is currently offering two types of products, cellulose acetate type (Celgreen PCA)
and polycaprolactone type (polyhydroxybutyrate (PHB)). The former type has a high
clarity, high strength and high viscosity, while the second exhibits good miscibility with
other polymers, high strength and elongation. Celgreen PCA can replace polystyrene
and Celgreen PH and PHB can replace PE and PP [10] in some application fields
Many different thermoplastic biopolymers have been developed and produced with
industrial procedures using current levels of technology. Thermoplastic starch (TPS) [13-
616
Biodegradable plastics
617
Biodegradable plastics
Handbook of Polymer Blends and Composites
618
Biodegradable plastics
Environmentally-Friendly Polymers and Blends
619
Starch/plastic bland
620
Starch/plastic bland
15], is an intermediate product for making biopolymer blends with film forming properties.
The research results for TPS bioplastics and their production processes are protected by
international patent law or have patents pending [16-21].
The EDP market in Japan consists of 50% PBS, 30% PLA, 10% (starch and PCL/starch)
and 10% other. It is expected that in 2003, the production of EPD will be 1% of total
plastics and in 2010 will reach 60,000 tonnes.
621
Handbook of Polymer Blends and Composites
Blends of starch and polyethylene (LDPE) have been studied but their biodegradability is
questionable, despite the use of additives [37-44]. Total Degradable Plastic Additive
(TDPA; Environmental Products, Inc.), could suffer an oxo-biodegradation process, so
their lifetime is controlled by varying the composition of the blend and the conditions of
the oxidative process [45].
Cargill Dow, one of the worlds largest agricultural processing companies, uses corn to
make PLA for its EcoPLA biopolymer. Like some other biodegradable materials, EcoPLA
can be made into film fibre and a paper coating, and can be thermoformed and injection
moulded. In 1997 Cargill marketed a 115 litre compostable bag under a recent
622
Environmentally-Friendly Polymers and Blends
arrangement with Duro Bag, based in Kentucky. Professor Narayan at Michigan State
University, and a longtime proponent of fully biodegradable polymers, is test marketing
a material called Envar, made, of an aliphatic polyester called PCL that has been alloyed
with a thermoplastic starch. In his opinion, regardless of whether it is made from synthetic
or natural materials, what matters to users of biodegradable bags is whether they will
disappear in compost without leaving a potentially harmful residue.
EnPac, a joint venture of DuPont and ConAgra, imports from Italy a biodegradable
plastic film called Mater-Bi, which is made into bags in the US. It has already been
successfully used for compostable bags in Europe and is now being used in the US market.
Mater-Bi is based on corn starch, but it would be misleading to think that is the only or
even the primary component. EnPac have been marketing Mater-Bi bags made for them
by Castle Bag company. Another firm, Biocorp of California is already promoting and
selling its resource-bag, also made from Mater-Bi film. Composting bags from Mater-Bi
grade Z are as biodegradable and compostable as paper; mechanical properties are similar
to LDPE, it is water resistant, act as an odour barrier, is transparent and has a low
volume.
ICI announced in the 1990s its first commercial application for its Biopol biodegradable
plastic. This is a linear polyester of 3-hydroxybutyric acid (HB) and hydroxy valeric acid
(HVA). The polyhydroxybutyrate-co-hydroxyvalerate (PHBV) is made by fermentation
with the naturally occurring bacterium, Ralstonia eutropha, which uses globules of PHB
homopolymer as an energy storage medium which is analogous to fat in animals or
starch in plants. After four weeks of development trials, Biopol has been accepted by the
hair care company Wella, and it went on the market in 1990 in West Germany as a bottle
for a product in Wellas Sanara shampoo range. ICI has been working on Biopol for
15 years [54]. Although the production capacity was small initially, ICI was able to
bring production up to 5000-10,000 tonnes/year by the mid 1990s. Starch itself is the
key to another type of biodegradable plastic that is making headway in some packaging
applications. A number of companies are involved in this work, including Epron Industries
of Oldham, UK. Eprons Polyclean is a masterbatch material that packaging manufacturers
incorporate into conventional PE at the film blowing stage, just as they would a colour
masterbatch. Polyclean is typically added at level of 15%. Epron has patents on delayed
action oxidation catalysts for PE, PS, and PP, which promote the requisite breakdown of
polymer chains (and reverse the action of conventional antioxidants). Although voluntary
recycling schemes are being launched, they are likely to cover less than 1% of the 10 million
tonnes/year of plastic packaging used.
623
Handbook of Polymer Blends and Composites
potential usage in the packaging of foods, pharmaceuticals and disposable clothing. The
biopolymer is degradable by a particular strain of soil bacteria which converts it to an
ecofriendly powder within 90 days. The scientists have also developed a proprietary soil
with the specific bacteria for disposing of the used packaging material.
Two of the biggest names in agriculture and chemicals have teamed up to commercially
to produce a polymer that they hope will become the basic raw material for fibres of the
future. Cargill Inc., and the Dow Chemical Company have formed a 50/50 joint venture
company, Cargill Dow Polymers LLC, to manufacture the family of NatureWorks PLA
polymers, the first family of polymers made out of annually renewable resources. The
partnership combines Cargills process technology and low cost manufacturing position
for lactic acid and PLA resins and Dows world-class polymer science, applications
technology and global customer base. Cargill and Dow have invested more than $300
million in a new manufacturing facility in Blair, Nebraska, which came on stream at the
end of 2001. It will produce 140,000 metric tons of NatureWorks PLA per year.
Cargill Dow Polymers currently operates a semi-commercial PLA polymer plant near
Minneapolis, MN. That plant has a capacity to manufacture 4,000 metric tons of PLA a
year and the company invested $8 million to double its capacity in 2000 to meet immediate
market development needs.
In addition to Cargill Dow Polymers plants in the USA, a plant will be built in Europe in
about two years time and it is expected that other plants will be built about every
18 months elsewhere in the world. Developed by Cargill Dow Polymers, the NatureWorks
process produces resin that will be used in everyday items such as clothing, cups, food
containers, sweet wrappers, as well as home and office furnishings.
Mitsui Chemical has been developing markets for two biodegradable resins, PLA and a
water absorbing amino acid polymer. The company considers value added package
uses as the most promising application for PLA, because products are not only highly
transparent and readily processable, but are also safe for both humans and the
environment as they are derived from natural plant matter. Target applications include
bottles for cosmetics that feature natural ingredients, and packaging film for premium
vegetables grown organically. The company has received awards in recognition of the
commercialisation of a PLA cutter blade attached to plastic wrapping film box in 2000.
Mitsui has an annual 500 ton capacity for resin production and test fabrication in Ohmuta
and is studying the possibility of resin exports.
624
Environmentally-Friendly Polymers and Blends
new microbe as part of a process that converts corn into the key intermediaries used to
make a range of industrial and consumer products, including polymers, clothing fibres,
paint, inks, food additives and automobile bumpers. Existing domestic markets for such
chemicals total more than $1.3 billion a year. A rise in the number of employees is also
expected over the next decade as the company builds manufacturing capacity and expands
into global markets. This is a typical example of combining chemosynthesis with
biosynthesis plants, an approach pursued in America to save energy.
Research being done suggests the possibility that in the near future the process might be
extended to use industrial organic wastes (such as from the sugar industry) to replace
corn. Metabolix (Cambridge, MA) has recently licensed the Massachusetts Institute of
Technologys (MIT) patents on the insertion of the genes for the production of the key
enzymes in the mechanism of production of PHB (an essential component of biodegradable
polyester thermoplastics) into bacteria and transgenic corn. The transgenic bacteria and
plants can also co-polyesterify -hydroxybutyrate with -hydroxyalcanoates up to C12
[56]. The potential of the production of biodegradable plastics, is a low cost, renewable
production system (using corn, cassava, soybean, etc.), and this is also apparent from the
spate of recent joint ventures as well as business purchases by big multinational commodity
firms, like Monsanto and Cargill [57].
Biodegradable plastics, which are decomposed by microorganisms in the soil and water,
have been attracting much attention worldwide, because they are incorporated into the
natural ecosystem and thus cause no environmental problems.
Biodegradable plastics should have the needed performance characteristic in their intended
use, but after use should undergo biodegradation, in an appropriate waste management
infrastructure, to environmentally compatible constituents. For example, a truly
biodegradable plastic will be converted to CO2, H2O, and compost in a composting
infrastructure, leaving no persistent or toxic residue. Degradability ensures that, after its
intended use, the polymers are subject to waste management and the relationship to the
carbon cycle of the ecosystem is self-evident.
625
Handbook of Polymer Blends and Composites
15.3.1 Japan
According to the Japanese Law on the Treatment of Industrial Wastes revised in 1997,
the used mulching films were designated as industrial waste to be treated at qualified
facilities and individual farmers were banned from burning their waste themselves.
Collection of used film by farmers cooperatives started, but tipping fees at the treatment
contractors were generally too high even if farmers washed the debris out of the used
films. Used biodegradable film does not need to be treated by such laborious and expensive
methods, but can be simply buried in corners of individual farms.
The product adopted was biodegradable Kie Maru manufactured by Topy Green. Kie
Maru was commercialised in 1997 and 1998 sales were 2,000 rolls, each 200 metres
long and sold for 6,000 yen per roll. Use of Topy Green, biodegradable mulching films
has begun to expand among environmentally conscious farmers in the major vegetable
production centres such as Nagano and Kyushu as well as Musashino.
Since 1998, Musashino City, one of the important vegetable growing areas supplying the
Tokyo metropolis, is subsidising the farmers purchase of biodegradable mulching films
from the general tax revenue (totalling 380,500 yen in 1998). This is done primarily to
prevent illegal field-burning of used films that may cause harmful dioxin emissions (and
draw strong complaints from surrounding residents).
The nationwide reporting in all media outlets surely strengthened the public consensus
to place incineration and landfill of waste as the last option and to reduce the volume of
waste by further introduction of their reuse according to the nature of the waste, including
biological treatment of organic waste by composting or anaerobic digestion. Municipalities
moved individually to reduce and limit the numbers of incinerators, and to accelerate
tests of composting treatment.
Package recycling law dictates that the municipalities and industries must have started
recycling all the waste from packaging from April of 2001. However, allowed recycling
methods and share of the burden between municipalities and industries/consumers are
still under political debate. Exemption of biodegradable plastic packages and containers
626
Environmentally-Friendly Polymers and Blends
from the current material recycling schemes is expected because the products are
compostable. If it could be realised, the market growth will reach a tremendous level.
15.3.2 Germany
The German Federal cabinet and the German Parliament have agreed on an amendment
to the Packaging Regulation with a rule for compostable plastic packaging. This special
provision is valid for compostable plastic packaging materials that consist mainly of
biodegradable materials based on renewable resources.
15.3.3 Asia
In the emerging and developing countries the legislation on EPD does not exist or is only
in the initial stages. This was found from discussion during two symposia dedicated to
the developments in this field. Future solutions for implementation of eco-friendly
materials were discussed in the ICS-UNIDO International Workshops, Environmentally
Degradable Plastics: Industrial Development and Application, 2000, Seoul, Korea and
Environmentally Degradable Plastics: Position of EDPs in Plastic Waste Management,
Lodz-Pabianice, Poland, 2001 The production of EPD in these countries is mainly PVA,
modified PVA blended with hydrolysed collagen, and polyolefins/starch blends as blown
film or fibres. A focal point is intended to be created to transfer know how and to
disseminate scientific and technological information in Middle and Eastern Europe. ICS-
UNIDO also supports development of environmentally friendly materials in China,
Indonesia, Malaysia, etc.
This decision is an important milestone for biodegradable materials and opens the market
for use of plastic packaging materials.
The obvious need for technical standards to unambiguously define what is meant by
such terms as biodegradable and compostable, and to lay out strict testing methods,
627
Handbook of Polymer Blends and Composites
The following definitions, which are now part of ASTM standard definitions, apply to
biodegradability [61]:
The problem has been compounded by the lack of any uniformly accepted definition of
the terms biodegradable and compostable. Waste bags made from a blend of PE and
starch, of which there are several on the market in Canada and the United States today,
are touted as biodegradable by their manufacturers because the bags will disintegrate
into minute particles under normal composting conditions. That residue, however, is PE.
Thus, the compost is contaminated with PE, which is often undetectable to the naked
eye. Is such a product entitled to call itself biodegradable? The Federal Trade Commission
(FTC) has been promulgating a definition of biodegradable that would exclude such PE
and starch blends. In the FTC, Guide for the Use of Environmental Marketing Claims
628
Environmentally-Friendly Polymers and Blends
[62], a product is biodegradable only if it will decompose into elements found in nature
within a reasonably short period of time after customary disposal.
The FTC definition presents a good base on which to build a uniform definition. Since
PE is not an element found in nature, a PE-starch bag that disintegrates into even invisible
PE fragments would not be biodegradable.
The modified MIT method, outlined next, is one of the analytical methods to determine
biodegradability of polymers, stipulated by the Law Concerning the Regulation of the
Study and Manufacture of Chemical Substances:
TOD: Theoretical oxygen demand when the sample is completely oxidised (mg).
15.4.1.2 Respirometry
One way to detect plastic film components that do not fully biodegrade during the
composting process involves a sophisticated laboratory technology called respirometry.
629
Handbook of Polymer Blends and Composites
Environment (soil, sea water, compost, sewage), time and breakdown products need to
be specified in all degradation testing because some of degradation intermediary products
could affect vegetation and living organisms [69-71].
The dominant commercial starch-based biodegradable polymers are marketed under the
name Mater-Bi (Table 15.2). The three major Mater-Bi classes differ in composition and
biodegradability (Table 15.3).
The first starch plastics in the marketplace were starch filled PE. They were only bio-
disintegradable and not completely biodegradable in practical time frames. Data showed
that only the surface starch biodegraded leaving behind a recalcitrant PE material. Products
made from these resins do not meet the criteria of complete biodegradability in defined
disposal systems (like composting) and within the operational time frames of the disposal
system.
630
Environmentally-Friendly Polymers and Blends
substitution had a Tg of 128 C. Plasticisers like glycerol triacetate and diethyl succinate
are completely miscible with starch esters and can be used to improve processability.
Water resistance of the starch esters is greatly improved over the unmodified starch. The
starch ester resin reinforced with biofibres has properties comparable to general purpose PS.
Appropriately formulated starch esters with plasticisers and other additives provide resin
compositions that can be used to make injection moulded products and for direct
lamination onto Kraft paper. Starch acetates up to degree of substitution = 2.5 undergo
complete and rapid biodegradation. In the case of starch triacetates, 70% of the carbon
is converted to CO2 at 580 C in 45 days. The National Starch and Chemical Company
offers intermediate degree of substitution starch esters for biodegradable plastics
applications.
Among the various biodegradable plastics available, there is a growing interest in the
group of PHA polymers made entirely by bacterial fermentation. One of the most
important characteristics of these microbial polyesters is that they are thermoplastics
with environmentally degradable properties. Potential useful products from these polymers
have been shown to be degraded in soil, sludge and even seawater. Some microorganisms
(such as bacteria and fungi) were shown to excrete extracellular depolymerases to degrade
these polymers and utilised the decomposed compounds as nutrients.
Of the PHA most extensively studied, two polymers show the potential for being applied
for the production of biodegradable plastic products such as agricultural mulching films,
packaging film, bottles and containers. These two polymers are poly(3-hydroxybutyrate)
(PHB) which have been produced from glucose by using a microorganism identified as
Ralstonia eutropha and its copolymer, PHB-co-3-hydroxyvalerate), P(3HB/3HV) from a
mixture of glucose and propionic acid with the use of the same microorganism. Other
microorganisms have been also used both for production of various kinds of PHA and
enzymic degradation [72-74]. P(3HB) has been commercialised by an Austrian company
Chemie Linz GmbH while P(3HB/HVA) [75] has been commercialised by Zeneca, UK,
through the use of large scale biotechnology processes in their mass production. PHB is
best made from palm oil because of the low cost.
631
Handbook of Polymer Blends and Composites
PLA with high molecular weight is a strong thermoplastic that can be injection moulded,
extruded, thermoformed, or used as spun or melt-blown fibres to produce nonwoven goods.
These polymers were first used commercially as medical sutures in 1970. Lactic acid polymers
and copolymers are offered in semi-commercial quantities for medical applications by
Hanley Chemicals (a Boehringer-Ingelheim subsidiary), by PURAC Biochemicals, Cargill
Dow Polymers LLC (NatureWorks). They are glassy and transparent, similar to
polycarbonates, polyesters, acrylic and general purpose PS, easily printed, readily metallised,
heat sealable, heat and UV-resistant, good barriers to the transmission of flavours and
odours found in food. Many microorganisms are capable of degrading PLA [81].
