Textbook of Polymer Science

TEXTBOOK OF POLYMER SC
THIRD EDITION
FRED W. BILLMEYER, JR.

Professor oj Analytzcal Chemzshy Rensselaer Polytechnic Instztute, Troy, New York
A Wiley-lnterscience Publication
John Wiley & Sons
New York
Chichester
Brisbane
Toronto
Singapore
Copyright Q 1984 by John Wiley & Sons, Inc All rights reserved Published simultaneously in Canada Reproduction or translation of any part of this work beyond that permitted by Section 107 or 108 of the 1976 United States Copyright Act without the permission of the copyright owner is unlawful Requests for permission or further information should be addressed to the Permissions Department, John Wiley & Sons, Inc
Library of Congress Cataloging in Publication Data:

Billmeyer , F ~ e d W Textbook of polymer science Includes bibliographies and indexes 1 Polymers and polymerization I Title QD381 B52 1984 ISBN 0-471-03196-8 668 9 83-19870
Printed in the United States of' America
PREFACE
''I am incllned to think that the development ojpolymer~zation S , I perhaps, the blggest thrng chemistry has done, where ~thas had the brggest effect on everyday life The world would place w~thoutartijiclal fibers, plastrcs, elastomers, etc Even in the be a totally d~rerent field of electronics, what would you do wlthout msulatron? And there you come back to polymers agaln " f
And indeed one does From the lowly thmwaway candy wrapper to the artificial heart, polymers touch our lives as does no other class of' materials, with no end to new uses and improved products in sight. Yet, many instances of' the need fbr better education in the polymer field, both in our universities and for the public, remain unchanged. Some of these were discussed at length in the Preface to the second edition of'this book, which fbllows, and I shall not repeat them The present revision has two major directions. The first is to improve its value as a textbook. To this end I have rearranged the text to consider polymerization before describing the properties of'polymers, a change that several of my colleagues feel has pedagogical advantages I have also drawn on my files from 2.5 y e a s of' teaching polymer science, at the University of' Delaware, the Massachusetts Institute of' Technology, and Rensselaer Polytechnic Institute, to provide material for a section on Discussion Questions and Problems at the end of' each chapter The second objective of' the revision is the more common one, to bring the contents up-to-date by judicious addition, deletion, and revision, and in this I hope I have been successful. Many sections have been changed little, reflecting the maturity of certain aspects of' polymer science, but the reader will find new material inserted in every chapter. A few additions of' particular note axe a section on polymerization reaction engineering in Chapter 6, a discussion of scaling concepts in Chapter 7, and expansion of' the sections on polymer processing in Chapter 17
i Lord Todd, president of the Royal Soclety of London, quoted in Chem Eng News 58(40), 29 (1980), m answer to the question, What do you think has been chemistry's biggest contribution to science, to Society?
vi
PREFACE
I have tried to include brief descriptions of the new polymer materials in the marketplace in Chapters 13-16 and in a section on composite materials in Chapter 17 Unfortunately, some discussion of less timely topics had to be eliminated to prevent undue expansion of the text My approach to referencing the literature had to remain essentially the same as that adopted in the second edition, despite some dissenting opinions The explosion of the literature in polymer science makes it totally impossible to provide full coverage of original articles, as was possible 20 or 25 years ago I have therefore cited many new books, and many articles from the Encyclopedza ojPolymer Science and Technology, the Kirk-Othmer Encyclopedza oj Chemzcal Technology, third edition, and the Modern Plastics Encyclopedia Each of these sources (save the last, which provides information on curlent commercial products and processes) was selected to provide detailed citation of the original literatwe, as well as more complete coverage of the topic for which it was cited With retirement imminent, I look back with pleasure on the preparation of this volume and its predecessors They have brought me much pleasure, more in the friendship of many readers and colleagues than in the accomplishment I hope that " the usefulness of this last revision will surpass that of those before it
Troy, New York January 1984
ACKNOWLEDGMENTS
I wish to thank many colleagues, both in the Polymer Science and Engineering Program at Rensselaer and elsewhere, for valuable suggestions that have been incorporated into this revision Seventy-five Rensselaer students who used a first draft of about two-thirds of the text in the course Introduction to Polymer Chemistry also provided helpful ideas and corrections The text was capably typed and retyped by Peggy Ruggeri To her, to my g~aduate students, and especially to my patient wife Annette, I owe many thanks
PREFACE TO THE SECOND EDITION
"Dear Colleague, Leave the concept o large molecules well alone f such thrng as a macromolecule "
there can be no
It is said7 that this advice was given to Hermann Staudinger just 45 years ago, after a major lecture devoted to his evidence in favor of the macromolecular concept Today it seems almost impossible that this violent opposition to the idea of the existence of polyme~molecules could have existed in relatively recent times NOW we take f o ~ granted not only the existence of macromolecules, but their value to us in food, clothing, shelter, transportation, communication, most other aspects of modern technology, and, last but far from least, the muscles, sinews, genes, and chromosomes that constitute our bodies and intellect Even within the years since the first edition of Textbook OJ Polymer Science (1962) was written, the use of synthetic polymers has proIiferated, as discussed in Chapte~ Not only has the annual production of plastics (for example) increased 7E some 250% in the last eight years, but on a volume basis it has already exceeded that of copper and aluminum, and is expected to surpass the production of steel by the mid-1980's. One consequence of this widening use of polymeric materials is that a substantial if not major fraction of all chemists and chemical engineers, to say nothing of those in other disciplines, is employed in industry related in some way to polymers Estimates vary, but this fraction appears to be one-third to onehalf or higher Education in polymer science has not kept pace By far the majority of colleges and universities in the United States have no courses in polymer science, no staff member conducting research in this area, and only cursory mention of polymers in other courses. There are, needless to say, many exceptions, ranging from the isolated effort of a single staff member to such major centers for polymer research as those
iRobert Olby, "The Macromolecula~Concept and the Origins of Molecuia~Biology," J Chem Educ 47, 168-174 (1970)
P E A E 'TO THE SECOND EDITION RFC
at Case-Western Reserve, the Universities of' Akson and Massachusetts, and Rensselaer, where 10 or more staff members constitute a fbrrnal or infbrmal polymer research center Fortunately, this scascity of education in polymer science is slowly diminishing, but it is still evident in many areas.. What is most unfbr.tunate is that it appears to exist, not because of'a lack of awareness but, rather, a lack of' interest..For example, on several occasions a graduating W . D . student in my own institution has dropped by to say that he was trained in another area of chemistry and was vaguely awase of' our polymer science program, but he had now accepted employment with a lasge company where he was told that he would be working in the polymer field, and please, was there any way he could learn all about polymers in the short time remaining before he started the job? Perhaps we could overlook a few individual instances of'this sort, but it is more disturbing to note that, while polymeric materials are widely used as previously indicated, most "materials" cur.r.icula ase merely renamed metallurgy departments giving only lip service to major families of materials other than metals, such as ceramics and polymers. Again, there are many exceptions, and I do not wish to leave the impression that this is always the case I have, however, yet to see an introductory materials course or textbook that treats polymers hirly in accord with their wide usage throughout the world. Another dichotomy that deserves mention, and which I wish I had the ability to overcome, is exemplified by the communications gap between the polymer scientist as now trained in the university or in industry and the biologist or biomedical scientist, whose concepts of' macromolecules ase vastly different. Many of' my colleagues seem to share my feeling that major advances can be made in the next few years by the application of' polymer physics and physical chemistry to biological materials The foregoing commentary emphasizes my feeling that the need for education in polymer science exists more than ever today, and that it must be filled by teaching at several levels in several disciplines. Clearly, no single book can serve all needs in this field, but I hope that in its new edition Te,xtbook oj'Polymer Science will continue to be valuable in many ways. In the revision I have attempted to keep in mind its use as supplemental reading material in such undergraduate chemistry courses as physical chemistry, organic chemistry, and instsumental analysis, where an effort is made to introduce polymer science into the chemistrly cur~iculumat appropriate places; as supplemental reading in other curricula such as biology and its interdisciplinary offshoots, materials in the broad sense, and the environmental sciences growing in popularity today; as the textbook in polymer science and engineering courses at the undergraduate and first-year graduate levels; as supplementary reading to broaden the background of the student in advanced polymer courses using appropriate specialized texts; in continuing education at the post,graduate level, in universities but particularly in industry; and as a reference and guide to the literature for the practicing polymer scientist and engineer. Just as the use of' polymers has proliferated in the past decade, so has its literature The leading journal in the field, Journal ofpolymer Science, has subdivided into several largely independent parts. New journals have appe&ed, some with national
'
~REFAC.'ETO T H E SEC.OND EDITION
xi
society sponsorship, such as Macromolecules, sponsored by the American Chemical Society, others as commercial ventures Polymer articles, and even polymer sections, appear in other journals, new and old. POST-J, described below, abstracts approximately 500 journals for articles on polymers Abstract journals devoted solely to polymers have appeared POST--J, acronym for Polymer Science and Technology-Journals, and its companion POST-P (Patents) are published biweekly, at a rate of 500-700 abstracts per issue in mid-1970, by the Chemical Abstracts Service of the American Chemical Society i Other such services exist, sponsored by universities and as commercial ventures Not only has the number of general and specialized books on polymers increased at a tremendous rate in recent years, but a number of valuable compilations has appeared. Outstanding among these are the Encyclopedza of Polymer Sclence and Technology and the Modern Plastrcs Encyclopedza, both referenced in this book One result of the appearance of many specialized books and encyclopedias is that it is no longer necessary for a textbook to cite the original literature in detall; indeed, it would be impossible to do so today Therefore I have limited references to specialized books and compilations where possible, on the assumption that most of them will be as readlly available as the original literature and can in their turn cover the subject more completely than is possible in the present volume The exceptions, aside from some key references of historical value, lie in the areas where adequate specialized coverage has not yet appeared The reader will recognize these areas by the extent and content of the bibliography at the end of each chapter In revising the Textbook of Polymer Science, I have reached the conclusion that many of the basic principles of polymer science are now well established Examples are the kinetics of condensation and free-radlcal addition polymerization In areas such as these, the reader will find only minor changes from the 1962 edition Elsewhere the revision has been more extensive Much new material has been added, particularly in areas still under rapid development This has had to be counterbalanced by the omission of an equivalent amount of material that no longer represents the current state of our knowledge, or was of lesser or only historical interest, in order to keep the length (and price) of the volume in hand. The actual arnngement of material has undergone little change Chapters 1--4 now comprise Part I, dealing with introductory concepts and the characterization of macromolecules Important additions in this section include discussion of solubility parameters, free-volume theories of polymer solution thermodynamics, gel permeation chromatography, vapor-phase osmometry, and scanning electron microscopy, with extensive revision of many other sections Part I1 (Chapters 5-7) deals with the structure and properties of bulk polymers and includes considerable revision of parts of Chapter 5, where a few of the concepts of crystallinity in polymers, new in 1962, have had to be modified as our knowledge in this area has grown Chapter 7 has been revised in order of presentation, with considerable new material added The format and content of Part 111, concerned with polymerization kinetics, have
?Post-J and Post-P ale now combined and issued as part of the Mac~omolecularSections of Chemical Abstracts (footnote added in 1983)
xii
PREFACE TO THE SECOND EDITION
been revised primarily for the citation of' recent advances and new references. The exception lies in Chapter 10, whose topic is ionic and coordination polymerization, in which field many of' the concepts new in 1962 have now reached a stage of further elucidation and acceptance. As in the earlier edition, the chapters of Part IV describe the polymerization, structure, properties, fkbrication, and applications of commercially important polymers, including those used as plastics, fibers, and elastomers.. The reader comparing old and new chapter titles will find that considerablerearrangement has been made, and the content is likewise extensively revised. Of particular note are new sections in Chapter 15 on aromatic heterochain, heterocyclic, ladder, and inorganic polymers. Part V, dealing with polymer processing, has in contrast been revised primarily by the addition of new references. It is my hope that the many readers whose kind comments on earlier editions have given me pleasure will continue to find this revision useful.
t
Troy, New York September I970
CONTENTS
PART ONE.
1.
IN'TRODUCTION
The Science of Large Molecules
A B C D
Basic Concepts of Polyme~Science 3 The Rise of Mac~omolecularScience 8 Molecular Forces and Chemical Bonding in Polymers 11 Molecular Weight and Molecular-Weight Distribution 16 Discussion Questions and P~oblems 19 Bibliography 19
- P R TWO. AT
2.
POLYMERIZATION
Step-Reaction (Condensation) Polymerization
B C D.
Classification of Polymers and Polyme~ization Mechanisms 25 Mechanism of Stepwise Polymerization 26 Kinetics and Statistics of Linear Stepwise Polymerization Polyfunctional Step-Reaction Polymerization 40 Discussion Questions and P~oblems 46 Bibliography 47
33
3.
Radical Chain (Addition) Polymerization
A. Mechanism of Vinyl Polymerization 49 B. Kinetics of Vinyl ~ a d i c a Polymerization 56 l C.. Molecular Weight and Its Distribution 68
xiii
CONTENTS
D.. Effects of Temperature and Pressure on Chain Polymerization 7 1 Discussion Questions and Problems 75 Bibliography 77
4.
Ionic and Coordination Chain (Addition) Polymerization A. Similarities and Contrasts in Ionic Polymerization 82 B . Cationic Polymerization 85 C. Anionic Polymerization 88 D . Coordination Polymerization 9 1 E Ring-Opening Polymerization 96 Discussion Questions and Problems 97 Bibliography 98
5. Copolymerization A. Kinetics of' Copolymerization 101 B . Composition of' Copolymers 110 C Mechanisms of Copolymerization 114 D . Block and Graft Copolymers 120 Discussion Questions and Problems 122 Bibliography 123
6.
A. B. C. D
Polymerization Conditions and Polymer Reactions Polymerization in Homogeneous Systems 127 Polymerization in Heterogeneous Systems 128 Polymerization Reaction Engineering 134 Chemical Reactions of' Polymers 136 Discussion Questions and Problems 144 Bibliography 145
PART THREE.
7.
C:HARACTERlZATlON
Polymer Solutions A Criteria for Polymer Solubilit,~ 1.51 B. Conformations of Dissolved Polymer Chains 154 C Thermodynamics of' Polymer Solutions 159 D . Phase Equilibrium in Polymer Solutions 168 E . Fractionation of' Polymers by Solubility 177 Discussion Questions and Problems 180 Bibliography 181
8
6
1
CONTENTS
8.
Measurement of Molecular Weight and Size A End-Group Analysis 186 B Colligative Property Measurement 187 C Light Scattering 198 D Ultracentrifugation 205 E Solution Viscosity and Molecular Size 208 F Gel Pe~meationChromatography 214 G Polyelectrolytes 218 Discussion Questions and Problems 219 Bibliography 223
9. Analysis and Testing of Polymers A Chemical Analysis of Polymers 229 B Spectroscopic Methods 230 C X Ray Diffraction Analysis 238 D Microscopy 241 E Thermal Analysis 242 F Physical Testing 245 Discussion Questions and P~oblems 252 Bibliography 253
PAR'T F O U R .
STRUCTURE A N D PROPERTIES
10. Morphology and Order in Crystalline Polymers A Configurations of Polymer Chains 261 of B Crystal Structu~es Polymers 266 C Morphology of Crystalline Polymers 273 D Crystallization and Melting 281 ' E Strain-Induced Morphology 289 Discussion Questions and Problems 295 Bibliography 29'7
11.
Rheology and the Mechanical Properties of Polymers A,. Viscous Flow 302 B . Kinetic Theory of Rubber Elasticity 307 C. Viscoelasticity 3 1 1 D. The Glassy State and the Glass Transition 320 E. The Mechanical Properties of Crystalline Polymers ,323 Discussion Questions and Psoblems 326 Bibliography 327
CONTENT'S
Polymer Structure and Physical Properties A. The Crystalline Melting Point 331 B.. The Glass Transition 337 C . P~operties Involving Luge Deformations 340 D.. Properties Involving Small Deformations 343 E.. Property Requirements and Polymer Utilization Discussion Questions and Problems 354 Bibliogsaphy 355
330
349
PART FIVE.
PROPERTIES OF COMMERCIAL POLYMERS

361
13. Hydrocarbon Plastics and Elastomers A . Polyethylene 361 B. Polypropylene 368 C. Other Olefin-Based Polymers and Copolymers 369 D.. Natural Rubber and Other Polyisoprenes 372 E Rubbers Derived from Butadiene 373 F,. Other Synthetic Elastomers 376 Discussion Questions and Problems 380 Bibliography 380
14. Other Carbon-Chain Polymers A. Polystyrene and Related Polymers 383 B . Acrylic Polymers 387 C Poly(viny1 esters) and Derived Polymers 391 D. Chlorine-Containing Polymers 395 E. Fluorine-Containing Polymers 398 Discussion Questions and Problems 404 Bibliography 404
15. Heterochain T'hermoplastics A . Polyamides and Polypeptides 407 B . Polyesters, Polyethers, and Related Polymers 413 C . Cellulosic Polymers 419 D. High-Temperature and Inorganic Polymers 426 Discussion Questions and Problems 43 1 Bibliography 43 1
xvii
16. Thermosetting Resins A. Phenolic and Amino Resins 436 B . Unsaturated Polyester Resins 442 C. Epoxy Resins and Polyurethanes 445 D . Silicone Polyme~s 448 E. Miscellaneous Thermosetting Resins 45 1 Discussion Questions and Problems 452 Bibliography 453
436
PART SIX.
POLYMER PR0C:ESSING
17. Plastics Technology A Molding 4.58 B. Extrusion 461 C . Other Processing Methods 464 D.. Multipolymer Systems and Composites 469 E . Additives and Compounding 47 1 F Tables of' Plastics Properties 47.5 Discussion Questions and Pxoblems 475 Bibliography 48 1
18. Fiber Technology A Textile and Fabric Properties 487 B Spinning 491 C Fiber After-Treatments 497 D Table of Fiber Properties 501 Discussion Questions and Problems 501 Bibliography 504
19. Elastomer Technology A. Vulcanization 507 B. Reinforcement 5 13 C. Elastomer Properties and Compounding 5 16 D . Table of' Elastomer Properties 518 Discussion Questions and Problems 5 18 Bibliography 521
APPENDIXES
Appendix 1 List of Symbols
Appendix 2 Table of Physical Constants Appendix 3 Trade Names and Generic Names
AUTHOR INDEX
SUBJECT INDEX
TEXTBOOK OF POLYMER SCIENCE
PART ONE
INTRODUCTION
CHAPTER ONE
THE SCIENCE OF LARGE MOLECULES
A.
BASIC CONCEPTS OF POLYMER SCIENCE
Over half a century ago, Wolfgang Ostwald (1917)t coined the term the land of neglected dimensions to describe the range of sizes between molecular and macroscopic within which occur most colloidal particles The term neglected dimensions might have been applled equally well to the world of polymer molecules, the highmolecular-weight compounds so important to man and his modern technology It was not until the 1930's that the science of high polymers began to emerge, and the major growth of the technology of these materials came even later Yet today polymer dimensions are neglected no more, for industries associated with polymeric materials employ more than half of all American chemists and chemical engineers The science of macromolecules is divided between biological and nonbiological materials Each is of great importance Biological polymers form the very foundation of life and intelligence and provide much of the food on which man exists This book, however, is concerned primarily with the chemistry, physics, and technology of nonbiological polymers These are the synthetic materials used for plastics, fibers, and elastomers, with a few naturally occurring polymers, such as rubber, wool, and cellulose, included Today these substances are truly indispensable to mankind, being essential to clothing, shelter, transportation, and communication, as well as to the conveniences of modern living Apolymer is a large molecule built up by the repetition of small, simple chemical units In some cases the repetition is linear, much as a chain is built up from its links In other cases the chains are branched or interconnected to form threei Parenthetical years or names and years refer to items in the bibliography at the end of the chapter
dimensional networks. The repeat unit of the polymer is usually equivalent or nearly equivalent to the monomer, or starting material from which the polymer is formed. its Thus (Table 1-1) the repeat unit of poly(viny1 chloride) is --cH,CHC1--; monomer is vinyl chloride, CH,=CHCI. The length of the polymer chain is specified by the number of' repeat units in the chain. This is called the degree ojpolymerization (DP). The molecular weight of the polymer is the product ofthe molecular weight of' the repeat unit and the DP. Using poly(viny1 chloride) as an example, a polymer of DP 1000 has a molecular weight of 63 x 1000 = 63,000, Most high polymers useful for plastics, rubbers, or fibers have molecular weights between 10,000 and 1,000,000 Unlike many products whose structure and reactions were well known befbre their indusbial application, some polymers were produced on an industrial scale long before their chemistry or physics was studied. Empiricism in recipes, processes, and control tests was usual.. Gradually the study of polymer properties began Almost all were first called anomalous because they were so different from the properties of low-molecular- , weight compounds. It was soon realized, however, that polymer molecules are many times larger than those of' ordinary substances. The presumably anomalous properties of polymers were shown to be normal for such materials, as the consequences of their size were included in the theoretical treatments of'their properties
Polymerization Processes. The processes of polymerization were divided by Flory (1953) and Carothers (Mark 1940) into two groups known as condensation
TABLE 1-1. Some Linear High Polymers, 'Their Monomers, and Their Repeat llnits
Polymer
-
Monomer
Repeat Unit
Polyethylene Poly(viny1 chlodde) Polyisobutylene
Polycaprolactarn (6-nylon) Polyisoprene (natural rubber)
H-N(CH2)tC-OH
' 0 H CHz=CH--C=CH2
II
I
"By convention, the symbol omitted
0is used
throughout to represent the benzene ring, double bonds being
BASIC CONCEPTS O F POLYMER SCrlENCE
and addztion polymerization or, in more precise terminology (Chap~v~. stepLA), and chain-reaction polymerization. Condensation or step-reaction polymerization is entirely analogo~lsto condensation in low-.molecular-weightcompounds In polymer formation the vc,,,densation takes place between two polyfunctional molecules to produce one l a ~ p . ~ , . polyfunc-, tional molecule, with the possible elimination of a small molecule s\lc\, water The reaction continues until almost all of one of' the reagents is used lll,; an equilibrium is established that can be shifted at will at high temperatures hy controlling the amounts of the reactants and products Addition or chain-reaction polymerization involves chain reactions ill which the chain carrier may be an ion or a reactive substance with one unp;tilV,i elecwon called a free radi'cal A free radical is usually fbrmed by the d e ~ ~ n ~ of.a ( , ~ i ~ i ~ ~ ~ ~ relatively unstable material called an initiator. The free radical is capable of'reacting to open the double bond of a vinyl monomer and add to it, with an elect1otr,xmaining unpaired. In a very short time (usually a few seconds or less) many mot,t*II,onome,, add successively to the growing chain. Finally two fiee radicals react annihilate each other's growth activity and form one or more polymer moleculcr~. With some exceptions, polymers made in chain xeactions often cc,l,t;lin only carbon atoms in the main chain (homochain polymers), whereas poly rr made in step reactions may have other atoms, originating in the monomer functiot\,,l groups, as p a t of' the chain (heterochain polymers ) !>olymeIiMolecular Weight and Its Distribution. In both chain and stepw~~lr zation, the length of a chain is determined by purely random events In slrl, leactions, the chain length is determined by the local availability of reactive gt(,\,l,s at the ends of the growing chains In radical polymerization, chain length r'r tlt\tcrmined by the time during which the chain grows before it diffuses into the vic.ii1ity of a second free radical and the two react In either case, the polymeric p~o(l~tc ( contains molecules having many different chain lengths Molecular weight anil tll,,lccularweight distribution in polymers is considered further in Section D.
Branched and Network Polymers. In contrast to the linear-chair, ,llolecules discussed so far, some polymers have bxanched chains, often as a rz.11~11 ,,f reactions during polymerization (Fig 1-la). The term branchzng i r n p l i ~ ~ the (hat individual molecules are still discrete; in still other cases crosslinked 01 ,letwork structures are formed (Fig 1-lb), as in the use of monomers containing III,,,, than two reactive g~oups stepwise polymerization If, for example, g1ycr.101is subin stituted fox ethylene glycol in the reaction with a dibasic acid, a three-rllrlrcllsionaj network polymer results In recent years, a variety of branched polymer Q / ~ u c t ~ ~ ~ ~ , some with outstanding high-temperature properties, has been synthe5i/~,l(chapter 15). In commercial practice crosslinking reactions may take place during lllc fabrication of articles made with thermosettzng resins The crosslinked network,t'xtending throughout the final article is stable to heat and cannot be made to f l o f ~ melt or In contrast, most linear or branched polymers can be made to soften ;1r,11 (like on
(b) FIG. 1-1. Schematic representation of' (a) branched and (b) network polymers
new shapes by the application of heat and pressure They are said to be thermoplastic
The Texture of Polymers.
The geometrical arrangement of'the atoms in a polymer chain can be divided conveniently into two categories: Ar~angementsfixed by the chemical bonding in the molecule, such as cis and trans isomers, or d and I fbrms. Throughout this book, such arrangements are described as configurations. The configuration of a polymer chain cannot be altered unless chemical bonds are broken and reformed.. b. Arsangements arising from rotation about single bonds. These arrangements, including the manifold forms that the polymer chain may have in solution, are described as conformations.
a.
In dilute solution, where the polymer chain is surrounded by small molecules, or in the melt, where it is in an environment of'similar chains, the polymer molecule is in continual motion because of its thermal energy, assuming many different conformations in rapid succession. As a polymer melt is cooled, or as this molecular motion so characteristic of' polymers is restrained through the introduction of strong interchain forces, the nature ofthe polymer sample changes systematically in ways that we important in determining its physical properties and end uses (Fig. 1-2). In the molten state, polymer chains move freely, though often with enonnous viscosity, past one another if' a force is applied This is the principle utilized in the
BASIC C.UNCEPTS OF POLYMER SC1ENC.E
LOW
INTERMOLECULAR FORCES
GH
FIG. 1-2. The intenelation of the states of bulk polymers The arrows indicate the directions in which changes from one state to another can take place (Billmeyex 1969)
fabrication of most polymeric articles and is the chief example of the plasticity from which the very name plastzcs is derived If the ir~eversibleflow characteristic of the molten state is inhibited by the introduction of a tenuous network of primary chemical-bond crosslinks in the process commonly called vulcanlzatzon (Chapter 19), but the local freedom of motion of the polymer chains is not restricted, the product shows the elastic properties we associate with typical rubbers If, however, the intetchain forces result from secondary bonds, such as the interaction of polar groups, rather than primary chemical bonds, the rubber is not one of high elasticity but has the properties of limpness and flexibility: A familiar example is the vinyl are film widely used alone or in coated fabrics Secondary-bond fo~ces capable of forming and breaking reversibly as the temperature is changed, as indicated by the arrows in Fig 1-2 Continued primary-bond crosslinking in the postpolymerization step of vulcanization converts rubber into hard rubber or ebonite, whereas ctosslinking concunent with polymerization produces a wide variety of thermosetting materials Common examples are the phenol-formaldehyde and amine-formaldehyde families widely used as plastics As the temperature of a polymer melt or rubber is lowered, a point known as the glass-transition temperature is reached where polymeric materials undergo a marked change in properties associated with the virtual cessation of local molecular motion Thermal energy is required for segments of a polymer chain to move with respect to one another; if the temperature is low enough, the required amounts of energy are not available Below their glass-transition temperature, amorphous polymers have many of the properties associated with ordinary inorganic glasses, including hardness, stiffness, brittleness, and transparency In addition to undergoing a glass transition as the temperature is lowered, some polymers can crystallize at temperatures below that designated as their crystalline
THE SC.IENCE O F LARGE MOLECULES
melting point. Not all polymers are capable of' crystallizing; to oversimplify somewhat, the requirements for crystallizability in a polymer are that it have either a geometrically regular structure or that any substituent atoms or groups on the spaced, they can still fit into backbone chain be small enough so that, if'ir~egularly an ordered structure by virtue of' their small size (see Chapter 10). The properties of' crystalline polymers are highly desirable. Crystalline polymers are strong, tough, stiff, and generally more resistant to solvents and chemicals than their noncrystalline counterparts..Further improvements in these desirable properties can be brought about in at least two ways First, by increasing intermolecular forces through the selection of' highly polx polymers, and by using inherently stiff' polymer chains, crystalline melting points can be raised so that the desirable mechanical properties associated with crystallinity are retained to quite high temperatures. The resulting plastics are capable of' c o m peting with metals A d ceramics in engineering applications (Chapters 12 and 15). Second, the properties of crystalline polymers can be improved for materials in fiber form by the process of' orientation or drawing. The result is the increased, strength, stiffness, and dimensional stability associated with synthetic fibers (Chapter 18).
GENERAL REFERENCES
M r 1966, 1977; Billmeyer 1972, 1982; Elias 1977, Part I; Bovey 1979. ak
B.
THE RISE OF MACROMOLECULAR SCIENCE
Early Investigations
Natural polymers have been utilized throughout the ages. Since his beginning man has been dependent on animal and vegetable matter for sustenance, shelter, warmth, and other requirements and desires.. Natural resins and gums have been used for thousands of' years.. Asphalt was utilized in pr.ebiblica1times; amber was known to the ancient Greeks; and gum mastic was used by the Romans. About a century ago the unique properties of' natural polymers were recognized The term colloid was proposed to distinguish polymers as a class from materials that could be obtained in crystalline fbrm. The concept was later broadened to that of the --colloidal state of matter," which was considered to be like the gaseous, liquid, and solid states. Although useful for describing many colloidal substances, such as gold sols and soap solutions, the concept of a reversibly attainable colloidal state of matter has no validity 'The hypothesis that colloidal materials are very high in molecular weight is also quite old, but before the work of Raoult and van't Hoff in the 1880's no suitable methods were available fbr estima~ing molecular weights When experimental methods did become available, molecular. weights ranging from 10,000 to 40,000 were
THE RISE OF MAC:ROMOLECUl.AR SC1ENC.E
obtained for such substances as rubber, starch, and cellulose nitrate. The existence of large molecules implied by these measurements was not accepted by the chemists of the day for two reasons. First, true macromolecules were not distinguished from other colloidal substances that could be obtained in noncolloidal form as well. When a material of well-known structure was seen in the colloidal state, its apparent high molecular weight was considered erroneous.. Thus it was assumed that Raoult's solution law did not apply to any material in the colloidal state. Second, coordination complexes and the structures in terms association of molecules were often used to explain polyme~ic of physical aggregates of' small molecules.. In the search by the early organic chemists fbr pure compounds in high yields, many polymeric substances were discovered and as quickly discarded as oils, tars, or undistillable residues. A few of' these materials, however, attracted interest. Poly(ethy1ene glycol) was prepared about 1860; the individual polymers with de.grees of polymerization up to 6 were isolated and their structures correctly assigned. The concept of extending the structure to very high molecular weights by continued condensation was understood.
The Rise of Polymer Science Acceptance of the Existence of Macromolecules. Acceptance of the macromolecular hypothesis came about in the 1920's, largely because of the efforts of Staudinger (1920), who received the Nobel Prize in 1953 for his championship of' this viewpoint. He proposed long-chain fbrmulas for polystyrene, rubber, and polyoxymethylene . IIis extensive investigations of' the latter polymers left no doubt as to their long-chain nature. More carefhl molecular-,weight measurements substantiated Staudinger's conclusions, as did x-ray studies showing structures fbr cellulose and other polymers that were compatible with chain formulas. The outstanding series of investigations by Carothers (1929, 1931) supplied quantitative evidence substantiating the macromolecular viewpoint. The Problem of End Croups. One deter~ent the acceptance of the macroto molecula theory was the problem of the ends of the long-chain molecules Since the degree of polymerization of a typical polymer is at least several hundred, chemical methods for detecting end groups were at first not successful Staudinger (1925) suggested that no end groups were needed to saturate terminal valences of the long chains; they were considered to be unreactive because of the size of the molecules Large ring structures were also hypothesized (Staudinger 1928), and this concept was popular for many years Not until Flory (1937) elucidated the mechanism for chain-reaction polymerization did it become clear that the ends of long-chain molecules consist of no~mal,satisfied valence structures. This was but one of many of his contributions to polymer science honored when Flory received the Nobel Prize in 1974. The presence and nature of end groups have since been investigated in detail by chemical and physical methods (Chapter 8A).
10
T H E SCIENCE OF LARGE M O L E C U L E S
Molecular Weight and Its Distribution. Staudinger (1928) was among the first to recognize the large size of' polymer molecules and to utilize the dependence on molecular weight of a physical property, such as dilute-solution viscosity (Staudinger 1930), for determining polymer molecular weights. He also understood clearly that synthetic polymers are polydisperse (Staudinger 1928). A few years later Lansing (193.5) distinguished unmistakably among the various average molecular weights obtainable experimentally. Configurations of Polymer Chain Atoms. Staudinger's name is also associated with the first studies (1935) ofthe configuration of polymer chain atoms. He showed that the phenyl groups in polystyrene are attached to alternate chain carbon atoms.. This regular head-to-tail configuration has since been established fbr most vinyl polymers. The mechanism fbr producing branches in normally linear vinyl polymers was introduced by Flory (1937), but such branches were not adequately identified and characterized fbr another decade (see Chapter SE).. Natta (1955a,b) first recognized the presence of sterospecific regularity in vinyl polymers He received the Nobel Prize in 1963 Early Industrial Developments Rubber. The modem plastics industry began with the utilization of natural rubber fbr erasers and in rubberized fibrics a few yeass befbre Goodyeas's discovery of vulcanization in 1839. In the next decade the rubber industry arose both in England and in the United States. In 1851 hard rubber, or ebonite, was patented and commercialized Derivatives of Cellulose. Cellulose nitrate, or nitrocellulose, discovered in 1838, was successfully commercialiied by Hyatt in 1870..His product, Celluloid, cellulose nitrate plasticized with camphor, could be formed into a wide variety of useful products by the application of heat and pressure. It has been superseded in almost all uses by more stable and more suitable polymers. Cellulose acetate was discovered in 1865, and partially acetylated products were commercialized as acetate rayon fibers and cellulose acetate plastics in the early 1900's.. Cellulose itself, dissolved and reprecipitated by chemical tseatment, was introduced still later as viscose rayon and cellophane. Synthetic Polymers. The oldest of' the purely synthetic plastics is the family of phenol-formaldehyde resins, of which Baekeland's Bakelite was the first commercial product.. Small-scale production of phenolic resins and varnishes was begun in 1907 (Baekeland 1909).. The first commescial use of styrene was in synthetic rubbers made by copoly-. merization with dienes in the early 1900's. Polystyrene was produced commercially in Germany about 1930 and successfully in the United States in 1937. Large-scale production of' vinyl chloride-acetate resins began in the early 1920's also. Table 1-2 shows the approximate dates of' introduction of some of the synthetic plastics of greatest commercial interest
THE RISE Of: MACROMOLECULAR SCIENCE

TABLE 1-2. Some Commercially important Polymers and Their Dates of introduction
11
- - - - . -
-------
Date
Polymer - - - - Styrene-butadiene rubber Poly(viny1 chloride) Polychloroprene (neoprene) Poly(rnethy1 methacrylate) Poly(viny1 acetate) Polysty~ene 66-Nylon Polytetrafluoroethylene Unsaturated polyesters Polyethylene, branched Butyl rubber 6-Nylon
Date
Polymer
Silicones Poly(ethy1ene ter.ephthalate) Epoxies ABS ~esins Polyethylene, linear Polyoxymethylene Polypropylene Polycarbonate Ionomer resins Polyirnides Thermoplastic elastome~s Aromatic polyamides
1930 1936 1936 1936 1936 1937 1939 1941 1942 1943 1943 1943
1964 1965 1970 1974
GENERAL REFERENCES
Flory 1953; Mark 1966, 1967, 1977, 1981; Staudinger 1970; Marvel 1981; Seymour 1982
C . MOL.ECUL.ARFORCES AND CHEMICAL. BONDING I N POLYMERS

The nature of the bonds that hold atoms together in molecules is explained by 'quantum mechanics in terms of an atom consisting of a small nucleus, concentrating the mass and positive charge, smrounded by clouds or shells of electrons relatively far away. It is among the outermost, more loosely bound electrons, called valence electrons, that chemical reactions and primary-bond formation take place.
Primary Bonds
lonic Bond. The most stable electronic configuration for most atoms (except hydrogen) important in polymers is a complete outer shell of eight electrons, called an octet In inorganic systems this structure may be obtained by the donation of an electron by one atom to another to form an lonic bond:
These bonds are not usually found in macromolecular substances except in the use of divalent ions to p~ovide "crosslinks" between carboxyl groups in natural resins, and in zonomers (Chapter 13)
12
THE S 1 N . OF LARGE MOL.EC'ULES C E CE
Covalent Bond. These bonds ase formed when one or more pairs of valence electrons are shared between two atoms, again resulting in stable electronic shells:
The covalent bond is the predominant bond in polymers
Coordinate Bond. This bond is similar to the covalent bond in that electrons ase shared to pmduce stable octets; but in the coordinate bond both of the shared electrons come from one atom. Addition compounds of boron trichloride are common examples:
where R is an organic group The coordznate or sem~polarbond has properties between those of the ionic and covalent bonds No polymers containing true coordinate bonds have reached commercialization
Metallic Bond. In the metalli'c bond the number of' valence electrons is far too small to provide complete outer shells fbr all the atoms. The resulting bonds involve the concept of' positively chasged atoms embedded in a permeating "gas" of' elec-, trons fiee to move about at will. The metallic bond is not utilized in polymeric systems Typical Primary-Bond Distances and Energies. From studies of the positions of atoms in molecules and the energetics of' moleculas formation and dissociation, it is possible to assign typical energies and lengths to primary bonds. Table 1-3 lists some of' these properties of interest in polymeric systems..The angles between successive single bonds involving the atomic arrangements usual in polymers range between 105" and 113", not fas from the tetrahedral angle of 109"28'. Secondary-Bond Forces
Even when all the primary valences within covalent molecules are saturated, there ase still forces acting between the molecules. These are generally known as secor ondary valence or intermolecula~forces, van der Waals forces. The following three types are recognized, and the first and third in particular contribute greatly to the physical properties of' polymers equal and opposite Dipole Forces. When different atoms in a molecule car~y electric charges, the molecule is said to be polar or to have a dipole moment. At
MO~ECULAR FORCES A N D CHEMICAL. BONDING IN POLYMERS
TABLE 1-3. Typical Primary-Bond lengths and Energies

--
--
Bond Length Bond

C-C C=C C-H C-N C=N C-0 C=O C- F C-Cl N -- II
--
(A>
1 54 1 34 1 10 1 47 1 15 1 46 1 21 1 35 1 77 101 0 96 1 32
Dissociation Energy - (M/moie)

347 61 1 414 305 89 1 360 749 473 339 389 464 146
0--H 0- 0
large distances such a molecule acts like an electrically neutral system, but at molecular distances the charge separation becomes significant and leads to a net intermolecular force of attraction The magnitude of the interaction energy depends on the mutual alignment of the dipoles Molecular or.ientation of this sort is always opposed by thermal agitation; hence the dipole force is strongly dependent upon temper atme
induction Forces. A polar molecule also influences surrounding molecules that do not have permanent dipoles The electric field associated with a dipole causes slight displacements of the electrons and nuclei of surrounding molecules, which lead to induced dipoles The intermolecular force between the permanent and induced dipoles is called the lnd~ict~on force The ease with which the electronic and nuclear displacements are made is called the polarlzabzlzty of the molecule The energy of. the induction force is always small and independent of temperature Dispersion Forces. The existence of'intermoIecular forces in nonpolar materials, plus the small temperature dependence of intermolecular fbrces even where the dipole e f t c t is known to far outweigh the induction effect, suggests the presence of a third type of intermolecular force. A11 molecules have time-dependent dipole moments that average out to zero and which arise from different instantaneous configurations of the electrons and nuclei. These fluctuations lead to perturbations of the electronic clouds of neighboring atoms and give rise to attractive forces called dispersion forces They are pr.esent in all molecules and make up a major portion of' the intermolecular forces unless very strong dipoles ase present. In nonpolar materials only the dispersion forces exist. They are independent of temperature Occasionally the term van der Waals' forces is applied to the dispersion fbrces alone.
14
THE SCIENCE OF LARGE MOL.EC:ULES
Interrelation of Intermolecular Forces. The energy of the intermolecular attractive forces varies as the inverse sixth power of' the intermolecular distance. As with primary-bond forces, repulsion arises when the atoms approach more closely than an equilibrium distance of' 3-5 A. The energy of typical secondary-bond attractive f o ~ c e s 8-40 kJ/mole, divided among the three secondary-bond types is according to the polarizability and dipole moment of the bonding molecules.. The Hydrogen Bond. The bond in which a hydrogen atom is associated with two other atoms is particularly important in many polymers, including proteins, and is held by many to be essential to life processes. Since the classical concepts of chemical bonding allow hydrogen to form only one covalent bond, the hydrogen bond can be considered electrostatic or ionic in character.. This model does not, however, account fbr all the properties of the hydrogen bond; it is appealing to consider the bond covalent in some cases. The hydrogen bond occurs between two functional groups in the same or different molecules. The hydrogen is usually attached to an acidic group (a proton donor), typically a hydroxyl, carboxyl, amine, or amide group The other group must be basic, usually oxygen, as in carbonyls, ethers, or hydroxyls; nitrogen, as in amines and amides; and occasionally halogens The association of such polar liquid molecules as water, alcohols, and hydrofluoric acid, the fbrmation of dimers of simple organic acids, and important structural effects in polar polymers such as nylon, cellulose, and proteins are due to hydrogen bonding. Typically, hydrogen bonds range between 2.4 and 3.2 A in length and between 12 and 30 kJ/mole in dissociation energy Only fluorine, nitrogen, oxygen, and (occasionally) chlorine are electronegative enough to form hydrogen bonds lntermolecular Forces and Physical Properties
Secondary-bond forces are not of great importance in the formation of stable chemical compounds. They lead, rather, to the aggregation of separate molecules into solid and liquid phases. As a result, many physical properties such as volatility, viscosity, surface tension and frictional properties, miscibility, and solubility are determined largely by intermolecular fbrces. The cohesive energy is the total energy necessary to remove a molecule from a liquid or solid to a position far from its neighbors. This is approximately equal to the heat of vaporization or sublimation at constant volume and can be estimated from thermodynamic data. The cohesive energy per unit volume, usually called the cohesive energy density, and its variation with molecular structure illustrate the effects of intermolecular forces on the physical properties of' matter.
Volatility and Molecular Weight. The tendency of a molecule to volatilize from its liquid is a function of its total translational energy and therefore of the temperature The boiling point depends on the relation of the translational eneIgy to the cohesive energy and thus is a fitnction of molecular weight in a homologous series At high molecular weights the total cohesive energy per molecule becomes greater
M0LEC:ULAR FORCES A N D CHEM1C:AI. B O N D I N G IN POLYMERS
15
than primary-bond energy, and the molecules decompose before they volatilize This point is reached at molecular weights f'ar below those of typical polymers The melting point is also related to the cohesive energy, but here another important factor comes into play. This is the influence of molecular order or entropy In thermodynamic terms, changes of' state take place only when the free-energy 1 change in the process
AG = AH -- T A S
is favorable, and the enthalpy term AH may easily be outweighed by the entropy term T AS whenever a radical change in molecular configurations occurs in the process. Thus, in general, a high boiling point is associated with a high melting point, but the relation between melting point and molecular structure is fairly complicated Symmetrical molecules, which have low entropies of'fusion, melt at higher temperatures than do similar but less symmetrical molecules
Effect of Polarity. A molecule containing strongly polar groups exerts corlespondingly strong attractive forces on its neighbors This is reflected in higher boiling and melting points and other manifestations of higher cohesive energy density Miscibility and Solubility. These properties are also determined by the intermolecular forces. The thermal effect on mixing or solution is the difference between the cohesive energy of the mixture and that of the individual pule components Again entropy considerations are important, but in general a negative heat of mixing favors solubility and a positive heat of mixing favors immiscibility. The intermolecular forces therefore lead directly to the solubility law of "like dissolves like." The role of the intermolecular forces and the cohesive energy density in determining the solubility of polymers is discussed further in Chapter 7A Intermolecular Forces and Polymer Types. Table 1-4 lists the cohesive energy densities of' some typical polymers. These data cor~oboratethe conclusions of' Section A regarding the texture of' polymers, for, in the absence of' primary-bond crosslinks, it is the intermolecular forces that provide the restraints on moleculru motion, which, as illustrated in Fig. 1-2, are a major determinant of the nature of bulk polymers. If the intermoleculx forces are small and the cohesive energy is low, and the molecules have relatively flexible chains, they comply readily to applied stresses and have properties usually associated with elastomers. Somewhat higher cohesive energy densities, accompanied in some cases by bulky side groups giving stiffer. chains, are characteristic of' typical plastics. If the cohesive energy is higher still, the materials exhibit the high resistance to stress, high strength, and good mechanical
i This book follows the convention of defining the (Gibbs) p e e energy (now sometimes called the free enrhalpy) as G = H - TS, and the work content or Helmfioltz free energy as A = E - TS
16
'TABLE 1-4. Cohesive Energy Densities of Linear Polymersa
- Polyethylene Polyisobutylene Polyisoprene Polystyrene Poly(rnethy1 methacrylate) Poly(viny1 acetate) Poly(viny1 chloride) Poly(ethy1ene terephthalate) Poly(hexarnethy1ene adiparnide) Polyacrylonitrile
"Walker (1952) and Small (1953)
Polymer
Repeat Unit
-CH2CH2-CH,C(CH,),--CH2C(CH3)=CHCH2.-CH2CH(C6H5)-CH,C(CH3) (COOCH,)-CH,CH(OCOCH,).--CHZCHCI-CH,CH,0COC6H,COO-.NH(CH2)6NHCO(CH2)4CO-CH,CHCN--
Cohesive Energy Density (J/cm3)

259 2'72 280 310 347 368 381 477 '774 992
properties typical of fibers, especially where molecular symmetry is favorable for crystallization Chain stiffness or flexibility, referxed to above, is largely determined by hindrance to free rotation about carbon-carbon single bonds in the polymer chain
GENERAL R F R N E EE E CS
Ketelaar 1953; Cottrell 1958; Pauling 1960, 1964; Pimentel 1960; Chu 1967; Phillips 1970; Elias 1977, Part I
D. MOLECULAR WEtGHT AND MOLECULAR-WEIGHT'DIS'T'RIBUTION
Perhaps the most important feature distinguishing polymers fiom low-molecularweight species is the existence of' a distribution of chain lengths and therefbre degrees of' polymerization and molecular weights in all known polymers (except possibly some biological macromolecules). This distribution can be illustrated by plotting the weight of' polymer of' a given molecular weight against the molecular weight, as in Fig. 1-3 Because of' the existence of' the distribution in any finite sample of polymer, the experimental measurement of' molecular weight can give only an average value Several different averages are important. For example, some methods of' molecularweight measurement in effect count the number of' molecules in a known mass of material Thr.ough knowledge of' Avogadro's number, this infbrmation leads to the
MOLEC:ULAR WEIGH7 AND MO1.ECULAR-WEIGHT DISTRIBUTION
Molecular weight +
FIG. 1-3. Distribution of molecular weights in a typical polymer
number-average molecular wezght M,, the sample. For typical polymers the of number average lies near the peak of the weight-distribution curve or the most probable molecular weight If the sample contains N, molecules of the zth kind, for a total number of molecules CtX=, and each of the zth kind of molecule has a mass m,, then the total mass of N,, all the molecules is C,"=, N,m, The number-average molecular mass is
and multiplication by Avogadro's number gives the number-ave~agemolecular weight (mole weight);
Number-average molecular weights of' commercial polymers usually lie in the range 10,000.-100,000, although some materials have values of M,, 10-fold higher, and others 10-fold lower. In most cases, however, the physical properties associated with typical high polymers are not well developed if' M,, below about 10,000. is After &, the next higher average molecular weight that can be measured by absolute methods is the weight-average molecular weight M,,. This quantity is defined as
18
THE SCIENCE O F LARGE MOLECULES
TABLE 1-5.
Typical Ranges of M,/M,, in Synthetic Polymersd
Polymer
Range
Hypothetical monodisperse polymer Actual "monodisperse" "living" polymers Addition polymer, termination b,y coupling Addition polymer, termination by disproportionation, or condensation polyme~ High conve~.sion vinyl polymers Polymers made with autoaccele~ation Addition polymers prepared by coordination polymerization Branched polymers
billm me ye^. (1977)
20
8-30 20-50
It should be noted that each molecule contributes to M, in proportion to the square of its mass: A quantity proportional to the first power of M measures only concentration, and not molecular weight. In terms of' concentrations c, = NiM, and weight fractions w i = CJC, where c = Cip_,ci,
Unfortunately, there appears to be no simple analogy for M, akin to counting molecules to obtain M,. Because heavier molecules contribute more to M, than light ones, M, is always greater than M,,except for a hypothetical monodisperse polymer. The value of' M, is greatly influenced by the presence of high-molecular-weight species, just as M,, is influenced by species at the low end of'the molecular-weight distribution curve. The quantity M,IM,is a useful measure of the breadth of the molecular-weight distribution curve and is the parameter most often quoted for describing this feature The range of' values of M,IM,, synthetic polymers is quite large, as illustrated in in Table 1-5. For some types of polymerization, the distribution of moleculax weights (more often expressed as degrees of' polymerization) can be calculated statistically; this topic is discussed in Chapter 3 E . Experimental methods for measuring the molecularweight averages defined above, among others, are the subject of Chapter 8.
GENERAL REFERENCES
Peebies 1971; Slade 1975; BilIingham 1977
DISCUSSIONQUES'TIONS AND PROBLEMS

1.
Define the following terms: polymer, monomer, repeat unit, network, degree of polymerization, homochain polymer, heterochain polymer, thermoplastic, thermosetting, configuration, conformation
2. Discuss some ofthe properties that make polymers useful materials, and show how they result from unique features of polymer structure such as high mo.lecular weight.. (This topic is amplified in later chapters.)
3. Consider three hypothetical monodisperse polymers, with M = 10,000, M = 100,000, and M = 1,000,000 For each, calculate M, and M,, after adding the following to 100 parts by weight c, of the polymer with M = 100,000: a 20 parts by weight of the polymer with M = 10,000. b 20 parts by weight of the polymer with M = 1,000,000
Calculate and M,, after adding the following to 100 parts by number (of molecules) of the polymer with M = 100,000: c 20 parts by numbe~of the polymer with M = 10,000 d 20 parts by number of the polymer with M = 1,000,000 Discuss the dependence of mo1ecula1-weight material
lMw
and M,, the presence of high- and lowon
BIBLIOGRAPHY
Baekeland 1909 L H Baekeland, "The Synthesrs, Constitution, and Uses of Bakel~te," J Ind Eng Chem 1, 149-161 (1909), reprinted rn Chemtech 6 , 40-53 (1976) B~ll~ngham 1977 N C Billrngham, Molar Mass Measu~ements zn Polymer Sczence, Halsted Press, " John Wiley & Sons, New York, 1977 Blllmeyer 1969 Fred W bill me ye^, Jr , "Molecular Structure and Polymer Properties," J Paint Techno1 41, 3-16; erratum, 209 (1969) nrllmeyer 1972 Fled W Billmeyer, 11 , S~nthetlc Polymers Buzld~ngthe Giant Molecule, Doubleday, Garden Crty, New York, 1972 Rlllmeyer 1977 Fred W Blllmeyer, Jr , "The S ~ z e Weight of Polymer Molecules," Chapter 4 in and to Herman S Kaufman and Joseph J Falcetta, eds ,Inttoduct~on Polymer Sczence and Technology An SPE Textbook, Wrley-Interscience, New York, 1977 Blllmeyer 1982 Fred W Billmeyer, 1 , '.Polymers," pp 745--755 In Martin Grayson, ed , K ~ r k 1 Othmer Encyclopedza of Chemzcal Technology, 3rd ed , Vol 18, Wiley-Interscience, New York, 1982 Bovey 1979 F A Bovey and F H Winslow, "The Nature of Macromolecules," Chapter 1 in F A Bovey and F H Wrnslow, eds , Macromolecules An Introductzon to Polymer Sczence, Academic Press, New York, 1979 Carothers 1929 W H Carothers, "An Introduction to the General Theory of Condensation Polymers," J Am Chem Soc 51, 2548-2559 (1929) Carothers 1931 Wallace H Carothers, "Polymenzation," Chem 'lev 8, 353-426 (1931)
20
THE SCIENCE OF LARGE MOLECUL.ES
Chu 1967 Benjamin Chu, Molecular Forces Based on the Baker Lectures of Peter 1 W Debye, WileyInterscience, New York, 1967 Cottrell I958 Tom L Cottrell, The Strengths of Chemrcal Bonds, 2nd ed ,Academic Press, New York, 1958 Elias 1977 Hans-Georg EIias, Macromolecules 1 Structure arid Propernes, Plenum Press, New York, 1977 Flory 1937 Paul J Fiory, "Mechanrsm of Vinyl Polymerization," J Am Chem Soc 59, 241-253 (1937) Flory 1953 Paul J Flory, Principles of Polymer Chemrstrv, Cornell Univers~tyPress, Ithaca, New York, 1953 Ketelaar 1953 J A A Ketelaar, Chemrcal Constrtutlon, Elsev~er, New York, 1953 Lanstng 1935 W D Lansing and E 0 Kraemer, "Molecular Weight Analysis of Mixtures by Sedimentation Equilibrium in the Svedberg Ultracentrifuge," I Am Chem Soc 57,1369-1377 (1935) Mark 1940 H Mark and G Stafford Whitby, eds , Collecred Papers of Wallace Hume Carothers on Hlgh Polvmerlc substance^, Intersc~ence, New York, 1940 Mark I966 Herman F Mark and the Editors of Lrfe. Glant Molecules. Time, New York, 1966 Mark 1967 H F Mark, "Polymers-Past, Present and Future," pp 19-55 in W 0 Milligan, ed , Proceedtngs of the Robert A Welch Foundation Conferences on Chemlcal Research, X Polymers, The Robert A Welch Foundation, Houston, 1967 Mark 1977 Herman F Mark and Sheldon Atlas, "Introduction to Polymer Science," Chapter 1 in Herman S Kaufman and Joseph J Falcetta, eds ,Introductron to Polymer Sclence and Technology An SPE Textbook, Wiley-Interscience, New Yolk, 1977 and Mark 1981 Herman Mark, "Polymer Chemistry in Eu~ope America-How it all Began," J Chem Educ 58, 527-534 (1981) Marvel I981 C S Marvel, "The Development of Polymer Chem~stry Amer~ca--The Early Days," In 1 Chem Educ 58, 535-539 (1981) Pino, Paolo Corrad~n~, Ferdrnando Danusso, Enrico Mant~ca,G I O I ~ I O Natta 1955a G Natta, P~ero Mazzant~,and Glovann~Morangl~o,"Crystalline H ~ g h Polymers of a-Olefins," 1 Am Chem SOC 77, 1708-1 710 (1955) Natta 1955b G Natta, "A New Class of a-Olefin Polymers with Exceptional Regularity of Structure" (in French), 1 Polym Sci 16, 143-154 (1955) Ostwald 1917 Dr Wolfgang Ostwald, An Introductror~to Theoretical and Applied Colloid Chemistry (The World oj Neglected Drmensions) (translated by DI Martin H Fischer), John Wiley & Sons, New York, 1917 Paulzng 1960 Linus Pauling, The Nature of the Chemltal Bond, 3rd ed , Cornell University Press, Ithaca, New Yolk, 1960 Pauling I964 Linus Pauling and Roger Hayward, The Archztecture oj Molecules, W H Freeman, San Francisco, California, 1964 Peebles 1971 Leighton H Peebles, Jr ,Molecular WetghrDlstrrbutionsrn Polymers, Wiley -Interscience, New York, 1971 Phillrps 1970 James C Phillips, Covalent Bondrng In Crystals, Molecules and Polymers, Chicago University Press, Chicago, Illinois, 1970 P~mentel1960 George C Pimentel and Aubrey L McClellan, The Hvdrogen Bond, W H Freeman. San Francisco, California, 1960 Seymour 1982 Raymond B Seymour, ed ,Hrstory ojPolbmer Scrertte and 7echnology, Marcel Dekker, New York, 1982 Slade 1975 Phllrp E Slade, Jr , Pol-vmer Molecular Werghts, Vol 4 of Ph111pE Slade, Jr , and Lloyd T Jenklns, eds , Tethnrques of Polymer Evaluation. Marcel Dekker. New York, 1975
Small 1953 P A Small, "Some Factors Affecting the Solubility of Polymers," I Appl Chem 3, 71-80 (1953) Staudlnger 1920 H Staudinger , "Polymerization" (in German), Ber Dtrch Chem Ger B 53, 10731085 (1920) Staudlnger 1925 H Staud~nger,"The Constitution of Polyoxymethylenes and Other High-Molecular Compounds" (In German), Helv Chlm Acta 8, 67-70 (1925) Startdinger 1928 H Staudinger, "The Constitution of High Polymers XIII" (in German), Ber Dtsch Chem Ges B 61, 2427-2431 (1928) Staudinger 1930 H Staudinger and W Heuer, "Highly Polymerized Compounds, XXXIII A Relation Between the Viscosity and the Molecular Weight of Polysty~enes" (in German), Ber Dtsch Chem Ges B 63, 222-234 (1930) Staudlnger 1935 H Staud~nger A Ste~nhofer,"Highly Polymenzed Compounds CVII Polystyand renes" (in German), lustus Llebzgs Ann Chem 517, 35-53 (1935) Staudlnger 1970 Herman Staud~nger, From Organcc Chemirrry to Macromolecules, Wiley-Intersc~ence, New York, 1970 Walker 1952 E E Walker, "The Solvent Action of O~ganlc Substances on Polyacrylonitnle," J Appl Chem 2, 470-481 (1952)
PART TWO
POLYMERIZATION
CHAPTER TWO
-
STEP-REACTION (CONDENSATION) POLYMERIZATION
A. CLASSIFICATION O F POLYMERS AND POLYMERIZATION MECHANISMS
In 1929 W. H. Carothers suggested a classification of polymers into two groups, condensation and addition polymers. Condensation polymers are those in which the molecular formula of'the repeat unit of the polymer chain lacks certain atoms present in the monomer from which it is formed (or to which it can be degraded). For example, a polyester is fbrmed by typical condensation reactions between bifunctional monomers, with the elimination of' water:
Addition polymers (Chapter 3) are those in which this loss of a small molecule does not take place. The most important group of addition polymers includes those derived from unsaturated vinyl monomers: CH2=CH
-.--,
-CH2--CH-CHZ--CH--,
etc.
Carother's original distinction between addition and condensation polymers was amended by Flory, who placed emphasis on the mechanisms by which the two types of' polymer are formed.. Condensation polymers are usually fbrmed by the stepwise intermolecular condensation of reactive groups; addition polymers osdinariiy ~esult from chain reactions involving some sort of active center
26
The classification adopted in this book is based on the reaction mechanism, dividing polymerization into step reactions, commonly producing step-reaction or. condensation polymers, and chain reactions, commonly producing chain,-reaction or addition polymers (Mark 1950). Polymerizations are classified without regard to loss of a small molecule or type of interunit linkage.. Wherever precise differentiation on the basis of' mechanism is required, the terms step reaction and chain reaction are used; in deference to well-,established tradition, the common terms condensation and addition are permissible where no confusion can result. Some of' the consequences of the differences between the mechanisms of' chain and stepwise polymerization are shown in Table 2-1
GENERAL REFERENCES
Mark 1940; Flory 1953; Lenz 1967; Vollmert 1973; Elias 1977.
B.
MECHANISM OF STEPWISE POLYMERIZATION
Types of Condensation Polymers
Table 2-2 lists some representative products of stepwise polymerization. Most of them are, stoichiometrically, condensation polymers. Proteins and cellulose are included on the basis that they can be degraded hydrolytically to monomers differing from their repeating units by the addition of the elements of a molecule of' water The type of products formed in a condensation reaction is determined by the functionality of the monomers, that is, by the average number of reactive functional groups per monomer molecule Monofunctional monomers give only low-molecular-weight products Bifunctional monomers give linear polymers, as illustrated
'TABLE 2-1.
Distinguishing Features of Chain- and Step-Polymerization Mechanisms
Chain Polymerization
Only growth reaction adds repeating units one at a time to the chain. Monomer concentration decreases steadily throughout reaction High polymer is formed at once; polymer molecular. weight changes little throughout reaction Long reaction times give high yields but affect molecular weight little Reaction mixture contains only monomer, high polymer, and about part of growing chains
"Degree of polymerization
Step Polymerization
Any two molecular species present can react Monomer disappears early in reaction: at DP" 10, less than 1% monomer remains Polymer molecular weight rises steadily throughout reaction Long reaction times are essential to obtain high molecular weights At any stage a11 molecular species are present in a calculable distribution
MECHANISM OF ST'EPWISE POLYMERIZATION
27
earlier in this chapter Polyf'unctional monomers, with more than two functional groups per molecule, give branched or crosslinked (three-dimensional) polymers The properties of the linear and the three-dimensional polymers differ widely The mechanism of' step-reaction polymerization is discussed in this section, following the classification scheme of Lenz (1967) Further infbrmation on the organic chemistry of' the major types of' condensation polymers having commercial utility is found in Part 4
Carbonyl Addition-Elimination Mechanism
The most important reaction that has been used fbr the preparation of condensation polymers is that of addition and elimination at the carbonyl double bond of' carboxylic acids and their derivatives. The generalized reaction is
where R and R' (below) may be alkyl or aryl groups, X may be OH, OR', NH,, NHR', OCR', o Cl; and Y may be R f O - , R'OH, RINH,, or R'COO- The species x in the bracket is considered to be a metastable intermediate, which can either return to the original state by eliminating Y or proceed to the final state by eliminating X The following paragraphs provide some typical examples of this reaction
Direct Reaction. The direct reaction of a dibasic acid and a glycol to form a polyestex, or a dibasic acid and a diamine to form a polyamide, works well and is widely used in practice In esterification, a strong acid or acidic salt often serves as a catalyst. The reaction may be carried out by heating the reactants together and removing water, usually applying vacuum in the later stages An important modification of the direct reaction is the use of a salt, as in the p~eparation of poly(hexamethy1ene adipamide) (66-nylon) by heating the hexamethylene diamine salt of adipic acid above the melting point in an inert atmosphere. The stringent requirement of stoichiometric equivalence to obtain high molecular weight is easily met by purifying the salt by recrystallization Interchange. The reaction between a glycol and an ester,
is often used to produce polyesters, especially where the dibasic acid has low solubility. Frequently the methyl ester is used, as in the production of poly(ethy1ene terephthalate) from ethylene glycol and dimethyl terephthalate. The reaction be-
TABLE 2-2. Typical Step-Reaction Polymers
TY pe
Polyester
Interunlt Llnkage
0
HO(CHz),COOH
Examples
HO[-(CH2)xCOO-]yH
+ Hz0
+ H20
Polyanhydride Poly acetal
-C-0-CH I
I 1
II
CHOH HOOC(CH2)xCOOH ---+ three-dimensional network I CH~OH HOOC(CH2)xCOOH HO[-CO(CH2),COO-]yH H20
HO(CH2),0H
+ CH2(0R)2
HO[-(CH~),OCHZO-],(CH~),~H
+ ROH
+
Polyurethane Polyurea
Silk fibroin
Cellulose
-C-NH0
II
NH2(CHz)xNHt HOOC(CHz)x,COOH H[-NH(CH2)xNHCO(CHz)x~CO-]yOH H20 HO(CHZ)~OH OCN(CH2)x,CNO + [-O(CHZ)~~CONH(CHZ)~~NHCO-]~

- - +
-0-C-NHNH2(CH2),NH2 4- OCN(CH2),.CNO -NH-C-NH0 II -C-NHNH2CH2COOH 4-NHzCHRCOOH C6Hi: 3 6

t -
[-NH(CH2),NHCONH(CH2),.NHCO-1, H[-NHCH2CONHCHRCO-IyOH
4-
4- H z 0
-c-0-c-
[C6Hio04]-0-[C6H!004]-
H20
Phenol-aldehyde
-CH2-CHzOCH2-
I + CH20
CH3
-+ -
+ Hz0
0
OH Uxa-aldehyde -NH-CHR-NH-CHR-v-CHRPolysulfide
H20 + three-dirnenslona network with C H ; and C H 2 0 C H 2 brldges between posltlons on rlngs o and p to hydroxyls NH2CONH2 C H 2 0 (1. I ratlo) -NHCONH-CH2-NHCONH-CHIH20 With excess CH20, three-dirnenslonal network H20
+ CH.0
+
-s-S-SS S -S-SR
CHR-
II II
I
CI(CH2),CI + Na2S CI[-(CH2),S--],Na NaCl HS(CHz),SH oxldatlon HS[-(CH2),SS-],(CH2).SH CI(CH2),Cl Na2S4 CI[-(CH2),S-S-],Na + NaCl II II
--
+ H 2 0 or H2S
Polysiloxane
-Si-0-SiR
R
-Si-0-Sii R
R R
0
!
I I
HO-&-OH I CH3
+ HO-~i-OH
OH
- - +
three-dirnenslonai network
+ H20
tween a carboxyl and an ester link is much slower, but other interchange reactions, such as amine-amide, amine-ester, and acetal-alcohol, are well known
Acid Chloride or Anhydride. Either of these species can be reacted with a glycol or an amine to give a polymer The anhydride reaction is widely used to form an alkyd resin from phthalic anhydride and a glycol:
The condensation in bulk of' an acid chloride with a glycol is not useful because of side reactions leading to low-molecular-weight products, but the reaction of' an acid chloride with a diamine is a valuable means of' preparing polyamides..
Interfacial Condensation. The reaction of' an acid halide with a glycol or a diamine proceeds rapidly to high-molecular-weight polymer if' car~ied out at the interface between two liquid phases, each containing one of the reactants (Morgan 1965). Very-high-molecular-weight polymer can be formed. Typically, an aqueous phase containing the diamine or glycol and an acid acceptor is layered at room temperature over an organic phase containing the acid chloride. The polymer fbrmed at the interface can be pulled off as a continuous film or filament The method has been applied to the fbrmation of' polyamides, polyurethanes, polyureas, polysulfbnamides, and polyphenyl esters. It is particularly useful for preparing polymers that are unstable at the higher temperatures usual in stepreaction polyrner.ization Ring Versus Chain Formation. In addition to polymer fbrmation, bifunctional monomers may react intramolecularly to produce a cyclic product. Thus hydroxy acids may give either lactones or polymers on heating,
amino acids may give lactams or linear polyamides, and so on.. The chief hctor governing the type of' product is the size of' the ring that can be formed. If' the ring contains less than five atoms or more than seven, the product is usually linear polymer. If a ring of five atoms can fbrm, it will do so to the
MECHANISM O F STEPWISE POLYMERIZATION
31
exclusion of linear chains If six-. or seven-membered rings can form, either type of product can result. Larger rings can sometimes be formed under special conditions The lack of fbrmation of' rings with less than five atoms is explained by the strain imposed by the valence angles of' the ring atoms Five-membered rings are virtually strain free, and all larger rings can be strain fiee if' nonplanar forms are possible.. As the ring size increases, the statistical probability of forming rings becomes smaller. The formation of five-membered rings to the co4plete exclusion of' linear polymer is not thoroughly understood The formation of high polymer from cyclic monomers ofthe type just discussed is often, kinetically, an anionic chain polymerization and is discussed in Chapter 4 This class of reaction includes the commercially important production of nylon from caprolactam.
Other Mechanisms Carbonyl Addition-Substitution Reactions. The reaction of aldehydes with algroups, is of cohols, involving first addition and then substitution at the ca~bonyl great practical and historic importance in stepwise polymerization. The general reaction, leading to acetal formation, is 0
R C H f R'OH -
11
I RCH I
OH
R'OH
0 R'
----+ R C H
0 R'
OR'
1 I
In addition to polyacetals, important polymers formed in this way include those of formaldehyde and phenol, urea, or melamine (Chapter 16)
Nucleophilic Substitution Reactions. These reactions are important from the standpoint of commercial organic polymers, primarily because of their use in the polymerization of epoxides The most common epoxide monomer is epichlorohydrin, which reacts with a nucleophile N: as follows:
Typically, the nucleophile is a bifunctional hydroxy compound such as bisphenol A, and the reaction proceeds as described in Chapter 16. A second nucleophilic substitution reaction of historic interest in polymer chemisQy is that used for the production of the early polysulfide rubbers from aliphatic dichlorides and sodium sulfide:
32
STEP..REAC'TION(CONDENSATION) POLYMERIZATlON
This type of' reaction can be used for the production of' semio~.ganic inorganic and polymers, and is probably also the basis for the formation of'natural polysaccharides and polynucleotides by polymerization in living organisms catalyzed by enzymes
Double-Bond Addition Reactions. Although addition reactions at double bonds are often associated with polymerization by chain mechanisms (Chapters 3 and 4), this is not necessarily the case, and many important stepwise polymerizations are based on this reaction. Of' major interest among these is the ionic addition of diols to diisocyanates in the production of' polyurethanes (Chapter 15):
The free-~adical addition of dithiols to unconjugated diolefins also proceeds by a stepwise mechanism
Free-Radical Coupling. These reactions lead to several polymerization schemes of interest, including the preparation of arylene ether polymers, polymers containing acetylene units, and arylene alkylidene polymers If [Ox] is an oxidizing agent, the first of these can be written as
Aromatic Electrophilic-Substitution Reactions. Reactions of this type, using standard Friedel-Crafts catalysts, produce polymers by a step-growth mechanism A typical example is the production of poly(p-phenylene):
GENERAL REFERENC,ES
Mark 1940; Marvel 1959; Morgan 1965; Lenz 1967, Chapters 4-8; Ravve 1967, Part IV; Sorenson 1968; Solomon 1972, Chapter 1; Collins 197.3, Exps. 1 and 2; Vollmert 1973, Chapter 2; Allen 1974, Chapter 5; Elias 1977, Chapter 17.1; Millich 1977; Bowden 1979, Chapter 2 4; Allcock 198 1, Chapter 2; Carlaher 1982
KINETICS A N D STATISTICS OF LINEAR STEPWISE POLYMERIZATION
C . KINETICS A N D STATISTICS OF LINEAR STEPWISE

p0LYMERlZATlON Reactivity and Molecular Size
The similarity between monof'unctional and polyf'unctional step reactions is illustrated in the chemical equations in this chapter Other similarities exist in the influence of' temperature and catalysts on the rates of the reactions. It is tempting to substitute the simple concept of reactions between functional groups for that of' the myriad separate reactions of' each molecular species with all the others. With this substitution, the kinetics of' stepwise polymerization is essentially identical with that of simple condensation; without it, analysis of the kinetics would be almost hopelessly difficult If' the simplification is to be made, the rate of' the reaction of a group must be independent of the size of the molecule to which it is attached. This assumption is amply justified by experimental evidence. The rate constants of' condensation re-, actions in a homologous series reach asymptotic values independent of chain length quite rapidly and show no tendency to drop off' with increases in molecular size The diluting effect of' the large chain must, of' course, be taken into account by adjusting to constant molar composition of reactive groups. In addition, the rate constants fbr monof'unctional and bifunctional reagents are identical for sufficiently long chains separating the reactive groups of' the bifunctional compound
Kinetics of Stepwise Polymerization
With the concept of functional group reactivity independent of molecular weight, the kinetics of stepwise polymerization becomes quite simple The formation of a polyester from a glycol and a dibasic acid (Solomon 1972) may be taken as an example It is well known that this reaction is catalyzed by acids In the absence of added strong acid, a second molecule of the acid being esterified acts as a catalyst The reaction is followed by measuring the rate of disappearance of carboxyl groups:i
!Stepwise polymerization typically involves equilibrium reactions of the type
axe where the rates of the fbrward and reverse ~eactions k,[A][B] and k,[C][D], respectively. At equi-. librium these rates rue equal, whence K = k,lk, = [A][B]I[C][D] If the system is f from equilib~ium, a as in the initial stages of' polymerization, the reverse reaction is negligibly slow, and changes in the concentrations of the reactants may be considered to result fiom the fbrwaxd reaction alone, as in Eq 2-.I and those fbllowing
34
STEP-REAC.TION(CONDENSATION) POLYMERIZA'TION
If' the concentrations c of' carboxyl and hydroxyl groups are equal,
and
It is convenient to introduce the extent of reaction p , defined as the fraction of the functional groups that has reacted at time t . Then
and
or a plot of 1/(1 - p)' should be linea in time Typical experimental data (Fig 2-1) bear this out over a wide range of reaction times, but in fact the linear portion corresponds to only a relatively narIow range of conversion, roughly 80-93%
200
400
600
800
1000
'I'irne, rnin
FIG. 2-1.
Reactions of (di)ethylene glycol (DE) with adipic acid (A) and caproic acid (C) (Flory 1946, 1949) Time values at 202'C have been multiplied by 2
KINETICS A N D STATISTICS O F LINEAR STEPWISE POLYMERIZATION
Deviations at lower conversions are attributed to changes in the polarity of the reaction medium as the monomers are converted to polymer (Solomon 1972), and at higher conversions to an increase in the rate of' the reverse reaction (Bowden 1979, pp. 179--180) If only bifunctional reactants are present and no side reactions occur, the number of unreacted carboxyl groups equals the total number of molecules N in the system If acid or hydroxyl groups separately (not in pairs) are defined as structural units, the initial number of carboxyls present is equal to the total number No of structural units present The number-average degree of' polymerization F,, is simply
The scale of' DP in Fig. 2- 1 shows that uncatalyzed esterif'ications require quite long times to reach high degrees of'polymerization. Greater success is achieved by adding to the system a small amount of' catalyst, whose concentration is constant throughout the reaction In this case the concentration of the catalyst may be included in the rate constant:
Hence Z , , increases linearly with reaction time (Fig 2-2) Similar reactions have been shown to be linear up to at least 2, = 90, corresponding to molecular weights of 10,000
Statistics of Linear Step-Reaction Polymerization