15.5.2.2 PHBV
15.5.2.3 PCL
Union Carbide in the United States, Solvay in the United Kingdom, and Daicel Chemical in
Japan are the three world producers of PCL resin. The total 1996 consumption in the United
States, Western Europe, and Japan is estimated to have been less than 4,500 tons. Most
typically, PCL is used to make film products, either alone or as a blend with starch. Other
applications are adhesive layers, orthopaedic casts, shoe components, and home crafts.
632
Environmentally-Friendly Polymers and Blends
reduced PLLAs brittleness, became more flexible and elastic, though its tensile strength
and modulus decreased a little. Interestingly, this PLLA copolymer showed degradation by
proteases that did not occur in the case of PLA homopolymer, which might show the
possibility of biodegradation of lactic acid containing polymer by microorganisms.
15.6 Plan to Produce L-Lactic Acid from Kitchen and Food Waste
Researchers are also developing other types of hydrophilic protein films. An active area
of research is using whey proteins to make edible and biodegradable films. Made from
633
Handbook of Polymer Blends and Composites
cheese whey, whey protein is separated from lactose using ultrafiltration, after which it
is evaporated and spray dried to a dry white powder. The California Dairy Foods Research
Centre, University of California, Davis, is one of the research centres working on the
development of these whey-protein-based films. Potential future applications include the
protection of oxygen-sensitive foods (such as nuts), films to block oil migration into
surrounding food components and alternatives to collagen for meat casings.
This reactive blend technology was developed for biodegradable film applications like
lawn and leaf collection compost bags, agricultural mulch film, etc. The technology
involves the following steps:
634
Environmentally-Friendly Polymers and Blends
By controlling the rheology in the extruder, one can obtain a morphology in which the
plastic starch is dispersed in a continuous PCL matrix phase. Good adhesion and
compatibilisation is promoted between the plastic-starch phase and the modified PCL
phase to obtain enhanced mechanical properties. Some of the advantages of using
plasticised starch instead of granular starch are:
All of the operations can be performed in the extruder. This eliminates the use of solvent,
reduces the number of steps and simplifies the process. From a life-cycle perspective, the
biodegradable product reduces waste and energy consumption and conserves resources.
635
Handbook of Polymer Blends and Composites
The properties of blends of starch and aliphatic biodegradable polyesters were determined.
The aliphatic polyesters used include: PCL, polybutylene succinate, and a butanediol
adipate and terephthalate copolymer. To improve the compatibility between the starch
and the synthetic polyester, a compatibiliser containing an anhydride functional group
incorporated onto the polyester backbone was used. The blends were melt-compounded
using a twin-screw extruder. The concentration of the starch in the blend was varied
from 10% to 70% by weight. The amylopectin content of the starches varied from 30%
to 99%. The addition of a small amount of compatibiliser increased the strength
significantly over the uncompatibilised blend. For the compatibilised blend, the tensile
strength did not vary with starch content when compared to the original polyester, while
it decreased with increase in starch content for the uncompatibilised blend. Blends
displayed a sharp intake of water and those containing Eastar Bio (see Table 15.1) had a
higher water absorption than those containing the other polyester. Each blend displayed
two Tg, one corresponding to the polyester and the other corresponding to starch. For
compatibilised blends, the Tg of starch in the blend is lower than that observed for the
uncompatibilised blend. For all polyester blends, those containing 99% amylopectin
starch at a 70% level had the lowest crystallinity, which otherwise decreased with a
decreased amylopectin level in the starch. Blend morphology indicates that the starch
phase became finer (increased melting) as the amylopectin in the blend increased. Also, a
higher starch content led to greater melting of the starch granules and at 70% starch by
weight, a co-continuous phase between the starch and the synthetic polymer exists. The
samples were found to be biodegradable under both aerobic and anaerobic conditions.
636
Environmentally-Friendly Polymers and Blends
Polylactic acid (PLA) is a promising synthetic biopolymer derived from starch through
bioconversion and polymerisation. However, PLA is expensive due to the complex
processing procedures. Another disadvantage of PLA is that it becomes too soft above its
Tg (60 C), which limits its applications. The objective of blending was to blend starch
with PLA to reduce the total material costs, to improve starchs compatibility with PLA
and enhancing the mechanical properties of the blend. A 45/55 (w/w) mixture of starch
and PLA was blended using an intensive hot mixer in the presence of various
concentrations of the compatibiliser. The tensile strength of the blend with optimum
concentration of the compatibiliser was about 66 MPa, which was slightly higher than
pure PLA (62 MPa). The crystallinity and microstructure of the blend was determined
by using differential scanning colorimetry, dynamic mechanical analysis (DMA) and scanning
electron microscopy. Results showed that the blend with the compatibiliser had higher
crystallinity and significant interaction at the interfacial between starch and PLA.
However since starch and PLA are immiscible, resulting in poor blend quality, PVA
containing an unhydrolytic residue group of poly (vinyl acetate) was used to enhance the
interaction in another study between starch and PLA [56, 83]. With fixed starch and
PLA ratio at 50:50 by weight, tensile strength of the blends increased slightly and linearly
as PVA concentration increased. The smaller molecular weight of PVA gave the blend
higher tensile strength and elongation than larger ones. PVA had little effect on the thermal
transition behaviour of the blends, which was almost the same as that of pure PLA.
DMA and water absorption are also discussed.
637
Handbook of Polymer Blends and Composites
Starch was blended with polyhydroxy ester ether (PHEE). Better knowledge of the
relationship between structure and properties in these blends is needed to design products
which can successfully meet the required properties. It is also important to understand
how these properties changes with the humidity and time. The blend structure was strongly
dependent on the processing conditions, especially the moisture level. This led to different
mechanical properties. The blends processed at high moisture levels possessed elongated
starch structures (fibre and sheet-like) which improved the strength and stiffness of the
blends. A model was used to predict the change in properties with the relative humidity
and the ageing time. Hence, an understanding was developed which would help determine
the manner in which these blends should be processed to tailor their end properties.
15.10 Applications
Polymers that form films can be composed of carbohydrates, protein, solid lipid/wax, or
resin. Examples of carbohydrate polymers include various forms of cellulose, such as
carboxymethylcellulose and hydroxypropyl cellulose (HPC), starch and dextrins, pectin,
and alginates. Proteins currently used include albumen, corn derivative, soy protein isolate,
638
Environmentally-Friendly Polymers and Blends
collagen and whey. Waxes include beeswax and carnuba wax, while shellac is the only food-
grade resin. Combinations of these materials can offer some of the benefits of each component.
Edible films are defined by two principles. First, edible implies that it must be safe to eat
or that it is generally recognised as safe (GRAS) by the FDA. Second, it must be composed
of a film-forming material, typically a polymer. You can dip, and you can coat, but the
term edible film refers to a continuous barrier that is formed as the film adheres to the
surface of the foodstuff. The key is to maintain that barrier for some period of time, in
the same way that an inedible package might.
Edible films have potential in a number of different areas. They can coat food surfaces,
separate different components, or act as casing, pouches or wraps. They can preserve
product quality by forming oxygen, aroma, oil or moisture barriers; carrying functional
ingredients, such as antioxidants or antimicrobials; and improving appearance, structure
and handling. Is there such a thing as the perfect package? From an environmental point
of view, its the edible variety that fits the bill. Long-term, edible films show potential in
eliminating synthetic oxygen and gas barriers now used in multi-layer packages,
minimising damage to the foods and extending their shelf life. Using edible films in place
of some of the current packaging might also make the packaging more recyclable. While
it is improbable that edible films will ever completely replace plastics, city landfills are
grateful for any fraction of the tons of plastic that they might replace today or in the future.
Novamont is selling polymers such as PCL/starch blend which generates 90% carbon
dioxide in a controlled composting test and disintegrates in real composting conditions
according to ASTM standards and criteria. They also have some other grades recently
which are completely biodegradable.
Also the polylactide type of polymers are completely biodegradable and have been shown
scientifically to biodegrade and disintegrate very fast, but require thermophilic conditions
to initiate the hydrolysis of the polymer structure. Polylactide does not biodegrade fast
enough in standard tests because of the test conditions, where the temperature is too
low. In a natural composting process the polymer will be biodegraded and disintegrated
very fast because of higher temperatures.
EarthShell Corporation, Santa Barbara, California, announced that after more than two
years in validation and testing, the product design of its hinged-lid containers has met
639
Handbook of Polymer Blends and Composites
performance targets and has received final approval by McDonalds Corporation, Oak
Brook, Illinois. Introduced in Chicago locations in April 2000, earth compostable
packaging has been used to serve more than 7 million Big Mac hamburgers.
Polymers like polylactides, polyglycolides, their copolymers and blends have excellent
properties and have thus been developed for the applications such as wound closure,
osteosynthesis, nets for support while wounds are healing, degradable bandages and
surgical cord for reinforcement, hollow fibres for treatment of damaged nerves as well as
prostheses for anastomoses.
640
Environmentally-Friendly Polymers and Blends
The 2000 Summer Olympic Games in Sydney, Australia, provided the greenest Games
ever thanks to Biocorp of the US, the leader in biodegradable industry. Biocorp was the
sole supplier of biodegradable, compostable, plastic utensils including knives and forks,
straws, cups, spoons, plates, and also biodegradable compostable garbage bags. Biocorp
supplied some 45 million biodegradable items to the Games.
Unlike conventional plastic cutlery, Biocorps products do not contain any PE or PS. The
resin used to make the products is a mixture of corn starch, water and other biodegradable
materials that can be broken down in the composting process.
Nihon Gosei Kagaku has been developing loose packaging film, garbage bags and the
other biodegradable plastic applications by importing Novamonts starch/polyester
polymer blend for ten years. Agricultural mulching film was considered as one of its
most promising and important applications. Sufficient initial strength of the film for
furrow application by machine, and reliable soil coverage during several weeks of plant
growth were the common and one of the most important challenges to be met by any
kind of biodegradable plastic film.
Recently, Nihon Gosei announced that Novamont had improved the film strength by
increasing the polyester content of Mater-Bi and is starting the final application tests in
fields in north-eastern Japan from February. As farmers have come to recognise more the
environmental and labour-saving values of biodegradable films in agriculture, the company
has high expectations from the field test results. This product can command a price of up
to 2.5 times conventional plastic prices for biodegradable mulching films, provided that
the government gives incentives to farmers for the use of biodegradable agricultural films.
Biodegradable fishing line Field Mate was recognised as the most environment friendly
fishing line in the Guinness Book of Records [84]. The product made from PLA by
applying Torays proprietary filament technology has been commercialised since 1996.
641
Handbook of Polymer Blends and Composites
The Asian market for biodegradable plastics is poised for massive growth, thanks to two
new developments in Japanese Environmental Law. Japan has been in the forefront of
early research in the field of state sponsored efforts over the past decade. Many Japanese
companies were among the earliest developers of biodegradable plastics materials. In
recent years, they have been outpaced by a US Company, Cargill Dow. Its 1,40,000 tonnes/
year plant in Nebraska (Japan) will supply polylactic acid for biodegradable and other
applications. In 2000 the Basic Law to create a Sustainable Society was passed, it will
oblige all companies, who produce more than 50 tonnes/year of organic waste, to divert
it, either to feed use or organic recycling. The result will be an explosion in composting
and consequently a demand for biodegradable bags in which to do it.
The revised Plastic Packaging Recycling Law (Japan), passed in 2000, imposed extended
responsibility on manufacturers, obliging them to pay for ultimate disposal of materials,
they place on the market, with annual reporting and auditing.
The Japan Organic Recycling Association (JORA) was created in 2000 to co-ordinate
the national organic recycling efforts relating to standardisation and technology exchange.
Japan Biodegradable Plastics Society (JBPS), which belongs to JORA, has established the
Green Pla logo mark for biodegradable plastics materials, that meet its existing standards.
Demonstration projects for composting using biodegradable plastic bags are also being
planned in Hong Kong and Shanghai, with the help of US Grains Council.
India, also has huge potential, Pakistan has not moved beyond the basic educational
stage. Thailand is looking at producing tapioca starch for biodegradable plastics
applications, while the Philippines is showing interest in the lactic acid route.
In the United States, the future of biodegradables will depend on developing applications
with competitive pricing and performance that demonstrate the economics of composting
organic residuals versus landfilling, from a systems perspective.
642
Environmentally-Friendly Polymers and Blends
According to Frost and Sullivan [50, 85] the market for degradable plastics should reach
7575 metric tons in 2004, representing growth of 35%/year, equivalent market values of
$185 M. This compares to market demand of 9072 metric tons ($23 M) in 1997.
Biodegradable polyesters and PET are widely available. Products include polylactic acid,
biodegradable PET, polybutylene succinate, polybutylene adipate, PHB, PHV, polyethylene
succinate, polyester amide and PCL and starch blends. Growth is being driven by
environmental concerns but high costs and established markets compete. Frost and Sullivan
report that there are 29 competitors in the US market, the main players are: National
Starch, Showa Highpolymer, Daicel, DuPont and Cargill-Dow.
These new biopolymers are priced higher than the commodity polymers typically in use
in plastics applications. However, producers are currently working toward bringing down
the price of degradable polymers by increasing production capacity and improving process
technology. Five years ago PLA and PHB sold for more than $55 per kilogram. Today
PLA, depending on the quantity, is between $3.30 USD and $6.60 USD per kg and PHB,
in large quantity is near $8.80 USD per kg.
The starch used to create the biodegradable plastic, typically wheat gluten, costs about
35 cents a kg. The cheapest commercial plastics cost about $2.20 a kg. Thus, when this
starch-based plastic becomes available to manufacturers, it could be the cheapest
biodegradable plastic. Now, most environment-friendly plastics cost about $5.50 per kg.
In the first stage of the project, researchers have brought the cost of their plastic down to
about $3.30 per kg. The ecological efficiency of natural polymers is very high.
NatureWorks PLA uses 30% to 40% less fossil fuel than is required to produce
conventional plastic resins. Also, overall carbon dioxide emissions are lower compared
with hydrocarbon-based resins because carbon dioxide is removed from the atmosphere
by growing corn.
643
Handbook of Polymer Blends and Composites
Although properties of the starch ester resin are comparable to PS, and can be injection-
moulded in cycle times comparable to PS, there are two problems: cost, and in some
applications, weight because of its higher density. If one could effectively foam the product
in an extrusion or injection moulding operation, less material would be required and so
the cost per article would be less. This would be true for all biodegradable plastics.
Therefore, generic process technology to make biodegradable foam plastics would be
very useful in the effort to successfully commercialise biodegradable plastic products.
US prices for biodegradable polymers range from $3.30 to $6.60/kg for PLA to $6.60 to
$8.80/kg for PHA. High prices have limited the use of biodegradable polymers to special
applications such as loose-fill packaging, compost bags, and medical devices. The total
1996 consumption of biodegradable polymers in the United States, Western Europe, and
Japan is estimated to have been approximately 14,000 tons (excluding PVA).
The problem with pricing is that, at least in the US, prices are unlikely to drop until a
larger market opens up, but mass market sales are hindered by high prices.
With all their advantages, why have Canada and the United States not used more
bioplastics? In Europe, especially in Germany, northern Italy, the Netherlands and
Scandinavia, the biodegradable bag is used extensively for the collection of organic wastes.
Use of the traditional PE bag is largely prohibited.
The first reason may be cost. Currently, biodegradable bags and food service utensils are
priced between two and three times the price of comparable products made from
conventional PE and other plastics. This price differential is the result of the very different
technologies and processes used for PE and for these biodegradable products.
When, however, the cost savings resulting from the elimination of debagging are taken
into consideration, the overall cost differential between conventional PE bags and
biodegradable bags vanishes. The biodegradable bag becomes as economical as the PE
bag. Similarly, if all the waste generated in a fast-food restaurant were compostable, the
restaurants overall waste disposal costs would be expected to drop significantly, since it
would no longer be necessary to separate the plastic and the Styrofoam from the food waste.
Daicel has succeeded in introducing structural branching into the PCL resin. Recently,
Daicel announced that an important effect of the technology was to raise the melt viscosity
644
Environmentally-Friendly Polymers and Blends
of the resin and thus reduce the extrusion costs of PCL film to the same level as PE. The
low melt viscosity of PCL resin had prevented the application of high speed extrusion
technology that is typical of PE film and resulted in comparatively higher converting
costs. Besides improving extrusion efficiency, the quality improvements in uniformity of
film thickness and transparency are so improved that truly commercial PCL film
manufacturing has been realised. Even so the sales price of PCL film remains up to three
times higher than that of PE due to very high production cost of the PCL resin.
With the help of this technology, Daicel intends to sell larger amounts of resin and to
reduce production costs. The company said that it would soon increase the production
capacity of PCL resin to an annual 5,000 tons from the current 1,000, targeting sales of
three billion yen in year 2000.