Rate equations similar to Eq. 2-6 may be written for each molecular. species in the reaction mixture. From these it is possible to derive the distribution of' molecular weights in a stepreaction polymer The same relation is, however, more readily derived from statistical considerations The statistical analysis of stepwise polymerization yields results that are equivalent to and in some cases more easily obtained than those of' kinetic considerations The method assumes independence of' reaction rate and molecular size. The extent of reaction p is defined as the probability that a functional group has reacted at time t. This definition is entirely equivalent to the previous definition of p as the fraction of'the functional groups that has reacted. It follows that the probability of' finding a functional group urreacted is 1 - p
Time, rnin FIG. 2-2. Reaction of ethylene glycol wlth adipic acid, catalyzed by 0 4 mol % of p-toluenesulfon~c acrd (Flory 1946, 1949)
It is now necessary to find the probability that a given molecule selected at random is an x-mer, that is, contains x repeating units Such a molecule contains x - 1 reacted functional groups (e g , carboxyls) and, on the end, one unreacted group of this type The probability of finding a single reacted carboxyl group i n the molecule is p , and that of finding x - 1 of them in the same molecule is p x - ' . The presence of one unreacted group has a probability of 1 - p Hence the probability of finding the complete molecule is px-'(1 -- p) This is also equal to the fraction of all the molecules that are x-mers If there are N molecules in all, the total number of x-mers is
If the total number of units present is No, N = No(l - p), and
N,
No(l - ~ ) ~ p " - '
(2-8)
This is the number-distribution function for a linear stepwise polymerization at extent of reaction p The weight fraction wx of x-mers is given by
KINETICS A N D ST;4TISTICS OF LINEAR S'TEPWISE POLYMERIZATION
FIG. 2-3.
Number- or mole-fraction distribution of chain molecules in a linear step-reaction polymer for several extents of reaction p (Flo~y 1946, 1949)
This is the weight-distribution function for a linear stepwise polymerization at extent of reactionp Equations 2-8 and 2-9 are illustrated in Figs 2-3 and 2-4, respectively On a number basis, monomers are more plentiful than any other molecular species at all stages of the reaction On a weight basis, however, the proportion of lowmolecular-weight material is very small and decreases as the average molecular
FIG. 2-4.
Weight-fraction distribution of chain molecules in a linear step-reaction polymer for sexera] extents of reaction p (Flory 1946, 1949)
38
weight increases. The maximum in the weight-.distribution curve occurs near. the number-average molecular. weight.. Various average degrees of' polymerization may be derived from the weight- 01 number-distribution functions according to the definitions of' the corxesponding molecular.~.~weight averages in Chapter 1 . The number-average degree of' polymerization .?,, calculated in this way is that given in Eq.. 2,-5 The weight-average degree of' polymerization is
The breadth of' the molecular-weight distribution curve ?,/in = 1 p; at large extents of' reaction ?,l.Z,, + 2.0..This has been confirmed experimentally fbr nylon (Howard 1961)
.+
Molecular-Weight Control
Although initial molecular-weight control in a stepwise polymerization can be achieved by stopping the reaction (e g , by cooling) at the desired point, the stability of the product to subsequent heating may not be adequate with respect to changes in molecular weight The easiest way to avoid this situation is to adjust the composition of the reaction mixture slightly away from stoichiometric equivalence, by adding either a slight excess of one bifunctional reactant or (as in molecular-weight stabilization of nylons by acetic acid) a small amount of a monofunctional reagent Eventually the functional group deficient in amount is completely used up, and all chain ends consist of the group present in excess. If only bifunctional reactants are present, the two types of groups being designated A and B and initially present in numbers NA < NB such that the ratio r = NAINB,the total number of monomers present is (NA + NB) = NA(l + l l r ) At extent of reaction p (defined for A groups; for B groups, extent of reaction = rp), the total number of chain ends is NA(l - p) + NB(l - rp) = NA[l - p (1 - rp)lr] Since this is twice the number of molecules present,
x,, =
NA(l + l / r ) / 2 1 NA[1 - p (1 - rp)lr]/2
l + r r - 2rp
(2-11)
x,, =
l + r 1 - r
KINETICS A N D STATIST1C:S OF LINEAR STEPWISE POLYMERIZATION
Thus if 1 mol.% of' stabilizing groups is added,
This illustrates the precision needed in maintaining stoichiometric balance in order to obtain high degrees of' polymerization. Loss of' one ingredient, side reactions, or the presence of monofunctional impurities may severely limit the degree of' polymerization that can be achieved The analysis can be applied to the case of' an added monofunctional reagent, retaining Eqs 2-1 1 and 2-12 unchanged if' r. is appropriately defined in terms of' the numbers of' functional g~oups present.
Interchange Reactions
In interchange reactions, which may occur freely at elevated temperature, as in polymer melts, the molecula weights of the reacting molecules can change, since, for example, two molecules of average length may react to give one longer and one shorter than average However, the total number of molecules and hence Z, do not change. It can be shown that if free interchange takes place, the final molecular-weight distribution is always the most probable distribution defined by Eqs 2-8 and 2-9 The change can be demonstrated by mixing two polymers of different 2, and following a weight-average property (such as melt viscosity) as a function of time at elevated temperature
Multichain Polymer
Another type of step-reaction polymer of interest is produced by polymerizing a bifunctional monomer of type A--Bwith a small amount of monomer of functionality f of the type R-Af If the reaction is carried nearly to completion, the resulting polymer consists off chains growing out from a central unit R Network structures cannot occur, since no units of the type B-B are present The caseJ = 1 cor~esponds to the most probable distribution of Eq 2-9 The polymer formed when f = 2 is also linear; those with higher functionalities are branched Since the length of each of the branches is statistically determined independent of all the others, the probability of having a molecule with several branches all longer or shorter than average becomes smaller as the number of branches increases The molecular-weight distribution thus becomes narrower with increasing functionality, as shown in Fig 2-5 (Schulz 1939) The distribution breadth is given by
For f
2, X,IX,
= 1 .5.
40
STEP-REACTION (CONDENSATION) POLYMERIZAIION
104
FIG. 2-5.
Molecular-weight dist~~butions rnult~chain for polymers with funct~onal~ty shown (Schulz f as 1939)
GENERAL. REFERENCES
Flory 1946, 1949, 1953; Howard 1961; Lenz 1967; Solomon 1972, Chapter 1; Collins 1973, Exp 10; Allen 1974, Chapter 5; Elias 1977, Chapter 17 2; Bowden 1979, Chapter 2 4; Odian 1981, Chapter 2 2
D. POLYFUNCTIONAL STEP-REACTION POLYMERIZATION
Three-dimensional step-reaction polymers are produced from the polymerization of reactants with more than two functional groups per molecule. The structures of these polymers are more complex than those of linear step-reactionpolymers Threedimensional polymerization is complicated experimentally by the occurrence of gelation, or the formation of essentially infinitely large polymer networks in the reaction mixture The sudden onset of gelation marks the division of the mixture into two parts: the gel, which is insoluble in all nondegrading solvents, and the sol, which remains soluble and can be extracted from the gel As the polymerization proceeds beyond the gel point, the amount of gel increases at the expense of sol, and the mixture rapidly transforms from a viscous liquid to an elastic material of infinite viscosity An important feature of the onset of gelation is the low number average molecular weight of the mixture at the gel point, where its weight-average molecular weight becomes infinite In the statistical consideration of three-dimensional step polymerization, it is assumed that all functional groups are equally reactive, independent of molecular weight or viscosity, and that all the reactions occur between functional groups on different molecules Er~orsresulting from these assumptions affect only the nu-
POLYFUNCTIONAL STEP-REACTION POLYMERlZATlON
41
merical agreement between theory and experiment, and not the overall considerations of' the theory
Prediction of the Gel Point. In order to calculate the point in the reaction at which gelation takes place, a branching coefficient a is defined as the probability that a given functional group on a brnnch unit (i e . , a unit of fhnctionality greater than 2) is connected to another branch unit. The value of'a at which gelation becomes possible can be deduced from statistical considerations to be just 1l ( f 1); hence the critical value of' a for gelation is
Here f is the fhnctionality of' the branch units; if' more than one type of' branch unit is present, an average J' over all types of' branch units may be used in Eq. 2-14. The relation between a and the extent of reaction can be shown to be
? '
where the extents of reaction for A and B groups are p, and p,, and the ratio of A groups on branch units to all A groups in the mixture is p. Either p, or pB can be eliminated from this expression by defining r = NAINB, whence p, = rp, Then
Simpler relations can be derived fbr several cases. When equal numbers of A and B groups are present, r = 1 and p, = pB = p:
When there are no A-.A units, p
1 and
If both conditions apply, that is, r = p = 1,

a = p2
42
Finally, with only branch units present the probability that a functional group on a branch unit leads to another branch unit is just the probability that it has reacted: (branch units only)
a =p
(2-20)
The equations hold for all branch-unit functionalities with the definitions given for r and p
Experimental Observations of the Gel Point. The gel point can be observed precisely as the time when the polymerizing mixture suddenly loses fluidity, for example, when bubbles no longer rise in it. If the extent of reaction has been followed as a function of time, say, by removing aliquots ofthe solution and titrating fbr the number of functional groups present, the value of p at the gel point can be determined. In several cases investigated by Flory, observed values of a, were always slightly higher than those theoretically requir.ed.This discrepancy is attributed to the reaction of some functional groups to form intramolecular links, which do not contribute to network structures The reactions must therefore be carried slightly further to reach the critical point Molecular-Weight Distributions in T'hree-Dimensional Step-Reaction Polymers
The distribution functions for three-dimensional polymers are derived with somewhat more difficulty than those fbr the linear case, Eqs. 2-8 and 2-9.. They depend upon the fhnctionality and relative amounts of all the units involved. Only one example is discussed, for simplicity that of' the reaction of equivalent quantities of' two trifunctional monomers, all three of the functional groups of' each monomer being equally reactive.. In this case the weight-distribution function is ( j x -. x)!j w x = [(X - l)!(fx - 21
+ 2)!
px-l(l - p)Jx-Z.r+2
(2-21)
This equation is analogous to Eq 2-9 The first factor (in brackets) arises because of the numerous geometric isomers of a polyfunctional x-mer (In the linear case this factor is just x ) The second term is the probability of finding x - 1 links in an x-mer, and the last term is the probability of finding (f - 2)x 2 = f x - 2r 2 unreacted links or ends Equation 2-21 is illustrated in Figs 2-6 to 2-8 In comparison with the linear case, the weight distributions of branched step-reaction polymers of increasing functionality are progressively broader at equivalent extents of reaction (Fig 2-6) Figure 2-7 shows that the distributions broaden out with increasing extent of reaction p (or, alternatively, a, in the trifunctional case) In Fig 2-8 the weight fraction of the various molecular species is plotted against
POLYFUNCTIONAL. STEP-REAC:T'ION PO1,YMERIZATION
r FIG. 2-6.. ~ o l e c u l a-weight dtstnbution tn stepwtse polymertzation for monomers with functtonallty f , a t p = 0 3 (Mark 1950)
OL = p . In contrast to the linear case (Figs.. 2-7 and 2-8 versus Fig. 2-4), the weight fiaction of monomer is always greater than the amount of any one of the other species, the weight fractions of' higher species being successively lower. The extent of reaction at which the weight fraction of' any species reaches its maximum shifts continuously to higher values fbr higher molecular weights. In no case, however, does the maximum occur beyond the gel point (a = 1). Up to the gel point the sum of the weight fractions of all the species present must equal unity; beyond this point the sum of' the weight fractions of' all finite species drops below unity as the weight fraction of gel increases
Number of units, x FIG. 2-7. Molecular-weight distribution in step-reaction polymer formed from trifunctional units at various stages in the reaction, denoted by a = p (Flory 1946, 1949)
44
C Y
FIG. 2-8. Weight fractions of various molecular species in a trifunctional stepwise polymerization as a function of a = p (Flory 1946, 1949) The we~ghtfractions of the finite species are calculated from Eq 2-21, and that of the gel from Eq 2-24
The number-average degree of polymerization is given by
and the weight-average degree by
At the gel point the weight-average degree of polymerization becomes infinite As may be seen in Fig 2-9, where both averages are plotted against a = p , F,, attains only a value o f 4 at this point. The very large values of' ?,I?, near the gel point illustrate the extreme breadth of the distributions. (The portions of'the curves in Fig. 2-9 above the gel point refer only to the sol fraction ) The increase in distribution breadth with increasing degree of branching is illustrated in Fig 2-10. These curves represent the calculated distributions fbr bifunctional step reactions in which the branching factor a was varied by varying the small amount of added trifunctional units The curves were calculated for ? = 50 , in each case. The curves are drawn so that the total area under each curve extended to infinite .xis the same Equation 2-21 is valid for finite species beyond the gel point If it is summed over all values of x and subtracted from unity, the weight fraction of gel results
pOLYFUNC,TIONAL. STEP-REACTION POLYMERIZATION
FIG. 2 m . 9 .
Weight- and number-average degree of polymerization as a function of a fbr a trifunctional step-reaction polymer (Flory 1946, 1949)
For trifunctional units this is
The distribution functions for a > 1 are identical with those at the conesponding lower value 1 - - a, except that they are reduced by a factor w,= 1 - w,,the total weight fraction of sol present at that value of a Thus the complexity of the
FIG. 2-70.
Weight distributions for step-reaction polymers made to = 50 with various degrees of branching a induced by the addition of trifunctional units: a = 0 indicates no branching, and a = 0 50 is the critical point for gelation (Flory 1946, 1949)
46
STEP-REAC:TION (C:ONDENSATION) POLYMERIZATION
distribution of' the sol fraction above the gel point is presumed to become less until, at a = 1, only monomer is left in a vanishingly small amount..The reduction of' ?,I?,, shown for values of a increasing above in Fig. 2-9 is another way of expressing the same phenomenon Although the discussion was confined to the case of' equivalent quantities of' two trifunctional monomers, and Figs. 2-6 to 2-10 refer specifically to this case, Eqs 2-21 to 2-23 are more broadly applicable if p is replaced with a, which is then determined in the case of interest by r , p, and p, as indicated in Eq. 2-16
GENERAL REFERENCES
Flory 1946, 1949, 1953; Howard 1961; Lenz 1967; Solomon 1972, Chapter 1; Allen 1974, Chapter 5; EIias 1977, Chapter 17 3; Bowden 1979, Chapter 2 4; Odian 1981
DISCUSSION QUESTIONS AND PROBLEMS

1.
Contrast step and chain polymerization with respect to types of' growth reaction possible, change in monomer concentration and molecular weight of' polymer with conversion, species present during the polymerization, and the effects of' long reaction times on molecular weight
2. Identify and justify the major assumptions in the kinetic treatment of' stepwise polymerization..
3. Show by graphs (a) the dependence of' 2, and 2, on conversion, (b) the appropriate function of'p versus time, and (c) w,and N, versus time in stepwise polymerization
4. Derive equations for (a) molecular-weight control by added stabilizer and (b) the weight distr$bution of' molecular weight in linear step polymerization. 5. Discuss briefly in step polymerization (a) interchange reactions, (b) ring versus chain fbrmation, (c) the advantages of' molecular-weight control by added stabilizer, (d) how to obtain high molecular weights, (e) why an ester is used as a starting material in the preparation of poly(ethy1ene terephthalate), and (f) why a salt is used as a starting material in the preparation of' 66-nylon
6.
Calculate .2, and 2, fbr an equimolar mixture of a diacid and a glycol at the following extents of' reaction: 0.500, 0.750, 0.900, 0.950, 0.980, 0.990, 0.995 Explain the significance of each term in the equation fbr the weight distribution of molecular weight in linear step polymerization b Calculate the weight fraction of' trimer in such a polymerization car~ied to 99% conversion , c . Calculate ? and 2, fbr the above case a
7.
8.
A laboratory preparation of 610-nylon was made with 0 2 mol % acetic acid (based on carboxyl groups) p~esent a viscosity stabilizer The reaction was as car~iedto completion a Calculate M, and M,, b In a subsequent preparation, by error 2 mol % acetic acid was used. This reaction was also carried to completion The two batches of polymer, of equal weight, were mixed and melted prior to spinning The enor was noted when the melt viscosity was found to change with time Describe qualitatively the dependence of the viscosity on time and discuss its source Calculate the final values of M ,and M,, the mixture for c. In an effort to salvage some high-molecular-weight polymer, the mixture was extracted with a solvent in which only low-molecular-weight species were soluble. Analysis of the extract showed that it contained species up through the tetramer Assuming all such species were ~emoved, calculat4 the weight fraction removed in the extract and the values of M, and &' for the remaining extracted polymer
9. A polyester, made with equivalent quantities of a dibasic acid and a glycol, is to be stabilized in molecular weight at 2, = 100 by adding methanol. a How much methanol is required? b Calculate 2, and the weight and number fractions of monomer in the resulting polymer c. By error, the same number of moles of hydroxyl was added as glycerol instead of methanol Will the mixture gel if the reaction is carried to completion?
10. When caprolactam is polymerized into 6-nylon, 10 wt % of the monomer is left unchanged, and the remaining 90% is converted into polymer having the Flory distribution of molecular weight and 2, = 600. Later the mixture was extracted by a process that removed all species up through that with x = 5 Calculate Z,, for the following: a. The polymer with E, = 600. b The mixture of 10% monomer and 90% polymer c The extract, excluding the cyclic monomer d The extracted polymer.
Allcock 1981 H a r ~ y Allcock and Frede~ickW Larnpe, Contemporary Polymer Chernzsrry, PrenticeR Hall, Englewmd Cliffs, New Jersey, 1981 Allen 1974 P E M Allen and C R Patrick, Kznerlcs and Mechantsms of Polymerlzatlon Reactions. John Wiley & Sons, New York, 1974
48
STEP-REACTION (C:ONDENSATION) POLYMERIZATION
Bowden 1979 M J Bowden, "Formation of Macromolecules," Chapter 2 in F A Bovey and F H Winslow, eds , Macromolecules An Introductzon to Polymer Sczence, Academic Press, New York, 1979 Carothers 1929 W H Carothers, "An Introduction to the General Theory of Condensation Polymers," J Am Chem Soc 51, 2548-2559 (1929) Carraher 1982 Charles E Carraher and Jack Preston, eds , Interfaclal Synthesis, Recent Advances, Vol 3, Marcel Dekker , New York, 1982 Collins 1973 Edward A Collins, Jan Bare:, and Fred W Billmeyer, JI , Experiments In Polymer Sclence, Wiley-Interscience, New York, 1973 Elias 1977 Hans-Georg Elias, Macromolecules 2 Synthesis and Materials (translated by John W Stafford), Plenum Press, New York, 1977 Flory 1946 Paul J Flory, "Fundamental Principles of Condensation Polymerization," Chem Rev 39, 137-197 (1946) Flory 1949 Paul J Flory, "Condensation Polymerization and Constitution of Condensation Polymers," pp 21 1-283 in R E Burk and Oliver Grummitt, eds ,Hlgh Molecular Weight Organrc Compounds (Frontiers m Chemistry), Vol 6, Interscience, New York, 1949 Flory 1953 Paul J Flory, Prrncrples of Polymer Chemistra, Comell University Press, Ithaca, New York, 1953 t Howard 1961 G J Howard, "The Molecular Weight Distribution of Condensat~onPolymers," pp 185-231 in J C Robb and F W Peaker, eds , Progress rn Hrgh Polymers, Vol 1, Academic Press, New York, 1961 Lenz 1967 Robert W Lenz, Organlc Chem~sny Syttthetrc H ~ g h of Polymers, Wiley-Interscience, New York, 1967 Mark 1940 H Mark and G Stafford Whitby, eds , Collected Papers of Wallace Hume Carothers on High Polymeric Substances, Interscience, New York, 1940 Mark 1950 H Mark and A V Tobolsky, Ph>srcal Chemlstrp o f f f ~ gPolymerrc Systems, Interscience. h New York, 1950 Marvel 1959 C S Marvel, An Introductron to the Organlc Chemistry of High Polymers, John Wiley & Sons, New York, 1959 Mrllrch 1977 Frank Millich and Charles E Carraher, Jr , eds , Interfac~alSynthesis, Vol 1; Fundamentals, Vol 2, Polymer Appllcatlons and Technology, Marcel Dekker, New York, 1970 Morgan 1965 Paul W Morgan, Condensation Polymers By Interfaclal and Solution Methods, WileyInterscience, New York, 1965 Odian 1981 George Odian, Prznczples of Polymerization, 2nd ed , John Wiley & Sons, New York, 1981 Ravve 1967 A Ravve, Organic Chemrstry of Macromolecules, Marcel Dekker, New York, 1967 Schulz 1939 G V Schulz, "The Kinetics of Chain Polymerization V The Effect of Various Reaction Species on the Polymolecularity" (in German), Z Phys Chem B43, 25-46 (1939) Solomon 1972 D H Solomon, ed , Step-Growth Polvmerrzations, Vol 3 in George E Ham, ed , Kinetics and Mechanrsms of Polymer~zanon,Marcel Dekker, New York, 1972 Sorenson 1968 Wayne R Sorenson and Tod W Campbell, Preparative Methods of Polymer Chemrstr~~. 2nd ed , Wiley-Interscience, New York, 1968 VoZlmert 1973 Bruno Vollmert, Polymer Chemrsfry (translated by Edmund H Immergut), SpringerVerlag, New York, 1973
CHAPTER
RADICAL CHAIN (ADDITION) POLYMERIZATION
A.
MECHANISM OF VINYL. POLYMERIZATION
The polymerization of unsaturated monomers typically involves a chain reaction for It can be initiated by methods typ~cal simple gas phase chain reactions, including the action of ultraviolet light It is susceptible to retardation and inhibition In a typical chain polymerization, one act of initiation may lead to the polymerization of thousands of monomer molecules The characterist~csof chain polymerization listed in Table 2-1 suggest that the - actlve center responsible for the growth of the chain is associated with a single polymer molecule through the addition of many monomer units Thus polymer molecules are formed from the beginning, and almost no species intermediate between monomer and high-molecular-weight polymer are found Of several postulated types of active center, three have been found experimentally: cation, anion, and free radical Free-radical polymerization is discussed in this chapter, and the related cases of ionic and coordination polymerization are described in Chapter 4 The concept of vinyl polymerization as a chain mechanism is not new, dating back to Staudinger's work in 1920. In 1937 Flory showed conclusively that radical the polymerization proceeds by and requi~es steps of initiation, propagation, and termination typical of chain reactions in low-molecular-weight species The carbon--carbon double bond is, because of its relatively low stability, particularly susceptible to attack by a free radical The reaction of the double bond with a radical proceeds well for compounds of the type CH,--CHX and CH2=CXY, called vinyl monomers (Monomers in which fluorine is substituted for hydrogen may be included in this class ) The polymerization of monomers with more than one double bond is considered in Chapter 6.
50
RADICAL. CHAIN (ADDITION) POLYMERIZATION
Not all vinyl monomers yield high polymer as a result of'radical polymesizatior~. Aliphatic hydrocarbons other than ethylene polymerize only to oils, and 1,2-tlisubstituted ethylenes not at all. Among compounds of the type CH,=CXY, thosc in which both groups are larger than CH, polymerize slowly, if' at all, by radic+ill mechanisms
Generation of Free Radicals. Many organic reactions take place through i n w mediates having an odd number of' electrons and, consequently, an unpaired elcctron. Such intermediates are known as free radicals. They can be generated in a number of' ways, including thermal decomposition of organic peroxides or hyclr{~.peroxides (Mageli 1968) or azo or diazo compounds (Zand 1965) Two reactions commonly used to produce radicals for polymerization are fhc thermal or photochemical decomposition of' benzoyl peroxide,
and of' azobisisobutysonitrile,
The stability of radicals varies widely Primary radicals are less stable and r f ~ reactive than secondary radicals, which are in turn less stable than tertiary mf5 (Some tertiary radicals, such as the triphenylmethyl, can be isolated in the V>!L state without decomposition ) The phenyl radical is more reactive than the b'.L;, radical, the ally1 radical is quite unreactive, and so on TO be useful in initiating polymerization, a compound undergoing therms. 0 7 composition to free radicals should have a first-order decomposition rate cc~.gzfJ' of to sec- at the desired polymerization temperature, usually 50- 1 5(,"'Although thermal decomposition is a common means of generating rad~cs:- r ' has a disadvantage in that the rate of generation of free radicals cannot be conrrf>l rapidly because of the heat capacity of the system Photo~nitzatedpoiymenz:Lfl on the other hand, can be controlled with high precision, since the generaf,fl.l radicals can be made to vary instantaneously by controlling the intensity rrg "'5 initiating light (Oster 1969) Light of short enough wavelength (i e , high C%PG energy per quantum) can initiate polymerization directly It is customary, h r , ~ x rfl to use a photochemical in~tzatorsuch as benzoin or azobisisobutyronitrile. 3r-C;' is decomposed into free radicals by ultraviolet light in the 3600-A region. &TI-"" direct initiation through decomposition of monomer does not occur In polymerization these initiators are used at temperatures low enough so that +fie not undergo appreciabIe thermal decomposition Convenient sources of < H f / light are available in the form of mercury-arc lamps or fluorescent l a m ~ g I': special phosphors, and lasers
'
"J'
-- ,# /'
'
MECHANISM OF VINYL POLYMERIZATION
51
High-energy radiation from a wide variety of' sources, including electrons, gamma rays, x-rays, and slow neutrons, is effective in producing free radicals that can initiate polymerization (Dole 1972, Wilson 19'74).The reactions involved are varied and nonspecific, resulting from the gross damage to molecular structures resulting from the transfer of' large amounts of'energy. Monomers can be polymerized in the solid as well as the liquid and gaseous states (Eastmond 1970). The reactions of' high-energy radiation with polymers are discussed in Chapter 6 0 . Radicals are also generated as a result of oxidation-reduction reactions (redox initiation). Appropriate to an aqueous medium, redox initiation is often used with emulsion polymerization, and is discussed with that topic in Chapter 6 8 Electrochemical initiation (Friedlander 1966, Funt 1967) results from the generation of' free radicals by electrode reactions. In all of' these methods it is only the means of generating the free radicals that varies; the remaining steps in the polymerization process are unchanged. \, The generation of radicals for the initiation of' polymerization has been reviewed (Bevington 196 1, North 1966, O'Dr~iscoll1969)..
Initiation. When free radicals are generated in the presence of' a vinyl monomer, the radical adds to the double bond with the regeneration of' another radical.. If the by radical f'o~med decomposition of the initiator I is designated by Re,
The regeneration of the radical is characteristic of chain reactions Evidence for the radical mechanism of addition polymerization comes not only from the capability of radicals to accelerate vinyl polymerization, but also from the demonstration that the polymers so formed contain fragments of the radicals. The presence of heavy atoms, such as bromine or iodine, or radioactive atoms in the initiator has been shown many times to lead to polymers from which these atoms cannot be removed The efficiency with which radicals initiate chains can be estimated by comparing the amount of initiator decomposed with the number of polymer chains formed. The decomposition of the initiator can usually be followed by analytic methods The most direct method of finding the initiator efficiency then depends upon analyzing the polymer for initiator fragments Most initiators in typical vinyl polymerizations have efficiencies between 0.6 and 1 0;that is, between 60 and 100% of all the radicals formed ultimately initiate polymer chains The major cause of low efficiency is recombination of the radical pairs before they move apart (cage effect) (Noyes 1965)
52
RADICAL CHAIN (ADDITION) POLYMERIZAT'ION
The chain radical fbrmed in the initiation step is capable of' adding successive monomers to propagate the chain:
Propagation.
Termination. Propagation would continue until the supply of monomer was exhausted were it not for the strong tendency of' radicals to react in pairs to fbrm a paired-electron covalent bond with loss of' radical activity.. This tendency is corn.pensated for in radical polymerization by the small concentration of radical species compared to monomers. The termination step can take place in two ways: combination or coupling;
or disproportionation;
in which hydrogen transfer results in the fbrmation of two molecules with one saturated and one unsaturated end group. Each type of' termination is known. For example, studies (Bevington 1954a,b) of the number of' initiator fragments per molecule showed that polystyrene terminates predominantly by combination, whereas poly(methy1 methacrylate) terminates entirely by dispropor.tionation at polymerization temperatures above 60C, and partly by each mechanism at lower temperatures The termination reaction, controlled by diffusion in many cases, has been reviewed (North 1974)..
Radical-Molecule Reactions
Although the three steps of initiation, propagation, and termination are both necessary and sufficient for chain polymerization, other steps can take place during polymerization As these often involve the reaction between a radical and a mo1.ecule, they are conveniently so classified.
MECHANISM O F VINYL POLYMERIZATION
53
Chain Transfer. It was recognized by Flory (1937) that the reactivity of a radical could be transferred to another species, which would usually be capable of continuing the chain reaction. The reaction involves the transfer of an atom between the radical and the molecule. If the molecule is saturated, like a solvent or other additive, the atom must be transf'er~ed the radical: to
If the molecule is unsaturated, like a monomer, the atom transferred (usually hydrogen) can go in either direction:
The major effect of' chain transfer to a saturated small molecule (solvent, initiator, or deliberately added chain-transfer agent) is the formation of additional polymer molecules for each radical chain initiated. Transfer to polymer and transfer to monomer with subsequent polyme~ization the double bond lead to the fo~mation of of' branched molecules. The latter reaction has a pronounced effect on molecularweight distribution (Section C) and is important in the production of' graft copol-, ymers (Chapter 5D). The efficiency of compounds as chain-transfer agents varies widely with molecular stxuctu1.e. Aromatic hydrocarbons are rather unreactive unless they have benzylic hydrogens. Aliphatic hydrocarbons become more reactive when substituted with halogens. Carbon tetrachloride and carbon tetrabromide ase quite reactive. The reactivity of va~ious polymer radicals to transfer varies widely. Transfer reactions offer a valuable means for comparing radical reactivities.
Inhibition and Retardation. A retarder is defined as a substance that can react with a radical to form products incapable of adding monomer. If the retarder is very effective, no polymer may be formed; this condition is sometimes called inhibition and the substance an inhibitor. The distinction is merely one of' degree The two phenomena are illustrated in Fig. 3-1, where the rate of polymerization is displayed in idealized fashion.. The action of' a retarder is twofbld: It both reduces the concentration of radicals and shortens their average lifetime and thus the length of' the polymer chains.
RADIClAL CHAIN (ADDITION) POLYMERIZATION
FIG. 3-1.
Time Idealized time-conversion curves for typical inhibition and retardation effects (Goldfinger 1967): curve A, normal polymerization; curve B, simple inhibition; curve C, simple rctardation; curve D, nonideal inhibition; curve E, inhibition followed by retardation
In the simplest case, the retarder may be a free radical, such as triphenylmethyl or diphenylpicrylhydrazyl, which is too unreactive to initiate a polymer chain The mechanism of retardation is simply the combination or disproportionation of radicals If the retarder is a molecule, the chemistry of retardation is more complex, but the product of the reaction must be a radical too unreactive to initiate a chain Inhibitors are useful in determining initiation rates, since their reaction with radicals is so rapid that the decomposition of inhibitor is independent of jts concentration but gives directly the rate of generation of radicals As a result, the length of the induction period before polymerization starts is directly prcprtlonal to the number of inhibitor molecules initially present This number then represents the the number of radicals produced dur~ng time of the induction period For this analysis to be valid it must be known that one molecule of inhibitor i s exactly equivalent to one radical There are a number of reasons why this may be So, and each case must be investigated separately
vt
Configuration of Monomer Units in Vinyl Polymer Chains
The addition of' a vinyl monomer to a free radical can take place in eitkiw of' two ways:
R-
+ CHI=CHX
R-CH2-CH. X I
(I)
MECHANISM OF VINYL POLYMERIZATION
55
the reaction leading to the more stable product being favored Since the unpaired electron can participate in resonance with the substituent X in structure I but not in 11, reaction I is favored Steric factors also favor reaction I The occurrence of reaction I exclusively would lead to a head-to.tall configuration in which the substituents occur on alternate carbon atoms:
Alternative possibilities are a head-to.head, tall-to-tail configuration:
--CH2-CH--CH-CH~-CHP-CH-CH--CH~,
X
etc.
or a random structure containing both arrangements The possibility of' obtaining a regular head-to-.head,tail-to-tail configuration exclusively by chain polymerization is remote, and it appears that only an occasional monomer unit enters the chain in a reverse manner to provide an isolated head-to-,head, tail-to-tail linkage
Laboratory Methods in Vinyl Polymerization
Small-scale laboratory polymerizations, carried to low conversion with pure monomer and minor amounts of initiator and other agents, are of great importance in obtaining kinetic data for substantiating the kinetics and mechanisms of polymerizations Laboratory manuals are available describing the methods and experiments that are most useful (Sorenson 1968, McCaffery 1970, Braun 1971, Collins 1973) The experimental data of inte~estinclude the overall rate of polymerization, the rate of initiation (and transfer where present), and the molecular weight of the polymer The determination of molecular weight is discussed .in Chapter 8 Industrial-scale polymerization is discussed in Chapte~ and Part 4. 6
Rate of Polymerization. A chemical method or isolation and weighing of polymers can be used to measure the rate of polymerization, but it is customary to work in an evacuated and sealed system to avoid extraneous effects due to the presence of oxygen Polymerization rate is then followed through changes in a physical property of the system, such as density, refractive index, or ultraviolet or infrared absorption Measurement of density is usually the most sensitive and convenient technique The reaction is carried out in a dilatometer, a vessel equipped with a capillary tube in which the liquid level can be measured precisely The decrease in volume on polymerization is relatively large, being 21% for methyl methacrylate and 27% for vinyl acetate In practice, a few hundredths percent polymerization can be detected The experiments are usually carried out in the
56
RADICAL C H A I N (ADDITION) POLYMERIZATION
region below 5% polymer so as to avoid deviations f'rom constant polymerization rate due to depletion of reactants and other fbctors.. The dilatometer is placed in a constant-temperature bath to ensure isothermal conditions throughout the expeximent.
Rate of Initiation. The number of' polymer chains initiated per unit time can be calculated f'rom the number of free radicals produced in the system, if' the initiator efficiency is known. In photopolymerization the number of' f'ree radicals produced is twice the number of' quanta absorbed, since one quantum of light decomposes a single initiator molecule into two free radicals.. The number of quanta absorbed in the reaction vessel can be measured in a separate experiment The initiation rate obtained in this way can be checked by computing the number of' polymer molecules fbrmed per unit time f'rom the overall rate of' polymerization and x,, provided that chain transfer is absent and that the mechanism of termination is known. Since this is usually not the case, an independent check on the rate of' initiation is desirable. This may be obtained by the use of an inhibitor or retarder; fbr the polymerization
GENERAL REFERENCES
Marvel 1959; Bevington 1961; Ham 1967; Lenz 1967; Collins 1973; Allen 1974; Tedder 1974; Bamfbrd 1976; Allcock 1981, Chapters 3 and 5..
B.
KINETICS OF VINYL. RADICAL POLYMERIZATION
Conversion of Monomer to Polymer
The chemicai equations of' Section A for initiation, propagation, and termination of vinyl radical polymerization can be generalized in the following mathematical scheme. Initiation in the presence of' an initiator I may be considered in two steps: first, the rate-determining decomposition of the initiator into f'ree radicals R.,
and, second, the addition of' a monomer unit to form a chain radical M,.,
KINETICS O F VINYL RADICAL POLYMERIZATION
57
where the k's in these and subsequent equations are rate constants, with subscripts designating the reactions to which they refer.. The successive steps in propagation,
are assumed all to have the same rate constant k,, since radical reactivity is presumed to be independent of chain length. The termination step involves combination
or disproportionation
Except where it is necessary to distinguish between the two mechanisms, the termination rate constant is denoted k, The rates of the three steps may be written in terms of the concentrations (in brackets) of the species involved and the rate constants The rate of initiation is
where the factor f represents the fraction of the radicals formed by Eq 3-1 that is successful in initiating chains by Eq 3-2 The rate of termination is
Fo1.man.ycases of' interest the concentration of free radicals [M,,] becomes essentially constant very early in the reaction, as radicals are formed and destroyed at identical rates. In this steady-state condition v, = v, and Eqs. 3-6 and 3-7 may be equated to solve fbr [M.]:
58
The rate of' propagation is essentially the same as the overall late of' disappearance of monomer, since the number of'monomers used in Eq..3-2 must be small compared to that used in Eq 3-3 if polymer is obtained..Then
01, with
substitution from Eq . 3-8,
Thus the overall rate of polymerization should, in the early stages of the reaction, be proportional to the square root of' the initiator concentration and, if f is independent of' [MI, to the first power of' the monomer concentration. This is true if' the initiator efficiency is high.. With very low efficiency, f may be proportional to, [MI, making up proportional to [MI3'' The proportionality of'the overalI rate to the square r.oot of' initiator concentration has been confirmed experimentally in a large number of' cases Typical experimental data are shown in Fig. 3-2 The straight lines are drawn with the theoretical slope of'*. The deviation fbr polystyrene at low initiator concentration (Fig. 3-2c) is due to the presence of' some thermal initiation. This phenomenon, peculix to very pure
II I
FIG. 3-2.
Dependence of rate of polymerization up on concentration of initiator [I] for (a) methyl methacrylate with azobisisobutyronitnle, (b) methyl methacrylate with benzoyl peroxide, and (c) styrene with benzoyl peroxide (Mayo 1959)
KINETICIS OF VINYL. RAD1C:AL POLYMERIZATION
59
styrene and a few of' its derivatives, without any initiator present, involves the thermal dimerization of' styrene fbllowed by a reaction with still another styrene to produce a radical capable of' initiating polymerization (Ebdon 197 1)
Evaluation of Rate C:onstants. If' the rate constant for initiator decomposition and the initiator efficiency are known, the ~.atio rate constants k;/kf can be of' evaluated from the overall polymerization rate: From Eqs. 3-6 and 3-10 we have
- = ---kf
kp2
2upZ
vi [MI
These rate constants can be separated only by non-steady-state methods, described below.. Measurements of kp2/kfas a function of' temperature lead to useful information about the thermochemistry of' polymerization (Section C)..
Overall Rate as a Function of Conversion. If the initiator concentration does not vary much during the course of polymerization and the initiator efficiency is independent of monomer concentration, polyme~ization proceeds by first-order kinetics; that is, the polymerization rate is proportional to monomer concentration In some systems, such as the benzyol peroxide-initiated polymerization of styrene, the reaction is accurately first order up to quite high conversions (Fig. 3 3) The polymerization of certain monomers undiluted or in concentrated solution is accompanied by a marked deviation from first-order kinetics in the direction of an increase in reaction rate and molecular weight termed autoacceleration or the
FIG 3-3.
Time, min Polymerization of vinyl phenylbutyrate in dioxane with benzoyl peroxide initiator at 60C plotted as a first-order reaction (Marvel 1940) At t = 0, monomer concentration was (a) 24gilOOm1, (b) 6O,o/100ml, a n d ( c ) 7 3 g/100rnl
60
gel effect (Norxish 1939, Schulz 1947, Trommsdorff 1948). The effect is particularly pronounced with methyl methacrylate, methyl acrylate, or acrylic acid (Fig.. 3-4) It is independent of initiato~ is due to a decrease in the rate at which the polyme~ and molecules diffuse through the viscous medium, thus lowering the ability of two long-chain radicals to come together and terminate. Because termination is in h c t diffusion controlled for most liquid-phase polymerizations, even at low conversion, the dependence of diffusion rate on the viscosity of the medium leads to the gel effect at high polymer concentrations, with high-.molecular-weight polymer, or in the presence of inert solutes increasing the viscosity of the medium (North 1974). The decrease in termination rate leads to an increase in overall polymerization Iate and in molecular weight, since the lifetime of the growing chains is increased. At quite high conversions (70-90%) the rate of polymerization drops to a very low value as the system becomes glassy and monomer can no longer be supplied to the growing chain ends
E
Degree of Polymerization and Chain 'Transfer
Kinetic Chain Length. The kinetic chain length v is defined as the number of monomer units consumed pe1 active center. It is therefbre given by uplui = u,lu,:
Time, rnin
FIG..3-4.
Autoacceleration of' polyme~.ization in poly(methy1 methacrylate) (Naylor 1953) rate
KINETICS OF VINYL RADICAL POLYMERIZATION
Eliminating the radical concentration by means of Eq 3-9,
For initiated polymerization, use of Eq. 3-8 leads to
If no reactions take place other than those al~eady discussed, the kinetic chain , length should be related to ? the degree of polymerization: For termination by n , combination i = 2v, and for disproportionation 5 = v This is found to be precisely true for some systems, but for others wide deviations are noted in the direction of more polyme~moIecules than active centers These deviations are the result of chain-transfer reactions:
P-
+ M -+
PM.,
etc
where P may be monomer, initiator, solvent, or other added chain-transfer agent.. (Transfer to polymer is omitted, since no new polymer molecule is produced.) The degree of polymerization is therefbre
xn
rate of' g~owth - - of' all reactions leading - -dead polymer Z rates to
where the terms in the denominator represent termination by combination and transfer to monomer, solvent, and initiator, respectively. If termination is by disproportionation, the first term becomes 2fkd[I]. If transfer constants are defined as
then (assuming termination by combination)
62
The above analysis assumes that the radical formed in the transfix process is approximately as reactive as the original chain radical, otherwise retardation or inhibition results
Transfer to Solvent. In the presence of a solvent, and by properly choosing conditions to keep other types of chain transfks to a minimum, Eq.. 3-17 reduces to
where (ll?,), combines the *polymerization and transfer to monomer terms The linear dependence of l/Tn on [S]/[M] is illustrated in Fig. 3-5 In experiments of this type the ratio U ~ I [ M ] ~ be held constant throughout if (?,),is to be identified must with the observed degree of polymerization in the absence of solvent s
Transfer to Monomer and initiator. If only the terms cor~espondingto polymerization, transfer to monomer, and transfer to initiator are kept in Eq. 3-17, the resulting equation for k,, is quadratic in up. This behavior has been observed, fbr
Oo
u
10
I
20
A Toluene
Benzene
30
lSl/IMl FIG. 3-5. Effect of chain transfer to solvent on the degree of polymerization of polystyrene (Gregg 1947)
KINETICS OF VINYL RADICAL. POLYMERlZATlON
Chain transfer
with monomer
0
0 02
0 04
0 06
0 08
[1 I
FIG. 3-6. Contxibution of various souxces of termination in the polymerization of styrene with benzoyl peroxide at 60C (Mayo 1951)
example, in the benzoyl peroxide-initiated polymerization of styrene The contribution of transfer to monomer is constant and independent of polymerization rate, and that of transfer to initiator increases rapidly with increasing rate, since high rate requires high initiator concentration (Fig 3-6)
"
Control of Molecular Weight by Transfer. Chain-transfer agents with transfer constants near unity are quite useful in depressing molecular weight in polymerization reactions This is often of great commercial importance, for example, in the polymerization of dienes to synthetic rubbers, where chain length must be controlled for ease of processing The choice of transfer constant near unity ensuxes that the transfer agent, or regulator, is consumed at the same rate as the monomer so that [S]/[M] remains constant throughout the reaction Too large quantities are needed of chain-transfer agents with constants much lower than unity, and agents with transfer constants greater than about 5 are used up too early in the polymerization Aliphatic mercaptans are suitable transfer agents for several common monomers Table 3-1 lists transfer constants for several solvents and radicals The effect of chain-transfer reactions on chain length and structure is summarized in Table 3-2. The chemistry and kinetics of chain transfer have been reviewed (Palit 1965, Jenkins 1974)
64
TABLE 3-1. Chain Transfer Constants for Various Solvents and Radicals at 60C Except as Noteda
c, x 104 Solvent
Benzene Cyclohexane Toluene Chloroform Ethylbenzene Triethylamine Tetrachloroethane Carbon btrachloride Carbon tetrabromide
Styrene
0 023 0 031 0 125 05 0 67 7 1 18 (80C) 90 22,000
Methyl Methacrylate
0 040 0 10 (80C) 0 20 1 77 1 35 (80C) 83 0 155 2 40 2700
-
Vinyl Acetate
12 70 22 150 55 370 107 9600 28,700 (70C)
--
'Young (1975)
Inhibition and Retardation