A major Japanese pen and pencil supplier, Mitsubishi Pencils started selling the Ecocorn
ball point pen from April 2000. It is said that each pen comprises up to forty grains of
corn. The company set the sales price at a competitive 100 yen per piece with the
expectation of selling five million pens in the first year.
Musashino City, one of the important vegetable growing areas supplying the Tokyo
Metropolis, is subsidising the farmers purchase of biodegradable mulching films from
last year. The subsidy from general tax revenue, totalled 380,500 yen in 1998, is
rationalised primarily to prevent illegal field-burning of used films that may cause harmful
dioxin emission (and draw strong complaints from surrounding residents).
Tables 15.5 and 15.6 summarise the target pricing of biodegradable polymers in USA,
Germany and Japan.
15.13 Conclusions
645
Handbook of Polymer Blends and Composites
646
Environmentally-Friendly Polymers and Blends
with these metals, which may impose significant long-term health risks as the compost is
used in agricultural applications. By inserting PE, an artificial substance, into soil we are
altering the fundamental property of soil. This alternation goes far beyond what human
beings can afford. The qualities and originality of soils are and remain the basis of our
planet. This repetitive alternation needs to come to an end until we are able to measure
the consequences. PE added to the soil daily, even in infinitesimally small quantities, is
unacceptable since it will affect ecology by altering the soil. Innovations in the development
of materials from biopolymers have led to the preservation of fossil-based raw materials,
the reduction in the volume of waste, complete biological degradability and compostibility
in the natural cycle, protection of the climate through the reduction of CO2 released, as
well as the application possibilities of agricultural resources for the production of bioplastics.
Up to 80% of all waste at sporting venues is organic waste. The inability to collect this
waste in an uncontaminated form has meant an enormous opportunity for recycling has
been missed.
By using biodegradable utensils and disposing of them in biodegradable bags the organic
waste stream will remain uncontaminated. This will result in thousands of tonnes of
waste being diverted from landfill.
References
4. S.H. Kim, J.Y. Lim, S. Lim and Y.H. Kim, Proceedings of the ICS-UNIDO
International Workshop, Environmentally Degradable Plastics: Industrial
Development and Application, Seoul, Korea, 2000, 146.
6. A.K. Kulshreshtha and S.K. Awasthi, Popular Plastics and Packaging, 1998, 43,
11, 53.
647
Handbook of Polymer Blends and Composites
8. Y.H. Kim and S.H. Kim, Proceedings of the ICS-UNIDO International Workshop,
Environmentally Degradable Plastics: Industrial Development and Application,
Seoul, Korea, 2000, 361.
15. C. Cordes and W-D. Jeserich, inventors; Basf AG, assignee; EP 0000397 B1,
1980.
16. G. Lay, J. Rehm, R.F. Stepto, M. Thoma, J-P. Sachetto, D.J. Lentz and J. Silbiger,
inventors; Warner-Lambert Company, assignee; US 5,095,054, 1992.
648
Environmentally-Friendly Polymers and Blends
23. R.P. Westoff, H.F. Otey, C.L. Mehltretter and C.R. Russell, Industrial
Engineering Chemistry: Product Research and Development, 1974, 13, 123.
24. G.J.L. Griffin in Emerging Technologies for Materials and Chemicals, Eds., R.M.
Rowell, T.P. Schultz and R. Narayan, ACS Symposium Series No. 476, American
Chemical Society, Washington, DC, USA, 1992, 203.
25. J.L. Willett, inventor; Fully Compounded Plastics, Inc., assignee; US 5087650,
1992.
26. U.R. Vaidya and M. Bhattacharya, Journal of Applied Polymer Science, 1994, 52,
617.
27. L. Nie, R. Narayan and E.A. Grulke, Polymer, 1995, 36, 2227.
29. Z. Yang, M. Bhattacharya and U.R. Vaidya, Polymer, 1996, 37, 2137.
30. M. Kim and A.L. Pometto, Journal of Food Protection, 1994, 57, 1007.
32. M.F. Koenig and S.J. Huang, Polymer, 1995, 36, 1877.
33. M.A. Kotnis, G.S. OBrien and J.L Willett, Journal of Environmental Polymer
Degradation, 1995, 3, 97.
34. H.F. Otey and M.W. Doane in Starch Chemistry and Technology, 2nd Edition,
Eds., R.L. Whistler, J.M. Bemiller and E.F. Paschall, Academic Press, New York,
NY, USA, 1984, 389.
35. J.W Lawton and G.F. Fanta, Carbohydrate Polymers, 1994, 23, 275.
37. A.C. Albertsson, Journal of Applied Polymer Science, 1978, 22, 3419.
649
Handbook of Polymer Blends and Composites
40. R. Narayan and W. Lafayette, Kunstoffe German Plastics, 1989, 79, 92.
41. G.J.L. Griffin, inventor; Epron Industries Limited, assignee; US 4,983,651, 1991.
42. W. Sung and Z.L. Nikolov, Industrial & Engineering Chemistry Research, 1992,
31, 2332.
44. G. Scott in Degradable Polymers, Principles and Applications, 1st Edition, Eds.,
G. Scott and D. Gilead, Chapman and Hall, London, UK, 1995, 169.
45. D.W. Wiles, J.F. Tung, G. Scott and G. Swift, Proceedings of the ICS-UNIDO
International Workshop, Environmentally Degradable Plastics: Industrial
Development and Application, Seoul, Korea, 2000, 298.
46. F.H. Otey, R.P. Westhoff and C.R. Russell, Industrial Engineering Chemistry:
Product Research and Development, 1997, 16, 305.
47. F.H. Otey, R.P. Westhoff and W.M. Doane, Industrial Engineering Chemistry:
Product Research and Development, 1980, 19, 1659.
49. G.F. Fanta, C.L. Swanson and W. Doane, Journal of Applied Polymer Science,
1990, 40, 811.
50. R.L. Shogren, R.V. Greene and Y.V. Wu, Journal of Applied Polymer Science,
1991, 42, 1701.
51. R.L. Shogren, A.R. Thompson, R.V. Greene, S.Y. Gordon and G. Cote, Journal of
Applied Polymer Science, 1991, 47, 2279.
52. G.F. Fanta, C.L. Swanson and W.M Doane, Carbohydrate Polymers, 1992, 17, 51.
650
Environmentally-Friendly Polymers and Blends
60. ASTM D5338, Standard Test Method for Determining Aerobic Biodegradation
of Plastic Materials Under Controlled Composting Conditions, 1998.
62. FTC Guide for the Use of Environmental marketing Claims, Federal Trade
commission, Washington, DC, USA, 1998.
64. ASTM D2096, Standard Test Method for Colorfastness and Transfer of Color in
the Washing of Leather, 2000.
651
Handbook of Polymer Blends and Composites
76. S.S. Im and E.S. Yoo, Proceedings of the ICS-UNIDO International Workshop,
Environmentally Degradable Plastics: Industrial Development and Application,
Seoul, Korea, 2000, 228.
77. G.E. Agafitei, M.C. Pascu, G. Cazacu, M. Popa, R. Hogea and C. Vasile,
Macromolecular Materials Engineering, 1999, 267, 44.
78. G.E. Agafitei, M.C. Pascu, G. Cazacu, M. Popa, R. Hogea and C. Vasile,
Macromolecular Materials Engineering, 2000, 283, 93.
652
Environmentally-Friendly Polymers and Blends
82. X.H. Wang, S. Susan and P. Seib, Proceedings of the 85th American Association
of Cereal Chemists Annual Meeting, Kansas City, MI, USA, 2000.
83. K. Tianyike and X.S. Sun, Proceedings of the 85th American Association of
Cereal Chemists Annual Meeting, Kansas City, MI, USA, 2000.
84. Guiness World Records 1999, Ed., M. Young, Gullane Publishing, London, UK,
1999.
653
Liquid Crystalline Polymers in Polymer Blends
16
Liquid Crystalline Polymers in Polymer
Blends
Dana-Ortansa Dorohoi
Liquid crystallinity [1-3] is a mesomorphic state, between the liquid and the crystalline
states, characterised by a long-range orientational order and either by partial positional
order or by complete positional disorder. Substances in the liquid crystalline state are in
fact ordered fluids.
The mesomorphic behaviour depends directly on the molecular shape and group
polarisability. A mesomorphic compound contains anisotropic units having a large axial
ratio, such as (a) rigid rods, (b) discs-shape, (c) planks or (d) helices (Figure 16.1a-d),
small molecules or macromolecules with anisotropic properties. These molecules must
also resist entropic forces which would otherwise tend, for example, to turn a rod into a
random coil. The result of this is that a substance exists as a mesophase only under
suitable conditions of temperature, pressure or concentration.
655
Handbook of Polymer Blends and Composites
The first detailed description of the nature of liquid crystals [4], was done in 1888 by
Reinitzer, for a melt of natural cholesteryl acetate and cholesteryl benzoate. Today many
examples of natural liquid crystals, such as DNA, cholesterol, or tobacco mosaic virus
are known.
There are also a lot of synthetic liquid crystals used in technical applications such as:
cyano-biphenyls, para-linked benzene rings stilbene, azo-links. Kelker and Hatz [3] list
many of the chemical substances that exhibit liquid crystallinity, in certain conditions.
The earliest classification of liquid crystals was made by Friedel [5]. He distinguished
between nematics, cholesterics and smectics mainly on the basis of optical studies. The
nematic mesophase is the least ordered liquid crystalline phase. One dimensional order,
shown in Figure 16.2a, characterises the nematic mesophase. Cholesterics [6]
(Figure 16.2b) are twisted structures usually due to the optically active molecules. The
smectic mesophase is two dimensional ordered, in layers (Figure 16.2c), hence the Greek
term smectic. However, Friedel argued that the cholesterics should be regarded as twisted
nematics, and so, one can consider that there are only two basically different types of
liquid crystals: nematics and smectics [5].
Figure 16.2 Main liquid crystal phases: a) nematic; b) cholesteric and c) smectic
mesophases
656
Liquid Crystalline Polymers in Polymer Blends
The liquid crystals that become ordered by heating are called thermotropic [2, 3] and
those that become ordered by dissolution are called lyotropic [7, 8].
It has been found that many compounds show [4, 6, 9] first-ordered phase transitions
within the smectic range, suggesting the existence of different types of smectics. Based on
the aspect of microscopic textures as well as of X-ray patterns and on the order in which
different phases appear in compounds with more than one smectic phase, Sackmann and
co-workers [10] introduced the identification of different smectic mesophases by capital
letters A, B, C, D E, F or H. The H-phase has been identified by de Vries and Fishel [11].
So, for example, smectics A, C and F are characterised by a disordered arrangement of
the molecular axes within the smectic layer, Bn have a hexagonal lattice in the smectic
layer, while a pseudohexagonal lattice in the plane perpendicular to the molecules
characterises Bt smectic. D smectics are characterised by a cubic arrangement of spherical
molecular aggregates.
Within each smectic layer, the molecules are arranged in a rather random manner and
have considerable freedom of translation and rotation around their long axis. The loose
organisation of the molecules makes the layers rather flexible and they often adopt complex
curved arrangements. The flat layers have been observed to slide very easily over each
other, but the viscosity perpendicular to the layers is much higher due to a strong tendency
to maintain the integrity of the layer.
The importance of this very interesting phase of matter is pointed out by various
applications in technics, in oil, food, detergent, or soap industries, in medicine, in biology
or environment science.
The great interest in the study of liquid crystals is justified by practical applications in:
numerical displays, information store, colour television, surface thermography of
microcircuits, in detection or visualisation of electromagnetic waves, ultrasounds or
vibrational fields, in aerodynamic tests, in chromatography, in the infrared interferometry.
Besides their technical applications, liquid crystals represent an exciting and challenging
area of research in biology, revealing the effect of molecular organisation on chemical
reactivity.
In the last decades, the liquid crystalline polymers included in blends were used to induce
or to improve the anisotropy of some materials and as in situ reinforcements.
The polymers exhibiting one or several mesomorphic phases are called liquid crystalline
polymers. The first findings in liquid crystalline polymers are attributed to Blumstein
657
Handbook of Polymer Blends and Composites
[12] and Finkelmann [13]. Polymer molecules having a stiff main-chain or a flexible
main-chain with mesogenic side groups (Figure 16.3) are known to form mesomorphic
phases as main-chain [14], or as sidechain types [15].
One can obtain main-chain systems by blending bifunctional monomers containing two
or more complementary groups, i.e., electron or proton donor and acceptor groups [16,
17]. Main-chain liquid crystals are known for their ability to form high tensile strength
fibres and moulding.
The principal approach in the design of the supramolecular side-chain liquid crystalline
polymers consists in the introduction of the mesogenic units via the interaction of the
polymers containing pendant binding side groups as complementary low molar mass
compounds [18-20]. Some new side-chain liquid crystalline polymers in which mesogenic
units are attached to the polymer backbones as pendant groups through flexible spacers
are reported in [15, 21].
658
Liquid Crystalline Polymers in Polymer Blends
The chemistry of the commercial liquid crystalline polymers has been dominated by
aromatic polyamides, aromatic polyesters, polycarbonates, polyphosphazines, siloxanes,
or by mixtures of such compounds.
The main liquid crystalline polymers commercially available are reviewed by Isayev [23].
As is shown in [23], thermotropic liquid crystalline polymers were firstly synthesised by
the DuPont Chemical Company in 1972 [24, 25] and commercialised as Ekkcel 1-2000.
This was an aromatic copolyester consisting of p-hydroxybenzoic acid (HBA), terephthalic
acid (TA) and 4,4-dioxydiphenol (DODP) with a high melting point. Liquid crystal
polymer Ekkcel 1-2000 became the originator of a new family of liquid crystalline
polymers called Xydar, firstly manufactured in 1985 by Dartco Manufacturing Company
and then by Amoco Chemical Company.
In 1973 the copolyester X7G, based on HBA and polyethyleneterephthalate (PET) was
commercialised by Tennessee Eastman Chemical Company [26, 27].
Since 1985 a number of other companies introduced new liquid crystalline polymers
[23] such as: Rodrun by Unitika (1986); KU by Bayer AG; Ultrax by BASF; Victrex by
ICI Company; Granlar by Granmont Inc. (1988); Zenite by DuPont (1995); Novacor by
Mitsubishi; Econol by Sumitomo and Rhodester by Rhone-Poulenc.
659
Handbook of Polymer Blends and Composites
Let us mention some commercially available main-chain liquid crystalline polymers and
their properties in Tables 16.1-16.3 [32]. Some of the main properties [32] of liquid
crystalline polymers are:
The liquid crystalline polymers formed by highly aromatic monomers are very stiff.
For example, Kevlar fibres have a specific modulus significantly higher than the one
of steel [33, 34] or carbon fibres produced from a liquid crystal pitch [35] which have
moduli up to 700 GPa.
The highly aromatic liquid crystalline polymers are very brittle, have very low failure
strains, but can be very strong [32].
The viscosity of liquid crystalline polymers aligned in the flow direction increases
with the molecular weight to the power 1, while measured under the same conditions,
the viscosity of normal flexible polymer depends on the superior powers of the
molecular weight. Thus, for equal molecular weights, liquid crystalline polymers have
a much lower viscosity compared with the normal flexible polymers. The most ordered
liquid crystalline polymers have the highest values of the viscosity and a pronounced
smectics anisotropy. It results that liquid crystalline polymers alter the rheology of
bulk polymers, reducing their viscosity.
Liquid crystalline polymers have low thermal expansion coefficients. This results in
low mould shrinkage and hence very high tolerance components can be moulded.
This is one of the most exploited property of liquid crystalline polymers. For example,
intricate electronic components that could not be moulded from traditional polymers
were achieved using liquid crystalline polymers.
Liquid crystalline polymers have been extensively studied from the point of view of
texture types, phase transitions [44, 45], cholesteric pitches [46, 47], electric or magnetic
field effects [48].
660
Liquid Crystalline Polymers in Polymer Blends
Liquid crystalline polymers offer the possibility of improving the melt processability and
of enhancing the mechanical properties of the flexible isotropic polymers. In order to
obtain high-performance materials, many mixtures of thermotropic liquid crystals with
isotropic polymers or with thermoplasts were prepared and carefully studied. Although
the addition of a small amount of liquid crystalline polymer into the isotropic polymer
has produced a significant reinforcement effect, a further improvement in mechanical
properties has been hampered by a poor interfacial adhesion between the liquid crystalline
polymer and the isotropic ones. Therefore the study of the miscibility as well as isothermal
phase separation processes are usually described in the specialised literature.
661
Liquid crystalline polymers
662
Liquid Crystalline Polymers in Polymer Blends
Some commercial main chain thermotropic liquid crystalline polymers have been mixed
with some polymers to obtain engineering plastics with improved mechanical properties.