The kinetics of' chain polymerization in the presence of' added inhibitor or retarder can be described (Goldfinger 1967; Odian 1981) by adding to the scheme of' Eqs 3-1 to 3-5 the reaction
where Z is the inhibitor It is assumed that the radical Z - does not initiate polymerization, and is terminated without regenerating Z . Application of' the steadyTABLE 3-2. Effect of Chain-Transfer Reactions on Chain Length and Structured
Effect on
-
Reaction
To small molecule, giving active radical To small molecule, retardation or inhibition To polymer, intermolecula~ To polymer, intramolecular
"Coll~ns (1973)
M,
None Decreases
Mw
Decreases
Structure
None
Decreases
May decrease or increase None None
May decrease or increase Increases Increases
None
None None
Produces long branches Produces short branches
K1NET'IC:S OF VINYL RADlCAl POLYMERIZATION
65
TABLE 3-3. Inhibitor Constants for Various inhibitors and Monomers at 50C Except as Notedd
Inhibitor
Styrene
0 326 64 2 518
-
Methyl Acrylate
------
Methyl Methacrylate
Vinyl Acetate
11 2 404
Nitrobenzene Trinit~obenzene p-Benzoquinone

DPPH
0 00464 0 204
5 5 (44C) 2000 (44C) 33,000
Oxygen
"Ulbricht 11975)
14,600
---
. -
----__.-___I
state assumption leads, in direct analogy with Eq.. 3-17, to the relation (again for termination by combination)
where C, = k,/kp. Table 3-3 lists inhibition constants fbr some common monomers and inhibitors..
Determination of Individual Rate Constants
It was shown above that the rate constants for propagation and termination occur together as the ratio kp2/k, the equations for both overall polymerization rate and in kinetic chain length Thus these rate constants cannot be evaluated separately through steady-state measurements Recourse must be had to the study of transient phenomena, such as the rate of polymerization before the free-radical concentration has reached its steady-state value The duration of this transient region depends on the length of time a free radical exists from its formation in the initiation step to its demise in termination If a photopolyme~~ization begun by turning on light of is intensity I, the concentration of free radicals grows gradually to its steady-state value over a time inter~al that 1s a function of the lifetime of the radicals When the light is turned off, the radlcaI concentration decays in a similar manner At the steady state the rate of polymerization is proportional to the square root of the light intensity: up = 1"' On the 0 t h hand, if the flashes are so rapid that the radical concentration changes very little during a single flash, the effect is the same as that produced by reducing the intensity by a constant factor but leaving it on all the tlrne: If the light and dark periods are equal, up = The rates for long and short flashes differ, in this case, by ~5(Fig 3-7).
The Sector Method.
RADICAL C H A I N (ADDI'TION) POLYMERIZATION
ON
OFF
0N
OFF
12 Time
18
24
[Minutes]
0
X
Z
C C -
[R
M'] Average
I 1
'I s
10 11 12 1 3 1 4 [ S e c o n d s ]
Time
(b) FIG.. 3-7. Radical concentration versus time in the sector method: (a) long flashes compared to T, for which the maximum radical concentration is, for example, 2 x molelliter and its average is 1 x lo-', and (b) short flashes, for which the average ~adical concentration is approximately fi X (Vollmert 1973, courtesy Springer-Ve~lag,New York)
The rate of' flashing at which the transition occurs can be related to the mean lifetime 7,~f the fxee radicals. This is defined as
Ts =
1 number of radicals -- -- [M-I number disappearing per unit time 2k,[M.I2 2 k t [ M . ]
(3-21)
or, if' the radical concentration is substituted from Eq. 3-9,
P = -k -
[MI
up
"kt
Thus T,, [MI, and the overall polymerization sate up suffice to calculate k,lk,. With the ratio k,2/kt obtained from other data, the individual xate constants may be evaluated The relation between 7,and the flashing rate is complex; in practice, the ex-
KINETICS O F VINYL. RAD1C:AL POLYMERIZATION
67
1 2 3 Log t/r, FIG. 3-8. Ratlo of polymer~zat~on rates for flashlng and steady ~llum~nation plotted agalnst log(tl7,) for several values of the ratlo T of dark to light per~ods (Bu~nett1947)
0
-1
perimental curves of polymerization rate (or radical concentration) against flashing rate are compared to the theoretical relation (Fig 3-8) In Figure 3-8 the ratio tl T,, where t is the time light is on, is plotted against E,l(v,),, where E, is the rate with flashing illumination and (v,), is the rate with steady illumination, for several values of the ratio r of dark to light intervals In practice, the light is usually flashed by placing a motor-driven rotatzng sector between the light source and the reaction vessel In the case of vinyl acetate, the radical lifetime is of the order of 4 sec, depending, of course, upon polymerization rate, according to Eq 3-22 The rate constants show that the growing chain adds about lo4monomers per second; the kinetic chain , length is the product of this figure and T , , or about 4 X lo4 The value of Z ,is about 10 times less than this because of chain transfer to monomer; thus about 10 transfers must occur during the life of a growing chain This represents a relatively large amount of transfer to monomer
Other Methods. The kinetic scheme for emulszon polymerzzatlon, described in Chapter 6, allows separation of the individual rate constants, since the growth of a single chain effectively takes place in an isolated system. In a vrscosity method (Bamford 1948), now little used, the polymerization is carried out in a viscomete~,measurement of molecula~weight by viscosity being substituted for measurement of rate The change in viscosity after photoinitiation is stopped provides data obtained under transient conditions. Dead-end polymerizatzon (Tobolsky 1960, Bohme 1966), which refers to polymerization with a limited amount of initiator, allows the determination of the efficiency f As the initiator is depleted, the initiation rate constant k, and initiato~ reaction stops short of completion at the point where the half-life of the propagating chain is comparable to that of the initiator.
68
TABLE 3-4.
-
Rate Constants for Propagation and Termination at 58C"
Monomer
k, (liter /mole~sec) 209 410 1000 2640 liOO0
k, (literlmole-sec x 10'
7,
Styrene Methyl methacrylate Methyl ac~ylate Vinyl acetate Vinyl chloride
115 24
35
117 2100
Values of the propagation and termination constants for several monomers are tabulated in Table 3-4 The rate constants vary over wide ranges Since they depend upon the reactivity of both the radical and the monomer, it is difficult to draw conclusions about the effect of structure on reactivity from these data alone Copolymerization experiments (Chapter 5) are more useful in this respect
GENERAL REFERENCES
Flory 1953; Walling 1957;Lenz 1967; Collins 1973, Exp. 3; Vollmert 1973, Chapter 211; Allen 1974, Chapter 7 ; Elias 1977, Chapter 20; Bamford 1976; Bowden 1979, Chapter 2.2; Allcock 1981, Chapter 12; Odian 1981, Chapter 3
I
I
I I
MOLECULAR WEIGHT AND ITS DISTRIBUTION
The distribution of molecular weights in vinyl polyme~s readily be calculated can if the reaction is restricted to low conversion, where polymer made throughout the reaction has the same average molecular weight If termination is by disproportionation (or transfer), let p be defined as the probability that a growing chain radical will propagate rather than terminate: p = vpl(vp v,) The probability of formation of an x-mer, as a result of x - 1 propagations and a termination, is px-'(1 - p ) The situation is in complete analogy with that for linear stepwise polymerization, and the expressions for 2,,, w,, and 2, given in N,, Eqs. 2-5 and 2-&40 2-10 apply If high polymer is formed, that is, p -+ 1 , ?,I = 2 Termination by combination is somewhat more complex, but it can be shown that, in analogy to the case of multichain polymer (Eq 2-13) with f = 2, Rwl f , = 1 5 if transfer is absent
x,
MOLECULAR WEIGHT A N D ITS DlSTRlBUTlON
69
The measurement of 5 % 1 would distinguish between the possibilities of termination by combination and by disproportionation However, this measurement is not easily car~ied with the desired precision; a preferable alternative procedure out involves measuring the number of' initiator fiagments per molecule of polymer For values of'p near 1, the most probable distribution of' Eq 2-9 reduces to
It can be shown that both this distribution, which describes low-conversion chain polymers terminating by disproportionation, and that describing termination by combination are special cases of a more general distribution function (Zimm 1948):
For z = I , Eq 3-24 reduces to Eq 3-23 with y = -In p For z = 2 it describes to the case of termination by combination Higher values of z cor~espond narTower distributions such as might be found in fractionated samples In contrast to the most probable distribution, the number distribution corresponding to Eq 3-24 does not indicate a large number of very small molecules; hence it is preferTed for describing polymers from which the small molecules have been removed, for example, by precipitation of the polymer from solution As polymerization proceeds, the degree of polymerization of the polymer being formed changes according to Eq 3-14 If the rate of initiation is constant as is usual, decreases throughout the reaction as [MI decreases, approaching zero at complete conversion At the same time f,l2,, increases with increasing conversion, approaching infinity as a limit Typical values of P,lb for high-conversion vinyl polymers seldom exceed 5 , however, except in special cases, such as autoacceleration in the polymerization of methyl methacrylate, where values of f,,,/2,,as high as 10 have been found The above distribution functions are not well suited to the description of these broader distributions Among empirical functions suggested for
70
these cases is the following fbrm, which has a high-molecular-weight "tail" when the exponent p is positive and less than 1 (here is the gamma function):
Where chain transfer to polymer is involved, extremely broad molecular weight distributions can risult The abnormal breadth arises from the fact that each transfer reaction adds a branch point to the molecule; after the branch has grown, thc molecule has increased in molecular weight. But the probability of branching is proportional (approximately) to weight-average molecular weight; hence branched molecules tend to become more highly branched and still larger The result is that the distribution of molecular weights has a long, high-molecula~-weight"tail." Among the empirical distribution functions suitable for describing fractions from which low-molecular-weight species have been removed is the so-called logarithmic normal distribution (Lansing 1935):
In this distribution the logarithm of the molecular weight follows a normal distribution function The ratio &If, is identical to f z / i , Other empirical equations have been proposed for cases of unusually broad distributions by Wesslau (1956), Tung (1956), Gordon (1961) and Roe (1961). Polymers with extremely narrow molecular-weight distribution are produced by certain types of anionic polymerization ("living" polymers, Chapter 4C) Their molecular -weight distribution is described by the Poisson distribution (Flory 1940):
=
$ ,,
I*
- I* 1 (x - I)!
X,-!J
1-2
.f,v
--=I+-Xn
CL
(I*
i
EFFECTS OF TEMPERATURE A N D PRESSURE O N C:HAIN POLYMERIZATION
71
GENERAL REFERENCES
Flory 19.53; Walling 1957; Eastmond 1967; Peebles 197 1; Bamf'ord 1976; Elias 1977, Chapter 20; Allcock 1981, Chapter 12; Odian 1981, Chapter 3
D. EFFECT'S OF TEMPERATURE AND PRESSURE O N CHAIN POLYMERIZATION
Thermochemical and thermodynamic data give valuable insight into the mechanisms of vinyl polymerization reactions. By the usual A ~henius treatment, the temper atuxe T dependence of the rates of the various steps in radical chain polymerization can be separated into an energy of activation representing the amount of energy that the reactant molecules must have to be able to react on collision, and a frequency factor which allows estimation of the fraction of the collisions that lead to reaction Enthalpies of polymerization are readily measured and may also be calculated from bond dissociation energies Differences between the two values give an indication of the amount of steric hindrance in the polymer Free energies of polymerization reactions can be calculated for some simple cases and give information on the thermodynamic equilibria involved in polymerization
Enthalpies and Free Energies of Polymerization Enthalpies of Polymerization. These may be obtained from the heats of combustion of monomer and polymer or directly by a calorimetric experiment; they may also be calculated from bond dissociation energies When the two results are compared, it is seen that the observed enthalpy of polymerization is equal to or somewhat lower than the calculated heat For vinyl compounds with two substituents on the same carbon atom the difference is 3-8 kJ/mole This is attributed to steric repulsion between the substituent groups in the polymer FOI monosubstituted vinyl derivatives the difference is much lower; only for styrene is it definitely greater than 4 kJ/mole. Enthalpies of polymerization (Table 3-5) vary over a wide range, reflecting differences in resonance stabilization and conjugation, steric strain in monomer and polymer, and (if present) dipole interactions and hydrogen bonding. Entropies and Free Energies of Polymerization. The entropy of polymerization is largely determined by the loss of the translational entropy of the monomer on polymerization Values of AS are therefore relatively insensitive to monomer type (Table 3-5) Both AH and AS are negative, but the exothermic nature of the polymerization reaction outweighs the influence of the entropy term in
AG = AH
TAS
72
TABLE 3,-5. Enthalpies and Entropies of Polymerization at 25C"
Monomer ---Tetrafluoroethylene Vinyl chloride Ethylene Vinyl acetate Propylene Acrylonitrile Isoprene Butadiene Styrene Methyl methacrylate Isobutylene a-Methylstyrene
(kllmole)
156 96 95 88 86 77 74 73 70 56 52 35
(Jlmol-deg)
112
100 110 116 109 101 86 105 117 120 104
so that the free energies are negative at usual temperatures Experimental methods for obtaining these quantities have been reviewed (.Joshi 1967) At elevated temperatures the entropy term increases, and when AH = T A S , AG = 0 and the polymerization reaction is in equilibrium with the reverse reaction of' depolymerization. This temperature is known as the ceiling tempernture for the polymer in question Some ceiling temperatures for. pure monomers are 6l'C fbr a-methylstyrene, 220C fbr methyl methacrylate, and 310C for styrene. The calculation of' the enthalpies and entropies of' reaction of' fiee radicals has been car~iedout for hydrocarbons and fluorocarbons and applied to the synthesis equations for polyethylene and polytetrafluoroethylene (Bryant 1951, 1962). lralues of enthalpy and free energy fbr the addition of' monomer to the radical W m ~ less e negative as the length of the chain radical increases in the first few prclpagation steps. After the growing free radical reaches a length of about eight carbr)ri atoms, all successive steps have the same enthalpy and free energy of' reaction, ti-i same as those for the overall polymerization step..
Activation Energies and Collision Factors
k jc (htained Overall Polymerization. The ratio of rate constants k,2/kr or ./, from the overall polymerization rate and the rate of initiation (Eq 3-1 1 ) E r s ~ e s s i n g a rate constant by an Arrhenius-type equation gives, for example,
E F C S OF TEMPERATURE AND P E S R ON CHAIN POLYMERIZATION FE T RSUE
73
where A, is the collision frequency factor fbr propagation and E, the energy of' activation for propagation Application to the sate-constant ratio gives
By plotting log(kplfi) against IIT, E, -- &Eland A , / ~ can be evaluated For , many polymerization reaction?, E, - iE, is about 20-25 kJ/mole This is equivalent to an increase in k,ldkI of about 30-35% for every 10C near room temperature Equation 3-29 refers only to the ratio of rate constants k,l+ and does not include the effect of temperaure on the rate of initiation In photopolymerization, where the rate of initiation is independent of temperature, Eq 3-29 represents the complete activation energy In an initiated polymerization, up is proportional to k , m , and the observed activation energy is ;Ed -t (E, - $El) The activation energy Ed for spontaneous decomposition of the initiator is about 125 kllmole for benzoyl peroxide or azobisisobutyronitrile The activation energy for initiated polymerization is therefore slightly greater than 80 kJ1mole This corresponds to a two- to threefold increase in rate for a 10C temperature change The activation energy for thermal polymer.ization is of the same order of magnitude, yet this reaction is much slower than normal because of the extremely low collision factor for thermal initiation The normal value of A for a bimolecular reaction is 10"- loi3;for the thermal initiation of polystyrene it is about lo4-lo6
Degree of Polymerization. If' transftr reactions are negligible, the temperature coefficient of' molecular weight depends on the initiation process. If an initiator is used, it fbllows from Eq 3-14, by virtue of'the differential fbrm d In kldT = E l RT2 of the Arrhenius equation, that
Since E, - iE, is 20- 25 kJ and Ed is about 125 kJ, this quantity is negative and the molecular weight decreases with increasing temperature The same is true in thermal polymerization In photopolyme~izationin the absence of transfer, d In Z,,ldT = (E, - iE,)IRT2, which is positive This is the only case in which molecular weight increases with temperature If transfe~reactions axe controlling, it is found that molecular weight decreases with increasing temperature By plotting log C, or -log Z ,against 1/T, El,, - E, , and A,,,IA, may be evaluated The quantity El,, usually exceeds E, by 20--60 kJ/ mole, higher activation energies going with poorer transfer agents Since the frequency factor for transfe~ usually exceeds that for propagation, the higher activation energy is responsible for the slower rate of transfer compared to monomer addition
74
Absolute Rates. The activation energies and frequency factors fbr propagation and termination can be separated if' the individual rate constants are known. For most monomers E, is near 30 kJ/mole, and E, varies from 12 to 20 kJ/mole.. The frequency factors vary rather widely, suggesting that steric effects are more important than activation energies in setting the values of the rate constants. For example, A, is much lower for methyl methacrylate with two substituents on the same ethylenic carbon atom that it is for less hindered monomers. All the values of' A, are considerably lower than normal, because it is likely of low probability of the transition leading to unpairing of' the double-bond electrons in the monomer.
'
Polymerization-.Depolymerization Equilibrium. Some polymers decompose by stepwise loss of' monomer units in a reaction that is essentially the reverse of' polymerization (see Chapter 6 ) The depolymerization reaction can be incorporated into the kinetic scheme for chain polymerization.. At ordinary temperatures the scheme is not changed because the rate constant for depolymerization is small. The activation energy of depolymerization is quite high (40-1 10 kJ/mole) compared to that of propagation, however, and at some elevated temperature the rates of' polyd merization and depolymerization become equal and the ceiling temperature is reached, above which polymer free radicals, in the presence of' monomer at I-atm pressure, depolymerize rather than grow
Effects of Pressure
The effects of high pressures (up to, say, several thousand atmospheres) on polymerization go far beyond the direct efftct of' pressure on the volume of'the system Both equilibrium constants (recall that AG = --RT In K ) and rate constants are affected, the former through the thermodynamic relation
and the latter through the Eyring rate theory (Glasstone 1941) by replacing K with k and A V with the volume change between the reactants and the transition state Since the volume change on polymerization is usually negative, an increase in pressure favors the fbrmation of' polymer and is accompanied by an increase in the ceiling temperature, often 15.-20C per 1000 atm The polymerization of' styrene at high pressures has been studied in more detail than that of' other monomers (Merrett 1951, Nicholson 1956a,b) The rate of' dissociation of' initiator is reduced, but the rate of polymerization is sufficiently increased to outweigh this, the overall polymerization rate being seven or eight times higher at 3000 atm than at ordinary pressures. The rate of' termination, being diffusion controlled, is decreased. All these factors are in the direction of' higher molecular weights, but the full extent of' the predicted increase is not observed, since the rate of transfer to monomer also increases with pressure and controls the molecular weight at 3000 atm and above
75
GENERAL REFERENCES
Flory 1953; Walling 19.57;Joshi 196'7; Sawada 1969; Allen 1974, Chapter 4; Ivin 1974; Weale 1974; Allcock 1981, Chapter 10; Odian 198 1 , Chapter 3
1. Write chemical equations for the following reactions in the benzoyl peroxideinitiated polymerization of vinyl chloride: initiation, propagation, termination by cornbination and by disproportionation, transfer to monomer and to polymer 2.
3.
State and justify the steady-state assumptions that are used in deriving the radical chain polymerization kinetic equations f o ~ Show by graphs the relations between the following quantities in radical chain polymerization: a. b. Conversion versus time, fbr the simple reaction and with (1) autoacceleration, (2) inhibition, and (3) retardation. Degree of polymerization and concentration of chain-transfer agent
4. Write kinetic equations for initiated radical chain polymerization showing (a) how rate is related to concenttations of initiator, radicals, and monomer and (b) how degree of polymerization is related to the same quantities 5. Discuss briefly the following phenomena in radical chain polymerization: (a) methods of generating free radicals; (b) inhibition and retardation; (c) the temperature dependence of rate and degree of polymerization; (d) absolute rate constants and radical lifetimes; (e) how to distinguish between termination by combination and by disproportionation; and (f) how to distinguish between transfer to monomer and to initiator
6. Rate the following steps in the polymerization of' ethylene in order of free energy, and indicate which ase thermodynamically favorable: propagation, termination, transfer to monomer, transfer to polymer.
7.
a,. Which of the following monomers would you expect to polymerize readily by a free-.radical mechanism? Why?
b c
Which of the above compounds would you expect to be most susceptible to attack by free radicals? Why? What relation do you see between the answers to (a) and (b)?
76
8. Each of' the fbllowing equations contains errors. Correct each equation and tell what it describes.. , [MI. a. v = k,1'2 (fkdlk,) b.. v = (kp/2k,) [M.] [MI.. c .?, = () X, ,, CS [S]/[M]. d. r, = (k,2/2k,) [Mllv,.
,
9. Show how degree of' polymerization depends on temperature in (a) polymerization initiated by thermal decomposition of' an initiator, (b) photopolymerization, and (c) when transfer reactions ase controlling.
10. If' the overall activation energy of an initiated polymerization is 90 kllmole and the activation energy of' decomposition of the initiator is 130 kJ/mole, how does the degree of'polymerization change with temperature between 50 and 100C? Assume no transfer reactions..
11.
Given that [MI = 10 molelliter, [Ma] = 1 x molelliter, kp = 150 liter/mole-sec, and vi = 3 X molelliter-sec, calculate (a) up in moles* per liter-second, (b) the rate of' polymerization in weight percent per hour,' (c) kt, (d) the radical lifetime, and (e) v . What can you say about the reaction if' the value of' .?, is (a) 10,000, (b) 5000, or (c) 1000?
12. A monomer with molecular. weight 100 and density 1 photopolymerizes in bulk at a rate of' 3.6 wt.% per hour when the sate of initiation is 1 X molellites-sec. The radical lifetime is 10 sec. Calculate [Me], k,, k,, M,, and M,, assuming termination by combination and no transfer
13. Mayo (1951) gave the following data for the benzoyl peroxide-initiated polyme~ization styrene at 60C Here f = 0 60, kd = 3 2 x 10-%ecC1, of and the molar volume of styrene is I20 cm3 Calculate the transfer constant to initiator up
X
lo4 molelliter-sec 0 05 0 07 0 09 0 13 0 20 0 26
Xn
V,
x lo4
0 40 0 60 0 80 1 25 1 60
Xn
8300 6700 5900 4500 3300 2200
1550 1170 770 510 340
14. Canale (1960) reported studies on the radical polymerization of methyl a-cyanoacrylate, CH2=C(CN)COOCH3, at 60C using az~bisisobut~ronitrile as initiator Here j kd = 1 1 x sec-' Monomer density at @J"C is 1.067 glml Polymer density at 60C is 1 289 g/ml The fractional rate of monomer conversion per hour determined at the beginning of each run was 0 035 when
I1
[I] = 4 x [I] = 1 52 made
0 125 when [I] = 5 04 x and 0 21 when molelliter Determine k,2/kl, indicating any assumptions
15. Derive kinetic equations for radical chain polymerization in the presence of
1
I
I
j1
an inhibitor such that essentially all the chains are terminated by reaction with the inhibitor Show how the rate of polymerization depends on the concentrations of monomer, initiator, and inhibitor at the steady state.
16. Polymerization by diradicals is initiated by the reaction M t - P . , where P is a diradical that can grow from both ends a Write all possible reactions for propagation (2), termination by combination (3), termination by disproportionation (3), and transfer to monomer (2) Note that reactions converting P to monoradicals Q. must be included b Write kinetic equations for each step, expxessing rate constants as multiples of k,, k,,, k,, and k, Remember that "cross-product" terms must be counted twice c. Using this kinetic scheme, show the following 1 The rate of initiation is
where [TI = [Q] + 2[P] is the total concentration of radical ends The overall polymerization rate is
'3. The r.ate of production of' polymer is
--d[ZMtl - k,, [TI [MI dt
+ krd[TI2
(Note: The steady-state condition says that d[P]ldt = 0 and d[Q]ldt = 0 The consequences of the latter are important ) d Derive an expression for 1I?,, and simplify it for the cases of termination by disproportion only and by combination only
BIBLIOGRAPHY
Allcock I981 Harry R Allcock and F~ederick Lampe, Contemporary Polymer Chemrrtry, PrenticeW Hall, Englewood Cliffs, New Jersey, 1981 Allen 1974 P E M Allen and C R Pat~ick, Krnetrcs and Mechanisms oj Polymer~zatzon Reactzons, John Wiley & Sons, New York, 1971
78
Bamford 1948 C H Bamford and M J S Dewar, "Studies in Polymerization I A Method for Determining the Velocity Constants in Polymerization Reactions and its Application to Styrene," Proc R Soc London A192, 309-328 (1948) Bamford 1976 C H Bamford and C F H Tippers, eds , "Free-Radical Polymerization," Vol 14A In Comprehensive Chemical Kinetics, Elsevier, Amsterdam, 1976 Bevington 1954a J C Bevington, H W Melville, and R P Taylor, "The Termination React~onin Radical Polymerization Polymerizations of Methyl Methacrylate and Styrene at 25O," 1 Polym Sci 12, 449-459 (1954) Bevzttgton 19546 J C Bevington, H W Melville, and R P Taylor, "The Termmation Reaction in Radical Polymerizations I1 Polymerization of Styrene at 60' and of Methyl Methacrylate at 0 and 60, and the Copolymerization of These Monomers at 60," 1 Polym Sci 14, 463 -476 (1954) Bevrngton 1961 J C Bevington, Radical Polymerization. Academic Press, New York, 1961 Bohme 1966 R D Bohme and A V Tobolsky, "Dead-End Polymerization," pp 594-605 In Herman F Mark, Norman G Gaylord, and Norbert M Bikales, eds , Encyclopedia oj Polymer Sc~ence and Te~hnology, Vo! 4, Wiley-Interscience, New York, 1966 Bowden 1979 M 1 Bowden, "Formation of Mac~omolecules,"Chapter 2 in F A Bovey and F H Winslow, eds , Macromoletules An Introductlon to Polymer Science, Academic Press, New York, 1979 Braun 1971 D~etrichBraun, Harald Cherdon, and Werner Kern, Techniques oj Polymer Synthesis andi Characterrzation, Wiley-Interscience, New York, 1971 Bryant 1951 W M D Bryant, "Free Energ~esof Format~onof Hydrocarbon Free Rad~cals I Appl~cation the Mechanism of Polythene Synthesis," J Polym St1 6, 359-370 (1951) to Bryant 1962 W M D Bryant, "Free Energies of Formation of Fluorocarbons and their Radicals Thermodynamics of Formation and Depolymerlzation of Polytet~afluoroethylene," I Polym Scl 56, 277-296 (1962) Burnett 1947 G M Burnett and H W Melville, "Determination of the Veloc~tyCoefficients for Polymerization Processes I The Direct Photopolymer~zationof Vlnyl Acetate," Proc R Soc London A189,456-480 (1947) Canale 1960 A J Canale, W E Goode, J B Kinsinger, J R Panchak, R L Kelso, and R K Graham, "Methyl a-Cyanoacrylate I Free-Radical Homopolyme~ization," I Appl Polym Sci 4, 231-236 (1960) Collins 1973 Edward A Collins, Jan BareS, and Fred W Billmeyer, Jr , Exper~ments in Polymer Science, Wiley-Interscience, New York, 1973 Dole 1972 Malcolm Dole, ed , The Radiatlon Chemistry oj Macromolecules, Aiadem~cPress, New York, 1972 Eastmond I967 G C Eastmond, "Free-Radical Polymer~zation," pp 361-431 in Herman F Mark, Norman G Gaylord, and Norbert M Bikales, eds , Encyclopedia of Polymer Science and Technology, Vol 7, Wiley.Interscience, New York, 1967 Eastmond 1970 G C Eastmond, "Solid-State Polymer~zat~on," Chapter 1 in A D Jenkins, ed , Progress in Polymer Science, Vol 2, Pergamon Press, New York, 1970 Ebdon 1971 J R Ebdon, "Thermal Polymerization of Styrene-A Crittcal Review," Br Polym J 3, 9-12 (1971) Ehas 1977, Hans-Georg Elias, Macromolecules 2 Synthesis and Materials (translated by John W Stafford), Plenum Press, New York, 1977 Flory 1937 Paul J Flory, "Mechanism of Vinyl Polymerizations," J Am Chem Soc 59, 241-253 (1937) Flory 1940 Paul J Flory, "Molecular Size Distribution in Ethylene Oxide Polymers," J Am Chem SOC 62, 1561-1565 (1940) Flory 1953 Paul J Flory, Prmciples of Polymer Chemistry, Cornell Un~versityPresn, Ithaca, New York, 1953
'
BIBLIOGRAPHY
79
Frredlander 1966 Henry Z Fr~edlander, "Electrochemical Init~ation," pp 629-641 in Herman F Mark, Norman G Gaylord, and Norbert M Bikales, eds , Encvclopedra of Polymer Scrente and Technology, Vol 5, Wiley-Interscience, New York, 1966 Funt 1967 B L Funt, "Electrolytically Controlled Polymenzations," Macromol Rev I, 35-56 (1967) Glasstone 1941 Samuel Glasstone, Keith J Laidler, and Henry Eyrlng, The Theory ojRate Processes, McGraw-Hill, New York, 1941 Goldfinger 1967 George Goldfinger, W~lliamYee, and Rlchard D Gilbert, "Inhibition and Retarda tion," pp 644-664 in Herman F Mark, Norman G Gaylord, and Norbert M Bikales, eds , Encyclopedra of Polymer Scrence and Technology, Vol 7, Wiley Intersclence, New York, 1967 Gordon 1961 Manfred Gordon and Ryong-Joon Roe, "Surface-Chemlcal Mechanism of Heterogeneous Polymer~zatlonand Derivation of Tung's and Wesslau's Molecular Welght Distnbution," Polymer 2, 41-59 (1961) Gregg 1947 R A Gregg and F R Mayo, "Chain Transfer in the Polymerizat~onof Styrene 111 The Reactivities of Hydrocarbons Toward the Styrene Radical," Drsc Faraday Soc 2, 328-337 (1947) Ham 1967 George E Ham, "General Aspects of Free-Radlcal Polyrnenzatlon," Chapter 1 In George E Ham, ed , Vinyl Polymerrzatzon, Vol 1, Part I, Marcel Dekker, New York, 1967 lvin 1974 K J Ivm, "The~modynamlcsof Addition Polymerization Processes," Chapter 16 In A D Jenklns and A Ledw~th,eds , Reactrvr~.Mechanzsm and Structure rn Polymer Chemrstry, John Wlley & Sons, New York, 1974 lenkrns 1974 A D Jenklns, "The Reactivity of Polymer Radicals m Propagation and Transfer Re actions," Chapter 4 In A D Jenklns and A Ledwith, eds , Reactrvrty, Mechanrrm and Structure In Polymer Chemistry, John Wlley & Sons, New York, 1974 Jorhr 1967 R M Josh1 and B J Zwolinskl, "Heats of Polymerlzatlon and their Structural and Mechanrst~cImplications," Chapter 8 in George E Ham, ed , Vrnyl Polymerlzatzon, Vol 1, Part I, Marcel Dekker, New York, 1967 Radlcal Korus 1975 R Korus and K F O'Dnscoll, "Propagat~on and Termination Constants In F ~ e e Polymenzat~on," pp 11-45 4 - 5 2 in J Brandrup and E H Immergut, eds , wlth the collaboration of W McDowell, Polymer Handbook, 2nd ed , Wlley-Interscience, New York, 1975 Lansrng 1935 W D Lansing and E 0 Kraemer, "Molecula~ Weight Analys~s M~xtures Sedlof by rnentation Equilibrium in the Svedberg Ultracentrlfuge," J Am Chem Soc 57,1369-1377 (1935) Lenz 1967 Robert W Lenz, Organrc Chemistry of Synthetrc Hrgh Polymers, W~ley-Interscience, New York, 1967 McCaffery 1970 Edward M McCaffery, Laboratory Preparatron for Macromolecular Chemrshy, McGrawHlll, New York, 1970 Magelr 1968 0 L Mageli and J R Kolcynskl, "Peroxy Compounds," pp 814-841 in Herman F Mark, Norman G Gaylord, and Norbert M Bikales, eds , Encyclopedra oj Polymer Sclence and New Technology, Vol 9, W~ley-lntersc~ence, York, 1968 Marvel 1940 C S Marvel, Joseph Dec, and Haiold G Cooke, Jr , "Optically Actwe Polymers from Active Vinyl Esters A Conven~entMethod of Studying the K~netics Polymenzation,'' J Am of Chem Soc 63, 3499-3504 (1940) Marvel 1959 Carl S Marvel, An lntroductron to the Organrc Chemzstry of Hcgh Polymers, John W~ley & Sons, New York, 1959 Mavo 1951 Frank R Mayo, R A Gregg, and Max S Matheson, "Cham Transfer in the Polyme~lzation of Styrene VI Cham Transfe~with Styrene and Benzoyl Peroxide, the Efficiency of Initiation and the Mechanism of Chain Termmation," J Am Chem Soc 73, 1691-1700 (1951) illavo 1959 Flank R Mayo, "Cont~ibutions of Vlnyl Polymenzation to Organic Chernlstry," J Chem Educ 36, 157-160 (1959) Merrett 1951 F M Menett, Ph D , and R G W Nor~lsh, D , "Aspects of Polymer~zatlon High Ph at Pressures," Proc R Soc London A206, 309-334 (1951)
80
Naylor 1953 M A Naylor and F W Billmeyer, Jr , "A New Apparatus for Rate Studles Applied to the Photopolymerization of Methyl Methacrylate," J Am Chem Soc 75, 21 81-2185 (1953) Nrcholson 19560 A E Nicholson and R G W Norrish, "The Decomposition of Benzoyl Peroxide in Solution at High Pressures," Dlsc Faraday Soc 22, 97-103 (1956) Nrcholron 1956b A E Nicholson and R G W Norrish, "Polymerization of Styrene at High Pressures using the Sector Technique," Drsc Faraday Soc 22, 104-113 (1956) Norrrsh 1939 R G W Norrish and E F Brookman, "The Mechanism of Polymerization React~ons I The Polymerization of Styrene and Methyl Methacrylate," Proc R Soc London A171, 147171 (1939) North 1966 Alastair M North, The Krnetrcr of Free Radrcal Polvmerrzat~on,Pergamon Press, New York, 1966 North 1974 A M North, "The Influence of Chain Structure on the Free-Rad~cal Terminat~on Reaction," Chapter 5 In A D Jenkins and A Ledw~th, , Reacrivrry, Mechanrsm and Structure rn Polymer eds Chemistry, John Wiley & Sons, New York, 1974 Noyes 1965 Richard M Noyes, "Cage Effect," pp 796-801 In Herman F Mark, Norman G Gaylord, and Norbert M Bikales, eds , Encvclopedra of Pohmer Scrertce and Technology, Vol 2, WileyInterscience, New York, 1965 Odlan 1981 George Odian, Prmcrpler oj Polymer~zation,2nd ed John Wlley & Sons, New York,1981 O'Dr~scoll 1969 Kenneth F O'Dr~scoliand Premamoy Ghosh, "In~t~at~onFree Radical Polymerl In zation," Chapter 3 in Telj~ Tsuruta and Kenneth F O'Dnscoll, eds , Structure and Mechanrrm m Vrnyl Polymerzzatron, Marcel Dekker, New York, 1969 Oster 1969 Gerald Oster, "Photopolymenzation and Photocrossllnk~ng," pp 145-156 in Herman F Mark, Norman G Gaylord, and Norbert M Bikales, eds , Encvclopedra of Polymer Sctence and New York, 1969 Technology, Vol 10, W~ley-Intersc~ence, Paltt 1965 Sant~ Pallt, Satya R Chatterlee, and Aslsh R Mukherjee, "Chain Transfer," pp 575R 610 In Herman F Matk, Norman G Gaylord, and Norbert M B~kales,eds , Encyclopedra oj Polymer Science and Technology, Vol 3, Wiley-Interscience. New York, 1965 Peebles 1971 Leighton H Peebles, Jr ,Molecular WeightDistribunons m Polymers, Wiley -Interscience, New York, 1971 Roe 1961 Ryong-Joon Roe, "A Test Between Rival Theories for Molecular Weight Distributions in Heterogeneous Polymerization: The Effect of Added Terminating Agent," Polymer 2, 60-73 (1961) Sawada, "Thermodynam~cs of Polymerization," J Macromol Scr Rev MacSawada 1969 H~deo romol Chem 3, 313--395 (1969) Schulz 1947 G V Schulz and G Harborth, "On the Mechantsm of the Explosive Course of the Polymerization of Methyl Methacrylate" (in German), Makroinol Chem 1, 106-139 (1947) Sorenson 1968 Wayne R Sorenson and Tod W Campbell, Preparatrve Methods ojPolymer Chemrctry, 2nd ed , Wiley-Interscience, New York, 1968 Staudinger 1920 H Staudinger, "Polymerization" (in German), Ber Dzcch Chem Grs B 53, 1073I085 (1920) Tedder 1974 J M Tedder, "The Reactlv~tyof Free-Radicals," Chapter 2 In A D Jenkins and A Ledwith, eds , Reactrvrrv, Mechanrrm and Structrrre rn Polvmer Chemrrrrv, John Wiley & Sons, New York, 1974 Tobolrkv 1960 A V Tobolsky, C E Rodgers, and R D Bnckman, "Dead-End Radical Polymerizatlon 11," I Am Chem Soc 82, 1277- 1280 (1960) Trommcdorff 1948 Emst Trommsdorff, Herbert Kohle, and Paul Lagally, "On the Polymerlzat~onof Methyl Methacrylate" (in German), Makromol Chem 1, 169-198 (1948) 7ung 1956 L H Tung, "Fract~onat~on Polyethylene," J Polvrn Scr 20, 495-506 (1956) of
BIBLIOGRAPHY
81
Ulbrrcht 1975 1 Ulbncht, "Inh~b~tors Inhlb~tlonComtants In Free Radical Polymer~zat~on," and pp
11 53-11 55 In J Brandrup and E H Immergut, eds , w~ththe collaborat~onof W McDowell,

I
Polymer Handbook, 2nd ed , W~ley-Intersclence,New York, 1975 Vollmert 1973 Bruno Vollmert, Polymer Chemrstry (translated by Edmund H Immergut), Spnnger-
Verlag, New York, 1973

iVallmg 1957 Cheves Walling, Free Radicals rn Solutron, John Wiley & Sons, New York, 1957 Weale 1974 K E Weale, "The Influence of Pressure on Polymerizat~onReactions," Chapter 6 in A D Jenkins and A Ledwlth, eds , Reacnvity, Mechanrsm and Structure in Polymer Chemrstry, John Wiley & Sons, New York, 1974 Wesslau 1956 Hermann Wesslau, "The Molecular Weight Distributions of Some Low Pressure Polyethylenes" (in German), Makromol Chem 20, 11 1-142 (1956) Wrlson 1974 Joseph E Wilson, Radratron Chemistry o ~%lonomerr, Polymers & Plastics, Marcel f
Dekker, New York, 1974

Yortng 1975 Lewis J Young, "Transfer Constants to Monomer, Polymer, Catalyst and Solvent in Free Radical Polymenzation," pp I1 57-11 104 in J Brandrup and E H Immergut, eds , w ~ t hthe collaboration of W McDowell, Polymer Handbook, 2nd ed , Wiley-Intersc~ence, New York, 1975 Zand I965 Robert Zand, "Azo Catalyst," pp 278-295 in Herman F Mark, Norman G Gaylord, and Norbert M B~kales,eds , Encycloped~ao j Polymer Science and Technology, Vol 2, Wlley
Interscience, New York, 1965

Zrmm 1948 Bruno H Zlmm, "Apparatus and Methods for Measurement and Interpretat~onof the Angular Varlatlon of Light Scattering. Prel~mlnaryResults on Polystyrene Solutions," J Chem Phys 16, 1099 -1 1 16 (1948)
CHAPTER FOUR
-
IONIC AND COORDINATION CHAIN (ADDITION) POLYMERIZATION
A.
SIMILARITIES AND CONTRASTS IN IONIC POLYMERIZATION
Chain-reaction polymerization is known to occur by several mechanisms other than those involving free radicals discussed in Chapter 3.. Prominent among these are reactions in which the chain carriers are carbenium ionst (cationic polymerization, Section B ) or carbanions (anionic polymerization, Section C ) . In addition, polymerization may take place by mechanisms involving coordination compounds among the monomer, the growing chain, and a catalyst, usually a solid (coordination or insertion polymerization, Section D ) Still other chain polymerization reactions include ring opening and ase discussed in Section E The mechanisms of' these polymerizations are less thoroughly understood than that of radical polymerization, for several reasons. Reaction systems are often heterogeneous, involving inorganic catalysts3 and organic monomers Usually large efftcts may be produced by a third component (cocatalyst) present in very low concentration.. Polymerization often leads to very-high-molecular-weight polymer at an extremely high rate, increasing enormously the difficulty in obtaining kinetic data or even reproducible results
tThe term carbenium ion replaces carbonium ion fbr. species like CH,+at the suggestion of Olah (1972) $The word catalyst should be used when the agent exerts its influence throughout the growth of thc chain and is not consumed in the reaction, and initiator when it is used up at the instant of initiation Because of uncertainty over the mechanisms of ionic polymerization, both terms are found In addition Kennedy (197'7) suggests that the terms cntalyst and cocatalyst be replaced by toinitiator and initiator respectively, in a complete reversal of' terminology from that used traditionally and in this book
SIMIL.ARITIES AND CONTRASTS IN IONIC POLYMERIZATION
83
As indicated in Table 4- 1, the type of monomer polymerizing best by each mechanism is related to the polarity of the monomer and the acid-base strength of the ion formed Monomers with electron-donatlng groups attached to the doublebonded carbons form stable carbenium ions and polymerize best with cationic catalysts. Conversely, monomers with electron-withdrawing substituents form stable anions and require anionic catalysts Free-radical polymerization may be considered an intermediate case that is favored by moderate electron withdrawal from the double bond plus conjugation in the monomer Many monomers can polymerize by more than one mechanism, including coordination polymerization. 1 Initiation of ionic polymerization usually involves the transfer of an ion or an electron to or from the monomer, with the formation of an ion pair The counterion of this pair stays close to the growing chain end throughout its lifetime, particularly in media of low dielectric constant Several possible structures, including bound or contact ion pairs, solvated or solvent-separated ion pairs, and free ions can sometimes be distinguished experimentally or may be in equilibrium In contrast to radical polymerization, termination in ionic polymerization never involves the reaction between two growing chains, but usually involves the unimolecular reaction of a chain with its counterion ox a t~ansferreaction leaving a species too weak to propagate An agent that retards radical polymerization, such as oxygen, often has little influence on ionic polymerization; but impurities that neutralize the catalysts prevent the reaction, and many aromatic, heterocyclic, olefinic, and acetylenic compounds retard or terminate ionic polymerization In addition to these differences, ionic polymerization can usually be distinguished from free-radical polymerization by its very high rate and low (often below 0C) reaction temperature
TABLE 4-1.
Types of Chain Polymerization Suitable for Common Monomersa
* -
--
Polymerization Mechanism*
-
Monomer. Type Ethylene Propylene and a-olefins Isobutylene Dienes Styrene Vinyl chloride Vinylidene chloride Vinyl fluoride Tetrafluoroethylene Vinyl ethers Vinyl esters Acrylic and methacrylic este~s Acrylonitrile
Radical
Cationic
Anionic
Coordination
- -- -
"Lenz (1967) t = high polymer formed, - = no reaction or ol~gomers only
IONIC AND C;OORDINATlON CHAIN (ADDITION) POLYMERIZAI'ION
Kinetics
It is convenient to consider some common features of cationic and anionic polymerization

Formation of Initiator. In many instances the species that initiates ionic polymerization is produced by ionization of' the added initiator, fbr. example, H ' from BF3 . H20 or C,H,,- from potassium amyl. Sometimes the added substancc reacts with the monomer or. with the solvent to produce the initiating species, as in the reaction of a strong base with dimethyl sulfbxide to form the initiating anion CH3SOCH2Initiation. In the step of initiation the initiating species can react with the monomer in one of two ways: transfer of one electron without the formation of a bond between the two and transfer of two electrons with bond formation The transfer of a single electron from initiator to monomer always produces a radical ion, two of which generally dimerize to produce a diion Three caws can be differentiated The first is electron transfer from an electron donor to thc monomer, for example, from the naphthalide ion (formed by the ionization of $odium naphthalide) to styrene:
d-c" The second and third cases of single-electron transfer are, respect~t-;J;~. transfer from a monomer to an electron acceptor and the formation of 8 :hxFtransfer complex between a monomer with electron-donor or ele~tron-2~~zpr:): r F P erties and another molecule When two electrons are transferred, the monomer and the initiaticg 5pficJfi zz bonded together Again three cases can be distinguished:
a. A cation can add to a monomer in a nucleophilic reaction to f o : v'~~::flloc='

cation, for example, in the initiation of isobutylene p o l y m r n 7 ~ ~!*- ~ ~ r : trifluoride
CATIONIC POLYMERIZATION
85
b. An anion can add to a monomer in an electrophilic reaction to form a

monomer anion, as in the initiation of styrene polymerization by potassium amyl:
c.
A neutral molecule and a monomer can add to produce a zwitterion, as in the case of initiation of polymerization of a lactone with a tertiary amine:
Propagation and Termination. Whether the initiation step has led to a monoion or a diion, propagation takes place by the further addition of monomers to the growing chain, accompanied by its counterion Termination involves the destruction of the growing anion or cation, possibly by recombination with the counterion or reaction with another species, but in many instances such a reaction does not take place and there is no termination These cases, known as "living polymers," are described in Section C
GENERAL REFERENCES
Elias 1977, Chapter I8 I ; Odian 1981, Chapters 5-1 and 5-5
B.
Typical catalysts for cationic polymerization include, in order of importance, aprotonic acids (Lewis acids and Fr~edel--Crafts halides), protonic (Bronsted) acids, and stable carbenium-ion salts All these are strong electron acceptors Many of them, particularly the Lewis acids, require a cocatalyst, usually a Lewis base or other proton donor, to initiate polymerization Those monomers with electron-donating 1,l-substituents that can form stable caxbenium ions are polymerized by cationic mechanisms. High rate of polymerization at low temperatures is a characteristic of ionic polymerizations. It is often difficult to establish uniform reaction conditions before the reactants are consumed The polymerization of isobutylene by AlCI, or BF, takes place within a few seconds at - 10O0C, producing polymer of molecular
86
weight up to several million. Both rate and molecular weight are much lower at room temperature. These considerations limit the industrial application of cationic polymerization to the above-mentioned example
Mechanism
As discussed in Section A , the most satisfactory theory of' cationic polymerization involves the carbenium ion as the chain car~ier For example, in the polymerization . of' isobutylene with boron trifluoride catalyst, the first step is the reaction of' the catalyst and a cocatalyst, fbr example, water, to form a catalyst-cocatalyst complex that donates a proton to an isobutylene molecule to give a carbenium ion, (CH,),C+ . This ion then reacts with monomer with the reformation of' a carbenium ion at the end of each step. The "head-to-tail" addition of' monomer to ion is the only one possible for energetic reasons. Since the reaction is in general car~iedout in a hydrocarbon medium of low dielectric constant, the anion and the growing cationic end form an ion pair ? The termination reaction can take place by the rearrangement of' the ion pair to yield a polymer molecule with terminal unsaturation, plus the original complex,
Here the catalyst-cocatalyst complex is regenerated, and many kinetic chains can be produced from each catalyst-cocatalyst species. An alternative termination mechanism, in which the catalyst or cocatalyst combines with the growing chain, is unlikely in this example, but can take place when a covalent bond is fbrmed, as in the polymerization of' styrene with trifluoroacetic acid:
Chain transfer to monomer can also take place, and chain transfer to polymer is known and leads to branched polymers The efficiency of the catalyst is dependent on the acid strength of' the catalystcocatalyst complex. Also, the efficiency of the catalyst as a terminator should be related to the base strength of' the complex anion. Thus the more active a molecule is as a catalyst, the less active it is as a terminator. Although the requirements of the isobutylene-BF, system are fully satisfied by the mechanism just given, gen-
87
eralization of the scheme should be applied with caution. The experimental data on other systems, for example, do not conclusively prove the necessity of a cocatalyst
Kinetics
Although many cationic polymerizations proceed so rapidly that it is difficult to establish the steady state, the following kinetic scheme appears to be valid.. Let the catalyst be designated by A and the cocatalyst by RH. Initiation, propagation, termination, and t~ansfer may be represented as follows:
The rate of initiation is
where [A] is the catalyst concentration (If the formation of H + - A R is the ratedetermining step, v, may be independent of [MI and the kinetic scheme should be appropriately modified ) Since, in strong contrast to radical polymerization, termination is first order,
where [Mf] is written as an abbreviation for [HM+-AR]. By the steady-state assumption (which appears to hold in most cases, despite uncertainty as to whether such a state is actually achieved),
Kk, [M'] = -- [A][RH][M] kt

The overall polymerization rate is
If termination predominates over transfer,
88
whereas if transfer predominates,
is Since up is proportional to kikplkf,and, if termination predominates, .in proportional to k,/k,, the overall rate of' polymerization increases with decreasing temperature if' E, > E, E,, and the molecular weight increases as the temperature of' reaction decreases if' E, > E, Since the propagation step involves the approach of'an ion to a neutral molecule in a medium of' low dielectric constant, no activation energy is necessary The termination step, on the other hand, requires the leu.rangement of' an ion pair and thus involves an appreciable activation energy. These conditions are exactly the reverse of those fbund in free-radical polymerization. The dielectric constant of the medium has a significant effect on the nature of' the ion pair, which can range between the extremes of complete covalency and completely free ions. It is often convenient to consider two types of' propagating species, an ion pair and a free ion, in equilibrium. The nature of the solvent and the counterion can alter the course of the polymerization greatly by shifting this equilibrium. Large increases in up and 2, are usually observed as the solvating power of the medium is increased: The concentration of' the free ion, which propagates more rapidly, is increased at the same time. Similarly, the larger and the less tightly bound the counterion, the more rapid is the propagation
Plesch 1963, 1968; Eastham 1965; Lenz 1967; Collins 1973, Exp 6 ; Baker 1974; Ledwith 1974a; Elias 1977, Chapters 18-2 and 18-4; Russell 1977; Bowden 1979, Chapter 2 2 3; Odian 1981, Chapter 5-2; Kennedy 1982a,b
C.
ANIONIC POLYMERIZATION
Anionic polymerization was carried out on a commercial scale for many years before the nature of' the polymerization was recognized, in the production of' the buna-type synthetic rubbers in Germany and Russia by the polymerization of' butadiene with sodium or potassium as the catalyst The first anionic chain reaction to be so identified was the polymerization of' methacrylonitrile by sodium in liquid ammonia at -75OC (Beaman 1948), In modern times, the growth of' commercial products of' anionic polymerization has been phenomenal (Hsieh 1981). Most of' these polymerizations are based on the use of'organometal catalyst systems, which allow unprecedented control over. polymer structure (Chapters 5D, 12E, and 13) In addition to the monomer types listed in Table 4-1, monomers that polymerize by ring scission (Section E ) may do so by anionic mechanisms..
f
;
Mechanism
The conventional method of initiation of ionic chains involves the addition of a negative ion to the monomer, with the opening of a bond or ring and growth at one end:
NH2- 5 CH2=CHX CH30- CH2CH2 \ / 0
---t
H2NCHzCHX-
---* CH30CHzCH20-
(In the equations of this section the counterions are omitted; however, their role is essentially the same as in cationic polymerization..)Simultaneous growth from more than one center can be obtained from polyvalent ions such as those derived from
The more basic the ion, the better it serves to initiate chains Thus OH- will not initiate the anionic polymerization of styrene, NE-I, initiates fairly well, but 0 CH2- is powerful Similarly, more acid monomers require less basic ions, with the acidity of the monomer depending on the strength of the X---M-bond formed in initiation and the stability of the resulting ion Initiation can also occur by the transfer of an electron to a monomer of high electron affinity If D or D- is an electron donor,
'
Presumably, M- is less reactive than a true carbanion or a free radical, but the addition of a monomer to M- gives a species that contains one radical end and one anion end:
CH~ZCHX -CH*=CHX ---+ I .
:CH~CHXCH~~HX
anion end radical end
This species can add monomer from the two ends by different mechanisms. Two radical ends may dimerize, however, leaving a divalent anion to propagate. Propagation in anionic polymerization may be conventional or may be more complex, as in the elimination of CO, from N-carboxyanhydrides In contrast to [r~dical polyme~ization,the P (unsubstituted) carbon atom at the end of the growing chain is the site of addition of the next monomer As in cationic polymerization, termination is always unimolecular, usually by transfer The recombination of a chain with its counterion or the transfer of a hydrogen to give terminal unsaturation, frequent in cationic polymerization, is
90
unlikely in anionic mechanisms, as may be recognized by considering the small likelihood of' transferring H- when the counterion is Naf . Thus in anionic polymerization termination usually involves transfer, and the kinetic chain is bmken only if the new species is too weak to propagate.. This leads to the unique situation in which, by careful purification to eliminate all species to which transfer can occur, the termination step is effectively eliminated and the growing chains remain active indefinitely.. This case is described below under "Living" Polymers..
Kinetics
The kinetics of anionic polymerization may be illustrated by the polymerization of' styrene with potassium amide in liquid ammonia (Higginson 1952, Wooding 1952):
The usual kinetic analysis leads for high degree of' polymerization to
and
"Living" Polymers
Since the termination step usually involves transfer to some species not essential to the reaction, anionic polymerization with carefully purified reagents r?ay lead to systems in which termination is lacking. The resulting species, called "living" polymers (Henderson 1968; Szwarc 1968b), can be prepared, fbr example, by polymerizing styrene with sodium naphthalene. Kinetic analysis shows that the polymer can have an extremely narrow distribution of' molecular weight and for all practical purposes be essentially monodisperse; if initiation is rapid compared to propagation, the molecular-weight distribution is the Poisson function (Chapter 3C) for which Z,/i,, = 1 Ili,,The polymer can be "killed" by addition of' a terminating agent, for example, water, at the end of the reaction The living polymer technique pr.ovides an unique opportunity fbr the preparation of block copolymers (Chapter 5D) of' precisely defined composition (Henderson 1968; Fetters 1969; Hsieh,L981).
COORDINATION POLYMERIZATION
91
GENERAL REFERENCES
Overberger 1965; Lenz 1967; Szwxc 1968a; Morton 1969, 1977, 1983; Collins 1973, Exp. 9; Dart 19'74; Parry 1974; Bywater 1975; Elias 1977, Chapters 18.2 and 18 4; Bowden 1979, Chapter 2.2 4; Odian 1981, Chapter 5-3.
D. COORDINATION POLYMERIZATION
In the early 1950's a major new polymerization technique was discovered that led to the production of polymers with unusual stereospecific structures (Chapters 10A and 13) Although there were earlier indications of this type of polymerization (Schildkne'cht 1947), the field actually came into existence with the work of Ziegler (1955) and Natta (1955), who developed new polymerization catalysts with unique stereoregulating powers. Their Nobel Prize addresses (Ziegler 1964; Natta 1965) form excellent introductions to the field. Since the earliest days, a number of names have been applied to the polymerization technique using what have become known as Ziegler or Ziegler-Natta catalysts. Some have emphasized the stereoregulating features of certain of these polymerizations, but that is not an essential feature The term poly~nsertzonis still used, suggestive of chain propagation by the insertion of a monomer between the catalyst and the polymer chain to which it is attached, but the term coordznatzon polyrnerizatzon, used early to suggest the essential feature of a coordination complex with the catalyst, has remained and has been justified by mechanistic studies Coordination polymerization is far from simple in terms of mechanism, kinetics, or application The most impo~tantcatalysts are solids, leading to heterogeneous polymerization systems, but soluble catalyst systems are known. Additional complications to the study of coordination systems arise because the same catalysts, conditions, can initiate polymerization by cationic, anionic, or even under o t h e ~ free-radical mechanisms. Quite often coo~dinationpolymerization is car~iedout using a catalyst in the form of a slurry of small solid particles in an inert medium (a'fluzd-bed process) or a supported solid catalyst Heterogeneous polyme~ization is not essential for coordination polymerization and the development of stereospecificity, however, for sufficiently polar monomers
Ziegler-Natta Catalysts
The Ziegler-Natta or, often, Ziegler catalysts have the remarkable property of' polymerizing a wide variety of monomers to linear and stereoregular polymers. Ethylene is polymerized to a highly linex chain, in contrast to the products of' radical polymerization (Chapter 13). Polypropylene may be made in either the isotactic or syndiotactic form, but higher or-olefins yield only isotactic polymers Dienes such as butadiene and isoprene may be polymerized to products that are almost exclusively cis-1,4, trans-1,4, isotactic-1,2, 01. syndiotactic-1,2, depending on choice of catalyst and conditions. Many other examples exist.
IONIC. AND C:OORDINATION CHAIN (ADDITION) POLYMERIZATION
TABLE 4-2. Components of Typical Ziegler-Natta C.atalystsa
Organometallic Compound Triethyl aluminum Diethyl aluminum chloride Diethyl aluminum chloride Diethyl aluminum chlo~.ide Diethyl aluminum chloride Butyl lithium Butyl magnesium iodide Ethyl aluminum dichloride
"Henrici-Oliv& (1969)
Transition-Metal Salt Titanium tetrachloride Vanadium trichloride Triacetyl acetone vanadium Triacetyl acetone chromium complex Cobalt chloride-py~.idine Titanium tetrachloride Titanium trichloride Dichlorodicyclopentadienyl titanium
Ziegler catalysts are complexes formed by the interaction of alkyls of' metals of groups 1-111 in the periodic table with halides and other derivatives of transition metals of groups IV-VIII. Some typical compositions are listed in Table 4-2. Therc are literally hundreds of' combinations that are active in polymerization. Cur~ent thinking is that the transition metal, with its d orbitals available for both a and n bonding, is the catalyst center (Cossee 1964).. A coordination complex is for.med with the organometallic part of the system supplying some of' the ligands. At least one empty coordination site accommodates the growing polymer chain; this is indicated here by the square box:
Here a monometallic catalyst is shown; where an organometallic compound play$ a role, the metal atoms may be bonded by chlorine or alkyl bridges (Fig. 4-11 7hc titanium has been reduced fiom Ti(1V) to Ti(II1) and the chlorine bridge is a socalled three-center bond
Mechanisms. Because mechanistic studies are so much easier when c a r ~ i d ~ ( in a homogeneous system, much information has been obtained from experiment>, with soluble Ziegler-Natta catalysts. Henrici-Olivt (1981) studied the system hi:., cyclopentadienyl titanium(1V) dichloride (Cp,TiCl,),
f
COORDINATION POLYMERIZATION
FIG. 4-1.
Structure postulated for the metal halide-metal alkyl complex of a typical coordination catalyst
and an aluminum alkyl, usually dichloroethyl aluminum The active catalyst species was identified as
The transition metal was identified as the active site The growth step in the polymerization of ethylene by catalysts of this general type is presumed by most investigators to be that shown in Fig 4-2 The ethylene is coordinated to the free site at the transition metal and inserted between the metal and the alkyl group R or, later, the growing chain Both the course and the rate of the reaction can be influenced by the nature of the other ligands present A chain-transfer reaction exists in which a 6-hydrogen is transferred from the growing chain to the transition metal, leaving the chain with a vinyl end group This is a molecular-weight-determining step, though in the commercial production of Ziegler polyethylene it is unimportant, as the molecular weights are usually quite high. Only heterogeneous Ziegler--Natta catalysts yield isotactic polymers from nonpolar monomers such as a-olefins The nature of the catalyst surface appears to be important in imparting stereoregularity in these cases Propagation probably takes place at active sites on the crystal surface of the transition-metal compound, TiCI,, for example The sites are formed by the reaction of the group 11 1 soluble me-1 talorganic compound, fox example, Al(C,H,), or AI(C,H,),Cl, with the crystal surface, yielding species of the type shown in Fig 4-1. The chemical and crystal structuxes of the catalysts determine the orientation of the monomer as it is added to the growing chain. The driving force for isotactic propagation thus results from
FIG. 4-2.
Growth step postulated for the polyme~ization ethylene by a soluble Ziegler-Natta catalyst of' (Reprinted with pe~mission from Henrici-Olive 1981 Copyright 1981 by the American Chemical Society )
interaction between the monomer and the ligands of'the transition metal at the active site (Boor 1967). Several other models have been proposed, but none appear to be as widely applicable
Kinetics. Considering here the important but complex case of heterogeneous polymerization, we denote the sur.face of the transition,-metalcompound by a heavy bar, E,the metal alkyl by AR, and the monomer by M . The fbllowing reactions me pertinent: Adsorption of AR fiom solution to form the active site, and of M fiom solution,
Initiation,
C:OORDINAT'ION POLYMERIZATION
95
Propagation,
Transfer with monomer,
where Mx is polymer. Other transfer reactions are possible. Termination of' the kinetic chain can occur by several reactions, an example being that with monomer to form an inactive site:
Rate equations can be dqrived if the adsorption law is known If it is the usual Langmui~ isothe~m, which the equilib~ia Eq 4-1 1 ale attained and maintained in of rapidly and in competition, the fractions of the available su~.face cove~edby AR , and M, 0 and OM, are given by
The overall rate is given by
where [C*] is the concentration of growing chains Assuming that [C*] reaches the steady state, we can equate the equations f o initiation and termination: ~
from which, by elimination of [C*], we obtain
can Similarly, an expression for in be written. These equations are oversimplified fbr clarity by assuming only one transfer and one termination reaction. Tait (197.5)
96
IONIC AND C.OORDINATIONCHAIN (ADDITION) POLYMERIZATION
has extended and confirmed these kinetics for the polymerization of' 4-methylpentane-1 by vanadium trichloride and triisobutyl aluminum.
Other Coordination Polymerizations Polymerization With Supported Metal Oxide Catalysts. Catalysts consisting of metals or metal oxides adsorbed on or complexed with the surface of a solid carrier are of commercial interest for the low-pressure polymerization of olefins Although often used in fixed-bed processes, the catalysts can be slur~ied and employed in fluid-bed operation like the Ziegler-Natta catalysts Common catalyst compositions include oxides of chromium or molybdenum, or cobalt and nickel metals, supported on silica, alumina, titania, zirconia, or activated carbon The mechanisms and kinetics of the reactions are still largely unknown, although many proposals like those made for Ziegler-Natta polymerization have been presented "Alfin" Polymerization. The "Alfin" catalyst (Morton 1964; Reich 1966) consists of a suspension, in an inert solvent like pentane, of a mixture of an alkylenyl sodium compound (such as ally1 sodium), an alkoxide of a secondary alcohol (such as sodium isopropoxide), and an alkali halide (such as sodium chloride) The catalyst is highly specific for the polymerization of dienes into the 1,4-forms
GENERAL REFERENCES
Gaylord 1959; Natta 1966; Ketley 1967; Lenz 1967; Ledwith 1974b; Witt 1974; Elias 1977, Chapter 19; Vandenberg 197'7; Boor. 1979; Bowden 1979, Chapter 2 2.5; Odian 1981, Chapter 8-4
E.
RING-OPENING POLYMERIZATION
Polymers may be prepared by routes involving ring opening, which must be classed stoichiometrically as addition, since no small molecule is split off in the reaction Some typical examples are given in Table 4-3 The polymerization of these compounds has some aspects of both chain and step polymerization as far as kinetics and mechanism are concerned It resembles chain polymerization in that it proceeds by the addition of monomer (but rlever of larger units) to growing chain molecules However, the chain-initiating and subsequent addition steps may be similar and proceed at similar rates; if so, these are not chain reactions in the kinetic sense As in stepwise polymerization, the polymer molecules continue to increase in molecular weight throughout the reaction Most of these cyclic compounds polymerize by ~onic cham mechanisms in the presence of strong acids or bases when water and alcohols are excluded These
DISCUSSION QUESTIONS A N D PROBL.EMS
97
TABLE 4-3.
Typical Ring-scission Polymers
TY pe
Example
Lactam Cyclic ether Cyclic anhydride N-Carboxyanhydride
HY(CH2)xqO
L.-.--J
r(CH2)xO-1
L - - J
CO(CH~),CO
--I
- ) -
-[HN(CH2)xCO---],
--[(CH2)xO-]y -[CO(CH2)xCOO--]y -[COCHRNH-1,
COCHRNHCO
L--oJ
----*
+ CO2
polymerizations are often very rapid. They are of commercial interest in the polymerization of caprolactam, ethylene oxide, and 3,3-61s-(chloromethyl)oxetane These are true chain reactions
GENERAL REFERENCES
Furukawa 1963; Lenz 1967; Bowden 1979, Chapter 2.3; Odian 1981, Chapter 7..
DlSC:USSlON QUESTIONS AND PROBLEMS