Relationships between morphology and physical properties of the blends obtained were
extensively analysed. Two phase structures were shown by the morphological studies,
showing the immiscibility of the liquid crystalline polymers with flexible chains. Obviously,
investigations such as phase separation and control of the supermolecular assembly are
very important for practical applications in order to prepare high-performance engineering
plastics.
Side-chain liquid crystalline polymers have been used as matrix resins in polymer dispersed
liquid crystals in order to improve the wide angle viewability and the optical transmittance
of the obtained blends [49].
Lately, polymer dispersed liquid crystals have contributed to sheets consisting of droplets
of nematic liquid crystals in an amorphous polymer with technical applications. The
orientation in the nematic droplets can be influenced by electric or magnetic fields, that
change the aspect of the sheet from cloudy to clear, opening up the possibility for
electrically switchable windows and panels and for a large area signs and advertising
boards [1].
In switchable panels the liquid crystal is constrained in a confined geometry [51] and the
ratio of surface contact area to bulk volume is high. The important aspects of the
confinement, such as porous polymer network assemblies in nematic liquid crystals,
polymer stabilised cholesterics with their implications for reflexive cholesteric displays,
filled nematics or anisotropic gels have been also reported in [51].
Initially, the liquid crystallinity has been controlled via specific interactions in binary
solutions of low molecular mass mesogens having unlike groups [52]. The clearing
temperature of these solutions was usually higher than the clearing temperatures of each
single component and often smectic behaviour was strongly enhanced or induced. Then,
663
Handbook of Polymer Blends and Composites
the liquid crystallinity of polymers or of polymer blends has been induced or modified
via specific noncovalent interactions [53, 54]. In the majority of these systems, liquid
crystallinity is induced or increased by charge transfer, such as electron donor-acceptor
interactions [54].
The balance between the inter- and intra- molecular interactions is responsible for the
morphology as well as for degree of order of the miscible blends. The dipolar interactions
may also be responsible for liquid crystallinity in the case of some mixtures [55].
Along with improved processing, the presence of the liquid crystalline polymers in blends
can determine an enhanced rigidity and strength or a viscosity variation versus
temperature. For example, when a sample was cooled from the isotropic phase, the
viscosity firstly increased and then it suddenly dropped at the isotropic-anisotropic
transition [32]. The decreasing in the viscosity was explained by the increase in the order
degree of the texture.
By reducing the viscosity of mixtures, liquid crystalline polymers have the ability to alter
the rheology of the bulk polymers. Properties of mixtures are usually modified when
compared to those of the individual polymers. For example the rheological properties of
polymer mixtures are not additive [56].
Studying the rheology-morphology relationships for several polymer mixtures, Han and
Kim [57] pointed out the influence of the shape of the droplets on the rheological
behaviour.
The viscosity of the polymer mixtures [58, 59] and their rheological behaviour [60] as a
function of the concentration were also studied by several authors. They pointed out
positive or negative deviations from the additivity rule, attributed to homogeneous and
to heterogeneous mixtures, respectively.
Decreasing the viscosity also improves the processability of the liquid crystalline polymer
mixtures [61] made for obtaining materials with high mechanical strength.
664
Liquid Crystalline Polymers in Polymer Blends
Specific interactions assist not only the polymer miscibility, improving their blend
processability, but also influence the structural, mechanical and thermal properties of
the blends. The specific charge transfer interactions themselves, or accompanied by dipolar
interactions, are responsible for induced liquid crystallinity in the majority of the systems
containing side-chains. This approach has been exploited for many years in the binary
mixtures of electron rich and electron poor low molecular mass mesogens. Recently,
similar effects have been observed for side-chain liquid crystalline copolymers, containing
electron donor and electron acceptor mesogenic groups. The same behaviour as in the
copolymers were shown in the majority of the corresponding homopolymer blends in
which liquid crystallinity is induced by specific interactions.
Extended studies on the specific interactions in polymers, copolymers and polymer blends
containing electron donor and electron acceptor side-chain groups were made by Kosaka
and Uryu [68-70]. A remarkable effect of the central linking groups on the thermal
properties of poly 4{[6-(methacryloyloxy)hexyl]oxy}-N-(carbazolylmethylene) aniline
(PM6Cz)/PM6XNO2 polymer blends, with the chemical structures of the components
from Scheme 16.1, has been reported [68].
665
Handbook of Polymer Blends and Composites
Although their components have electron donor and electron acceptor groups, the PM6Cz/
PM6PBNO2 blend exhibited, over the entire composition range, two glass transitions
(Tg1 = 38 C and Tg2 = 75 C) and two mesophase-isotropic transitions (Ti1 = 100 C and
Ti2 = 165 C), corresponding to those of respective homopolymers. The immiscibility of
this blend has been explained by the steric effects of the carboxyl groups that disturbed
the overlapping between the mesogenic groups, the energy of the specific interactions
being smaller than those corresponding to the steric effects.
X-ray diffraction showed smectic phases for the miscible blends [68]. Measured from the
X ray pattern of quenched PM6Cz/PM6SBNO2 blends, the smectic layer spacing was d =
31.5A (2 = 2.80) and the distance between the mesogenic side groups was d = 4.41A
(2 = 20.14).
666
Liquid Crystalline Polymers in Polymer Blends
The three times increase of the isotropisation enthalpy for 1:1 PM6Cz/PM6SBNO2 blend,
compared with those of the individual homopolymers, proved an increasing degree of
order, induced in blends by specific interactions.
The blends made by homopolymers with similar central linking groups have remarkable
thermal stability. To demonstrate this assertion, Kosaka and Uryu [69] used the
homopolymers from Scheme 16.2. All PM6Cz/PM6BA-Y blends with Y = -NO2; -CN;
-OCH3 exhibited homogeneous smectic phase when the proportion of the mesogenic
carbazolyl units was in the 0.3-0.6 range, though the smectic phase did not appear in
both homopolymers.
667
Handbook of Polymer Blends and Composites
The 1:1 PM6Cz/PM6BA-NO2 miscible blend had a single glass transition at 69 C and a
single mesophase-isotropic transition at 190 C. This blend exhibited a focal conic texture
smectic phase by optical polarising microscopy (Figure 16.5). X-ray pattern [69] gave a
thickness of the smectic layer d = 30.9A (2 = 2.86) and a distance between the mesogenic
groups d = 4.43A (2 = 20.0), smaller than those in the pure homopolymers (d = 4.64A
for PM6Cz and d = 4.63A for PM6BA-NO2).
The thickness of the smectic layer determined in the blend was equivalent to the maximum
length of the monomeric units in the mesogenic copolymers, showing that electron donor
(carbazolylmethylene)aniline groups overlapped the electron acceptor
(nitrobenzylidene)aniline groups, as it is illustrated in Figure 16.6.
Kosaka and Uryu [69] studied the influence of specific interactions on the blend miscibility
by replacing nitro group from PM6BA-NO2 (Scheme 16.2) with cyano, or methoxy
groups. While PM6Cz/PM6BA-CN was a miscible blend, phase separation occurred in
PM6Cz/ Poly 4{[6-(methacryloyloxy)hexyl]oxy}-N-(4-methyloxybenzilidene) aniline
(PM6BA-OMe) blend (characterised by two glass transitions at Tg1 = 63 C and Tg2 =
80 C and two mesophase-isotropic transitions at Ti1 = 139 C and Ti2 = 175 C,
corresponding to the respective polymers). So, using substituents with different capacity
to accept electrons, Kosaka and Uryu provided that the blend miscibility was caused by
intermolecular electron donor-acceptor interactions. For the systems formed with stronger
668
Liquid Crystalline Polymers in Polymer Blends
interactions, smectic phase was shown, showing the capacity of these interactions to
induce a superior degree of order. They suggested that electron interactions between
electron rich and electron poor units is a weak charge transfer between the electron
donor carbazolyl groups and the electron acceptor nitrophenyl groups.
Kosaka and Uryu [70] also examined the structural and thermal properties of the side-
chain liquid crystalline polymers containing both (2- or 3-quinolinylmethylene)aniline
and (4-nitrobenzylidine)aniline groups as well as of their blends. They prepared 4-{[6-
(methacryloyloxy)hexyl]oxy}-N-(4-nitrobenzylidene)aniline (M6NO2) as an electron
acceptor monomer and methacrylate monomers containing (quinolinylmethylene)aniline
groups such as: (3-quinolinylmethylene)aniline (M6Q3); (2-quinolinylmethylene)aniline
(M6Q2); (2-naphthylmethylene)aniline (M6N); (6-methoxy-2-naphthylmethylene)aniline
(M6NOMe) and (4-methoxybenzylidene)aniline (M6OMe) as electron donor monomers.
Concomitantly, the corresponding copolymers and homopolymers [Poly 4{[6-
(methacryloyloxy)hexyl]oxy}-N-(4-R-benzilidene) aniline (PM6R)] having: R = -NO2;
-Q3, -Q2, -N NOMe; -OMe, were prepared in order to study the specific interactions in
their mixtures and blends.
The PM6R homopolymers with average molecular weights ranging between 2.42 x 104
to 5.23 x 104 were obtained in high yields, all of them exhibiting liquid crystallinity
between the glass- and the isotropic-temperatures. The isotropic temperature of the
669
Handbook of Polymer Blends and Composites
polymers with methoxy group was higher, compared with that of the other homopolymers,
showing that the methoxy substituent stabilises the mesophase.
Liquid crystallinity of pure polymers depends on the R- structure. So, the Poly 4{[6-
(methacryloyloxy)hexyl]oxy}-N-(2-quinolinylmethylene) aniline (PM6Q2) homopolymer
exhibited a nematic phase from 49 C to 72 C, whereas PM6Q3 gave a smectic phase
from 55 C to 143 C.
670
Liquid Crystalline Polymers in Polymer Blends
The morphology of the binary PM6R/PM6NO2 blends was shown by differential scanning
calorimetry (DSC) curves and by optical polarising microscopy. So, DSC curves of
PM6OMe/PM6NO2 blends exhibit three endothermic peaks at 121 C, 140 C and
162 C, corresponding to the isotropic transitions of PM6NO2, PM6OMe and to the
1:1 PM6OMe/PM6NO2 miscible blend, while the PM6Q3/PM6NO2 miscible blend had
a single glass transition, at 52 C and a single crystal liquid isotropic phase transition at
166 C. These results suggest that the electron donor acceptor interactions acted between
equimolar electron donor and acceptor groups.
PM6Q3, being a structure with a stronger electron donor nature as compared PM6Q2,
it forms miscible blends with PM6NO2, while PM6Q2 does not. The miscible PM6Q3/
PM6NO2 blend has a smectic focal-conic fan texture (Figure 16.7).
As is shown in Table 16.4, for PM6Q3/PM6NO2 blend, the smectic layer spacing was
28.9 A, almost the same as that of the (M6Q3- M6NO2) ([M6Q3] = 0.55) copolymer,
indicating that the polymer blend has a mesophase structure similar to that of the
copolymer containing the same mesogens.
The phase diagram of the PM6Q3/PM6NO2 blends is given in Figure 16.8. A single
smectic phase appeared in the polymer blends with approximately equimolar
compositions. In the blends containing PM6Q3 in the proportions smaller than 0.3 as
671
Handbook of Polymer Blends and Composites
well as higher than 0.7, the phase separation occurred. For the molar fractions of PM6Q3
outside the range 0.3-0.7, optical polarising microscopy showed the existence of mixtures
consisting from PM6NO2, PM6Q3 and the PM6Q3/PM6NO2 (1:1) miscible blend.
Studying the effect of specific interactions on the thermal behaviour of the polymers P1
and P2 as well as of the copolymers P3 containing the same mesogenic groups
(Scheme 16.3), and of their P1/P2 blends, Imrie and Paterson [53] observed that the relative
modest interactions between two unlike mesogenic species should increase the clearing
temperatures above the glass temperature, revealing liquid crystalline behaviour.
The significant difference between the colour of the substances P1 and P2, permitted
optical polarising microscopy studies referring to phase separation in the P1/P2 blends
(with different proportions of the polymer P1) versus temperature. Thus, when cooling
the isotropic phase of each blend at 0.2 C/min, from 10 degrees above the clearing
672
Liquid Crystalline Polymers in Polymer Blends
Figure 16.8 Phase transition temperatures on the proportion of the PM6Q3 units in
the PM6Q3/PM6NO2 blend [70]
Reproduced with permission from Y. Kosaka and T. Uryu, Macromolecules, 1995, 24,
8295. Copyright 1995, American Chemical Society
temperature, clear, characteristic focal conic fan textures appeared, indicating the presence
of a smectic A phase. The phase separation was not complete, but there were regions rich
or poor in each component, the separation being less pronounced for the equimolar
blends. The smectic A-isotropic transitions were essentially independent on the blend
composition.
These aspects suggest that the specific interactions between the unlike mesogenic units
can assist the miscibility of the polymers.
The manner in which the liquid crystalline phase develops in the blends also states that a
specific interaction between the unlike mesogenic groups leads to the mesophase
formation.
673
Handbook of Polymer Blends and Composites
Scheme 16.3 The homopolymers and copolymers studied by Imrie and Paterson [53]
The interactions between the unlike groups were maximised in the regions containing a
large relative fraction of the component having a minor proportion in the blend. Thus
the blend with 25% of the compound P1 contained focal conic fans in the regions rich in
the component P1, whereas for 75 wt% of the compound P1, the focal conic fans developed
initially from the domains rich in component P2.
Phase behaviour and phase separation in blends containing a liquid crystalline polymer
and an amorphous polymer, interacting by hydrogen-bonds were analysed by Chen and
674
Liquid Crystalline Polymers in Polymer Blends
Even PS/PBPF blends, containing pure polystyrene (PS), were phases totally separated
over all composition and temperature ranges, the incorporation of -C(CF3)2OH groups
into polystyrene led to a partial miscibility of PS(OH) in PBPF.
After drying at 60 C under vacuum, the cast films of 10/90 and 30/70 (wt/wt) PS(OH)/
PBPF were transparent and homogeneous. Upon annealing at a temperature above the
glass temperature, the sample showed a faint anisotropy under crossed polarisers. After
annealing at very high temperatures, phase separation took place with formation both of
an anisotropic phase and an isotropic one. Two-phase morphology with a periodic distance
of about 10 m and a bicontinuous feature, corresponding to a spinodal decomposition,
has been shown in the micrographs of the annealed PS(OH)/PBPF blends. In the anisotropic
domains [71], a certain nematic texture of liquid crystalline polymer has been shown.
On the contrary, the cast films with a larger weight fraction of PS(OH) were translucid.
An irregular two-phase structure with anisotropic and isotropic phases was observed
distinctly under optical polarising microscope. Upon annealing at high temperatures
above glass temperature, the domains of anisotropic and isotropic phases grew in size
and a certain nematic texture of liquid crystalline polymer was observed in the anisotropic
phase. By temperature increasing above glass temperature, the PS(OH)/PBPF blends with
90/10 and 70/30 (wt/wt) compositions showed a transition of the nematic phase into an
isotropic one.
In Figure 16.9 the observations described previously are summarised. An arbitrarily drawn
line delimitates the single- and two-phase regions. Similar results were obtained for the
PS(OH)-4/PBPF and PS(OH)-8/PBPF blends containing 4 mol% and 8 mol% of
-C(CH2)3OH in polystyrene, as shown in Figure 9A and 9B. Although the miscibility
increased with increasing numbers of C(CF3)2OH groups in polystyrene, a total miscibility
was not obtained over the entire composition range.
The phase boundary for PS(OH)-4/PBPF and PS(OH)-8/PBPF blends, shown in the phase
diagrams from Figures 9A and 9B, suggests a separation into two phases from miscible
blend, when the weight molar fraction increases from 0.3 to 0.5. This result was also
shown by the optical polarising microscopy.
675
Handbook of Polymer Blends and Composites
In order to study the molecular origin of the miscibility in the PS(OH)/PBPF blends,
FTIR technique was used. As is shown in [71, 72], in such blends, the most probable are
intermolecular hydrogen bonds between the ester carbonyl groups of PBPF and OH
groups of PS(OH), i.e., CF3)2OH..O=C-. Hydrogen bonds are also expected between:
676
Liquid Crystalline Polymers in Polymer Blends
vibrations of -OH..-OH dimer within the -C(CF3)2OH groups. The shifted peak to the
corresponding frequency of pure components instead of the split of the peaks by the
modification of the blend composition, implies the intermolecular hydrogen bonding
between OH groups of PS(OH) and phenolic OH end-groups of PBPF.