1.
Unlike free-,radical polymerization, both cationic and anionic polymerizations show a marked dependence on the type of' solvent used. Discuss the causes and nature of' these eff'ects.
2. Write chemical equations for the major steps in the polymerization of the following: a Isobutylene by stannic chloride b Styrene by sodium naphthalene c Ethylene by titanium tetrachloride and diethyl aluminum chloride
3. Write kinetic equations, as far as they are known, for each of the polymerizations of Question 2
4. Discuss the influence of ion pairs on polymerization by ionic mechanisms. 5. The Nobel Prize was awarded to Karl Ziegler and Giulio Natta for the discovery and developinent of what are now called Ziegler-Natta catalysts.. Discuss the structure of these catalysts and the mechanisms of' the coordination polymerization in which they are used
98
IONIC AND C:OORDINATION CHAIN (ADDITION) POLYMERIZATION
6. Anionic polymerizations often do not have f'acile termination steps. Discuss the consequences of this fact and the resulting "living" polymers
7. Contrast termination mechanisms in cationic and free-radical polymerization, and indicate an easy way, based on kinetics, to distinguish between the two
BIBLIOGRAPHY
Baker I974 R Baker, "Carbonium Ions," Chapter 8 in A D Jenkins and A Ledwith, eds ,Reactrvrty, Mechanism and Structure In Polymer Chemistry, John Wiley & Sons, New York, 1974 Beaman 1948 Ralph G Beaman, "Anionic Chain Polymerization," J Am Chem Soc 70, 31 153118 (1948) Boor 1967 J Boor, Jr , "The Nature of the Active Site in the Ziegler-Type Catalyst," Macr;omrl Revs 2, 115-268 (1967) Boor 1979 John Boor, Jr , Zlegler-Natta Catalysts and Polymerizat~on,Academic Press, New York, 1979 Bowden 1979 M J Bowden, "Formation of Macromolecules," Chapter 2 in F A Bovey and F H Wlnslow, eds ,Macromolecules-Anlntroductlon to Polymer Sc~ence, Academic Press, New York, 1979 Bywater 1975 S Bywater, "Anionic Polyme~ization,"Chapter 2 in A D Jenkins, ed , Progress in Polymer Sc~ence, Vol 4, Pergamon Press, New York, 1975 Collins 1973 Edward A Collins, Jan Bare$, and Fred W bill me ye^, Jr , Experlmentr in Polymer Sclence, Wiley-Interscience, New York, 1973 Cossee 1964 P Cossee, "Ziegler-Natta Catalysis I Mechanism of Polymerization of a-Olefins with Ziegler-Natta Catalysts," I Catal 3, 80-88 (1964) Dart 1974 E C D a ~ t , "Caxbanions," Chapter 10 In A D Jenkins and A Ledwith, eds , Reactivity, Mechanism and Structure zn Polymer Chemistry, John Wlley & Sons, New York, 1974 Eastham 1965 A M Eastham, "Cationic Polymerization," pp 35-59 in He~man Maxk, Norman F G Gaylord, and Norbert M Bikales, eds , Encyclopedia of Polymer Science and Technology. Vol 3, Wiley -Interscience, New York, 1965 Ehas I977 Hans-Georg Elias, Macromolecules 2 Syntherlr and Maferialr (translated by John W Stafford), Plenum Press, New York, 1977 Fetters 1969 L J Fetters, "Synthesis of Block Polymers by Homogeneous Anionic Polymerization," J Polym Scl C26, 1-35 (1969) Furukawa 1963 Junji Furukawa and Takeo Saegusa, Polymerizat~onof Aldehydes and Oxides, WileyInterscience, New York, 1963 Gaylord 1959 Norman G Gaylord and Herman F Mark, Linear andStereoregular Addition Polymers, Interscience, New York, 1959 Henderson 1968 1 F. Henderson and M . Szwarc, "The Use of Living Polymers in the P~eparationof Polymer Structures of Controlled Architecture," Macromol. Rev 3, 317-401 (1968) Henricr-OhvP 1969 Gisela Henrici-Olivt and Salvado~ Olivt, Polymerisation-KaraEyse, Kmetlk, Me chanismen (in German), Verlag Chemie, Weinheim, West Germany, 1969 Henricl-Ohvk 1981 G Henrici-Olivt and S Olivt, "Mechanism for Ziegler-Natta Catalysis," Chem tech 11, 746-752 (1981) Hlggrnson 1952 W C E Higginson and N S Woodlng, "Anion~c Polyme~isation Part I The Polymerisation of Styrene in Liquid Ammonia Solution Catalysed by Potassium Amide," I Chem SOC 1952, 760--774 (1952) Hszeh 1981 Henry L Hsieh, Ralph C Farrar, and Kishore Udipi, "Anionic Polymerization: Some Commercial Applications," Chemterh 11, 626-633 (1981)
BIBLIOGRAPHY
99
Kennedy I977 J P Kennedy, S Y Huang, and S C Feinberg, "Cat~onicPolymer~zation with Boron Hal~des I11 BCI, Coinitiator for Olefin Polymerization," J Polym Scr Polym Chem Ed 15, 2801-2819 (1977) Kennedy 1982a Joseph P Kennedy and Ernest Marechal, Carbocatronic Polymerrzation, John Wiley & Sons, New York 1982 Kennedy 19826 Joseph P Kennedy, Tibor Kelen, and Rrenc Tudos, "Quasiliving Carbocationic Polymerizatron I: Classification of Living Polymerizations in Carbocationic Systems," J Macromol Sci Chem A18, 1189-1207 (1982); R Faust, A Fehervari, and Joseph P Kennedy, "I1 The Discovery: The a-Methylstyrene System," J Macromol Scr Chem A18, 1209.- 1228 (1982); and following papers Ketley 1967 A D Ketley, ed , The Stereochemistry of Macromolecules, Vol 1, Marcel Dekker, New York, 1967 Ledwrth 1974a A Ledwith and D C Shemngton, "Reactivity and Mechanism In CatIonic Polymerizat~on,"Chapter 9 in A D Jenkins and A Ledwith, eds ,Reactrvity, Mechanism and Structure in Polymer Chemistry, John Wiley & Sons, New York, 1974 Ledwrth 19746 A Ledwlth and D C Shenington, "Reactivity and Mechanism in Polymerization by Complex Organometalllc Derivatlves," Chapter 12 in A D Jenkins and A Ledwith, eds ,Reactivrty, Mechanism and Structure rn Polymer Chemistry, John Wiley & Sons, New York, 1974 Lenz I967 Robert W Lenz, Organrc Chemrstry of Synthetic High Polymers, Wiley.Interscience, New York, 1967 Morton 1964 Avery A Morton, "Alfin Catalysts," pp 629-638 in Herman F Mark, Norman G Gaylord, and N0rber.t M Bikales, eds , Encyclopedia of Polymer Science and Technology, Vol 1, Wiley-Interscience, New York, 1964 Morton 1969 Maurice Morton, "Anionic Polymerization," Chapter 5 in George E Ham, ed , Vinyl Polymerization, Part I , Vol 2, Marcel Dekker, New York, 1969 Morton 1977 Maurice Morton and Lewis J Fetters, "Anionic Polymerizations and Block Copolymers," Chapte~ in Calvin E Schildknecht, ed , w ~ t h 9 Irving Skeist, Polymerlzatzon Processes, WileyInterscience, New York, 1977 Morton I983 Maurice Morton, Anronic Polymerrzatron Principles and Practice, Academic Press, New York, 1983 Natta 1955 G Natta, Piero Pino, Paolo Corradini, Ferdinand0 Danusso, Enrico Mantica, Giorgio Mazzanti, and Giovanni Moranglio, "Crystalline High Polymers of a-Olefins," J Am Chem Soc 77, 1'708-1710 (1955) Narta 1965 Giuiio Natta, "Macromolecular Chemistry: From the Stereospecific Polymerizat~on the to Symmetric Autocatalytic Synthesis of Molecules," Sc[ence 147, 261-272 (1965) Natta 1966 Giulio Natta and Umberto Giamini, "Coordinate Polymerization," pp 137-150 in Herman F Mark, Norman G Gaylord, and Norbert M Bikales, eds , Encyclopedia of Polymer Scrence and Technology, Vol 4, Wrley-Interscience, New York, 1966 Odran 1981 George Od~an, Prznciples of Polymerrzatron, 2nd ed , John Wrley & Sons, New York, 1981 Olah 1972 George A Olah, "The General Concept and Structure of Carbocatrons Based on Drffer ent~atlon Trivalent ('Classrcal') Carben~um of Ions from Three-Center Bound Penta or Tetracoordrnated ('Nonclass~cal') Carbonmm Ions The Role of Carbocatlons rn Electroph~l~c Reactrons," J Am Chem Soc 94, 808 820 (1972) Overberger 1965 C G Overberger, J E Mulvaney, and Arthur M Schiller, "Anionic Polymerization," pp 95-137 in Herman F Mark, Norman G Gaylord, and Norbert M Bikales, eds , Encyclopedra oj Polymer Scrence and Technology. Vol 2, Wiley-Interscience, New York, 1965 Parry I974 A Pany, "Anionic Polymerization," Chapter 11 in A D Jenkins and A Ledwith, eds , Reacrrvzty, Mechanrsm and Structure in Polymer Chemistry, John Wiley & Sons, New York, 1974 Plesch 1963 P H Plesch, ed , The Chemistry of Cationrc Polymerization. Macmillan, New York, 1963
100
Plesch 1968 P H Plesch, "Cationic Polymerization," pp 137-188 in 3 C Robb and F W Pc:~l,t.~, eds , Progress tn Hlgh Polymers, Vol 2, CRC Press, Cleveland, Ohio, 1968 Rerch 1966 Leo Reich and A Schindler, Polymerrzat~onby Organomelallrc Comporn~ds,\\'ilc! Interscience, New York, 1966 Russell 1977 Kenneth E Russell and Geoffrey J Wilson, "Cationic Polyme~izations," Chapter 10 in Calvin E Schildknecht, ed , with Irving Skeist, Polymer~zationProcesses, Wiley-Interscicrr~,~, New York, 1977 Schildknecht 1947 C E Schildknecht, A 0 Zoss, and Clyde McKinley, "Vinyl Alkyl Ethers." ir1,j Eng Chem 39, 180-186 (1947) Siwarc 1968a Michael Szwarc, Carbanlons, Lzvlng Polymers and Electron-Transfer Procerres. J o l i ~ ~ Wiley & Sons, New York, 1968 Szwarc 19686 M Szwarc, " 'Living' Polymers," pp 303-325 in Herman F Mark. Norman C; Gaylord, and Norbert M Bikales, eds , Encyclopedia of Polymer Sclence and Technolog\. Yo\ 8, Wiley-Interscience, New York, 1968 Tart 1975 Peter J T Tait, "Ziegler-Nstta Polymerization--Model, Mechanism, and Kinedcs.' Chcnr tech 5, 688-692 (1975) Vandenberg 1977 Edw~n Vandenberg and Ben C Repka, '-Ziegler-Type Polymenzatiom ' Ch.iprsr 3 11 in Calvln E Schildknecht, ed ,with Irving Skeist, Pol\merrzatron Processes, Wlley htenciencc, New York, 1977 Wrtt 1974 D R Witt, "Reactivity and Mechanism with Chromium Ox~de Poiyme~lzation Catal\\l\." Chapte~13 In A D Jenktns and A Ledwith, eds , Reactn lty, Mechanlcm and Strucrurr m Pohtctc*~ Chemrstry, John Wiley & Sons, New York, 1974 Woodrng 1952 N S Wooding and W C E Higginson, "Anronic Polyrne~~sation I IDI Tlis ~ Polymensation of Styrene in Liquid Ammonia Catalysed by Potasslum," / Chem 5CIE 1952, 1178-1 180 Zregler 1955 Karl Z~egler,E Holzkamp H Breil, and H Mart~n, "Polymnzation of E ? \ l e n r arid Other Olefins" (in German), Angew Chem 67, 426 (1955) Zlegler 1964 Karl Ziegler, "Consequences and Development of an Invention" (in Gemzc.3. 4n:e,! Chem 76, 545-553 (1964)
CHAPTER
A.
KINETICS OF COPOLYMERIZATION
Early Experiments in Copolymerization
Although the polymerization of organic compounds has been known for over 100 years, the simultaneous polymerization (copolymerzzatzon)of two or more monomers was not investigated until about 1911, when copolymers of olefins and diolefins were found to have rubbery properties and were more useful than homopolymers made from single monomers In the 1930's it was found that monomers differed markedly in their tendencies to enter into copolymers. Staudinger (1939) fractionated a vinyl chloride-vinyl acetate copolymer made from a mixture of equimolar quantities of the two monomers He found no polymer containing equal amounts of each monomer but, instead, found vinyl chloride: vinyl acetate rdtios of 9: 3, 7: 3, 5: 3, and 5: 7 among the fract~ons At about the same time, acrylic esters were found to enter copolymers wlth vinyl chloride faster than did the second monomer The first polymer formed was rich in the acrylate; later, as the amount of acrylate in the monomer system was depleted, the polymer became richer in vinyl chloride. Maleic anhydride and other monomers that homopolymerize with great difficulty were found to copolymerize readily with such monomers as styrene and vinyl chloride. Pairs of monomers, such as st~lbenemaleic anhydride and isobutylene-fumaric ester, in which neither monomer poiymerizes alone, readily gave high-molecular- weight copolymers in which the monomers appeared in a 1 : 1 ratio, no matter which was in excess in the monomer feed More recently, the ability to produce polymers contaming long sequences of two or more different monomers (block and graft copolymers) has led to new products with unique and valuable properties
The kinetic scheme for chain-reaction copolymerization is developed fbr radical reactions; extension to ionic systems is straightfbrwasd..
The C.opolymer Equation
In 1936 Dostal made the first attack on the mechanism of copolymerization by assuming that the rate of addition of' monomer to a growing free radical depends only on the nature of' the end group on the radical chain. Thus monomers M I and M2 lead to radicals of types MI. and M2. . There ase four possible ways in which monomer can add:
-
Reaction
MI,. + M I .+ M2ML. .+
MI MI. M2.M2. MI -+ M I . M2 --4 M2,
Rate
~II[MI.I[MI] k12[M1,1[M21 k21[M2.][M11 k22[M~.I[M21
(5-1)
The kinetics of'copolyme~ization was more fully elucidated in 1944 by Alfrey, Mayo, Simha, and Wall. To Dostal's reaction scheme they added the assumption of'the steady state applied to each radical type separately, that is, the concentrations of M I - and M2,.must each remain constant It follows that the rate of conversion of' MI - to M2- must equal that of conversion of' M2. to MI.:
The rates of' disappearance of the two types of monomer ase given by
By defining r., = klllk12 ri = k221k21 combining Eqs. 5-2 and 5-3, it can and and b shown that the composition of' copolymer being fbrmed at any instant is given e
by
This is known as the copolymer equation, it has been verified by many experimental investigations of copolyme~composition
103
Monomer Reactivity Ratios
The monomer reactivity ratios rl and r, are the ratios of the rate constant fbr a , given radical adding its own monomer to the rate constant for its adding the other monomer. Thus rl > 1 means that the radical M I .prefers to add M I ;r, < 1 means that it prefers to add M,. In the system styrene (MI)-.methyl methacrylate (M,), for example, r, = 0.52 and r2 = 0..46; each radical adds the other monomer about twice as fast as its own. Since the rate constants for initiation and termination do not appea in Eq. 5-4, the composition of'the copolymer is independent of overall reaction rate and initiator concentration.. The reactivity ratios are unaffected in most cases by the presence of inhibitors, chain transftr agents, or solvents. Even in heterogeneous systems they remain unchanged, unless the availability of the monomers is altered by their distribution between phases A change from a free radical to an ionic mechanism, however, changes rl and ri markedly. A few typical values of monomer reactivity ratios are given in Table 5-1.. An extensive compilation is given by Young ( 1 9 7 5 ~ ) .
Types of Copolymerization Ideal. A copolymer system is said to be ideal when the two radicals show the = same preference for adding one of the monomers over the other: k l l l k I 2 k21/k22, or r = I , 01 r = 1 In this case the end group on the growing chain has no influence on the rate of addition, and the two types of units are arranged at random along the chain in relative amounts determined by the composition of the feed and
TABLE 5-1. Typical Monomer Reactivity Ratiosa Monomer 1

Acrylon~tr~le
,
Monomer 2
1,3-Butadlene Methyl methacrylate Styrene Vinyl acetate Vinyl chloride Methyl methacrylate Styrene Vinyl chloride Styrene Vinyl acetate Vinyl chloride Vinyl acetate Vinyl chloride Vinyl chloride
rI
0 02 0 15 0 04 42 27 0 75 1 35 88 0 46 20 10 55 17 0 23
12
T("C)
-
1,3-Butadiene
Methyl methacrylate
Styrene Vinyl acetate

"Young (1975~)
03 1 22 0 40 0 05 0 04 0 25 0 58 0 035 0 52 0 015 0 1 0 01 0 02 1 68
40 80 60 50 60 90 50 50 60 60 68 60 60 60
104
COPOLYMERIZATION
the relative reactivities of' the two monomers. The copolymer equation reduces to d[M,lld[M,l = ~ , [ M I I I [ M , I
Alternating. Here each radical prefers to react exclusive1,y with the other monomer: r., = r., = 0 . The monomers alternate regularly along the chain, regardless of the composition of' the monomer feed. The copolymer equation simplifies to d[Ml]ld[M2] 1 = Most actual cases lie between the ideal and the alternating systems: 0 < r ,r2 < 1 A third possibility, with both r , and r, greater than unity, cor~esponds the tendency to to form block copolymers (Section D) instantaneous Composition of Feed and Polymer
Let F , and F2 be the mole fractions of' monomers 1 and 2 in the polymer being formed at any instant:
If f , and
f2
similarly represent mole fractions in the monomer feed,
the copolymer equation can be written as
It is apparent that in general F , does not equal f , , and both J l and F , change as the polymerization proceeds The polymer obtained over a finite range of conversion consists of' a continuous distribution of varying composition, described in Section B * Equation 5-7 may be used to calculate curves of' feed versus instantaneous polymer composition for various monomer reactivity ratios.. Such curves for a series of ideal copolymerizations (r.,r., = 1) are shown in Fig. 5.-1. Except fbr pairs of' monomers having very similar reactivities, only a small range of' feeds gives copolymers containing appreciable amounts of' both components Several curves fbr nonideal cases are shown in Fig 5-2 These curves show the effect of increasing tendency toward alternation. As alternation increases, more and more feeds yield a copolymer containing a good deal of each component. This tendency makes practical the preparation of' many important copolymers If one of' the monomers is very much more reactive than the other, the first polymer fbrmed contains mostly the more reactive monomer. Later in the poly-
r1
FIG. 5-1.
Instantaneous composition of copolymer (mole fraction F,) as a function of monomer composition (mole fraction f ,) for ideal copolymers with the values of r, = I l b indicated
.-
FIG. 5-2.
Instantaneous composition of copolymer F , as a function of monomer composition the values of the reactivity ratios r , / r , indicated
for
merization this monomer is largely used up, and the last polymer formed consists mainly of' the less reactive monomer. Styrene and vinyl acetate form such a system For cases in which both r, and r2 are less (or greater) than unity, the curves of' Fig. 5-2 cross the line representing Fl = f , . At the point of intersection polymerization proceeds without a change in the composition of' feed or polymer.. This is known as azeotropic copolymerization. Solution of' Eq. 5-4 with d[Ml]ld[M2] [MI]/ = [M,] gives the critical composition fbr the azeotrope:
or, from Eq. 5-7,
Evaluation of Monomer Reactivity Ratios

The usual experimental determination of r, and r2 involves polymerizing to low conversion fbr a variety of' feed compositions..The polymers are isolated and their compositions measured. One of' several methods of' analysis of' the data may be used: Direct curve fitting on polymer:-monomer composition plots. This is a poor method, since the composition curve is rather insensitive to small changes in r, and r2 b. Plots of' r, versus r,. The copolymer equation may be solved for one of' the reactivity ratios:
a.
c.
Each experiment with a given feed gives a straight line; the intersection of several of these allows the evaluation of' r, and r , (Fig. 5-3) If' the experimental errors are high, the lines may not intersect in a single point; the region within which the intersections occur gives infbrmation about the precision of' the experimental results.. Plots of' F, versus f,.. Equation 5-7 can be rear~angedto
FIG. 5-3. Evaluation of monomer reactivity ratios by graphic solution of the copolymer equation (Burnett 1954, data of Mayo 1944)
This is the equation of a straight line with slope r , and intercept r, Each experiment gives one point on the line The least-squares treatment of a series of such points gives the best values of r, and r , in a straightforward way d. Analysis of data giving the copolymer composition as a function of conversion, using the integrated foim of the copolymer equation given in Section B (Meyer 1966)
Rate of Copolymerization
The discussion so far has considered only the relative rates of' the four possible propagation steps in a binary system. The overall rate of'copolymerization depends also on the rates of initiation and termination If the initiator releases radicals that have a high efficiency of combination with each monomer, the two types of' initiation step need not be considered separately.. Steady-state conditions apply both to the total radical concentration and to the separate concentrations of the two radicals. Three types of termination must be considered, involving all possible pairs of radical types The steady-state condition for all the radicals requires that the initiation and termination rates be equal:
108
COPOLYMERIZATION
Here ktll and k12* the termination sate constants fbr pairs of like radicals, and are k,,, is the constant for the cross-reaction If this condition is applied to the equation for the overall polymerization rate,
where
and
as The 6 values are simply related to quantities of the type kp2/2k,, seen in Chaffer 3; compares the cross-termination rate constant to the geometric mean of termination rate constants for like pairs of radicals A value of greater tha: 1 means that cross termination is favored, and conversely If r , and r2 are known, measurement of the rate of copolymerization allov/* 4 to be determined A typical result is shown in Fig 5-4 The value of grcz*zr
FIG. 5-4
Mole fract~onof styrene Copolymerization rate data for the system styrene-methyl methacrylate zr @j'C f7A~-.- '"; 1949) The broken line assumes d, = > I , the s o l ~ d 11ne 1s calculated for r6 = I ?
109
than unity is typical of' most copolymerizations; indeed, cross termination is markedly favored, except for nearly ideal copolymerizations. It is likely that differences in polarity are responsible fbr the preference fbr cross termination, as well as for deviations from ideal copolymerization.. As a result, the copolymerization of pairs of monomers of quite different polarities is likely to be slower than the mean of' the sates for the separate homopolymerizations. The above treatment assumes that the termination process is controlled by the chemical nature of the growing-chain radical ends It is known, however, that termination in radical polymerization is generally diffusion controlled. It is often useful to express the rate equation in terms of a single termination constant k, (Atherton 1962, North 1963):
'
Ideally, kr might be expected to be the mole-fraction-weighted average of the termination rate constants for the two homopolymerizations, but this appears to be an oversimplification, possibly due to penultimate effects (see below)
Extensions of Copolymerization Kinetics Multicomponent Copolymerization. The mathematics of the copolymer equation has been extended to the cases of three or more monomers (Ham 1964), but the quantitative treatment is quite complex; for example, three monomers require the specification of nine propagation and six termination reactions, and six reactivity ratios Some simplification results from rather liberal assumptions about the steady state and the probabilities of finding certain sequences of monomers in the resulting polymer The predicted results are in good agreement with experiment Penultimate and Remote-Unit Effects. Ham (1964) and co-wo~kershave developed the early suggestion of Merz (1946), that units before the last may affect the reactivity of a chain radical This is often the case, the effect being particularly pronounced with monomers of type CHX=CHX, where X is a polar substituent It has been detected in copolymers of styrene with maleic anhydride and fumaronitrile. When the penultimate monomer unit does have some influence on the subsequent addition of monomer, four reactivity ratios, corresponding to the four possible types of ~adicals,must be considered Effects of units still farther back along the chain have been detected experimentally in some cases Depropagation During Copolymerization. If one or both of the monomers has a tendency to depropagate during a copolymerization, an additional complication is introduced (Lowry 1960). This deviation from the simple treatment can be detected by the dependence of the copolymer composition on the absolute concentration of the monomers, even though their relative concentration is invariant
110
COPOLYMERIZATION
Complex Participation. The presence of' complexes between comonomers, fbr example, chasge-transfer complexes (Section B) also can cause deviations from normal copolymerization kinetics, often accompanied by an enhanced tendency toward alternation. Four additional propagation steps and six reactivity ratios ase required in the analysis. As in the case of depropagation, copolymer composition is predicted to vary with temperature and monomer concentration. Without the availability of complete and accurate experimental data it is often difficult to assign the cause of deviations from simple copolymer kinetics to one or another of' the above models.
GENERAL REFERENCES
Ham 1964, 1966; Bamfbrd 1976; Collins 1973, Exp. 5; Elias 1977, Chapter 22; Odian 1981, Chapter 6.
B.
COMPOSITION O COPOLYMERS F
The copolymer equation (Eq 5-4) predicts the average composition of the polymer formed at any instant in the polymerization. Not only may statistical fluctuations of composition about this average occur, but also polymer fbrmed during a finite interval may contain a range of' compositions resulting from variation of' feed composition.. In this case, the overall polymer composition can be calculated by integrating the copolymer equation
Statistical Fluctuation of Instantaneous Copolymer Composition
Stockmayex (1945) showed that the spread in copolymer composition due to statistical fluctuations is small in practical cases.. Alternating copolymers give the minimum spread in composition; even in ideal copolymers, where the spread is greater and is a function of the degree of' polymerization, the effect is not large. For a copolymer containing on the average 50% of' each monomer, only 12% of , the polymer molecules contain less than 43% of either monomer at ? = 100, while only 12% ofthe molecules contain less than 49.3% of either monomer at 2, = 10,000. The distribution of' compositions follows a Gaussian curve. This small spread in composition is neglected in the discussion to follow. The probability of occurrence of sequences of n similar monomer units can be calculated statistically for an ideal copolymer (Alf'rey 1952): The number N(n) of such sequences is
COMPOSIT'ION OF COPOLYMERS
111
where p is the probability of' addition of' that monomer,
Except for a monomer present in large excess, the sequences are always short. Any tendency toward alternation clearly leads to shorter sequences than those in an ideal copolymer.
Integration of the Copolymer Equation

The direct integration of the copolymer equation gives a result that is not convenient for calculations The most convenient method (Skeist 1946) for calculating copolymer composition and distribution involves the use of Eq 5-7. For a system in which F, > fl, when dM moles of monomer have polymerized, the polymer contains F, dM moles of monomer 1, and the feed contains ( f , - dj,) (M - dM) moles of monomer 1 For a material balance,
Combining this with Eq. 5-.7 gives
by which can easily be integ~ated numerical or graphic means to give the desired composition of polymer as a function of conversion One convenient closed-form result of this integration is (Meyer 1965, 1967)
log
(z)
E
FT
I2
log
(A)+ -log ( 2 1 2.
1
1-
71
(fi)~
where
= (1
- r2)/(2
r1
+ r2).
Variation of Copolymer Composition With C.onversion Ideal System. The system styrene (MI)-2-.vinylthiophene (M,) is nearly ideal, with r., = 0.35and r, = 3.10. The variation of the composition of'polymer fbrmed at any instant in this system with conversion and initial feed composition is shown in Fig. 5-5.. The greater reactivity of the vinylthiophene causes it to be depleted until at the last only styrene remains, and pure polystyrene is the final product
FIG. 5-5.
Variation in instantaneous composition of copolymer with initial feed and conve~sionfor the system styrene-2.-vinylthiophene (Mayo 1950)
fbrmed The overall polymer composition varies less with conversion and ultimately approaches the initial feed composition as required fbr a material balance. The polymer composition distribution for 100% conversion for several initial feeds is shown in the bar graphs in Fig. 5-6. Each block represents a 5% interval in copolymer composition. The U-shaped distribution curves arise whenever the monomer reactivity ratios are less than 0 . 5 and greater than 2, respectively. They indicate that appreciable quantities of two distinct compositions predominate, with little material of intermediate composition
Alternating System. The system styrene (M,wiethyl fumatate (M,) with Y , = 0 30
and r, = 0.07 has a strong tendency to alternate and an azeotropic composition at 57 mol.% styrene.. The diagrams for this copolymer aIe more complex than those for the previous system Feeds near the azeotropic composition remain almost unchanged up to rather high conversion, but eventually all feeds containing more than 57 mol.% styrene drift toward pure styrene, and those containing less styrene than the azeotropic amount yield pure diethyl fumarate. Thus (Fig. 5-7) pure polystyrene is formed at the last from all initial fteds higher in styrene than the azeotropic, and pure poly(diethy1 fumarate) fiom all initial feeds containing less than 57 mol% styrene. The overall polymer composition shows markedly less change with conversion for all feeds and is of course invariant at the azeotsopic composition
Attainment of Homogeneity in C:opolymers. It is often desirable to produce a polymer that is homogeneous in composition rather than one having a broad distribution of compositions. The fbregoing discussion suggests two ways of doing this: (a) The reaction can be stopped short of complete conversion.. For example, the styrene.-diethyl fumarate copolymer from feed containing 40% styrene, when
0 Mole per cent styrene in copolymer FIG. 5-6. Dlstnbutlon of copolymer cornposlt~onat 100% conversion, for the lnd~catedvalues of lnltlal mole percent styrene In the feed, for the system styrene-2-vlnylthlophene (Burnett 1954, after Mayo 1950)
FIG. 5-7. Variation in instantaneous cornpositlon of copolymer with initial feed and conversion for the system styrene--dlethyl fumarate (Mayo 1950)
114
COPOLYMERIZATION
stopped at 75% conversion, contains only polymer having between 44 and 52% styrene. (b) The polymerization can be started with a feed giving the desired polymer and the feed composition maintained by adding an increment of' more reactive monomer as the polymerization proceeds Both methods find commercial use
GENERAL REFERENCES
Alf'rey 19.52; Odian 1981, Chapter 6
C.
MECHANISMS OF COPOLYMERIZATION
Free-Radical Copolymerization
The order of reactivity of' monomers toward free radicals not only is a function of the reactivity of' the monomers, but also depends on the nature of' the attacking radical. This is illustrated by the tendency of' many monomers to alternate in a copolymer chain. The two factors, general reactivity and alternating tendency, are predominant in determining the behavior of monomers in copolymerization The reactivity of monomers and radicals in copolymerization is determined by the nature of'the substituents on the double bond of'the monomer.. These substituents influence reactivity in three ways: They may activate the double bond, making the monomer more reactive; they may stabilize the resulting radical by resonance; or. they may provide steric hindrance at the reaction site
Reactivity of Monomers. The relative reactivities of' monomets to a reference radical can be derived f'tom the monomer reactivity ratio. The inverse of this ratio is the rate of reaction of' the reference tadical with another monomer, relative to that with its own monomer. If the latter rate is taken as unity, relative monomer reactivities can be examined. A few such rates are listed in Table 5-2; more extensive tabulations are given in the General References. As a different reference point is
TABLE 5-2. Relative Reactivities of Monomers to Reference Radicals at 60C
Reference Radical Monomer Styrene Methyl methacrylate Acrylonitrile Vlnylidene chloride Vinyl chloride Vinyl acetate Styrene
(1 0) 19 25 54 0 059 0 019
Methyl Methacrylate
22 (1 0) 0 82 0 39 0 10 0 05
Acrylonitrile
25 67 (1 0) 11 0 37 0 24
Vinyl Chloride
50 10 25 5 (1 0) 0 59
Vinyl Acetate
100 67
20 10 44 (1 01
MECHANISMS OF COPOLYMERlZATlON
115
taken for each radical, values in different columns are not compa~able,although ratios of values can be compared from column to column. Although there are some exceptions, the effectiveness of substituents in enhancing the reactivity of the monomer lies in about the order
The effect of' a second substituent on the same carbon atom is usually additive. This order of' reactivity corresponds to the resonance stabilization of' the radical formed after the addition In the case of styrene the radical can resonate among forms of the type
The radical is stabilized with a Iesonance energy of about 80 Wimole At the other extreme, substituents that have no double bonds conjugated with the ethylenic double bond give radicals having very low resonance energies (5-15 kJ/mole), since only polar or unbonded forms can contribute to the resonance Substituents that stabilize the product radical tend also to stabilize the monomer, but the amount of stabilization is much smaller than for the radical; in styrene the monomer is stabilized to the extent of about 12 kJ/mole Thus the stabilizing effect of a substituent on the product radical is compensated for only to a limited extent by stabilization of the monomer
Resonance Stabilization. Resonance stabilization of monomers and radicals can be considered in terms of a potential energy diagram of the type shown in Fig 5-8 The energy changes shown Iepresent situations before and after each of four possible reactions in which a resonance-stabilized (M,) or unstabilized (M) monomer adds to a stabilized (R;) or unstabilized (R ) radical The set of four curves labeled with the four reaction possibilities represents the combined energy of the growingchain radical and the monomer before reaction, while the two curves labeled "new radical" represent the potential energy of the new bond formed by their combination Activation energies are represented by solid arrows, and heats of reaction by broken arrows If the entropies of activation are all about the same (a good assumption in the absence of steric hindrance), the order of reaction rates is predicted to be the opposite of that of the activation energies:
This series summarizes much of the rate information available for both copolymerizations and homopolymerizations
Radical monomer (before bond formation)
w
C
m .-. +
+ .
'
1
I
I
--
Y-L-Separation
FIG. 5-8.
Potential eneIgy of a radical-monome1 pair as a function of ~eactioncoordinate; see text fo~ further explanat~on (Walling 1957)
Reactivity of Radicals. The reactivity of' radicals to a reference monomer can be derived by combining the absolute propagation rate constants k , , with reactivity ratios to obtain the rate constants k,, for radical 1 adding monome1 2 . Typical data are listed in Table 5-3 As expected, resonance stabilization depresses the reactivity of radicals so that their order of reactivity is the reverse of that of their monomers; styrene is one of the least reactive radicaIs and vinyl acetate one of'the most reactive. The effect of a substituent in depressing the activity of' a radical is much greater than its effect in enhancing the activity of'the monomer. For example, the styrene radical is about 1000 times less reactive than the vinyl acetate radical to a given monomer if' alternation effects are disregarded, but styrene monomer is only about 50 times more reactive than vinyl acetate to a given radical Steric Effects. The effect of' steric hindrance in reducing reactivity may be demonstrated by comparing the reactivities of 1,l- and 1,2-disubstituted olefins to reference radicals. As indicated in the preceding discussion, the addition of a second I-substituent usually increases reactivity 3- to 10-fold; however, the same substituent in the 2-position usually decreases reactivity 2- to 20-fold The extent of' reduction in reactivity depends on energy differences between cis and trans forms and on the possibilities :br resonance Charge-Transfer Complexes. Two monomers with widely different polarities, for example, an electron donor and an electron acceptor, can fbrm a charge-transfer complex that may act as a single species in polymerization, leading to 1 : 1 alternating
MEC:HANISMS OF COPOLYMERIZATION
117
TABLE 5-3. Radical-Monomer Propagation Rate Constants at 60C (LiterIMole-sec)

p -
---
Radical Monomer Butadiene Styrene Methyl methacrylate Methyl acrylate Vinyl chloride Vinyl acetate
_____
Methyl Methyl Vinyl Vinyl Butadiene Styrene Methacrylate Acrylate Chloride Acetate
-
100 74 134 132 11
-. ---
--
250 145 278 206 8 26
2,820 1,520 705 370 70 35
42,000 14,000 4,100 2,090 230 520
350,000 600,000 230,000 123,000 150,000 200,000 23,000 12,300 10,000 7,300 2,300
polymers The formation of such complexes is confirmed by spectroscopic evidence Typical electron acceptors are methyl methacrylate, methyl acrylate, and acrylonitrile; typical donors for these complexes are vinyl chloride, vinyl acetate, propylene, and isobutylene In many instances the presence of a Lewis acid enhances the tendency for complex fo~mation It is in this way that monomers such as isobutylene and propylene, which do not homopolymerize because of degradative chain transfer, can participate in free-radical polymerization
Polarity and Alternation. By examining the tendency for alternation as given by the product r 1 r 2 , it is possible to tabulate monomers in a series arranged so that two monomers farther apart in the series have a greater tendency to alternate Such an arrangement is shown in Table 5-4 Values of r,r2generally decrease as monomers farther apart in the table are compared There are some exceptions, suggesting specific interactions, probably steric in natuIe For example, vinyl acetate alternates more than styrene with vinylidene chloride, and less than styrene with acrylonitrile There is little doubt that alternation is primarily due to the polarity of the double bond in the monomer: The order of monomers in Table 5-4 closely parallels the order of the tendency of substituents around the double bond to donate electrons to the bond (hydrocarbon, acetoxy) or to withdraw them (carbonyl, cyano) As might be expected, the tendency toward alternation (if not too great) parallels the tendency toward decreased reactivity induced by substitution of electron-with drawing groups at the double bond T'he Alfrey-Price Treatment (Alfrey 1964).. Alfrey and Price (Alfrey 1947) attempted to express the factors of' general reactivity and polarity quantitatively. They wrote the rate constants for copolymerization in the form
where exp denotes the exponential, P, is the general reactivity of the radical MI-, (2, is the reactivity of' the monomer M7, and e, and e, are proportional to the electrostatic interaction of' the permanent charges on the substituents in polarizing
118
COPOLYMERIZAT'ION
TABLE 5-4. Product of Reactivity Ratios With Monomers Arranged in Order of Alternating Tendency
Vinyl acetate
-
Butadiene
0 98 O 31 O l9 <01 00006
0 55 0 39 0 30 06 0 21
Styrene
O 34
O 24 016 0016 096 011
Vinyl chloride Methy1 methacrylate

O 0 18
Vinylidene chloride
0 35
Acrylonitrile Diethyl fumasate
the double bond. This fbrm is analogous to Hammett's (1940) equation for. the effects of' nuclear substituents on the reactivity of' aromatic compounds. From the definitions of' the monomer reactivity ratios it fbllows that
and
r1r2= exp[- (el - e2)2]

The Alfrey-Price equation assumes that the same value of e applies to both monomer and radical This assumption is justified only by the success of the @e scheme in interpreting relative reactivities. It should be possible to assign Q and e values to a series of monomers from the data on a few copolymerization experiments and compute values of r , and r, for all the possible arrangements of the monomers in pairs (Price 1948). Unfortunately, this practical application of the scheme is hampered by large uncertainties in the values of Q and e that can be assigned with the present experimental erTor The scheme is not entirely sound from the theoretical standpoint and is best regarded as semiempirical In a sense, it represents the transcription of the reactivity and polarity series of Tables 5-2 and 5-4 to equation form Young (1975b)provides an extensive tabulation of values of Q and e
MECHANISMS OF COPOLYMERIZATION
119
tonic Copolymerization
The order of monomer ~eactivityin cattonic copolymerization is quite ddferent from that in free-radical polymerization and is distinguished by the high reactivities of vinyl ethers and isobutylene However, the reactivities cor~espond the anticto ipated effect of substituents upon the reactivity of double bonds toward electrophilic reagents The differences of reactivity arise from the change in electron availability in the double bond and the resonance stabilization of the resulting carbenium ion Compared with radical-induced copolymerization, the carbenium-ion reactions show a much wider range of reactivity among the common monomers, and no tendency toward alternation in the copolymer These characteristics mean that relatively few monomer pairs easily yield copolymers containing large proportions of both monomers The data on anionic copolymerization suggest that the order of monomer reactivity is much different from both free-radical and carbenium-ion polymerization As anticipated, the order of reactivity in the anionic process is determined by the ability of substituents to withdraw electrons from the double bond and to stabilize the carbanion formed Anionic and coordination polymerizations involving monomers of like elechonegativity are usually ideal, since the unsubstituted carbon atom at the end of the growing chain has little influence on the course of the reaction When the electronegativities of the monomers are widely different, however, the rates of the cross-propagation ~eactionsmay be quite different also In extreme cases, only one monomer can add at all. Both cationic and anionic reactivity ratios can be profoundly affected by solvent and counterion effects These are both complicated and variable, but it appears that if the catalyst and its counterion form a tightly bound pair, solvent effects can be large, and conve~sely The contrast among the three methods of polymerization is brought out very clearly by the results of copolymerization studies using the three modes of initiation Figure 5-9 shows the composition of the initial polymer formed from a feed of styrene and methyl methacrylate using different initiators These results verify the differ en^ polymerization mechanisms and illustrate the extreme variations of reactivities that monomers exhibit with different type initiators Again, the results reflect the pvticular selection of catalyst, solvent, and counterion almost as much as the difference between mechanisms
Step-Reaction Copolymerization
Since the reactivities of all functional groups in simple bifunctional stepwise polymerization are essentially identical, irrespective of the length of the molecule to which they are attached, comonomers in such systems are randomly distributed along the chain in amounts proportional to their concentrations in the feed. Thus the "reactivity ratios" for stepwise polymerization are unity. Departures from this generalization are occasionally encountered, however; fbr example, the esterification rate of succinic acid HOOC(CH,),COOH, differs from that of' adipic acid,
COPOLYMERIZATION
Per cent styrene in monomer