The presence of an additional very broad band centred at 3362 cm-1, assignable to the
OH stretching vibration of OH..OH multimers within -C(CF3)2OH groups in the 50/50,
70/30, 90/10 PS(OH)-8/PBPF blends, as well as in the pure PS(OH)-8, indicated that
there are voluminous aggregations of the PS(OH)-8 component to form PS(OH) rich
phase. Concomitantly, this band did not emerge in the spectra of the single-phase blends
(10/90 and 30/70 PS(OH)-8/PBPF blends). In other words, the miscibility in the PS(OH)/
PBPF has been explained by a balance between the intermolecular hydrogen bonds between
OH from PS(OH) and phenolic -OH groups from PBPF (favourable to the miscibility)
and intramolecular hydrogen bonds within PS(OH) or PBPF (unfavourable to the
miscibility).
The changes from the 1150-1050 cm-1 and 1300-1100 cm-1 ranges of FTIR spectra,
induced by modifying the PS(OH) content in blends, argued the replacement of the
intramolecular hydrogen bonds between C-OH groups in PS(OH)-8 by the intermolecular
hydrogen bonds between C(CF3)2OH and phenolic OH end groups of PBPF.
With the increase of PBPF content, the intermolecular hydrogen bonds between the
phenolic OH end-groups of PBPF and the OH groups of PS(OH) increased, while the
intramolecular hydrogen bonds within PS(OH) decreased. So, at high composition in
PBPF, single-phase blends were obtained. Also, as the degree of intermolecular hydrogen
bonding in the PS(OH)/PBPF system is dependent on the concentration of phenolic OH
end-groups, the further increase in OH content of PS(OH) might be ineffective for an
additional improvement of the miscibility.
No pronounced differences are observed between the blends containing the different
isomers of poly(vinyl pyridine). Two competing processes can influence the miscibility
and structural features of the above mentioned blends [73]:
677
Handbook of Polymer Blends and Composites
the specific interaction between the acid and the pyridine groups, yielding hydrogen
bonded complexes.
Studies on the polymer systems have shown, however, that the equilibrium constant for
the self-association of carboxylic acids is greater than that for complex formation, despite
the fact that the hydrogen bond between the unlike components is a stronger interaction.
This fact suggested that the entropic contribution to hydrogen bond formation could
play a decisive part to determine the phase behaviour. The occurrence of hydrogen bond
formation for all the blend compositions indicates that the interphase region is stabilised
via hydrogen bonding and hence a high compatibility is observed.
Alder and co-workers [73] suggested that in order to use hydrogen bonding in the
construction of supramolecular side-chain liquid crystalline polymers, in which the side-
chain is noncovalently bonded to the backbone, the strength of the hydrogen bond should
be increased. However, simply increasing it, a proton transfer can result. The authors
recommended an alternative approach, using complementary units containing multiple
binding sites, as it has been used to construct mesogenic units [74].
Liquid crystalline polymers offer to the blend technologist considerable possibilities for
improving the processability as well as the mechanical and thermal properties of the
polymer blends, by controlling their viscosity.
The factors influencing the viscosity of the blends containing liquid crystalline polymers are:
the morphology (droplets, ellipsoids, fibres or layers) of the melt, determined by the
previously applied processing;
the processing history (previous processing, previous stress applied to the melt, or
thermal history);
the chemical structure of the components, the content in liquid crystalline polymer, as
well as the temperature.
The processability improvement can be realised either by temperature reducing [75, 76]
or/and by lessening the die swell [77, 78] or the mould shrinkage [79], using a suitable
liquid crystalline polymer in corresponding proportions.
678
Liquid Crystalline Polymers in Polymer Blends
Blizard and Baird [80] studied the viscosity reduction for the blends made of polycarbonate
(PC) and a liquid crystalline copolyester 60-HBA-PET, containing 60 wt% HBA and
40 wt% PET. They measured, at 260 C, a viscosity reduction from 1000 Pa-s
corresponding to neat PC, to 20 Pa-s at 20 wt% liquid crystalline polymer. The increase
over 20 wt% of the liquid crystalline polymer content in PC/60-HBA-PET blend
determined a slow decreasing of the blend viscosity until 6 Pa-s, corresponding to the
measured viscosity of the neat liquid crystalline copolyester.
Studying the viscosity of amorphous polyamides (PA) blended with liquid crystalline
HBA-HNA, at 260 C and different shear rates, Siegmann and co-workers [81] observed
a steep viscosity reduction at small content of copolymer; the greatest reduction of viscosity
being seen at 5 wt% (See Figure 16.10). The viscosity increased with the growth of
liquid crystalline copolymer concentration, the greatest increasing has been reported for
the PA/(HBA-HNA) blends of a high content in liquid crystalline copolymer. Figure 16.10
illustrates the same behaviour of PA/HBA-HNA blend for four shear rates (54 s-1; 135 s-1;
540 s-1 and 2700 s-1).
When the viscosity of liquid crystalline polymer (LCP) was almost the same of the pure
PA, the PA/LCP blends exhibited a viscosity of an order of magnitude lower than those
of the components. Examples of viscosity reduction can also be found in the review of
Dutta and co-workers [75].
Viscosity increasing has been reported for some liquid crystalline polymer blends [32],
such as for the blend containing 20 wt% of HBA-HNA liquid crystalline copolymer in
polyethylene terephthalate-12 (PET12) that exhibited an increased viscosity in the 270-
290 C temperature range.
Weiss and co-workers [82] also reported a rise in viscosity of a blend obtained by mixing
4,4-dihydroxydimethylbenzalazine liquid crystal with polystyrene, but only at low shear
rates (of about 1 s-1).
For the blends obtained from polysulfone Amocos-3500 (PS) and a liquid crystalline
polyester BASFs Ultrax-4002 (PE), Kulichikhin and co-workers [83] noticed the
dependence of viscosity on the liquid crystalline polymer content, at three temperature
values, as plotted in Figure 16.11. The measured viscosity at 280 C, gave a linear fall
with liquid crystalline polymer content throughout the composition range. At 260 C,
the fall was initially faster, but increased at higher liquid crystalline polymer
concentrations, giving a minimum value at about 50 wt%. At 240 C, two viscosity
minima appeared near 30 wt% and 70 wt% content in liquid crystalline polymer.
679
Handbook of Polymer Blends and Composites
Figure 16.10 The PA/(HBA-HNA) blends viscosity versus the HBA-HNA content at
260 C at four shear rates [81]
Reproduced with permission from A. Siegmann, A. Dagan and S. Kening, Polymer, 1985,
26, 1325. Copyright 1985, Elsevier
At the same temperature, different shear rates can induce increasing or decreasing of the
blend viscosity, as it was shown by Nobile and co-workers [84] in the case of the BPPC/
PET-OB blend containing bulk bisphenol-A polycarbonate (BPPC) and a liquid crystalline
copolyester (PET-OB) with ethylene terephthalate (PET) and oxybenzoate (OB) units
(see Figure 16.12).
At very low shear rate (10-2 s-1), the viscosity was found to increase with liquid crystalline
polymer content, at a shear rate of 0.17 s-1, the viscosity seems to be a constant over all
680
Liquid Crystalline Polymers in Polymer Blends
Figure 16.11 Viscosity versus composition for polysulfone blended with a liquid
crystalline polyester (Ultrax). After Kulichikhin and co-workers [83]
Reproduced with permission from V.G. Kulichikhin, O.V. Vasileva, I.A. Litinov, E.M.
Antopov, I.L. Parsamyan and N.A. Plate, Journal of Applied Polymer Science, 1991, 42,
363, Figure 2. Copyright 1991, Wiley
concentrations, whereas at higher shear rate (700 s-1), a significant drop appeared (about
a factor of 10% at 50 wt% liquid crystalline polymer content).
Theories have been developed to explain the behaviour of the viscosity of the blends
containing a liquid crystalline polymer.
681
Handbook of Polymer Blends and Composites
In the case of deformable droplets of one fluid in another, when no chemical interactions
occur between the dispersed phases, the viscosity has been expressed by Schowalter [85],
using the extended Einstein equation:
[
= 1 1 + (51 / 2 + 2)( 21 / 2 + 1) 2 ] (16.1)
where 1, 2 are the viscosities of the components 1 and 2 and 1, 2 are the relative
volume fractions. From Equation (16.1) it results only an increase of the blend viscosity
for all compositions and all viscosity ratios.
682
Liquid Crystalline Polymers in Polymer Blends
Taylor [86, 87] extended the concept of shear flow behaviour for the rigid spheres in
Newtonian fluids, giving for the blends viscosity b, in the case of component 2 dispersed
in component 1, the following equation:
[ {( + 2.5 ) /( + )} ]
b = 1 1 + 1 2 1 2 2 (16.2)
Equation (16.2) was obtained when the droplets retain their spherical shape in the shear
flow field (the dispersed phase forms uniform spherical droplets). The effect of the droplet
size on the viscosity is not reflected in Equations (16.1) and (16.2).
Equations (16.1) and (16.2) do not explain the experimentally observed reduction in the
blend viscosity at different liquid crystalline polymer content.
According to Chuang and Han [88], if no chemical interactions occur between the phases,
the experimental viscosities would generally be lower than those theoretically predicted.
They compared Taylors prediction (16.2) with the observed shear viscosity of poly(amide
imide) (PAI)/LCP system [89], at two shear rates (3.93 s-1 and 118.1 s-1), both when PAI
is taken as component 2 (predicted values though low, closely followed the experimental
values) and when PAI is taken as component 1 (the predicted values were higher and are
different from the observed values.
Heitmiller and co-workers [90] predicted a dispersion for the viscosity of mixtures based
on inverse volume-weighted rule and assuming concentric layers of component 1 in the
component 2. For a large number of layers, the viscosity is given by:
1 1 2
= + (16.3)
1 2
with the same significance of the parameters as in Equations (16.1) and (16.2).
A similar relation has been established by Lees [91] for the viscosity of perfect cylindrical
layers following through a circular die. The possibility of forming such layers with
decreased viscosity at high liquid crystalline polymer concentrations is mentioned by Lee
[77, 78].
Equation (16.3) can in fact explain large viscosity reduction with small addition levels, if
1 >> 2. According with Equation (16.3), the blend viscosity, b, varies monotonically
with the volume fractions 1 and 2. Equation (16.3) under-predicts the viscosity values,
but the predicted values are closer to experimental ones at higher shear rates, than at
lower shear rates.
683
Handbook of Polymer Blends and Composites
Hashin [92] extended the model for the prediction of the upper and lower bound for the
elastic modulus of a composite material and obtained a viscosity envelope for polymer blends.
a) upper bound:
2 2
b = 2 + 1 / 1 / ( 1 2 ) +
5 2
b) lower bound:
2 1
b = 1 + 2 / 1 / ( 2 1 ) +
5 1
a) upper bound:
1 2
b = 2 + 1 / 1 / ( 1 2 ) +
2 2
b) lower bound:
1 1
b = 1 + 2 / 1 / ( 2 1 ) +
2 1
The experimental values for the viscosity of the PAI/LCP blends as function of PAI
concentration superpose above the predicted results according to the Hashin model for
two shear rates, all the values being closed to the lower bound of the envelope with the
higher shear rate [89].
De Meuse and Jaffe [93] applied a linear dependence to express natural logarithm of the
viscosity for miscible blends containing two liquid crystalline polymers of different
composition, using the following equation:
ln = 1 ln 1 + 2 ln 2 (16.4)
684
Liquid Crystalline Polymers in Polymer Blends
The reduced viscosity of the bulk polymer blended with a liquid crystalline polymer
characterised by a low viscosity, could qualitatively be explained by involving polymer
blend theory developed by treating the two fluids as isotropic fluids approaching Lees
layers [91]. However the blend viscosity which is much smaller than the viscosities of its
both components, can not be explained by the traditional theories discussed previously.
When a blend melt containing droplets begins to flow several events occur. The droplet
can be deformed into a fibre, the fibres could break into droplets again, as described
Utracki [94], or droplets could coalesce. It is even possible that fibres coalesce to form
platelets. These phenomena could influence the viscosity behaviour.
During the flow, a continuous velocity profiles within a liquid crystalline blend are more
probable, but, as it has been shown by Kulichikhin and co-workers [83], these profiles
might be highly discontinuous.
The problem is complicated by the shear thinning shown by liquid crystalline molecules,
that become oriented in the flow direction and hence the decrease of the viscosity occurs.
When the applied shear is realised, the molecules will remain oriented in the low viscosity
state. This shear thinning memory effect may well be accentuated in a blend, self
perpetuating the viscosity reducing.
Beery and co-workers [95] and Berry [96] pointed out that the elongational viscosity of
liquid crystalline polymers and their blends, that can be 100 to 400 times greater than
their shear viscosity, was not systematically studied. Most practical polymer processing
involves a considerable elongational component to the flow, e.g., at the entry in an extruder
die. Berry concluded that elongational flow, rather than the shear component dominates
the die entry pressure drop. Thus, the measured shear viscosities could be in fact the
result of differences in elongational viscosity. The importance of elongational flow
producing a desirable morphology was also noted by Dutta and co-workers [75]. They
considered that the extensional deformation of the liquid crystalline polymer domains,
during the flow in the entrance region of the viscosimeter, could explain the decreasing
of the viscosity at high shear rates.
685
Handbook of Polymer Blends and Composites
the mechanical properties of the blends. The rheology of the phase separated fluids with
complex interfaces, created by shearing of two immiscible fluids, was studied from point
of view of the modifications in the area and the orientation of the dispersed phase. A
review is given in [96].
A model has been developed [97, 98] to describe the rheological behaviour of a liquid in
which the dispersed phase changes its area as well as its orientation under the flow. The
system considered comprised two isotropic, non viscoelastic components, with equal
viscosities. The flow field enlarges and orients the interface between the phases, driving
the minor phase from an otherwise spherical shape, whereas the interfacial tension opposes
to these tendencies. Under these conditions, the stress tensor s and the total interfacial
area per unit volume Q, for a flow with velocity gradient tensor k(t), are related to the
same properties for a flow with velocity gradient ck(ct) by the expressions:
[ ] [ ]
t, ck(ct) = c ct, k(ct) (16.5)
[ ] [
Q t, ck(ct) = cQ ct, k(ct) ] (16.6)
where c is a constant.
Thus, the stress tensor and the interfacial area at the time t under a velocity gradient
ck(ct) are c times higher than their values at the moment ct, under a velocity gradient
k(ct). From equation (16.5) for a velocity gradient independent on time, the steady-state
stress is given by:
Under a steady shearing flow, the transient stress growth is given by:
Rheological studies [99, 100] on immiscible blends of two flexible chain polymers gave
results in accordance with the equation mentioned above, showing that the ratio of the
transient to steady-state stress depends only on kt, and not on the individual values t and k.
In this model, the rheological behaviour is traced to the dependence on k of the structural
feature and the consequent lack of characteristic time or length scales. For example, the
dependence on k of the size and shape of the disperse phase in an immiscible blend can be
described using the results of this model. Rheological models similar to that predicted for
immiscible blend of isotropic fluids might be applied to an immiscible liquid crystalline
686
Liquid Crystalline Polymers in Polymer Blends
phase dispersed into an isotropic matrix, with the flow tending to elongate and orient the
liquid crystalline phase. But, an extensional flow that orients the dispersed phase more
effectively does not verify the relationships in Equations (16.5)-(16.8). The deviations [97,
98] could be explained by the component viscoelastic behaviour, neglected in this model.
At the room temperature, the ternary phase diagram exhibits large biphasic regions with
complex zones of: isotropic-isotropic; isotropic-anisotropic and anisotropic-anisotropic
phase equilibrium [103, 104].
Figure 16.13 illustrates the viscosity dependence on the shear rate in the case of HPC/
EC/acetic acid mixture with a total polymer concentration of 40 wt% in acetic acid.
Depending on HPC content in HPC/EC polymer, three morphological types were seen
[56] in HPC/EC/acetic acid solutions:
A monophase anisotropic structure for HPC/EC with high content in HPC (100-
90 wt%);
687
Handbook of Polymer Blends and Composites
The decrease of viscosity from the predicted values by the additivity rule has been explained
by phase separation, while the increased viscosity values, with very small deviations
from the additive rule, characterised the homogeneous phase. Near the isotropic-
anisotropic transitions, the solutions exhibited an anisotropy induced by shear.
Lee and co-workers [105] studied the phase behaviour and rheology of binary blends of
PC and a semiflexible liquid crystalline polyester (LCPE) having the structure from
Scheme 16.4. A distinctive nematic- to isotropic-transition in the pure polymers and in
LCPE/PC blends has been seen.
The phase diagram of LCPE/PC binary blends, obtained by DSC and optical polarising
microscopy, exhibited isotropic (I), nematic (N) and glass (G) phases, depending on the
temperature and on the LCPE content in the blends.