FIG.. 5-9. Instantaneous composition of copolymer as a function of monomer composition for the
system styrene-methyl methacrylate polymerized by cationic (SnCI,), fiee-radical (benzoyl peroxide, BzlOL), and anionic (Na) mechanisms (Landler 1950, Pepper 1954)
HOOC(CH,),COOH, because of' the proximity of' the carboxyl groups in the former compound For similar reasons, ethylene glycol and 1,4-.butanediol have different esterification rates. Also, 1,'-butanediol differs in reactivity from 1,4-butanediol because one of the hydroxyl groups of' the former is secondary, whereas both of the latter are primary. Where such reactivity differences exist, the tendency is fbr the more reactive species to enter into the polymer first.
GENERAL REFERENCES
Ham 1964; Lenz 1967; Bamford 1976; Elias 1977, Chapter 22; Schulz 1979; Odian 1981, Chapter 6
D. BLOCK AND GRAFT' COPOLYMERS
It is shown in Section I3 that the probability of' finding long sequences of' one monomer in an ordinary random copolymer is very small, except in the trivial case where one monomer is present in large excess Methods of synthesis of' polymers containing such long sequences are of' interest, however, since they may lead to polymers with properties widely difftrent from those of either homopolymers or random copolymers (as discussed in Chapter 12). Polymers with long sequences
BL.OC,K A N D GRAFT COPOLYMERS
121
of two monomers can have two arrangements of chains: In block copolymers the sequences follow one another along the main polymer chain, ---AABB-BBAA--AABB.-., etc
whereas in graft copolyme~ssequences of' one monomer are "grafted" onto a "backbone" of the second monomer type, -.AAA-AAA-AAA-, B B B etc
The ultimate aim in preparing either type is to obtain the desired copolymer free from homopolymer or other unwanted species, on a scale that allows evaluation of' physical properties. Few of the techniques described in the literature satisfy these requirements The industrial importance of' both block and graft copolymers has increased markedly in recent years. These commerical polymers are discussed in Part 4.
Block Copolymers
Polymers With Labile End Croups. The usual method of preparing block co-. polymers utilizes a polymer with end groups that can be made to react under different conditions. In stepwise polymerization the method is trivially simple: Two polyesters, for example, of' different type are separately prepared, mixed, and allowed to polymerize further. Ring-scission polymerization can be used effectively By means of' free.-radical techniques, block copolymers can be made in several ways. Labile end groups (as, fbr. example, bromine incorporated through termination by transftr to CBr,) can be activated by thermal or ultraviolet.-light scission of the C--Br bond. Difunctional initiators can be used to produce block copolymers if' the two initiator groups can be decomposed independently, or polymer radicals can be formed by milling or masticating the polymer. In each case, of' course, the polymer radical must be produced in the presence of' a second monomer to yield the copolymer By far the most important synthesis of' block copolymers is that of' S z w m (1968a,b) fbr producing "living" polymers by unterminated anionic polymerization (Chapter 4C). Addition of a second momomer to the still active polymer. leads to block copolymer uncontaminated with homopolymer and with blocks of accurately known and controlled length. The process can be repeated, with care, to produce multiblock polymers Several commercial block copolymers are produced in this way and are discussed in Part 4; see also Fetters (1969).
COPOLYMERIZATION
Graft Copolymers
Graft copolymerization results from the formation of an active site at a point on a polymer molecule other than its end, and exposure to a second monomer. Most graft copolymers are formed by radical polymerization. The major activation reaction is chain transfer. to polymer (Chapter 3). In many instances the transfer reaction involves abstraction of' a hydrogen atom. An important commercial application is the grafting of polystyrene or styrene-acrylonitrile copolymer onto butadiene or acrylonitrile-butadiene copolymer rubber in the production of ABS resins (Chapter 14A). Ultraviolet or ionizing radiation, or redox initiation, among other methods, can also be used to produce the polymer radicals leading to graft copolymers
GENERAL REFERENCES
Burlant 1960; Hoffman 1964; Ceresa 1965, 19'73, 1976; Battaerd 1967; Aggarwal 1970; Allport 1973; Noshay 19'7'7; Morton 1977; Schildknecht 1977
1.
Define the reactivity ratios r, and r2, and indicate their values for (a) ideal, (b) alternating, and (c) azeotropic copolymerization.
2. Derive the copolymer equation, stating the assumptions used
3. Discuss the chain structures obtained in alternating, random, block, and graft i copolymers
4.
Consider the copolymerization of methyl acrylate (1) and vinyl chloride (2) The following compositions were found:
Mole Percent Methyl Acrylate in Feed Mole Percent Methyl Acrylate in Polymer
BIBLIOGRAPHY
123
a b
Calculate the reactivity ratios Explore the three methods described in the text and comment on their relative merits. For a 50:50 mole peIcent feed, and using for uniformity the lite~ature values r, = 9.0 and r , = 0.083, calculate and plot curves of the following: 1 Remaining feed composition versus conversion. 2 Instantaneous copolyme~composition versus conversion 3. Overall copolymer composition versus conversion 4 Distribution of copolymer compositions.
5 . Using the reactivity ratios given in Table 5-1, check the findings of Staudinger (1939) quoted in the first section ofthis chapter for the compositions of' fractions isolated from a copolymer of vinyl acetate and vinyl chloride.
6 . Given the fbllowing values of Q and e, calculate monomer reactivity ratios for the copolymerization of' the monomers in pairs.. Compare your results with those given in Table 5-1 and discuss the merits of the Q-e scheme for these monomers.
Monomer 1,3-Butadiene Methyl methacrylate Styrene Vinyl chloride
Q
2 39 0 74 1 00 0 044
e
-- 1 05
0 40 - 0 80 0 20
BIBLIOGRAPHY
Aggarwal1970 S L Aggarwal, ed , Block Polymers, Plenum Press, New York, 1970 Alfrey 1944 Turner Alfrey, JI , and G Goldfinger, "The Mechanism of Copolymerization," J Chem Phys 12, 205-209 (1944) Alfrey 1947 Turner Alf~ey,Jr , and Charles C Price, "Relative Reactivities in Vinyl Copolymerization," J Polym Scz 2, 101-106 (1947) Alfrey 1952 Turner Alfrey, 11 , John J Bohrer, and H Mark, Copolymerrzatzon, Interscience, New York, 1952 Alfrey 1964 T Alf~ey,JI , and L J Young, "The Q-e Scheme," Chapter 2 in George E Ham, ed , Copolymerization, Wiley-Intersc~ence,New York, 1964 Allport 1973 D C Allport and W H Janes, eds ,Block Polymers, Halsted Press, John Wiley & Sons, New York, 1973 Atherton 1962 J N Atherton and A M North, "Diffusion-Controlled Termination in Free Radical Copolyme~ization," T T ~ Faraday Soc 58, 2049-2057 (1962) ~ J Bamford 1976 C H Bamford and C F H Tippers, "Free-Radical Polymerisation," Vol 14A in Comprehenrrve Chemrcal Kznetzcs, Elsevier, Amsterdam, 1976 Battaerd 1967 H A J Battaerd and G W Tregear, Graft Copolymers, John Wiley & Sons, New York, 1967
Burlant 1960 William J Burlant and Allan S Hoffman, Block and Graji Polymers, Reinhold, New York, 1960 Burnett 1954 G M Burnett, Mechanism of Polymer Reactions, Interscience, New York, 1954 Ceresa 1965 R J Ceresa, "Block and Graft Copolymers," pp 485-528 in Herman F Mark, Norman G Gaylord, and Norbert M Bikales, eds ,Encyclopedza ojPolymer Science and Technology, Vol 2, Wiley-Interscience, New York, 1965 Ceresa 1973 R J Ceresa, ed , Block and Graji Polymerization, Vol 1, Wiley-Intersc~ence,New York, 1973 Ceresa 1976 R J Ceresa, ed , Block and Graft Polymerizot~on, Vol 2, Wiley-Interscience, New York, 1976 Collins 1971 Edward A Collins, Jan BareS, and Fred W Billmeyer, Jr , Experiments rn Polymer Science, Wiley-Interscience, New York, 1973 Dostal 1936 H Dostal, "A Basis for the Reaction Kinetics of Mixed Polymerization" (in German), Monatsh Chem 69,424-426 (1 936) Elras 1977 Hans-Georg Elias, Macromolecules 2 Synthesrs and Marercals (translated by John W Stafford), Plenum Press, New York, 1977 Fetters I969 L J Fetters, "Synthesis of Block Polymers by Homogeneous Anlonic Polymenzat~on," J Polym Scc C26, 1-35 (1969) Ham 1964 George E Ham, ed , Copolymerizat~on, Wiley-Interscience, New York, 1964 Ham I966 George E Ham, "Copolymerization," pp 156-244 In Herman F Mark, Norman G Gaylord, and Norbert M Bikales, eds , Encyclopedia oj Polymer Science and Technology, Vol 4, Wiley-Intersc~ence,New York, 1966 Hammett 1940 Louis P Hammett, Physical Organic Chemrstry, McGraw-Hill, New York, 1940 Hoffman 1964 Allan S Hoffman and Robert Bacskai, "Block and Graft Copolymerizat~ons," Chapter 4 in George E Ham, ed , Copolymerrzation. W~ley-Interscience,New York, 1964 Landler 1950 Yvan Landler, "On Ion~c Copolymerization" (in French), C R Acad Sci 230, 539541 (1950) Lenz 1967 Robert W Lenz, Organic Chemisny oj Synthetic High PoOrners, Wiley-Interscience, New York, 1967 Lowry 1960 George G Lowry, "The Effect of Depropagation on Copolymer Composition I General Theory for One Depropagating Monomer," J Polym Scl 42, 463-477 (1960) Mayo 1944 Frank R Mayo and Frederick M Lewis, "Copolymerization I A Basis for Comparing the Behavior of Monomers in Copolymerization; The Copolymerization of Styrene and Methyl Methacrylate," J Am Chem Soc 66, 1594-1601 (1944) Mayo 1950 Frank R Mayo and Cheves Walling, "Copolymerization," Chem Rev 46, 191--287 (1950 ) Merz I946 E Merz, T Alfrey, and G Goldfinger, "Intramolecular Reactions in Vinyl Polymers as a Means of Investigation of the Propagation Step," J Polym Sci 1, 75-82 (1946) Meyer 1965 Victor E Meyer and George G Lowry, "Integral and Differential Binary Copolymerization Equations," J Polym Sci A3, 2843-2851 (1965) Meyer 1966 Victor E me ye^, "Copolymerization of Styrene and Methyl Methacrylate Reactivrty Ratios from Conversion-Composttion Data," J Polym Sci A-1 4, 2819-2830 (1966) Meyer 1967 Victor E Meyer and Richard K S Chan, "Computer Calculations of Batch-Type Copolymerization Behavior," Am Chem Soc Div Polym Chem Prepr 8 (I), 209-215 (1967) Morton 1977 Maur~ce Morton and Lew~s Fetters, "An~onlcPoiymenzat~ons Block Copolymers," J and Chapter 9 In Calvln E Schlldknecht, ed , wrth Irving Skelst, Pol,mer~zat~on Processes, W~leyIntersclence, New York, 1977 North 1963 A M North, "The +-Factor In Free-Radical Copolymenzatlon," Polymer 4, 134-135 (1963)
;
1
BIBLIOGRAPHY
125
Noshay 1977 Allen Noshay and James E McGrath, Block Copolymers-Overvrew and Crrtrcal Survey, Academic Press, New York, 1977 Odlan 1981 George Odlan, Prrntrples of Polymerrzatron, 2nd ed , John Wtley & Sons, New York, 1981 Pepper 1954 D C Pepper, "Ionlc Polymensat~on," Chem Soc Q Rev 8, 88-121 (1954) Price 1948 Charles C Pnce, "Some Relatrve Monomer Reactrvlty Factors," I Polym Sci 3, 772775 (1948) Schtldknecht 1977 Calvln E Schlldknecht, "Industnal G ~ a f Copolymer~zat~ons," t Chapter 8 In C E Schrldknecht, ed , wlth Irvtng Skelst, Polymerrzatron Processes, W~ley-Interscrence,New York, 1977 Schulz 1979 D N Schulz and D P Tate, "Copolymers," pp 798 818 in Martin Grayson, ed , KrrkOrhmer Encyclopedra of Chemlcal Technology, 3rd ed , Vol 6, Wlley Intelsctence, New York, 1979 Simha 1944 Robert Srmha and Herman Branson, "Theory of Chain Copolymenzatron Reactrons," I Chem Phys 12, 253-267 (1944) Skerst 1946 Irvlng Skeist, "Copolymer~zat~on Composition Distrlbutlon Curve," J Am Chem The Soc 68, 1781-1 784 (1946) Staudrnger 1939 H Staudrnger and J Schneiders, "Macromolecula~ Compounds CCXXXI Polyvtnylchlondes" (In German), Iustus Lrebigr Ann Chem 541, 151-195 (1939) Stotkmayer 1945 W H Stockmayer, "D~stnbut~on Chain Lengths and Composltlon rn Copolymers," of J Chem Phys 13, 199-207 (1945) Szwarc 1968a Michael Szwarc, Carbanrons, Lrving Polymers and Electron-Transfer Processes, John Wiley & Sons, New York, 1968 Szwarc 1968b M Szwa~c," 'Ltvlng' Polymers," pp 303-325 In Herman F Mark, Norman G Gaylord, and Norbert M Bikales, eds , Encyclopedra of Polymer Scrence and Technology, Vol 8, W~ley-Interscrence,New York, 1968 Wall 1944 Freder~ckT Wall, "The Structure of Copolymers 11," I Am Chem Soc 66, 2050- 2057 (1944) Wallrng I949 Cheves Walling, "Copolymenzat~on XI11 Over-all Rates rn Copolymer~zatton Polar Effects In Chatn Inltiatlon and Termtnatron," J Am Chem Soc 71, 1930-1935 (1949) Wallrng 1957 Cheves Walling, Flee Radrcals rn Solutron, John Wlley & Sons, New York, 1957 Young 1975a Lewls J Young, "Copolymer~zat~on Reacttvtty Ratlos," pp I1 105-11-386in J Brandrup and E H Immergut, eds , with the collaborat~onof W McDowell, Polymer Handbook, 2nd ed , Wtley-Intexsctence, New York, 1975 Young 1975b Lewis J Young, "Copolyme~~zatron Reacttvlty Ratlos," pp I1 387-11-404 In J Brandrup and E H Immergut, eds , with the collaboration of W McDowell, Polymer HandbootC, 2nd ed , Wtley Intersc~ence,New York, 1975
CHAPTER SIX
POLYMERIZATION CONDITIONS AND POLYMER REACTIONS
The discussion of' polymerization kinetics in Chapters 2--5 has dealt, with the exception of the systems discussed in Chapter. 40, with polymerization of pure monomer or of' homogeneous solutions of monomer and polymer in a solvent. Certain other types of' polyme~izingsystems ase of' great interest because they offer practical advantages in industrial applications. Among these are polymerizations in which the monomer is carried in an emulsion or in usp pension in an aqueous phase.. Some of the advantages and disadvantages of' these systems are itemized in Table 6,-1.For purposes of' discussion, it is convenient to classify polymerization systems as homogeneous or heterogeneous..These two categories of polymerization proc.esses are discussed in Sections A and B, respectively As the industrial production of' polymers has increased over the years, there has developed an increasing interest in the application of'chemical engineering principles to polymerization. The first books and review articles on this subject are just beginning to appear. Polymerization reaction engineering is discussed in Set.tion C Applications based on chemical reactions of' polymers, during or after polymerization, ase also increasing.. Section D includes discussion of several of these, including crosslinking, both during and after polymerization; two other important postpolymerization phenomena, degradation (which frequently involves just the reverse reactions of' polymerization) and reactions induced by high-energy radiation (which often involve radicals and lead to crosslinking or degradation); and the use of' polymers as reagents or substrates for chemical reactions
1
i
1
i
POLYMERIZATION IN HOMOGENEOUS S S E S YTM
127
TABLE 6-1.
Comparison of Polymerization Systems
----Homogeneous Bulk (batch type)
Type
Advantages
Minimum contamination. Simple equipment for making castings Lower conversion per pass leads to better heat control and narrower molecular-weight distribution Ready control of heat of polymerization Solution may be directly usable Ready control of heat of polymerization Suspension or resulting gxanu la1 polymer may be directl y usable Rapid polymerization to high molecular weight and narrow distribution, with ready heat control Emulsion may be directly usable
Disadvantages
Str ongly exothermic Broadened molecular-weight distribution at high conversion Complex if small particles required Requires agitation, material transfer, separation, and recycling
ii
1
I
Bulk (continuous)
Solution
Not useful for dry polymer because of difficulty of complete solvent removal Continuous agitation ~equired Contamination by stabilizer possible Washing, drying, possibly compacting required Contamination with emulsifier, etc , almost inevitable, leading to poor color and color stability Washing, drying, and compacting may be required --
Heterogeneous Suspension
Emulsion
L .
.A.
POLYMERlZATlON IN HOMOGENEOUS S S E S YTM
Bulk Polymerization. Polyme~ization bulk, perhaps the most obvious method in of synthesis of polymers, is widely practiced in the manufactu~e condensation of polymers, where the reactions are only mildly exothermic, and most of the reaction occurs when the viscosity of the mixture is still low enough to allow ready mixing, heat transfer, and bubble elimination. Control of such polyme~izations relatively is easy. Bulk polymerization of vinyl monomers is more difficult, since the reactions are highly exothermic and, with the usual thermally decomposed initiators, proceed at a rate that is strongly dependent on temperature This, coupled with the problem in heat transfer incurred because viscosity increases early in the reaction, leads to difficulty in control and a tendency to the development of localized "hot spots" and "runaways " Except in the preparation of castings, for example, of poly(methy1 methacrylate), bulk polymerization is seldom used commerciaily for the manufactme of vinyl polymers
128
POLYMERIZATION CONDITIONS AND POLYMER REAC.TlONS
Solution Polymerization. Polymerization of' vinyl monomers in solution is advantageous from the standpoint of' heat removal (e..g., by allowing the solvent to reflux) and control, but has two potential disadvantages. First, the solvent must be selected with care to avoid chain transfer and, second, the polymer should preferably be utilized in solution, as in the case of poly(viny1 acetate) to be converted to poly(viny1 alcohol) and some acrylic ester finishes, since the complete removal of solvent from a polymer is often difficult to the point of' impracticality
GENERAL REFERENCES
Ringsdorf' 1965; Lenz 1967; Matsumoto 1977; Schildknecht 1977a,b
B.
POLYMERIZATION IN HETEROGENEOUS SYS'T'EMS
Polymerization From Gaseous Monomers

The polymerization of' gaseous monomers can take place with the formation of' a liquid phase (polymer melt) or a solid polymer. In each case the polymerization of' ethylene provides the most important industrial example. Both polymerizations are discussed in Chapter 13A.. The high-pressure polymerization of' ethylene to branched polyethylene takes place by a free-radical mechanism in the presence of a liquid phase, at temperatures above the melting point o f t h e polymer.. The polymerization is carried only to low conversion, with the remaining ethylene recovered when the pressure is lowered and recycled The low-pressure polymerization of' ethylene to linear polyethylene takes place by coordination polymerization using a catalyst suspended in gaseous ethylene in a fluid-bed reactor. The ethylene polymerizes to the solid phase, with the small amount of' the catalyst required remaining in the solid polymer
Emulsion Polymerization
In emulsion polymerization two immiscible liquid phases are present, an aqumu5 continuous phase and a nonaqueous discontinuous phase consisting of' monomer and polymer, as described below. The initiator is located in the aqueous phase, and the monomer-polymer particles are quite small, of' the order of' 0. I pm in diameter. The kinetics of emulsion polymerization differs importantly from th;F: of bulk polymerization, fbr which a limitation is set by Eq. 3- 13-for combination,
POLYMERIZATION IN HETEROGENEOUS SYSTEMS
TABLE 6-2. "Mutual" Recipe for the Emulsion Copolymerization of Styrene and Butadiene at 50C
Ingredient Butadiene Styrene Water

Soap
Parts by Weight 75 25 180 5.0 0.50 0.30
----
Lorol mercaptana Potassium persulfate
"Crude dodecyl mercaptan
-on the molecular weight obtainable at a given rate of polymerization With this limitation removed, emulsion systems allow higher-molecular-weight polymer to be produced at highel rates than do bulk or suspension systems A typical recipe for an emulsion polymerization is given in Table 6-2 This is the famous "Mutual" recipe used in the polymerization of styrene-butadiene synthetic ~ubber during World War I1 The ingredients, in addition to monomexs and water, are a fatty-acid soap, a mercaptan-type chain-txansfer agent, and the watersoluble per sulfate initiator. The soap plays an important role in emulsion polymerization At the beginning of the reaction it exists in the form of micelles, aggregates of 50- 100 soap molecules probably having a layered structure like that shown in Fig 6-1 Part of the monomex enters the micelles, but most of it exists as droplets a micrometer or more in diameter In the ideal case no polymer is f o ~ m e in the monomer droplets. Polymerization d can take place (at a very low rate) in the homogeneous phase in the absence of soap, but this cannot account for the bulk of the polymer formed At the beginning of the reaction, polymer is formed in the soap micelles; these represent a favorable 'environment fox the free radicals generated in the aqueous phase, because of the relative abundance of monome1 and the high surface/volume ratio of the micelles compared to the monomer droplets As polymer is formed, the micelles grow by the addition of monomer from the aqueous phase (and ultimately from the monomer droplets)
(b) FIG. 6-1. Idealized structure of a soap micelle (a) without and (b) with solubilized monomer
(a)
130
Soon (2-3% polymerization) the polymer particles grow much larger than the original micelles and absorb almost all the soap from the aqueous phase. Any micelles not already activated disappear; further polymerization takes place within the polymer particles already formed. The monomer droplets are unstable at this stage; if' agitation is stopped, they coalesce into a continuous oil phase containing no polymer. The droplets act as reservoirs of monomer, which is fed to the growing polymer pa~ticlesby diffusion through the aqueous phase. The polymer particles may contain about 50% monomer up to the point at which the monomer droplets disappear, at 6 0 4 0 % polymerization. The rate of' polymerization is constant over most of the reaction up to this point, but then fills off as monomer is depleted in the polymer particles Rate increases with increasing soap (and initial micelle) concentration..
Smith-Ewart Kinetics (Smith 1948a,b).. In an ideal emulsion system, free radicals are generated in the aqueous phase at a rate of about lOI3per cubic centimeter per second. T h e ~ e about lOI4 polymer particles per cubic centimeter. Simple are calculations show that termination of the free radicals in the aqueous phase is negligible and that diffusion currents are adequate fbr the rapid diffusion of free radicals into the polymer particles-on the average, about one per particle every 10 sec It can also be calculated from the known termination rate constants that two free radicals within the same polymer particle would mutually terminate within a few thousandths of' a second. Therefore each polymer particle must contain most of the time either one or no free radicals. At any time half of'the particles (on the average) contain one free radical, the other half none.. The rate of polymerization per cubic centimeter of emulsion is
where N is the number of polymer particles per cubic centimeter. Since the monomer concentration is approximately constant, the rate depends principally on the number of' particles present and not on the rate of generation of' radicals The degree of polymerization also depends upon the number of' particles:
of where p is the rate of gene~ation radicals. Unlike up, .?, is a function of the rate of free-radical fbrmation.. In bulk polymerization rate can be increased only by increasing the rate of initiation; this, however, causes a decrease in the degree of polymerization In emulsion polymerization the rate may be increased by increasing the number of polymer particles If the rate of initiation is kept constant, the degree of' polymerization increases rather than decreases as the rate rises Since the number
POLYMERIZA'IION IN HETEROGENEOUS S S E S YTM
131
of' polymer particles is determined by the number of' soap micelles initially present, both rate and molecular weight increase with increasing soap concentration. The Smith-Ewart kinetics require that
up x N, [II04, [El0
N a [IIo4, [Elo6
.x, x N, [El0
6,
[I] --O
where [El is the soap or emulsifier concentration.

Deviations From Smith-Ewart Kinetics. The above kinetic scheme is highly idealized, though valuable for its simplicity.. It explains adequately only a small portion of the vast literature on emulsion polymerization, though it works well for monomers such as styrene, butadiene, and isoprene, whose water solubility is very low, less than 0.1%.. Among the circumstances under which the Smith-Ewart kinetics fails to apply quantitatively are the fbllowing:
Larger particles (>O 1 - 0 15 k m in diameter), which can accommodate more than one growing chain simultaneously b. Monomers with higher water solubility (1--lo%), such as vinyl chloride, methyl methacrylate, vinyl acetate, and methyl acrylate Here initiation in the aqueous phase, followed by precipitation of polymer, becomes important These particles may absorb emulsifier, decreasing [El and thus N, or they may serlre as sites for polymerization, increasing N. c. Chain transfer to emulsifier This often takes place in large enough amount to suggest that growing chains are localized near the surfaces of the particles, where the soap exists
a.
inverse Emulsion Systems. Emulsion polymerization can also be carlied out in systems using an aqueous solution of a hydrophilic monomer, such as acrylic acid 01 acrylamide, emulsified in a continuous oil phase using an appropriate water-inoil emulsifier Either oil- or water-soluble initiators can be used The mechanism seems to be that of normal emulsion polymerization, but the emulsions are often less stable. "Redox" Initiation. The decomposition of peroxide-type initiators in aqueous systems is greatly accelerated by the presence of a reducing agent This acceleration allows the attainment of high rates of radical formation at low temperatures in emulsion systems A typical redox system is that of ferrous iron and hydrogen peroxide In the absence of a polymerizable monomer the peroxide decomposes to free radicals as follows:
\ .
132
POLYMERlZATlON CONDITIONS AND POLYMER REAC:TlONS
Another widely used peroxide-type initiator is the persulfate ion. With a reducing agent R the reaction is
The reducing agent is often the thiosulfate ion,
or the bisulfite ion,
S20g2-
+ HS03- --+SO,2-
t-SO,-,,
+ HSO,
Many other redox initiator systems have been used

Other Heterogeneous Liquid Systems Suspension Polymerization. The term suspension polymerization refers to polymerization in an aqueous system with monomer as a dispersed phase, resulting in polymer as a dispersed solid phase. The process is distinguished from superficially similx emulsion polymerization by the location of' the initiator and the kinetics obeyed: In typical suspension polymerization the initiator is dissolved in the monomer phase, and the kinetics is the same as that of bulk polymerization.. The dispersion of' monomer into droplets much larger than those found in em& sion polymerization, typically 0..01-0..5 cm in diameter, is maintained by a combination of agitation and the use of water-soluble stabilizers. These may include finely divided insoluble organic or inorganic materials that interfere with agglorneration mechanically, electrolytes to increase the interfacial tension between the phases, and water-soluble polyme~s increase the viscosity of the aqueous phase to The tendency to agglomerate may become critical when the polymerization has advanced to the point where the polymer beads become sticky. At the completion of the reaction the polymer is freed of' stabilizer by washing and is dried.. For some applications (Chapter 17), the polymer beads can be used directly, whereas for others compaction is required. A typical recipe for suspension polymerization is given in Table 6-3. The method is used commercially to prepare hard, glassy vinyl polymers such as polystyrene, poly(methy1 methacrylate), poly(viny1 chloride), poly(viny1idene chloride), and polyacrylonitrile Precipitation Polymerization. In the preparation of a polymer insoluble in its monomer, or in polymerization in the presence of a nonsolvent for the polymer, marked deviations from the kinetics of homogeneous radical polymerization may occur. The normal bimolecular termination reaction is not effective, as the result of trapping or occlusion of radicals in the unswollen, tightly coiled precipitating polymer The theory has been confirmed by the demonstration of the presence of radicals in the polymer both by chemical methods and by electron paramagnetic
POLYMERIZATION IN HETEROGENEOUS S S E S YTM
133
TABLE 6-3. Typical Recipes for Suspension Polymerization
Parts per 100 Parts of Monomer Component

Peroxide initiator Water Stabilizers"
--
Methyl Methacrylate
-0 5 -350 001-1
-
-----
Vinyl Chloride
0 1-0 5 150- 350 0 01-1
-
"Typical stabilizers include methyl cellulose, gelatin, poly(viny1 alcohol), and sodium polyacrylate Minor amounts of emuisifiers and buffers are also used
resonance (Chapte~ B ) The lifetime of the trapped radicals is many hours at room 9 temperature, and if polymer containing such radicals is heated in the presence of monomer to a temperature where the mobility of the radicals is increased, extremely rapid polymerization takes place The polymerization of vinyl chloride in bulk or in the presence of a nonsolvent (Mickley 1962) follows a rate equation of the form
where the first term inside the braces arises from the normal rate equation for polymerization in the homogeneous liquid phase, and the second term represents the increment in rate due to polymerization in the precipitated polymer particles The function f (P) is proportional to the polymer concentration [PI at low conversion and to [PI2l3at later times Radical occlusion occurs, but the depth of radical penetration into the polymer particles is small, limiting radical activity to thin surface layers in larger particles present at higher conversions Termination occurs primarily in the liquid phase, probably as a result of transfer to monomer within the polymer particles, and subsequent diffusion of the monomer radicals to the liquid phase In conbast, transfer to monomer in the polymerization of acrylonitrile is so slow that permanent radical occlusion occurs
'.
Solid-Phase Polymerization
A large number of olefin and cyclic monomers can be polymerized from the crystalline solid state. Among those that react rapidly under these conditions are styrene, acrylonitrile, methacrylonitrile, formaldehyde, trioxane, 0-propiolactone, diketene, vinyl stearate, vinyl carbazole, and vinyl pyr~olidone Polymerization always appears to be associated with defects in the monomer crystals, most likely line defects (Chapter 10D); othe~wise,it would be required that the monomer and the polymer be isomorphous, that is, that they have the same crystal structure, lattice parameters, and so on This seems extremely unlikely and has not been observed
134
POLYMERIZATION CONDITIONS AND P L M R REACTIONS OY E
GENERAL REFERENCES
Lenz 1967; Schildknecht 1977~; Odian 1981; and specifically the following:
Emulsion Polymerization. Duck 1966; Vanderhoff' 1969; Collins 1973, Exps 4 and 8; Cooper 1974; Blackley 1975; Gardon 1977; Piirma 1979; Eliseeva 1981; Odian 1981, Chapter 4. Suspension Polymerization. Fasber 1970; Munzer 1977.. Precipitation Polymerization. Jenkins 1967 Solid-Phase Polymerization Lenz 1967; Bamfbrd 1969
C.
POLYMERIZATION REACTION ENGINEERING
While polymerization and the reactions of polymers are in many respects similar to ordinary chemical reactions, there are some significant differences that make the former unique in the sense of reactor and reaction engineering These are high viscosities and low diffusion rates associated with concentrated polymer solutions and polymer melts This section describes some engineering aspects of polymerization related to these unique polymer properties; brief descriptions of the production of important polymers are given in Chapters 13-16
Reactor Design Theory
Although the high viscosity and low difusivity associated with polymer melts and concentrated solutions lead to complications in designing polymerization reactors, they actually simplify the associated theoretical analysis. The reason for this is that there is no turbulence in most polymer systems, and the consequent steady laminar flow can be analyzed more easily The goal of such an analysis is to describe the flow rates, temperatures, and compositions at any point within the reactor This is done by solving the partial differential equations that govern momentum, heat, and mass transfer.. Three simultaneous nonlinear. partial differential equations of motion, the Navies--Stokes equations, descr.ibe fluid velocities, but often the geometry is such that one or at most two of these suffice. Temperatures are found by solving one partial differential equation involving the viscosity, and there is a single partial differential equation for mass transfer fbr each component leading to infbrmation on concentrations Usually it suffices to consider only two components, monomer and polymer. The equations ase fully described in textbooks on transport phenomena (e .g. ,Bird 1960), and for polymer systems they can often be solved exactly, Important results of the solutions, depending strongly on the type of mixing in the reactor, ase distributions of residence times and thermal times.
!
~oLYMERIZATION REACTION ENGINEERING
I
135
Types of Reactors
Batch Reactors. Some of' the polymerizations carried out in batches are those
in emulsion or suspension, most condensation reactions, and the early stages of' a few bulk or solution vinyl polymerizations.. The reactants a1.e usually well mixed by agitation, and the reaction initiated by heating the vessel externally; later, the reaction exotherm supplies the needed heat, and heat control may be required, for example, by refluxing solvent. Bulk systems are usually terminated, quenched, or transferred to other equipment before the viscosity becomes too high Runaway reactions must be guarded against, and the safest practice is to design the reactor to contain the maximum temperature and pressure that could result.
Tubular-Flow Reactors. The design objective of a tubular-flow reactor is to provide progressive flow with little or no internal recycle, achieving a uniform distribution of thermal and residence times. To do this it is necessary to promote radial mixing or flow redistribution, and this is not easy to do without turbulence A number of stationary techniques can be used, such as coiling the tube into a helix, introducing flow redistribution devices (so-called motionless mixers), or utilizing diffe~ences temperature or pressure along the tube. The polymerization of of ethylene at high pressures is an example of the use of a stationary-wall tubular reactor Radial mixing can also be achieved mechanically, as with the use of extruders or similar devices as reactors. Especially in cases where polymer must be melted or volatiles removed, extruders p~ovidea very favo~able reaction environment Stirred-Tank Reactors. A major design consideration for stirred-tank reactors (sometimes, autoclaves) is the removal of the heat of polymerization This is usually achieved by running the reaction with a small positive exotherm, and maintaining . the desired operating temperature by control of the amount of cooling. The usual design objective of these reactors is to promote mixing throughout the volume of the reactor in order to minimize temperature and composition variations An almost inevitable result of this recirculation is an exponential distribution of reaction times and a high recycle rate The kinetics of polymerizations are sufficiently complex to be affected by mixing at the molecular scale Yields can be affected, but only when there is a gel effect and termination is diffusion controlled The molecular-weight distribution is more readily affected, but the magnitude of the effect depends upon the type of mixing achieved in the reactor With a reasonably designed agitation system, perfect mixing can be achieved on the macro scale in a stirred-tank reactor, while low diffusivities lead to poor mixing on the micro scale A reactor operating in this way is termed segregated The perfect mixer, segregated stirled-tank reactor, and batch or tubular reactor represent three extremes of macro and micro mixing behavior (Fig. 6-2) The effect of micromixing on the molecular-weight distribution has been
' .
bl
or tubular reactor
'
Perfect mixer
-6atch
Segregated st~rred tank

- -- t
Increasing macromixing FIG..6-2.
Mixing space at the micro and macro levels, showing the locations of typical polymerization reactors (Reprinted with permission fiom Nauman 1982 Copyright 1982 by John Wiley & Sons )
studied by Tadmor (1966) and Biesenberger (1966). They showed that when the lifetime of a growing chain is long compared to the residence time, as in condensation polymerization, the molecular-weight distributions obtained with perfect mixing or in segregated stirred-tank reactors are extremely broad compared to those obtained in batch reactors (Fig. 6-3) On the other hand, in free-radical polymerization where the chain lifetime is short, the distribution is narrowest with perfect mixing (Fig 6-4)..
GENERAL REFERENCES
Nauman 1982; Biesenberger 1983.
D. CHEMICAL REACTIONS OF POLYMERS
The concept of the reactivity of functional groups being independent of molecular weight, used in developing the kinetics and statistics of stepwise polymerization (Chapter 2), applies to all functional groups, regardless of their location on the polymer chain As a result, polymers undergo chemical reactions and serve as reagents much as do low-molecular-weight compounds, provided that reactants can be supplied to the sites of reaction To accomplish this, most polymer reactions of importance are carlied out in solution Many of these reactions are discussed elsewhere in this book, as indicated by cross-references A well-known sequence of polymer reactions is the conversion of poly(viny1 acetate) through poly(viny1 alcohol) to a poly(viny1 acetal) (Chapter 14C) In the
CHEMICAL. REACTIONS OF POLYMERS
I\
Seoreaated st~rred tank
Batch reaccor
I/
FIG. 6-3.
Perfect mixer Molecular weight
Molecular-weight distributions obtained with various types of mixing in condensation polymexization (Reprinted with permission from Nauman 1982 Copyright 1982 by John Wiley & Sons )
latter reaction, as in the reaction of metals to remove chlorine from poly(viny1 chloride), functional groups react in pairs along the chain, with occasional groups isolated and incapable of reaction Polyesters ale readily hydrolyzed unless low solubility or steric hindrance interferes Thus, linear condensation polyesters and poiyacrylates hydrolyze readily, whereas poly(ethy1ene terephthalate) and crosslinked alkyd and "polyester" resins are insoluble Polymethacrylates are inert because of steric hindrance The nitration, sulfonation, and reduction of styrene polymers are widely used to produce ion-exchange resins (Chapter 14A) The acetylation, nitration, and xan-
Molecular weight
FIG. 6-4.
Molecular weight distributions obtained with various types of mixing in free-~adical poly. merization (Reprinted with permission from Nauman 1982 Copyxight 1982 by John Wiley & Sons )
138
POLYMERIZATION C:ONDITIONS AND POLYMER REACTIONS
thation of' cellulose (Chapter 15C) are important reactions, as are vulcanization and other reactions of' natural and synthetic rubbers (Chapters 13D-F and 19). In addition, unsaturated polymers can undergo such reactions as isomerization, cyclization, addition, epoxidation, and hydrogenation, while saturated polymeric hydrocarbons can be substituted on the main chain, the side chain, or the nucleus if aromatic. Terminally reactive polymers can produce block copolymers, whereas branching reactions can lead to graft copolymers (Chapter 5D).. Coupling reactions can promote increases in molecular weight, as in the polyurethanes (Chapter 15B), or crosslinking, as in a variety of' thermosetting resins (Chapter 16). The class of oxidation--reduction or redox polymers is discussed by Cassidy (1965) The advantages of polymers as reagents have been reviewed (Kraus 1979). They include the ease of' separation of' the (polymeric) reagent from the reaction mixture; the mutual inaccessibility of' reagents attached to crosslinked polymers, so that otherwise incompatible reagents can be used simultaneously ("wolf-and-lamb chemistry"); the ability of polymers to immobilize attached species, providing high local concentrations at high gross dilutions; and the possibility of tailoring the polymer to provide a special microenvironment, for example, steric or' electric in nature, at the site of reaction.. Other major classes of' polymer reactions are considered in the sections that fbllow
Crosslinking Reactions
Monomers with two double bonds can be arbitrarily divided into 1,3-diene and divinyl compounds. With the exception of cases in which the entrance of one double bond in the monomer into a polymer chain leaves the second double bond so unreactive that it will not polymerize, these monomers can produce branched or crosslinked polymers when present during polymerization.. It is usual to reduce the effect of the crosslinking reaction by utilizing only small quantities of' the dif'unctional monomer in a copolymerization By the proper choice of relative reactivity of'the two double bonds, it is possible to reduce the reactivity of' one just enough so that it will not enter polymerization under. the same conditions as the other, but can be made to react under more drastic conditions.. This leads to postpolymerization crosslinking reactions of' which vulcanization reactions are an example If the two double bonds are well separated, the reactivity of' one is not affected by the polymerization of' the other: An example is 2,11-dodecadiene. If the two double bonds are close enough together that the polymerization of one can shield the other sterically, or if' they are conjugated, as in the dienes, a difference in reactivity can be expected. Some examples are given in Table 6-4
Crosslinking During Polymerization. Consider the copolymerization of vinyl monomer A with divinyl monomer B-B, where all groups are equally reactive, as in the copolymerization of' methyl methacrylate and ethylene glycol dimethacrylate.. Let the concentrations of' A, B-B, and B groups be [A], [BB], and [B],
CHEMICAL. REAC.TIONS OF POLYMERS
139
TABLE 6-4.
-
Examples of Monomers With Two Double Bonds