688
Liquid Crystalline Polymers in Polymer Blends
Scheme 16.4 Liquid crystalline polyester (LCPE) studied in [105] by Lee and co-workers
Dynamic oscillatory measurements show that there are some interactions between the
separate isotropic and anisotropic phases. The complex viscosity of the blends is between
those of pure components and show a significant deviation from the logarithmic rule of
mixtures expressed by Equation (16.4). In Figure 16.14a and 16.14b, the storage and
loss moduli versus frequency for various blend compositions at 180 C and 200 C
respectively, are plotted.
From Figure 16.14 it results that the loss moduli G() in the pure liquid crystalline
polymer are about two to five times higher than the storage moduli, G() over the
entire frequency range. In the case of the LCPE/PC blends, G() was found depending
on the frequency as, G()0.8, while the storage modulus G() changed slope over
the entire frequency range as well as in other thermotropic melts [106, 107].
The blend with a weight fraction, LCP, (of about 0.1) exhibits values for G() and
G() lower than those in the pure LCPE. The cross point of G() and G() shifts to
a higher frequency in comparison to that for neat PC, indicating a decrease in the longest
relaxation time of the blend. When the weight fraction of LCPE exceeds 0.5, moduli
decrease monotonically with increasing LCP, but do not follow a simple mixing rule,
suggesting an influence of the interfacial tension on the rheological properties. At 200 C
(Figure 16.14b) as the LCPE concentration increases, the slope of G() becomes smaller
than in the pure PC, but still larger than in pure LCPE, proving the partial miscibility of
the two components.
689
Handbook of Polymer Blends and Composites
Figure 16.14 Storage and loss moduli at the various composition of LCPE/PC blends
measured in biphasic region at (a) 180 C and (b) 200 C. Numbers of the figure
denote the LCPE weight fraction. Black and white symbols denote the G and G,
respectively, at each composition [105]
Reproduced with permission from S. Lee, P.T. Mather and D.S. Pearson, Journal of
Applied Polymer Science, 1996, 58, 2, 243, Figure 6. Copyright 1996, Wiley
690
Liquid Crystalline Polymers in Polymer Blends
Figure 16.15 Dynamic viscosity at the various blend composition measured at 1 rad/s
during cooling after annealing 20 minutes at 270 C. Numbers on the figure denote
the LCPE weight fraction [105]
Reproduced with permission from S. Lee, P.T. Mather and D.S. Pearson, Journal of
Applied Polymer Science, 1996, 58, 2, 243, Figure 7. Copyright Wiley
and the viscosity begins to rise with decreasing temperature. The finite temperature range
over which viscosity decreases during cooling is due to the polydispersity of the liquid
crystalline polymer.
In the case of LCPE with a weight fraction is about 0.7, a similar, but less dramatic
viscosity change was observed. This experimental fact was in accordance with the optical
polarising microscope observations which showed that, at cooling, the isotropic-nematic
phase separation occurs in the blends, below the isotropic-nematic transition temperature
of the pure liquid crystalline polymer. With temperature decreasing, the reduction in
viscosity of the LCPE-rich phase is compensated by an increasing viscosity of the PC-rich
phase. In these conditions, the viscosity of LCPE/PC blends appears to remain unchanged
for a certain range of temperatures.
This behaviour is diminished as the LCPE content decreases, since the viscosity of the
PC-rich phase dominates at low LCPE composition. So, the temperature dependence of
viscosity for the weight fraction of about 0.1 LCPE in the LCPE/PC blend is similar to
that of pure PC, but shows considerably reduced values.
691
Handbook of Polymer Blends and Composites
Lee [77, 78] has injection moulded a blend of chlorinated poly(vinyl) alcohol (PVC) and
HBA-PET as liquid crystalline copolymer. He found that the liquid crystalline polymer
increased the spiral mould flow length and was able to model the improvement in injection
moulding processing using a simple power law. He also noted that extrudate die swell
was reduced by addition of the liquid crystalline polymer.
In order to establish optimum gas barrier performance from injection moulded parts,
Toshikazu [108] investigated the liquid crystalline polymer/ethylene vinyl alcohol
copolymer blend in a 75 ton injection machine with a 20 cm diameter extruder, observing
that screw torque and melt pressure decreased in all blend systems and established a
relationship between morphology, viscosity and surface tension.
Dutta and co-workers [75] have also discussed the implications of rheology modifications
arising from liquid crystalline polymers incorporation. They highlighted the benefit of
reduced melt temperature which may imply reduced energy costs or less degradation.
The rod-like molecular conformation and chain stiffness give their self-reinforcing
properties to the liquid crystalline polymers.
The thermotropic liquid crystals have become available because they can be produced in
the melt to give highly orientated structures that are largely retained on cooling.
692
Liquid Crystalline Polymers in Polymer Blends
The good, but anisotropic, mechanical properties of the relatively expensive liquid
crystalline polymers recommend their use to reinforce the less expensive, isotropic, but
mechanically weaker engineering plastics, such as polyethylene, polypropylene, PS, PC
or polyether sulfones.
The most common reinforcing commercial liquid crystalline polymers are Vectra, Econol,
Rodrun, Sumicosuper. Composites containing Kevlar are examples of liquid crystalline
blends in commercial use. Kevlar itself is a liquid crystalline polymer and a solution
spun in its lyotropic state. Another available thermotropic melt spun, manufactured by
Celanese, the heat treated Vectra [109] have a tensile strength comparable with that of
Kevlar 49.
the high para-linked aromatic content gives samples with good moduli;
the very low melt viscosity allows good flow properties and processability;
they readily form fibrous structures, for example injection moulded samples of a neat
liquid crystalline polymers resins having an appearance similar to wood.
Because of the incompatibility of isotropic and liquid crystalline materials, the reinforcing
is difficult to realise with commercial engineering plastics and liquid crystalline polymers.
The molecular reinforcement needs a perfect molecular dispersion of the liquid crystalline
polymer in the matrix [110, 111].
The generation of reinforcing species in situ offers advantages over the addition of solid
fibres and fillers. The use of techniques such as in situ crystallisation and polymerisation
have been reviewed by Kiss [114].
The mechanical properties of in situ composites, were listed by Crevecoeur and Groeninckx
[115].
693
Handbook of Polymer Blends and Composites
Some results in the reinforcement of the blends using liquid crystalline polymers will be
presented next.
Siegmann and co-workers [81] firstly reported in situ composite formation in the systems
of amorphous polyamide and Vectra with improvements in processing and reinforcement
as well.
Kiss [117] investigated the effect of mixing Vectra A or Vectra B grades with wide variety
of amorphous and crystalline polymers, including polyether sulfones, polyetherimide
(PEI), polyarylate (PAr), polyacetal (PAc), polyamide-6 (PA-6), polybutylene terephthalate
(PBT), PC, poly(chlorotrifluoroethylene) (PCFE) and poly(etheretherketone) (PEEK). The
blends were obtained by compounding techniques, by extrusion into strands, or by
injection moulding. Differences in adhesion of the liquid crystalline polymer phase in the
various systems determined the morphology of the studied blends and the extension of
the elongated liquid crystalline domains resulting from extended droplets to fibres [117].
Dramatic increases in the tensile and flexural moduli (two to three times), with smaller
increases in the flexural and tensile strength, accompanied by a substantial decrease in
elongation to break were observed for all blends studied by Kiss.
Swaminathan and Isayev [118] showed that the extent of property improvement is
dependent on the method of compounding of PEI/Vectra A959 blends. They compared
the PEI/Vectra A959 blends obtained by using a static mixer and a co-rotating twin-
screw extruder as well. The blends obtained at static mixer had a higher degree of
fibrillation and corresponding improvements in Youngs modulus over those produced
using the twin-screw extruder.
694
Liquid Crystalline Polymers in Polymer Blends
Figure 16.16 Elastic modulus and ultimate elongation of injection moulded LCP/PA
blends versus LCP content [81]
Reproduced with permission from A. Siegmann, A. Dagan and S. Kening, Polymer, 1985,
26, 1325. Copyright 1985, Elsevier Science
The interphase adhesion between the PEI and liquid crystalline polymer was found to be
poor, skin-core morphology being in injection moulded specimens. SEM of PEI/Vectra
A959 blends showed droplets or elongated structures of the liquid crystalline phase at
concentrations below 10 wt% and only fibrils (typically 2-5 m diameter) at
concentrations of the liquid crystalline polymer greater than 20 wt%.
The PEI/Vectra A959 extrudates from a capillary die with shorter length:diameter (L/D)
ratios had the modulus and strength higher than those obtained with longer L/D ratios.
695
Handbook of Polymer Blends and Composites
Blizard and Baird [80] studied the extent of fibrillation in the PA-6/(PET-HBA;40:60)
blends prepared by extrusion through capillary dies of various L/D ratios and immediate
quenching. They obtained blends with higher mechanical properties in which fibrils
appeared when the liquid crystalline concentration exceeded 30 wt%.
The influence of L/D ratio on the fibril formation was studied using a capillary die with
a short band [80]. Capillary rheometry showed that fibrils are readily produced with a
die of L/D = 7.82, but not with one having L/D = 21.4.
La Mantia and co-workers [119] obtained similar results for the (PA-6)/Vectra B950
blends. SEM analyses of a 20 wt% Vectra B950 blend, extruded at a shear rate of 1200 s-1
through a capillary die (L/D = 40), showed no fibrillation, but when the blend passed,
under identical conditions, through a die having L/D = 0 and conical inlet, fibrils were
produced. These results showed that the fibrils can be lost during the flow in a long
capillary, when the orientational relaxation time of the liquid crystalline polymer is less
than the time necessary to flow through the capillary.
The importance of extensional flow on fibril formation for the PC/(PET-HBA) blends
has been underlined by Jung and Kim [120]. SEM analyses for strands of blends containing
30 wt% liquid crystalline polymer produced through capillary rheometry, showed the
lack of fibrillation, although a 5 wt% liquid crystalline polymer blend, having a draw
ratio of 15, gave an extruded fibril morphology.
For the PC/Vectra A900 and PC/(PET-HBA, 40:60) blends, as well as for PPS/(PET-HBA
40:60 or 20:80) blend having polyphenylene sulfide (PPS), Ramanathan and co-workers
[121, 122] reported an improvement in tensile strength and modulus, but did find that
the fibril formation for all cases, except the 1:1 PPS/(PET-HBA) for both ratios of PET-
HBA (40:60 or 20:80) extruded with a draw ratio of up to 9. The lack of fibrillation in
the case of Vectra A900/(PET-HBA) blend may be explained by the higher temperature
of processing (260 C) at which the viscosity of liquid crystalline polymer component is
likely to be higher than those of PC. For the PPS/Vectra A900 blend, the lack of fibrillation
was attributed to a possible chemical reaction between the two components under
processing conditions.
Lee and Dibenedetto [123] used the chemical reactions to improve the adhesion between
incompatible aromatic fibres and thermoplastic matrix. In composites LCP/LCP such as
KU-9211(BayerA6)/(PET-HBA; 40:60), KU-9211 acted as reinforcing phase. Extruded
and drawn samples properties were influenced by chemical interactions between the two
thermotropic liquid crystalline polymers.
696
Liquid Crystalline Polymers in Polymer Blends
The PET-HBA (40:60) also bonds well with conventional thermoplastics [124, 125],
such as PET.
Zhang and co-workers [126] studied the correlation between the miscibility of the blend
components, the liquid crystalline polymer concentration and the morphology for the
PET-HBA (40:60) with PS, PC and PET blends. By DSC, SEM and dielectric thermal
analysis methods, they demonstrated the dependence of the blend morphology on the
component miscibility. So, liquid crystalline polymer was immiscible with PS and gave
the most coarse and defined morphology, while for the most miscible system PET/(PET-
HBA; 40:60) the obtained blend had a less distinctive, but the finest morphology.
Transition from ellipsoidal to fibrous liquid crystalline polymer domains with the
increasing shear rate in capillary extrudates was also show [126].
Compatibility between PET and a LCP, such as PET-HBA (40:60), was investigated by
Brostow and co-workers [127]. In the case of the PET/LCP blend with high content of
liquid crystalline polymer, they observed a limited degree of solubility of LCP in PET.
Adding a small amount of LCP (2-5 wt%) the observed mechanical properties decreased,
then increased to maximum and fall again as the LCP content increased.
Friedrich and co-workers [128] studied the blends PC/(PET-HBA) that usually had two
values for Tg (the lower associated with the PET-rich phase of liquid crystalline polymer
and the higher with HBA-rich phase), indicating the immiscibility. Concomitantly, DSC
measurements gave a single Tg value on annealing of the PC/(PET-HBA) blend, interpreted
as a partial miscibility through transesterification.
For the same system, PC/(PET-HBA; 40:60), Jung and Kim [120] reported that Tg of PC
decreased by a small addition of PET-HBA (40:60) although the glass transitions of the
liquid crystalline polymer remain unchanged. They interpreted this fact as a possible
exclusion of the PC from liquid crystalline polymer phase, while some partial mixing in
the PC phase occurred. Good interface adhesion was also observed between the two
components in PC/(PET-HBA; 40:60) blends [104].
Shin and Chung [130] showed that PET was nucleated by an unusual liquid crystalline
polymer [melting temperature (Tm) = 272 C] containing long methylene spacer units.
697
Handbook of Polymer Blends and Composites
While the blend components were found to be immiscible, the strong interface adhesion
between liquid crystalline fibrils and PET determined a good fibre reinforcement.
An important finding of Kricheldorf and co-workers [131] was the formation lyotropic
blends from rigid-rod polyesters (RRP), which is not liquid crystalline as a neat materials,
and commercial poly(-caprolactone) (PCL), acting as solvent. The RRP studied in [131]
were those from Scheme 16.5.
Despite the non-symmetrical substitution patterns, most polyesters, such as RRP derived
from monosubstituted terephthalic acid [132], are semicrystalline materials. The glass
transition temperatures of RRP are so low, that drying of the freshly precipitated samples
at 115 C has an annealing effect.
The melts of RRP showed a Schlieren texture and mobility typical for a nematic liquid
crystal at optical polarising microscopy. Only the RRP derived from phenylhydroquinone
do not form a liquid crystalline melt because their very low melting temperatures.
698
Liquid Crystalline Polymers in Polymer Blends
Scheme 16.5 Rigid-rod polymers (RRP) used by Kricheldorf and co-workers [131] to
obtain RRP/LCP blends
699
Handbook of Polymer Blends and Composites
The lyotropic blends of 2, 3, 4 and 5 RRP in PCL (PCL acted as solvent) that could not
be achieved by blending in an extruder or kneader, were obtained by coprecipitation of
both components from the co-solvent. A significant mechanical reinforcement effect,
was shown in the case of RRP/PCL blends when RRP had aromatic substituents, such as
2c, 3b, 4d and 5c.
The elastic modulus (E) reflecting the elasticity, the storage modulus (E) reflecting the
transformation of mechanical energy into heat, and their quotient, tan , were recorded.
Figure 16.17a displays the measurement results for RRP-2a blend. Compared to neat
PCL (Figure 16.17b), the loss modulus was almost unchanged, but the storage modulus
showed a 80% gain at low temperatures.
The normalised storage moduli E/EPLC for the samples: neat PCL; 4 wt% in RRP in
RRP/PCL blends with RRP-2a and RRP-1; PMMA/PCL blend with 4 wt% polymethyl
methacrylate (PMMA) and P3-HBA/PCL blend, with 4 wt% poly(3-hydroxybenzoic acid)
(P3-HBA) in PCL are listed in Table 16.6.
From the data in Table 16.6 it can be seen that the highest values for the normalised
storage moduli are obtained for blends containing RRP 1 and 2a, while for those with
PMMA and P3-HBA, the corresponding values are near unity.
700
Liquid Crystalline Polymers in Polymer Blends
(a)
(b)
Figure 16.17 DMA measurement of a blend containing 4 wt% of RRP 2a. Neat PCL
measured under identical conditions gave a storage modulus (E) of 4.6 x 102 MPa
(E = loss modulus) [131]
Reproduced with permission from H.R. Kricheldorf, L.H. Wahlen, C. Friedrich and T.J.
Menke, Macromolecules, 1997, 30, 9, 2642. Copyright 1997, American Chemical Society
701
Handbook of Polymer Blends and Composites
The normalised elastic moduli (stress-chain measurements) for some RRP/PCL blends of
8 wt% RRP in PCL are listed in Table 16.8. The highest values from this table correspond
to the blends containing the -5c, -2a, and 1 RRP.
Between the 8 wt% RRP/PCL blends with RRP derived from methylhydroquinone (-2a,
-1, -3a, -5a, -4b), the most effective from point of view of normalised elastic are those
containing 2a, -1 and 3a RRP (see Table 16.8).
All RRP-4/PCL blends have normalised elastic moduli near the unity, excepting RRP-4c/
PCL blend that exhibited a value below the unity.
In the 8 wt% RRP-5/PCL blends, the high normalised elastic moduli are induced by a
high rigidity of the repeating units (see Scheme 5 and Table 16.8). The highest value of
the normalised elastic moduli, of about 2.40, has been obtained for RRP-5c/PCL blend.
702
Liquid Crystalline Polymers in Polymer Blends
The reinforcing effect of some RRP in blends with PCL was explained by electronic
interactions between the aromatic -electrons and the polar ester groups of the PCL.
16.6.2.1 Rheology
From the rheological measurements it resulted that the time sweeps at a frequency of
1 Hz generally show a slow increase of both G and G and reached constant values
after 2 hours. The increase of G for 1 wt% RRP in PCL, at 80 C was 30% from the
initial value. The comparable long relaxation times of RRP/PCL blends could even indicate
that the lyotropic blends posses a distinct anisotropy, characteristic to a nematic liquid
crystalline phase. As it can be seen in Figure 16.18, the 1 wt% RRP-4b/PCL blend has a
viscoelastic behaviour, with G closed to that of the neat PCL. At lower frequencies
(wat <1 Hz), the storage modulus (G) considerably deviates from that of PCL, reaching
a secondary plateau at around 100 Pa.
Inducing morphological changes, the shear significantly influences the properties of these
materials. Shearing of the samples at elevated temperatures (time sweep for 2 h at 180 C,
1 Hz and 10% strain) prior to rheological investigation leads to significant changes in
the rheological response of the blends.
16.6.2.2 Morphology
Connections between the transition temperature and the breakdown of orientation and
thus, the reinforcement in the RRP/PCL blends are shown in [131]. PCL is a semicrystalline
polymer above its melting point Tm, but its blends containing 1 wt% of a RRP showed
birefringence only in 15-25% of the total volume, while the concentrations of 4 wt% or
8 wt% of RRP in RRP/PCL blends induced birefringence in the entire volume. The
birefringent phase can be sheared and oriented above Tm of the PCL and thus, it must be
considered as a mobile anisotropic phase [131].
The homogeneous birefringence is stable at temperature below 150 C for many hours
(its final disappearance has never been observed), but slowly vanishes when the
temperature approaches the Tm of the RRP. Above Tm of the neat RRP the Schlieren
texture of the neat RRP may become detectable.
It results that an initially homogeneous finely dispersed phase of RRP in PCL, undergoes
an irreversible phase separation at temperatures around the Tm of RRP.
For all the blends the phase separation was irreversible if the cooling rate was very
high or extremely slow (10 C/min). The existence of a rather stable mobile birefringent
703
Handbook of Polymer Blends and Composites
Figure 16.18 Rheological measurements (80 C) of the neat PCL and the blends
containing 1 wt% of 2a: (A) storage modulus of the blend; (B) storage modulus of
PCL; (C) loss modulus of the blend; (D) loss modulus of PCL [131]
Reproduced with permission from H.R. Kricheldorf, L.H. Wahlen, C. Friedrich and T.J.
Menke, Macromolecules, 1997, 30, 9, 2642. Copyright 1997, American Chemical Society
phase below the Tm of RRP but above Tm of the polymer matrix, is characteristic of
these blends.
The formation of the anisotropic phase is dependent on the electronic interactions between
-electrons and the polar ester groups of the PCL. Only the blends containing aromatic
substituents show a significant reinforcing effect.
704
Liquid Crystalline Polymers in Polymer Blends
The coprecipitation of PCL with an RRP having more or fewer aromatic substituents
yields a blend with a nearly molecular dispersion of the RRP. The individual RRP chains
of the stacks of few chains are solvated by PCL chains which are more or less aligned
parallel along the RRP
The anisotropy in the matrix surrounding the RRP chains (or stacks) is induced by the
orientation of the PCL chains along the RRP chains. The induced anisotropy of the
matrix contributes to the mechanical reinforcement. The birefringent mobile phase in
RRP/PCL blends encompasses a much larger volume than the neat RRP itself, allowing
shearing above Tm of PCL.
Figure 16.19 (A) Schematic illustration of the double stacks formed by polyester
derived from monosubstituted terephthalic acids. (B) Schematic illustration of the
stacks and layer structure of polymers derived from 2,5-disubstituted terephthalic
acids [131]
Reproduced with permission from H.R. Kricheldorf, L.H. Wahlen, C. Friedrich and T.J.
Menke, Macromolecules, 1997, 30, 9, 2642. Copyright 1997, American Chemical Society
705
Handbook of Polymer Blends and Composites
Because the compatibilising role of aromatic substituents for various matrix, a lyotropic
blend can be obtained even from two isotropic components, for example the RRP-2c/
PCL blend reported in [131].
The Ballauff and Flory theory [133] predicts that the formation of the mesophase strongly
depends on the free volume of the neat liquid crystalline polymer. The melt of RRP-2c
containing too much free volume above the Tg, as well as the intensive solvation by more
or less parallel aligned PCL chains improve the stiffness of this aggregate and reduce its
free volume. So, from two isotropic components a nematic blend can appear.
La Mantia and co-workers [134-136] measured the melt viscosity, the spinnability
evaluated through the melt strength and the breaking stretching ratio (BSR), such as the
mechanical properties for several blends of flexible polymers with wholly aromatic and
semi-aromatic liquid crystalline polymers spun under different conditions. The
thermoplastic polymers and liquid crystalline polymers listed in Table 16.9 were used to
obtain blends with improved mechanical properties.
Blends containing 0.5 wt%, 10 wt%, 20 wt% LCP and a thermoplastic polymer were
prepared with a laboratory single screw extruder (D = 19 mm, L/D = 25, Brabender,
Germany) equipped with a die assembly for ribbon extrusion. The shear viscosities of
some investigated polymer/LCP blends, measured at shear rate of 24 s-1 generally decreased
with the growth of LCP content.
The blends PET/Va and PA/VB have minimum of shear viscosity, that usually can appear
[133, 134] when the viscosity of LCP is similar or higher than that of the matrix.
Concomitantly, the addition of LCP in the studied blends [134] leads to an increase of
elongational viscosity as it is shown in Figure 16.20.
Figure 16.20 contains the values of the elongational viscosity as a function of extrusion rate
for the VB, PA, and PA/VB (90:10) blend. The elongational viscosity of the 90:10 PA/VB
blend is considerably higher than that of the pure matrix. Similar behaviour, reported in
[95, 137], has been attributed to the considerable amount of energy required for deforming
706
Liquid Crystalline Polymers in Polymer Blends
Table 16.9 Characteristics of the polymers used for the blend preparation
[134]
Reproduced with permission from F.P. La Mantia, A. Roggero, U. Pedretti and P.L.
Magagnini, Liquid Crystalline Systems, Technological Advances, Eds., A.I. Isayev, T. Kyu
and S.Z.D. Cheng, American Chemical Society, Washington, DC, 1996, Chapter 8, 110
Sample Mw Name Manufacturer
PP 680000 D60P Himont
PC 36000 Sinvet 301 Enichem Polimeri
PET Enichem Polimeri
PA 62000 ADS40 SNIA
VA Vectra A-900 Technical Polymers Ticoria
VB Vectra B-950 Technical Polymers Ticoria
SBH SBH 1:1:2 Eni Tecnologie
SBHN SBHN 1:1:3:5 Eni Tecnologie
Chemical composition of the liquid crystalline polymers is: Va=[(4-hydroxybenzoic acid
(HBA) (73%), 2-hydroxy-6-naphthoic acid (HNA) (27%)]: VB= [HNA (60%),
terephthalic acid (TA) (20%), 4-aminophenol (4AP) (20%)]; SBH= [sebacic acid (SA)
(25%), 4,4-dihydroxybiphenyl (DHBP) (25%), HBA (50%)], SBHN = [SA (10%),
DHBP (10%), HBA (30%), HNA (50%)]
and orienting the liquid crystalline polymer domains in the convergent flow at the die
entrance.
Melt strength and BSR measurements (carried out under non-isothermal conditions of
the spinning line), were made by La Mantia and co-workers in order to obtain information
on the spinnability of the studied polymer/LCP blends. The results are shown in
Figure 16.21 for melt strength and in Table 16.10 for BSR.
Decrease in melt strength for some blends such as PC/VB, PC/SBH, PET/SBH, PET/
SBHN and PA/SBH concomitantly with the increases in melt strength for the other were
reported. Under non-isothermal conditions, the viscosity of some blends may be higher
than that of the neat matrices, in some temperature intervals, if the viscosity of the two
components has different activation energies.
707
Handbook of Polymer Blends and Composites
The BSR values, i.e., the ratios of the wind up speeds to the extrusion speeds to filament
breaking, measured at a shear rate 24 s-1, are listed in Table 16.10 for some of the blends
studied.
By adding LCP to the polymer/LCP blends, the BRS decreased for all investigated blends,
indicating the two phases nature of these blends. The extent of BSR reduction significantly
depends on the chemical structure of the two blended components. The decrease of BSR
is very small in the case of PET blended with semi-crystalline LCP such as SBH or SBHN,
whereas it significantly drops when PET is blended with wholly aromatic ones, such as
Va. The same reduction of BSR exhibited the PA/LCP and PC/LCP blends with aromatic
or semi-aromatic LCP. The spinning behaviour of the PP/SBH blend is characterised by a
marked increase of melt strength and a fairly small BSR reduction with the increase of
LCP content.
The strongest modulus improvement is found for PA especially when a wholly aromatic
copolyesteramide (VB) is added into it (Eb/ELCP = 2.75, for 20 wt% LCP). Good
reinforcement was also obtained for the blends PA/SBH, containing a semi-aromatic
copolyester.
708
breaking stretching ratios
Figure 16.21 Melt strength of some LCP/polymer blends as a function of the LCP
content [117]
Reproduced with permission from C. Kiss, Polymer Engineering and Science, 1987, 27,
410. Copyright 1987, Society of Plastics Engineers
709
Handbook of Polymer Blends and Composites
An enhancement of mechanical properties was obtained for the PC-based blends but Eb/
ELCP was smaller comparatively with PA based blends. The results are explained by the
difference between the tensile modulus of PC fibres and PA fibres prepared under
comparable conditions (60 GPa for PC and 0.8 GPa for PA).
For PET-based blends, addition of a semi-aromatic LCP, such as SBH or SBHN, leads to
remarkable modulus enhancement, whereas a stiffer, wholly aromatic, LCP such as Vectra
A, does not influence the value of the elastic modulus of the blends. On the contrary, for
compression moulded blends, the modulus improvement was found to be higher for
PET/Va blends than for PET/SBH blends. Although PP/SBH blends show good values of
melt strength and BSR, the improvement in mechanical properties was negligible. The
authors explained this fact by a strongly depressed interphase adhesion between PP and
SBH in the solid state compared with molten state.
Despite the fact that the LCP usually reduce the shear viscosity of molten flexible polymers,
the melt strength and BRS of the polymer/LCP blends may still be high enough to grant
good spinnability, indicating the fibrillation of the LCP particles and the orientation of
the matrix macromolecules. High values of melt strength and BRS indicate a good phase
compatibility in the molten state.
A considerable reinforcing effect has been shown only for the blends with good spinnability
and good interphase adhesion.
Isayev and Varma [89] studied the blends of PAI, high performance polymer having high
viscosity and melt reactivity, with a LCP having easy processability and self-reinforcing
capacity. The blends were obtained using two installations; one containing a co-rotating
twin screw extruder and another one with a single screw extruder attached to a static
mixer. The blends were injection moulded and characterised both before and after heat
treatment. The thermal, flow and mechanical properties of the blends and of the pure
components were studied. The PAI used were Torlon 4000Tf-40 (PAI-1) and Torlon 4203L
(PAI-2) supplied by Amoco Performance Products. The thermotropic LCP was Vectra A950
a random HBA-HNA copolyester from Celanese Company.
The DMTA studies indicated that PAI-1/LCP and PAI-2/LCP are immiscible blends. After
heat treatment, the molecular stiffness and Tg increased concomitantly with a decrease of
the loss tangent. Viscosity studies [89] revealed a complete segregation of the components,
the lower viscosity component (LCP) forming the outer layer.
710
Liquid Crystalline Polymers in Polymer Blends
After heat treatment, the tensile modulus and elongation of the blends and of the pure
components increased concomitantly, with an increasing in the flexibility and toughness
of the liquid crystalline polymer, due to the loss of the orientation. Tensile strength modulus
of LCP decreased on the heat treatment. The water loss by heating leads to the decrease
of flexibility of the PAI chains.
The general tendency after heating of PAI/LCP blends with different proportions of PAI,
is an increasing of the break stress and of the secant modulus. No significant improvement
in the modulus and other tensile properties were observed on curing.
Mechanical studies of the blends PAI/LCP show that the properties of the pure components
are better than those of the blends, hence no reinforcement due to the fibre formation is
evident by morphological studies. Voids and destruction of fibres on heat treatment
were observed. Agglomerations and segregation, indicating a poor mixture, combined
with the lack of interaction between the phases, lack of reinforcing fibres and the inhibition
of the imidisation reaction of PAI in the blends contributed to the poor mechanical
properties of the blends.
Because the high-performance polymers, such as LCP, are often quite costly, there is an
incentive to obtain materials with relative small amounts of liquid crystalline polymers
into commercially available polymers.
Main-chain or side-chain liquid crystalline polymers can be mixed with flexible engineering
polymers in order to obtain systems as polymer-dispersed liquid crystalline polymer, or
polymer-stabilised liquid crystalline polymer, used in electro-optical applications (liquid
displays, light shutters, etc).
Increasing applications of liquid crystalline polymers in many fields are based on their
capacity to modify compatibilisation, morphology, as well as mechanical and thermal
711
Handbook of Polymer Blends and Composites
performances of the blends. These results are very useful to those industries using liquid
crystalline polymer blends for electronics, telecommunications, fibre optics, consumer
goods, automotive, marine or aerospace applications.
Acknowledgments
I gratefully acknowledge Professor Cornelia Vasile, who read this chapter and offered
many helpful suggestions to improve it.
References
1. Liquid Crystals: The Fourth State of Matter, Ed., F.D. Saeda, Marcel Dekker,
New York, NY, USA, 1979.
2. Handbook of Liquid Crystals, Eds., D. Demus, J.W. Goodby, G.W. Gray, H.W.
Spiess and V. Vill, Wiley-VCH, Weinheim, Germany, 1998.
4. A. de Vries in Liquid Crystals: The Fourth State of Matter, Ed., F.D. Saeda,
Marcel Dekker, New York, NY, USA, 1979, Chapter 1, 1.
6. H.W. Gibson in Liquid Crystals: The Fourth State of Matter, Ed., F.D. Saeda,
Marcel Dekker, New York, NY, USA, 1979, Chapter 3, 99.
9. S.E.B. Petrie in Liquid Crystals: The Fourth State of Matter, Ed., F.D. Saeda,
Marcel Dekker, New York, NY, USA, 1979, Chapter 4, 163.
712
Liquid Crystalline Polymers in Polymer Blends
10. H. Sackmann and D. Demus, Molecular Crystals and Liquid Crystals, 1973, 21,
239.
11. A. de Vries and D.L. Fishel, Molecular Crystals and Liquid Crystals, 1972, 16, 311.
14. Liquid Crystalline Polymers, Ed., C. Carfagna, Pergamon Press, Oxford, UK,
1994.
15. Side Chain Liquid Crystals Polymers, Ed., C.B. McArdle, Blackie, Glasgow, UK,
1989.
16. M. Kotera, J.M. Lehn and J.P. Vigneron, Chemical Communications, 1994, 197.
17. C. Alexander, C.P. Jariwala, C.M. Lee and A.C. Griffin, Die Makromolekulare
Chemie - Macromolecular Symposia, 1994, 77, 283.
20. U. Kumar, J.M.J. Frechet, T. Cato, S. Ujiie and K. Timura, Angewandte Chemie,
International Edition, 1992, 31, 1531.
21. Recent Advances in Liquid Crystalline Polymers, Ed., L.L. Chapoy, Elsevier,
London, UK, 1985.
23. A.I. Isayev in Liquid Crystalline Polymer Systems Technological Advances, Eds.,
A.I. Isayev, T. Kyu and S.Z.D. Cheng, American Chemical Society, Wasighton,
DC, USA, 1996, Chapter1, 1.
24. M. Jaffe, P. Choe, I.S. Chung and S. Makhija, Advances in Polymer Science,
1994, 117, 297.
25. S.G. Cottis, J. Economy and L.C. Wohrer, inventors; The Carborundum
Company, assignee; US Patent, 3,975,487, 1976.
26. W.J. Jackson, Jr. and H.F. Kuhfuss, inventors; Eastman Kodak, assignee; US
Patent 3,778,410, 1973.
713
Handbook of Polymer Blends and Composites
27. W.J. Jackson, Jr. and H.F. Kuhfuss, Journal of Polymer Science, Chemistry
Edition, 1976, 14, 2043.
29. A.J. East and G.W. Calundann, inventors; Celanese Corporation, assignee; US
Patent, 4,318,841, 1982.
30. A.J. East and G.W. Calundann, inventors; Celanese Corporation, assignee; US
Patent, 4,318,842, 1982.
31. G.W. Calundann and M. Jaffe in Proceedings of the Robert A. Welch Foundation
Conference on Chemical Research: Synthesis of Polymers, 1982, 36, 247.
32. C.S. Brown and P.T. Alder in Polymer Blends and Alloys, Eds., M.J. Folkes and
P.S. Hope, Blackie, Glasgow, UK, 1993, Chapter 8, 195.
33. M.G. Dobb and J.E. McIntyre, Advances in Polymer Science, Springer, New
York, NY, USA, 1984, 61.
34. C.B. Hansson, Physical Data Book, Pergamon Press, Oxford, UK, 1972.
35. A.A. Bright and L.S. Singer, Carbon 1979, 17, 57.
36. P.G. Hedmark, J.M.R. Lopez, M. Westdahl, P. Werner, J. Janseen and U.W.
Gedde, Polymer Engineering and Science, 1988, 28, 19, 1248.
37. F. Roussel, J.M. Buisine, U. Maschke and X. Coqueret, Molecular Crystals and
Liquid Crystals, 1997, 299, 321.
40. D. Dorohoi, Analele Stiintifice Universite Al .I. Cuza Iasi, s.Chimie, 2000, 8, 1, 35.
42. V.N. Tsvetkov, E.J. Riumtsev, I.N. Shtennikova, E.V. Korneeva, B.A. Krentsel and
Y.B. Amerik, European Polymer Journal, 1973, 9, 6, 481.
714
Liquid Crystalline Polymers in Polymer Blends
43. H.R. Kricheldorf, L.H. Wahlen, C. Friedrich and T.J. Menke, Macromolecules,
1997, 30, 9, 2642.
45. M.J. Seurin, A.T. Bosch and P. Sixou, Polymer Bulletin, 1983, 9, 8-9, 450.
46. F. Fried, J.M. Gilli and P. Sixou, Molecular Crystals and Liquid Crystals, 1983,
98, 209.
47. R.S. Werbowyj and D.G. Gray, Macromolecules, 1984, 17, 8, 1512.
48. J.M. Gilli, J.F. Pinton, P. Sixou, A. Blumstein and O. Thomas, Molecular Crystals
and Liquid Crystals, 1985, 1, 123.
49. J. West in Liquid-Crystalline Polymers, Eds., R.A. Weiss and C.K. Ober, ACS
Symposium Series, No. 435, American Chemical Society, Washington, DC, USA,
1990, Chapter 32, 475.
50. J.W. Doane, in Liquid Crystals: Applications and Uses, Ed., B. Bahadur, World
Scientific, River Edge, NJ, USA, 1990, 361.
51. Liquid Crystals in Complex Geometries, Eds., G.P. Crawford and S. Zumer,
Taylor and Francis, London, UK, 1996, Chapter 2, 21.
53. C.T. Imrie and B.J.A. Paterson, Macromolecules, 1994, 27, 22, 6673.
56. S. Ambrosino and P. Sixou, Journal of Polymer Science, Part B, 1994, 32, 1, 77.
57. C.D. Han and Y.W. Kim, Transactions of the Society of Rheology, 1975, 19, 245.
58. L.A. Utracki, A.M. Catani, G.L. Bata, M.R. Kamal and V. Tan, Journal of
Applied Polymer Science, 1982, 27, 6, 1913.
59. G.V. Vinogradov, B.V. Yarlykov, M.V. Tsebrenko, A.V. Yudin and T.I. Ablazova,
Polymer, 1975, 16, 7-8, 609.
715
Handbook of Polymer Blends and Composites
60. Y.S. Lipatov, A.F. Nesterov, V.F. Shumsky, T.D. Ignatova and A.N. Gorbatenko,
European Polymer Journal, 1982, 18, 11, 981.
61. Thermotropic Liquid Crystal Polymer Blends, Ed., F.P. La Mantia, Technomic
Publishing Company, Inc., Lancaster, PA, USA, 1993.
62. U. Kumar, T. Kato and J.M. J.Frechet, Journal of the American Chemical Society,
1992, 114, 6630.
67. T. Schleeh, C.T. Imrie, D.M. Rice, F.E. Karasz and G.S. Attard, Journal of
Polymer Science, Part A: Polymer Chemistry, 1993, 31, 7, 1859.
68. Y. Kosaka and T. Uryu, Journal of Polymer Science, Polymer Chemistry Edition,
1995, 33, 13, 2221.
73. K.I. Alder, D. Stewart and C.T. Imrie, Journal of Materials Chemistry, 1995, 51,
12, 2225.
74. C. Fouquey, J.M. Lehn and A-M. Levelut, Advanced Materials Journal, 1990, 2,
5, 254.
75. D. Dutta, H. Fruitwala, A. Kohli and R.A. Weiss, Polymer Engineering and
Science, 1990, 30, 17, 1005.
716
Liquid Crystalline Polymers in Polymer Blends
76. F.N. Cogswell, B.P. Griffin and J.B. Rose, inventors; Imperial Chemical
Industries, assignee; US Patent 4,433,083, 1984.
78. B. Lee, Proceedings of Antec 88, Atlanta, GA, USA, 1988, 1088.
80. K.G. Blizard and D.G. Baird, Polymer Engineering Science, 1987, 27, 9, 653.
82. R.A. Weiss, W. Huh and L. Nicolais, Polymer Engineering and Science, 1987, 27,
9, 684.
83. V.G. Kulichikhin, O.V. Vasileva, I.A. Litinov, E.M. Antopov, I.L. Parsamyan and
N.A. Plate, Journal of Applied Polymer Science, 1991, 42, 2, 363.
85. W.R. Schowalter, C.E. Chaffey and H. Brenner, Journal of Colloid and Interface
Science, 1968, 26, 152.
86. G.I. Taylor, Proceedings of the Royal Society, London, 1932, A138, 431.
87. G.I. Taylor, Proceedings of the Royal Society, London, 1934, A146, 501.
88. H.K. Chuang and C.D. Han, Journal of the American Chemical Society, 1984,
106, 171.
89. A.I. Isayev and T.R. Varma in Liquid-Crystalline Polymer Systems, Eds., A.I.
Isayev, T. Kyu and S.Z.D. Cheng, ACS Symposium Series, No. 632, American
Chemical Society, Washington, DC, USA, 1996, Chapter 10, 142.
90. R.F. Heitmiller, R.Z. Maar and H.H. Zabusky, Journal of Applied Polymer
Science, 1964, 8, 873.
91. C. Lees, Proceedings of the Physical Society, London, 1990, 17, 460.
717
Handbook of Polymer Blends and Composites
93. M.T. De Meuse and M. Jaffe, Molecular Crystals and Liquid Crystals, 1988, 157,
535.
94. L.A. Utracki, Polymer Alloys and Blends, Carl Hanser Verlag, Munich, Germany,
1989, 192.
97. M. Doi and T. Ohta, Journal of Chemical Physics, 1991, 95, 1242.
98. H.M. Lee and O.O. Park, Journal of Rheology, 1994, 38, 5, 1405.
99. G.K. Guenther and D.G. Baird, Journal of Rheology, 1996, 40, 1, 1.
101. F. Fried and P. Sixou, Journal of Polymer Science, Polymer Chemistry Edition,
1984, 22, 1, 239.
102. P. Navard, Journal of Polymer Science, Polymer Physics Edition, 1986, 24, 2, 435.
103. S. Ambrosino, T. Khallala, M.J. Seurin, A. Ten Bosch, F. Fried, P. Maissa and P.
Sixou, Journal of Polymer Science, Polymer Letters, 1987, 25, 8, 351.
105. S. Lee, P.T. Mather and D.S. Pearson, Journal of Applied Polymer Science, 1996,
59, 2, 243.
106. D.S. Kalika, L. Nuel and M.M. Denn, Journal of Rheology, 1989, 33, 7, 1059.
107. D.S. Kalika, D.W. Giles and M.M. Denn, Journal of Rheology, 1990, 34, 21, 139.
109. R.L. Brady, and R.S. Porter, Journal Thermoplastic Composite Materials, 1990,
3, 253.
718
Liquid Crystalline Polymers in Polymer Blends
111. H.H. Chuah, T. Kyu and T.E. Helminiak, Polymer, 1987, 28, 12, 2130.
112. C. Taesler and H.R. Kricheldorf and J. Petermann, Journal of Materials Science,
1994, 29, 11, 3017.
114. G. Kiss, A.J. Kovacs and J.C. Wittmann, Journal of Applied Polymer Science,
1981, 26, 8, 2665.
118. S. Swaminathan and A.I. Isayev, Polymer Materials Science and Engineering,
1987, 57, 330.
119. F.P. La Mantia, A. Valenza, M. Paci and P.L. Magagnini, Polymer Engineering
and Science, 1990, 30, 1, 22.
120. S.H. Jung and S.C. Kim, Polymer Journal (Japan), 1988, 20, 1, 73.
121. R. Ramanathan, K.G. Blizard and D.G. Baird, Proceedings of Antec 87, Los
Angeles, CA, USA, 1987, 1399.
122. R. Ramanathan, K.G. Blizard and D.G. Baird, Proceedings of Antec 88, Atlanta,
GA, USA, 1988, 1123.
123. W-C. Lee and A.T. Dibenedetto, Polymer Engineering and Science, 1992, 32, 6, 400.
124. E.G. Joseph, G.L. Wilkes and D.G. Baird in Polymeric Liquid Crystals, Ed., A.
Blumstein, Plenum Press, New York, USA, 1985, 197.
125. J-F. Croteau and G.V. Laivins, Journal of Applied Polymer Science, 1990, 39, 11-
12, 2377.
126. P. Zhang, T. Kyu and J.L. White, Polymer Engineering and Science, 1988, 28, 17,
1095.
719
Handbook of Polymer Blends and Composites
129. L.I. Minkova, M. Paci, M. Pracella and P. Magagnini, Polymer Engineering and
Science, 1992, 32, 1, 57.
130. B.Y. Shin and I.J. Chung, Polymer Engineering and Science, 1990, 30, 1, 13.
131. H.R. Kricheldorf, L.H. Wahlen, C. Friedrich and T.J. Menke, Macromolecules,
1997, 30, 9, 2642.
132. H.R. Kricheldorf and J. Engelhart Journal of Polymer Science, Part A: Polymer
Chemistry, 1990, 28, 9, 2335.
134. F.P. La Mantia, A. Roggero, U. Pedretti and P.L. Magagnini in Liquid Crystalline
Polymer Systems, Eds., A.I. Isayev, T. Kyu and S.Z.D. Cheng, ACS Symposium
Series, No. 632, American Chemical Society, Washington, DC, USA, 1996,
Chapter 8, 110.
136. F.P. La Mantia and A. Valenza, Polymer Networks and Blends, 1993, 3, 125.
137. D.E. Turek, G.P. Simion, F. Smejkal, M. Grosso, L. Incarnato and D. Acierno,
Polymer Communications, 1993, 34, 204.
720
INDEX
Applications (Cont.)
electronics 91
energy sector 98
engineering 91
housing sectors 98
railways 117
shipbuilding 510
societal technology sector 98
sports and leisure goods 98 120
structural 508
transportation 97
vehicle 467
Aramid fibres 34 74
Kevlar 74
Aromatic diphenols 336
ARPANET 493
ATLAC 580 361
Autoclave curing 398
resin flow 398
tool plate 398
Automated Tape Placement 484
Automotives 117
plastics in 227
use of GRP in 117
Automotive frame 464
EB-cured 464
Bakelite 22
Bamboo curtain plywood 166
floor 166 171
mat plywood 165 175
mechanical properties 175
physical properties 175
parallel strip lumber 165
plate plywood 165
strip board 171
plywood 175
mechanical properties 175
physical properties 175
Barium sulphate 257
azo groups 257
Bicycle
frame 126
Bikerman model 248
Biofibres 255
lignocelluloses 255
Birch plywood 173
mechanical properties 173
Bisphenol-A 332
Bleeder cloth 399
Bleeder layer 398
Blockboard 164
solid 170
Blockboard 164
solid 170
mechanical properties 174
physical properties 174
BMC 57
Board 182
wheat straw 182
Boat
laminated 26
Bone 6 8
Boron fibres 14
properties 69
Borsic fibres 69
Brake disc 134
composite 134
Bridge deck 130
glass fibre reinforced 130
Brominated diphenyl oxide 332
Brominated polystyrene 339
Bulk system 370
Butadiene acrylonitrile 423
carboxy terminated 423
Composites (Cont.)
additives 215
advanced polymer 27 206
aircraft/aerospace industry 221 223
aminoplasts 212
aramid fibres 218
automotive industry 29 221 224
bonding 451
boron 27
building and construction 228
carbon fibres 218
carbon tubes 205
cement and wood particle 178
civil engineering 517
coil spring 129
concept vehicle 455 466
electron beam bonding 466
consumer pattern 493
consumption 200 202 207 221
coupling agents 215 220
curing 451 454
electron beam 454
damage repair 114
decorative 159
deepwater 34
developments 237
new trends 238
fabrication 24 25 456 486
aerospace 456
VARTM 486
epoxy resin 214
Composites (Cont.)
equipment 234
fibre type 159
fillers 215 220
flake 155
garden furniture 228
glass fibres 216
reinforced 317
glass mats 216
glastic 506
in Asia 89
in bicycles 120
Indian market 91
interface 275
LCP 219
liquid crystal polymer 505
macromolecular 7
manufacturing processes 229 390 452
developments 229
resin type 390
marine sector 223
markets 222
materials polymer-based 510
mixtures 281
mechanical tests 281
mechanical properties 275
microfibrillar 267
new resins 235
non-aerospace industry 89
overlaid 159
particle type 158
PEEK 211 212 503
Composites (Cont.)
phenolic resin 212
plank 155
plywood 164
polyamide 208
polyimides 215
polymer matrix 451
polyolefin-polypropylene 208
polyurethanes 215
portable electric tools 228
pressure vessels 34
processing 461
aerospace 461
properties 472
production 200 207 221
and sales in Korea 137
properties of 315
raw material production
in Korea 133
recent developments 230 231 232
recycling 235 236 506
reinforcements 215 315
short history 199
silicones 215
special 159
special fibres 219
structural lumber 159
structures 393
fabrication 393
Composites (Cont.)
structures laboratory
in India 98
in Korea 123
technology 94 123
missile 94
technology transfer 203
technologies 234
thermoplastic 207
thermosets 212
unsaturated polyesters 214
veneer 155
wood-based 155 159
worldwide consumption 90 202
Composites design centre 97
in India 97
Composites products 233
developments in 233
Compression moulding 96
Connective tissue 9
Contact moulding 96
Cooling towers 100
GRP in 100
Copolyesters 343
Core-shell particles 262
Cotton 260 261
Coupling agents 22 221 246 252
277 315
silane 246 252 277
DGEBA 291
structure of 291
DGEBPA 510 511
Diamino diphenyl sulphone 409
cure data 409
2,5-Dichlorostyrene 339
Dielectric measurements 416
Dielectric traces 430 433
Differential scanning calorimetry 501
Diffusion 247
Diglycidyl ether of bisphenol A 409
cure data 409
Drum chipper 181
DryAdd 434
Dupres equation 249
DuraSpan floor deck 132
E-glass 64
Electron beam
accelerator 457
bonding 456 483
curing 452 456 460 462
483
limitations 453
processing 453
curable adhesives 483
lap-shear strength 483
curable resins 472 478
properties 472
cured components 463
processing 451
Elastomers 57
Electrostatic adhesion 246
Epoxy composites 309
fillers 312
glass-fibre reinforced tetrafunctional 309
paste adhesives 480
EB cured 480
polymers 315
prepreg laminates 105
mechanical values 105
resin 385 389
cure 389 408
resin systems 386
cure chemistry 386 387
Epoxy resin 22 56 255 291
387
5-amino-1-naphthol 293
applications 314
characterisation 306
curing agent 307
curing of 307
dihydroxynaphthalene-based 304
diluents 310
epoxy content 306
flame retardant 300
fortifiers 311
modifiers for 310
multifunctional 292
naphthol-based 304
naphthol-based pentafunctional 305
novalac 292 302 303
Fibre glass 10 11 22
coupling agents 23
reinforced polyesters 22
Fibre optics 501
Fibre reinforced plastics 357 499
Fibre reinforcements 39 61
aluminium oxide