-
Dependency of Reactivity
Independent Intermediate Interdependent
Symmetrical
Ethylene glycol dimethacrylate Divinylbenzene Butadiene
Unsy mm&ical
Ally1 acrylate p-Isopropenylstyrene Chloroprene
----.-
respectively, where [B] = 2[BB]. Since the A and B groups are equally reactive, their ratio bla in the copolymer is
a simplification of the copolymer equation, Eq. 5-4. At extent of reaction p , here the fraction of the double bonds reacted, there are present p[A] reacted A groups, (1 -. p)[A] unreacted A goups, p2[BB] doubly reacted BB molecules, 2p(l - p)[BB] singly reacted BB molecules, and (1 -. P)~[BB]unreacted BB molecules. In te~,ms of' B groups, the latter quantities become (1 - p)*[B] (unreacted) B groups on unreacted BB molecules, p ( l - p)[B] unreacted or reacted B groups on singly reacted BB molecules, and p2[B] (reacted) B groups on doubly reacted BB moleculcs. In this mixture the number of crosslinks is p2[BB], and the number of chains is derived in terms of' the degree of polymerization x : number of chains = (total number of' reacted units)l.x = ([A] -t.[B])plx. At the critical extent of' reaction pc fbr the onset of' gelation, the number of' crosslinks per chain is 4, whence
Equation 6-7 holtls since Flory has shown that the appropriate average of x is i,.. except where gelation occurs at very low extent of' reaction; here the distribution of' crasslinks is not random. Similar considerations apply to unsymmetrical divi~~yl monomers and dienes. In either case the important practical result of delaying the onset of' gelation until high conversions can be achieved in the same ways: By reducing the mole fraction of'the divinyl or diene monomer. b. By reducing the weight-average chain length, possibly by the use of' chni~ltsansftr agents By reducing the relative reactivity of the second double bond. c.
a.
Nonconjugated dienes that have the double bonds separated by three atoms can be polymerized to give soluble, noncrosslinked polymers by an alternating intra -
140
POLYMERIZATION CONDITIONS AND POLYMER REAC:llONS
intermolecular cyclopolymerization mechanism (Butler 1966, Gibbs 1967). Successive propagation steps involve, alternately, the addition of' a monomer and the formation of a six-membered ring:
where X is one of' a variety of groups. Five- and seven-membered rings form from the appropriate polymers with more difficulty, and some open-chain unsaturated groups occur in the polyme~
Crosslinking After Polymerization: Vulcanization. This section is concerned only with the statistics of vulcanization and the distribution of' molecular weights and crosslinks in the resulting polymer. Chemical aspects of' vulcanization are discussed in Chapter 19. The random crosslinking of double bonds bears fbrmal resemblance to the fbrmation of three-,dimensional networks from polyfunctional monomers by stepwise polymerization. In keeping with previous nomenclature, the fraction of'the monomer units on a chain that can be crosslinked is defined as q and the degree of polymerization of the chain as x The "functionality" of the chain is its total number of' vulcanizable groups qx The extent of reaction p is the fraction of the total available crosslinks that have been fbrmed Gelation occurs at a critical value of p, p,, where there is on the average one crosslink for every two chains:
As in the case of' three-dimensional step-reaction polymers, after gelation the finite species constitute only part of the material, the rest being in the form of' gel networks. As p increases from p, to 1, the weight fraction of' the finite species drops fiom unity to zero. The case above does not correspond to actual vulcanization because it neglects the molecular-weight distribution ofthe chains initially present. If this is considered, the detailed analysis becomes complicated, but some important features remain about the same as fbr the monodisperse case. The gel point is given by the equation
where .iw the weight-average degree of' polymerization of the starting polymer is Differences in chemical composition introduced by copolymerization or mixing may have more effect on the type of network structure than does molecular weight distribution. If', for example, a small amount of high-functionality polymer is mixed
CHEMlC,AL REAC,TlONS O F POLYMERS
141
with a polymer of' low functionality, the resulting vulcanizate will have regions of very tight crosslinking as inclusions in a looser network, leading to pronounced nonuniformity of' physical properties. The foregoing treatment also assumes that the points of crosslinking are distributed at random. This is not always true; for example, if vulcanization is carried out in dilute solution, intramolecular crosslinking may occur before two molecules come near enough to form intermolecular links. Thus gelation is delayed considerably beyond the point predicted by random statistics.
Degradation
In this section the term degradation is taken to mean reduction of molecular weight. There are two general types of polymer degradation processes, cor~espondingroughly and to the two types of polymerization, step ~eaction chain reaction. Random degradation is analogous to stepwise polymerization Here chain rupture or scission occurs at random points along the chain, leaving fragments that are usually large compared to a monomer unit. Chain depolymerrzatzon involves the successive release of monomer units from a chain end in a depropagation or unzlppering reaction that is essentially the reverse of chain polymerization These two types may occu~ separately or in combination, may be initiated the1mally or by ultraviolet light, oxygen, ozone, or other foreign agent, and may occur entirely at random or preferentially at chain ends or at other weak links in the chain It is possible to differentiate between the two processes in some cases by following the molecular weight of the ~esidue a function of the extent of reaction as Molecular weight drops rapidly as random degradation proceeds, but may remain constant in chain depolymerization, as whole molecules are reduced to monomer that escapes from the residual sample as a gas Examination of the degxadation .products also differentiates between the two processes: The ultimate product of random degradation is likely to be a disperse mixture of fragments of molecular weight up to several hundred, whereas chain depolyrnerization yields large quantities of monomer
Random Degradation. The kinetics and statistics of' random degradation can be treated in exact analogy with the kinetics of linear stepwise polymerization. If p is defined, as before, as the extent of reaction, then 1 -- p = (number of' broken bonds)/(total number of bonds) is the extent of degradation. If' the degradation is random, the number of' bonds broken per unit time is constant as long as the total number of' bonds present is large compared to the number broken. It follows that the number of chain ends increases linearly with time; hence 1I.?, increases linearly with time.. The acid-catalyzed homogeneous degradation of cellulose is a random degradation of this type. An example of random degradation initiated by attack at a "weak link" in a chain is the ozonolysis of the isobutylene--isoprene copolymer butyl rubber. Here the initial attack is at the double bonds of' the isoprene residues.
142
POLYMERIZAT'ION CONDIT'IONS AND POLYMER REACTIONS
Chain Depolymerization. Chain depolymerization is a free-radical process that is essentially the reverse of chain polymerization The point of' initial attack may be at the chain end or at a "weak link" that may arise from a chain defect, such as an initiator fragment or a peroxide or ether link arising from polymerization in the presence of oxygen. The slightly higher activity of a tertiary hydrogen atom may be enough to provide a site for the initiation of' the degradation process Poly(methy1 methacrylate) degrades thermally by this process. The yield of monomer is 100% of' the weight of polymer lost over a lasge fraction of the reaction. Polystyrene shows an intermediate behavior: The depropagation reaction ceases before the chain is completely destroyed. In some cases, such as olefin-SO2 copolymers, an equilibrium can easily be reached between propagation and depropagation
Kinetics of Degradation. A general kinetic scheme has been formulated that appears to cover all types of depolymerization. It is based on the concept of inverse chain polymerization and includes the steps of initiation, depropagation, termination, and chain transfer. The important feature of this scheme is the inclusion of' the chain-transfer step, fbr it can be shown that the kinetics of random degradation result if the kinetic chain length before transfer is just the breaking of' one bond The transfer reaction probably occurs rapidly by the abstraction of a hydrogen atom from a polymer.. The chain that was attacked is likely to split into a radical and one or more inactive fiagments at the elevated temperatures where degradation is rapid. Evidence for the transfer reaction includes the observations that tertiary hydrogen atoms at branch points in polyethylene are pref'erentially attacked and that the degradation of' poly(methy1 methacmlate) that has been copolymerized with a little acrylonitrile is quite different fiom that of' pure poly(methy1 methacrylate) because of the activity of' the a-hydrogens on the acrylonitrile units. The two factors that appear to be important in dete~miningthe course of' degradation are the reactivity of the depropagating radical and the availability of'reactive hydrogen atoms for transfer. With the possible exception of' styrene, where the radical is stabilized by resonance, all polymers containing a-hydrogens, such as polyacrylates and polyacrylonitrile, give poor yields of monomer. Conversely, the methacrylates give high yields of monomer because of' the active radical and the a-methyl group that blocks the possibility of chain transfer. Polytetrafluoroethylene gives high yields of' monomer because the strong C-F bond is not easily broken to allow transfer. The scheme does not apply to polymers such as poly(viny1 acetate) and poly(viny1 chloride), where degradation results from the removal of' side groups rather than from chain scission. Degradation Products. Study of the pmducts of' thermal depolymerization in vacuum has shown that the chemical nature and relative amounts of' these products are remaskably independent of' the temperature and extent of' the degradation reaction Polystyrene, fbr example, degraded to 40% styrene, 2.4% toluene, and other products having an average molecular weight of' 264, at temperatures between 360 and 420C, and extents of' degradation from 4 to 100% The amounts of monomer produced by various polymers are shown in Table 6-5
CHEMICAL REACTIONS OF POLYMERS
TABLE 6-5. Percent Monomer Resulting From Thermal Degradation

-. ---
Percent Monomer Mole

100 100 78 65 57 52 44 21
Polymer Poly(methy1 methacrylate) Poly(a-methylstyrene) Polyisobutylene Polystyrene Polybutadiene SBR (butadiene-styrene copolymer) Polyisopsene Polyethylene
Weight
100 100 32 42 14 12 11
------3
Environmentally degradable polymers have received much publicity in recent years and have been reviewed (Taylor 1979)..
Radiation Chemistry of Polymers
The interaction of high-energy radiation with molecular substances involves the following sequence of events, regardless of the source of energy (photons, protons, electrons, neutrons, etc ) The molecules are first excited and ionized Secondary electrons are emitted with relatively low speeds and produce many more ions along sec or so, molecular rearrangements take place in the their tracks Within ions and excited molecules, accompanied by thermal deactivation or the dissociation of valence bonds As far as subsequent reactions are concerned, bond dissociation is the more important It leads to the production of ions or radicals whose lifetimes depend on diffusion rates and may be weeks or months in solids at low temperatures The major effects in polymers arise from the dissociation of primary valence bonds into radicals, whose existence can be demonstrated by chemical methods or by EPR spectroscopy (Chapter 9B). The dissociations of C---C and C---H bonds lead to different results, degradation and crosslinking, which may occur simultaneously
Degradation. The major result of radiation is degradation by chain scission in
1,l-disubstituted polymers, such as poly(methy1 methacrylate) and its derivatives, polyisobutylene, and poly(u-methylstyrene), and in polymers containing halogen, such as poly(viny1 chloride), poly(viny1idene chloride), and polytetrafluoroethylene. The tendency toward degradation is related to the absence of tertiary hydrogen than average C--C bond (low heat of polymerization), or unusually atoms, a weake~ strong bonds (such as C--F) elsewhere in the molecule Degradation is, of course, evidenced by decrease in molecular weight, the weightaverage molecular weight being inversely proportional to the amount of radiation received In polymers with bulky side chains, such as poly(methy1 methacrylate), extensive degradation of the side chains to gaseous products also occurs
144
POLYMERIZATION C:ONDITIONS AND P L M R REAC:TIONS OY E
of Crosslinking. Crosslinking is the predominant reaction on the ir~adiation polystyrene, polyethylene and other olefin polymers, polyacrylates and their derivatives, and natural and synthetic rubbers. It is accompanied by the formation of gel and ultimately by the insolubilization of' the entire specimen. Radiation crosslinking has a beneficial effect on the mechanical properties of some polymers and is carried out commercially to produce polyethylene with enhanced form stability and resistance to flow at high temperatures (Chapter 13A).
Other Reactions. Radiation crosslinking is often accompanied by the formation of trans-vinylene unsaturation, both reactions resulting in the fbrmation of hydrogen gas. If' the irradiation is car~ied in the presence of air, surface oxidation results out The resulting peroxides may be decomposed later in graft copolymerization (Chapter 5D).
GENERAL REFERENCES
Lenz 1967; Elias 1977; Loan 1979; Allcock 1981, Chapter 9; Odian 1981, Chapter 9; and specifically the following: Crosslinking Alfrey 1952; Temin 1966; Elias 1977, Chapter 23 5; Odian 1981, Chapters 6-6 and 9-2 Degradatzon. Sirnha 1952; Madorsky 1964; Grassie 1966; Jellinek 1966, 1977; Pinner 1967; Hawkins 1972; Elias 1977, Chapters 23 6 and 24; Kelen 1982; Schnabel 1982. Radzation Chemistry Bovey 1958; Cha~lesby1960; Chapiro 1962, 1969

1.
Discuss methods of controlling the heat evolved in vinyl polymerization, comparing the merits of bulk, solution, suspension, and emulsion polymerization and of the use of' batch, tubular, and stirred-tank reactors.
2. Discuss the merits of the above methods and equipment with respect to the purification of the polymer produced.
3. Define and describe rnicelles and discuss their role in emulsion polymerization
4. List all the necessary ingredients, and describe their functions, for the recipe fbr emulsion polymerization used in Collins (1973, Exp. 8).
5. Contrast the kinetics of suspension and emulsion polymerization and discuss the relative merits of the two methods
6. A 20 wt % aqueous solution of acrylamide is polymerized adiabatically with a redox initiator, using a starting temperature of 30C. The heat of polymer-
BIBLIOGRAPHY
145
ization is - 74 Jlmole-deg. Assume that the heat capacity of the reactor and its contents is 4 Jldeg. What is the final temperature? What is the maximum solution concentration that could be used without danger of a runaway reaction? An emulsion polymerization of styrene is run at 30C with lOI4 particles (0.2 p m in diameter) per milliliter and loi3radicals produced per milliliter per second. a. Starting with Fick's law for the diffusion flux I across a surface of' area A ,
where c is concentration and x is distance, der.ive an equation for the diffusion flux of small species into the monomer-polymer particles State your assumptions. b. Calculate whether or not monomer can be adequately supplied to the monomer-polymer particles c Justify the assumption that appreciable termination does not take place in the aqueous phase Styrene has a density of 0 90 glml and its solubility is in wate~ 0 034 wt % Assume that for small molecules D = 1 x cm2/sec At 30C, k, = 2 5 x lo7 literlmole-sec and k, = 55 liter1 mole-sec
8. Discuss batch, stirred-tank, and tubular reactors with respect to mixing, res-
idence time, thermal time, and the molecular-weight distribution produced for condensation and for free-radical polymerization
9. If a vinyl monomer is copolymerized with 2 mol % (based on vinyl groups) of a symmetrical independent divinyl monomer to 2 , = 1000, at what conversion will gelation take place? Comment on the accmacy of you1 result 10. Discuss causes fbr the initiation of', and methods for preventing, degradation and chain depolymerization.. 1 1 . Discuss the factors favoring crosslinking and degradation in the effect of ionizing radiation on polymers
Alfrey 1952 Turner Alfrey, Jr , John J Bohrer, and H Mark, Copolymer~zatzon,Interscience, New York, 1952 Allcock I981 Harry R Allcock and Frederick W Larnpe, Contemporary Polymer Chemistry, PrenticeHall, Englewood Cliffs, New Jersey, 1981 Bamford 1969 C H Barnford and G C Eastrnond, "Solid-Phase Addit~on Polymerization," Q Rev Chem Soc 23, 271 -299 (1969)
146
POLYMERIZAT'ION CONDITIONS AND P L M R REACTIONS OY E
Biesenberger 1966 J A Biesenberger and Z Tadmor, "Residence Time Dependence of Molecular Weight Distributions in Continuous Polymerizations," Polym Eng Sci 5, 299-305 (1966) Biesenberger 1983 Joseph A Biesenberger and Donald H Sebastian, Principles of Polymerizatron Engmeering, Wiley-Interscience, New York, 1983 Brrd I960 R Byron Bird, Warren E Stewart, and Edwin N Lightfoot, Transport Phenomena, John Wiley & Sons, New York, 1960 Blackley 1975 D C Blackley, Emulsron Polymerrsatron Theory and Practrce, Halsted Press, John Wiley & Sons, New York, 1975 Bovey 1958 Frank A Bovey, The Effects of lonlzrng Radiation on Natural and Synthetic High Polymers, Interscience, New York, 1958 Butler 1966 George B Butler, "Cyclopolymerization," pp 568-599 in Herman F Mark, Norman G Gaylord, and Norbert M Bikales, eds , Encyclopedra of Polymer Science and Technology, Voi 4, Wiley-Inteiscience, New York, 1966 Cassrdy 1965 Harold G Cassidy and Kenneth A Kun, Oxrdatron-Reduction Polymers (Redox Polymers), Wiley-Interscience, New York, 1965 Chapiro 1962 Adolphe Chapiro, Radratron Chemrstry of Polymerrc Systems, Wiley-Interscience, New York, 1962 Chaprro 1969 A Chapiro, Robert B Fox, Robert F Cozzens, W Brennei, and W Kupfer, "RadlationInduced Reactions," pp 702-783 in Herman F Mark, Noiman G Gaylord, and Norbert M Bikales, eds , Encyclopedra of Polymer Scrence and Technology, Vol 11, Wiley-Interscience, New York, 1969 Charlesby 1960 Arthur Charlesby, Atomic Radiation and Polymers, Pergamon Press, New York, 1960 Collrns 1973 Edward A Collins, Jan Bare;, and Fred W Billmeyer, h , Experiments in Polymer Science, Wiley-Interscience, New York, 1973 Cooper 1974 W Cooper, "Emulsion Polymenzabon," Chapter 7 in A D Jenkins and A Ledwith, eds , Reactrvity, Mechanrsm and Structure rn Polymer Chemistry, John Wiley & Sons, New York, 1974 Duck 1966 Edward W Duck, "Emulsion Polymerization," pp 801-859 in Herman F Mark, Norman G Gaylord, and Norbert M Bikales, eds , Encyclopedra of Polymer Scrence and Technology, Vol 5 , Wiley-Interscience, New York, 1966 Elias 1977 Hans-Georg Ellas, Macromolecules 2 Synthesrs and Materrais (translated by John W Stafford), Plenum Press, New York, 1977 Elrseeva 1981 V I Eliseeva, S S Ivanchev, A I Kuchanov, and A V Lebedev, Emulsron Poly merrzatron and its Applrcations in Industry (translated by Sylvia 1 Teague), Plenum Press, New York, 1981 Farber 1970 Elliott Farber, "Suspension Polymerization," pp 552-571 in Herman F Mark, Norman G Gaylord, and Norbert M Bikales, eds , Encyclopedra ofPolymer Science and Technology, Vol 13, Wiley-Interscience, New York, 1970 Gardon 1977 J L Gardon, "Emulsion Polyrnenzation," Chapter 6 in Calvin E Schildknecht, ed , with Irving Skeist, Polyrnenzation Processes, Wiley-Intersclence, New York, 1977 Grbbs 1967 William E Gibbs and John M Barton, "The Mechanism of Copolymerization of Nonconjugated Diolefins," Chapter 2 in George E Ham, ed , Vrnyl Polymeftzatron, Part I, Marcel Dekker, New York, 1967 Grassie 1966 N Grassie, "Degradation," pp 647-716 in Herman F Mark, Norman G Gaylord, and Norbert M Bikales, eds , Encyclopedra of Polymer Science and Technology, Vol 4 , WileyInterscience, New York, 1966 Hawkns 1972 W Lincoln Hawkins, ed , Polymer Stabrlrzation, Wiley-Interscience, New York, 1972 Jellrnek 1966 H H G Jellinek, "Depolymerization." pp 740-793 in Herman F Mark, Norman G Gaylord, and Norbert M Bikales, eds , Encyclopedia of Polymer Scrence and Technology, Vol 4 , Wiley-Intersc~ence, New York, 1966
BIBLIOGRAPHY
147
Jellrnek 1977 H H G Jellinek, ed , Aspects of Degra&ion and Stabrlrzatron of Polymers, Elsevier, New York, 1977 jenkms 1967 A D Jenkins, "Occlusion Phenomena in the Polymerization of Acrylonitrile and Other at Monomers," Chapter 6 in George E Ham, ed , Vinjf Polymertzatlon, P r I, Marcel Dekker, New York, 1967 Kelen 1982 Tibor Kelen, Polymer Degradatron, Van Nostrand Reinhold, New York, 1982 Kraus 1979 M Kraus and A Patchomik, "Polymeric Reagents," Chemtech 9, 118-128 (1979) Lenz 1967 Robert W Lenz, Organic Chemrstry of Syntheric High Polymers, Wiley-Interscience, New York, 1967 Loan 1979 I, D Loan and F H Winslow, "Reactions of Macromolecuies," Chapter 7 in F A Bovey and F H Winslow, eds , Macromolecules An Introductton to Polymer Sclence, Academic Press, New York, 1979 Mador sky 1964 Samuel L Madorsky, Thermal Degradatton of Organrc Polymerr, Wiley-Interscience, New York, 1964 Mat~umoto 1977 M Matsumoto, K Takakura, and T Okaya, "Radical Polymerizations in Solution," Chapter 7 in Calvin E Schildknecht, ed , with Irving Skeist, Polymerrzatron Processes, WlleyInterscience, New York, 1977 Mickley 1962 Harold S Mickley, Alan S Michaels, and Albert L Moore, "Kinetics of Precipitation Polymerization of Vinyl Chloride," / Polym k 60, 121-140 (1962) r Munzer 1977 M Munzer and E Trommsdorf, "Polymerizations in Suspension," Chaptel 5 in Calvin E Schildknecht, ed , with Irving Skeist, Polymerrzation Processes, Wiley-Interscience, New York, 1977 Nauman 1982 E B Nauman, "Synthesls and Reactor Design," Chapter 10 in Mahendra D Baijal, ed , Plastlcs Polymer Scrence and Technology, John Wlley & Sons, New York, 1982 Odran 1981 George Odlan, Prrnctples of Polymerrzatron, 2nd ed , John Wiley & Sons, New York, 1981 Prrrma 1979 Irja Piirma, ed , Emulston Polymerrzatron, Academic P~ess, New York, 1979 Prnner 1967 S H Pinner, ed , Weathering and Degradation of Plastrcs, Gordon and Breach, New York, 1967 Rrngsdorf 1965 H Ringsdorf, "Bulk Polymerization," pp 642-666 ~n Herman F Mark, Norman G Gaylord, and Norbert M Bikales, eds , Encyclopedia of Poljmer Scrence and Technology, Vol ' 2, Wiley Interscience, New York, 1965 Schlldknecht I977a C E Sch~ldknecht, "Cast Polymerizations," Chapter 2 in Calvln E Schlldknecht, ed , with Irvlng Skeist, Polymerrzatzon Procerser, Wiley.Interscience, New York, 1977 ~ckildknecht 1977b C E Schildknecht, "Other Bulk Polymer~zations,'. Chapter 4 in Calvln E knecht, ed , with Irving Skeist, Polymerczation Processes, Wiley-Interscience, New York, 1977 Schlldknechr 1 9 7 7 ~ Calvin E Schildknecht, ed , with Irving Skeist, Polymertzarion Processes, WileyInterscience, New York, 1977 Schnabel1982 Wolflam Schnabel, Polymer Degradatron Prrnctples and Practrcal Applrcatrons, Macmillan, New York, 1982 Srmha 1952 Robert Simha and L A Wall, "Klnetlcs of Chain Depolyme~lzation," 1 Phys Chem 56, 707-715 (1952) Smzth 1948a Wendell V Smith, "The Kinetlcs of Styrene Emulsion Polymerization," J Am Chem Soc 70, 3695-3702 (1948) Smtth 1948b Wendell V Smith and Roswell H Ewart, "Kinetics of Emulsion Polymerization," J Chem Phys 16, 592-599 (1948) Ihdmor 1966 Zehev Tadmor and Joseph A Biesenberger, "Influence of Segregation on Molecular Weight Distribution in Continuous Linear Polymerizations," Ind Eng Chem Fund 5, 336-343 (1966)
/ child-
148
POLYMERIZATION CONDITIONS AND P L M R REACTIONS OY E
Taylor 1979 Lynn Taylor, "Degradable Plastics: Solution or illusion^," Chemtech 9, 542-548 (1979) Temin 1966 Samuel C Temin and Allan R Shultz, "Crosslinkmg," pp 331-414 in Herman F Mark, Norman G Gaylord, and Norbert M Bikales, eds , Encyclopedia oj Polymer Science and Technology, Vol 4, Wiley-Interscience, New York, 1966 Vande~hoff 1969 John W Vanderhoff, "Mechanism of Emulsion Polymerization," Chapter 1 in George E Ham, ed , Vinyl Polymerizat~on, Vol I, Part 11, Marcel Dekker, New York, 1969
PART THREE
CHARACTERIZATION
CHAPTER SEVEN
POLYMER SOLUTIONS
A.
CRITERIA F R POLYMER SOLlJBlLlTY O
The Solution Process. Dissolving a polymer is a slow process that occurs in two stages First, solvent molecules slowly diffuse into the polymer to produce a swollen gel. This may be all that happens---if', for example, the polymer-polymer intermolecular fbrces are high because of'crosslinking, crystallinity, or strong hydrogen bonding. But if these forces can be overcome by the introduction of' strong polymersolvent interactions, the second stage of solution can take place. Here the gel gradually disintegrates into a true solution Only this stage can be materially speeded by agitation. Even so, the solution process can be quite slow (days or weeks) fbr 'materials of' very high molecular weight.
Polymer Texture and Solubility. Solubility relations in polymer systems are more complex than those among low-molecular-weight compounds, because of' the size differences between polymer and solvent molecules, the viscosity of' the system, and the effects of the texture and molecular. weight of the polymer. In turn, the presence or absence of solubility as conditions (such as the nature of' the solvent, or the temperature) are varied can give much information about the polymer; this is in fact the topic of most of' this chapter. From what has already been said, it is clear that the topology of'the polymer is highly important in determining its solubility..Crosslinked polymers do not dissolve, but only swell if indeed they interact with the solvent at all. In part, at least, the degree of'this interaction is determined by the extent of crosslinking: Lightly crosslinked rubbers swell extensively in solvents in which the unvulcanized material would dissolve, but hard rubbers, like many thermosetting resins, may not swell appreciably in contact with any solvent..
152
POL.Y&iER SOLUTIONS
The absence of solubility does not imply crosslinking, however. Other features may give rise to sufficiently high intermolecular forces to prevent solubility. The presence of' crystallinity is the common example. Many crystalline polymers, particulasly nonpolar ones, do not dissolve except at temperatures near their c~ystalline melting points. Because crystallinity decreases as the melting point is approached (Chapter 10) and the melting point is itself' depressed by the pr.esenceof the solvent, solubility can often be achieved at temperatures significantly below the melting point. Thus linear polyethylene, with crystalline melting point T, = 135"C, is soluble in many liquids at temperatures above 100C, while even polytetrafluoroethylene, Tm = 325"C, is soluble in some of' the few liquids that exist above 300C More polar crystalline polymers, such as 66-nylon, Tm = 265"C, can dis,solve at room temperature in solvents that interact strongly with them (for example, to form hydrogen bonds).. There is little quantitative information about the influence of branching on solubility; in gene~al, branched species appear to be more readily soluble than their linear counterparts of the same chemical type and molecular weight.. Of' all these systems, the theory of solubility, based on the thermodynamics of polymer solutions, is highly developed only fbr linex polymers in the absence of crystallinity. This theory is described in Sections C and D . Here the chemical nature of the polymer is by far. the most important determinant of'solubility, as is elucidated in the remainder of this section.. The influence of' molecular weight (within the polymer range) is fir less, but it is of great importance to fractionation processes (Sections D and E), which yield information about the distribution of molecular weights in polymer samples,
Solubility Parameters. Solubility occurs when the free energy of' mixing
AG = AH - TAS
is negative. It was long thought that the entropy of' mixing AS was always positive, and therefore the sign of AG was determined by the sign and magnitude of the heat of mixing AH.. For reasonably nonpolar molecules and in the absence of hydrogen bonding, AH is positive and was assumed to be the same as that derived for the mixing of small molecules.. For this case, the heat of' mixing per unit volume can be approximated (Hildebr'and 19.50) as
where v is volume fraction and subscripts 1 and 2 refer to solvent and polymer, respectively. The quantity is the cohesive energy density or, for small molecules, the energy of vaporization per unit volume. The quantity 6 is known as the solubility parameter. (This expression for the heat of mixing is one of'several alternatives used in theories of the thermodynamics of polymer solutions; in Section C, AH is written in a different but equivalent way.) The value of the solubility.-parameter approach is that 6 can be calculated for both polymer and solvent As a first approximation, and in the absence of' strong
CRITERIA FOR POL.YMER SOLUBILITY
153
interactions such as hydrogen bonding, solubility can be expected if 6, - 6, is less

than 3.5-4.0, but not if it is appreciably larger
This approach to polymer solubility, pioneered by Bur~ell(1955), has been extensively used, particularly in the paint industry A few typical values of 6, and 6, are given in Table 7- 1; for polymers, they are the square roots of the cohesiveenergy densities of Table 1-4 Extensive tabulations have been published (Bur~ell 1975, Hoy 1970). Perhaps the easiest way to determine 62 for a polymer of known structure is by the use of the molar-attraction constants E of Table 7-2,
where values of E are summed over the structural configuration of the repeating unit in the polymer chain, with repeat molecular weight M and density p The original solubility parameter approach was developed for nonpolar systems Modifications to include polarity and hydrogen bonding have led to three-dimensional solubility-parameter schemes, which lack the simplicity of the single-parameter method but are more widely applicable Despite its shortcomings, the concept is nevertheless still extremely useful and should not be abandoned without test In contIast to the above considerations of the thermodynamics of dissolution of polymers, the Iate of this step depends primarily on how rapidly the polymer and the solvent diffuse into one another (Ueberyeiter 1962, Asmussen 1962). Solvents that promote rapid solubility are usually small, compact molecules, but these kinetically good solvents need not be thermodynamically good as well. Mixtures of a kinetically good and a thermodynamically good liquid are often very powerful and rapid polymer solvents
TABLE 7-1. Typical Values of the Solubility Parameter 6 for Some Common Polymers and Solventsd
Solvent
n-Hexane Carbon tetrachloride Toluene 2-Butanone Benzene Cyclohexanone Styrene Chlorobenzene Acetone Tetrahydrofuran Methanol Water
"Coll~ns (1973)
6, [ ( J / c ~ ~ ) " ~ ]
14 8 17 6 18 3 18 5 18 7 19 0 19 0 19 4 19 9 20 3 29.7 47 9
Polymer
Polytetrafluoroethylene Poly(dimethy1 siloxane) Polyethylene Polypropylene Polybutadiene Polystyrene Poly(methy1 methacrylate) Poly(viny1 chloride) Poly(viny1 acetate) Poly(ethy1ene te~ephthalate) 66-Nylon Polyacrylonitrile
62[(J/cm3)'/2]
12 7 14 9 16.2 16 6 17 6 17 6 18 6 19 4 21 7 21 9 27 8 31 5
--
154
POLYMER SOL.UTIONS
TABLE 7-2. Molar Attraction Constants Ea
Group
E [(J-~m~)"~/rnolt]
Group
E [(J-cm3)"*/mole]
259 249 173 -CH=aromatic ,C=a omatic -0-ether , acetal .-0-epoxide -COO\ ,C=O
\
.-sclz
NCO
C1 primary C1 secondary C1 aomatic F Conjugation
cis
trans Six-membered ring ortho meta Pala
-CHO (C0)20 -OH+ OH aromatic -H acidic dimes
GENERAL. REFERENCES
Hildebrand 1950; Gardon 1965; Hoy 1970; ASTM D3132; Burrell 1975; Morawetz 1975, Chapter 2; Olabisi 1979, Chapter 2.3; Snyder 1980.
B.
CONFORMATIONS O F DISSOLVED POLYMER CHAINS
As specified in Chapter 1, those arrangements of' the polymer chain differing by reason of' rotations about single bonds are termed conformations.t In solution, a polymer molecule is a randomly coiling mass most of' whose conformations occupy many rimes the volume of its segments alone. The average density of' segments within a dissolved polymer molecule is of' the order of' 10-.4-10-5 g/cm3. The size of the molecular coil is very much influenced by the polymer-solvent interaction forces. In a thermodynamically "good" solvent, where polymer-solvent contacts
?Here we differ from some authorities, notably F l o ~ y (1969), who prefers the use, well established in statistical mechanics, of the term configuration here We follow the convention of organic chemistry in which configu~ationdesignates stereochemical arrangement (see Chapter IOA)
CONFORMATIONS O DISSOLVED POLYMER CHAINS F
FIG. 7-1. Model of one of the many conformations of a random-coil chain of 1000 links (T~eloar 1958)
are highly favored, the coils are relatively extended In a "poor" solvent they are xelatively contracted It is the purpose of this section to describe the conformational properties of both ideal and real polyme~chains The conformations of polymer chains in the crystalline state are considered in Chapter 10B The importance of the random-coil nature of the dissolved, molten, amorphous, and glassy states of high polymers cannot be overemphasized. As the following chapters show, many important physical as well as thermodynamic properties of high polymers result f ~ o m characteristic structural feature The random coil this (Fig 7-1) arises from the relative freedom of rotation associated with the chain bonds of most polymers and the formidably large number of conformations accessible to the molecule One of these conformations, the fully extended chain (often an all-trans planar zigzag carbon chain; Fig 7-2), has special interest because its length, the contour length of the chain, can be calculated in a straightforward way In all other cases the size of the random coil must be expressed in terms of statistical parameters such as the root-mean-square distance between its ends, ( 7 ) ' l 2 , or its radius of gyration, the-root-mean-square distance of the elements of the chain from its center of gravity, ( s ' ) " ~ . For linear polymers that are not appreciably extended beyond their most probable shape, the mean-square end~o-end distance and the square of the radius of gyration are simply related: 7 = 6s2 For extended chains 7 > 6s2 The use of the radius of gyration is sometimes prefer~ed because it can be determined experimentally, as described in Chapter 8E
The Freely Jointed Chain. A simple model of a polyme~chain consists of a series of x links of length 1 joined in a linear sequence with no restrictions on the angles between successive bonds The probability W that such an anay has a given end.to-end distance r can be calculated by the classical random-flight method (Ray-
156
POLYMER SOLUTIONS
FIG. 7-2.
The fully extended all-t~ans conformation of the carbon-carbon chain (Billmeyer 1969)
leigh 1919). The most important result of' the calculation is that the end-to-end distance is proportional to the squase root of' the number of' links:
(The subscript f indicates the random-flight end-to-end distance ) Thus ( r f 2 ) ' 1 2 is proportional to MIt2, or r f / M is a characteristic property of the polymer chain structure, independent of' molecular weight or length The distribution of' end-to-end distances over the space coordinates W(.r,,y,z)is given by the Gaussian distribution function shown graphically in Fig. 7-3. This is the density distribution of' end points and shows that if' one end of'the chain is taken at the origin, the probability is highest of finding the other end in a unit volume
FIG. 7-3.
Gaussian distribution of end-to-end distances in a random coil of lo4 links each 2 5 A long; W ( x ,y,z) is expressed in angstroms to the inverse third power and r in angstroms
C.ONFORMATIONS OF DISSOLVED P L M R CHAINS OY E
157
near the origin This probability decreases continuously with increasing distance from the origin On the other hand, the probability of finding a chain end within the volume of a spherical shell between distances r and r dr from the origin has a maximum, as shown in Fig 7-4. It should be noted that the Gaussian distribution does not fall to zero at large extensions and so must fail to describe the actual conformations of the chain where r nears the contour length. Over most of the range of interest for the dilute-solution properties of polymers, this is of little consequence; better approximations are available where needed, as in the treatment of rubber elasticity (Chapter 11B). The freely jointed or random-flight model seriously underestimates the true dimensions of real polymer molecules for two reasons: First, restrictions to completely free rotation, such as fixed valence angles, the corIect weights for trans and gauche conformations, and statistical deviations from ideal trans and gauche states, and other short-range interactions described below, lead to larger dimensions than calculated above; and second, long-range interactions resulting from the inability of chain atoms far removed from one another to occupy the same space at the same time result in a similar effect
Short-Range Interactions. One of the triumphs of modern polymer chemistry is the extent to which Flory and his collaborators (Flory 1969), among others, have calculated completely and accurately the effect of short-range interactions on the dimensions of random-coil polymers Several effects are involved The restriction to fixed bond angle 0 expands the chain by a factor of [(I - cos 0)/(1 cos 0)]"*, equal to for carbon-carbon bonds Restricted rotation, whether resulting from steric hindrances and resulting potential-energy barriers or from resonance leading to rigid plana conformations, increases dimensions still more Finally, conformations that would place two atoms close together along the chain too close to one another are not allowed, leading to further expansion. The most important of these is the so-calledpentane znterference (Taylor 1948) between the first and fifth chain atoms in a sequence Bovey (1979) summarizes these conformational features The net results of these short-range interactions can be expressed as a charac-
FIG. 7-4.
Radial distribution of end-to-end distances for the coil of Fig 7-3 H e ~ e ( r ) is expressed W in angstrom-' units
158
POLYMER SOLUTIONS
teristic ratio of' the square of' the actual chain dimensions in the absence of longrange interactions (called the unperturbed dimensions and given the symbol ?) and the square of'the random-flight end-to-end distance, 12x Typical values of' the characteristic ratio for some common polymers are given in Table 7-3. It should be pointed out that despite short-range interactions, the chains are still random in overall conformation if long enough..
Long-Range Interactions and the Excluded Volume. Correction of' polymer chain dimensions for short-range interactions still fails to eliminate conformations in which two widely separated chain segments occupy the same space. Each segment of' a real chain exists within a volume from which all other segments are excluded The theoretical calculation of' the excluded volume and its effect on the dimensions of the polymer chain has remained a major unsolved problem in polymer science for many years.. A statistical approach has been more fruitful and has allowed the simulation of'chain confbrmations by digital computer up to chain lengths of several hundred segments. The results of these calculations of' "self avoiding walks" and experimental studies (light scattering, dilute-solution viscosity, sedimentation; Chapter 8C-E) lead to the result that the mean-square end-to-end distance is proportional to . x o 6 . This exponent was first calculated many years ago by Flory (19.53). It is also predicted by the application of scaling concepts (de Gennes 1979, Sections C and D) . Practically, the effect of'the long-range interactions is to cause a further.expansion of' the chain over its unperturbed dimensions, since more of the compact conformations, with small values of' 7, must be excluded. The actual dimensions of. the real chain can be expected to exceed its unperturbed dimensions by an expansion hctor a ; thus (v2)112 a ( 3 ) ' l 2 The value of' a depends on the nature of. the =
TABLE 7-3. The Characteristic Ratio r,2/i2x for Some Common Polymers, Evaluated at the Limit of High Chain Lengthd 102 -
Polymer
Polyethylene Polystyrene (atactic) Polypropylene (isotactic) Poly(methy1 methacrylate) Atactic Isotactic Syndiotactic Poiy(ethy1ene oxide) 66-Nylon
"Flory(1969)
12x
67 10 0 57 69 93 -7 40 59
THERMODYNAMICS OF POLYMER SOLUTIONS
159
solvent: A solvent for which a is large is said to be a thermodynamically "good" solvent for that polymer, and vice versa In a sufficiently poor solvent, or at a sufficiently low temperature (since solvent power and a vary with the temperature), it is possible to achieve the condition a = 1, where the chain attains its unperturbed dimensions This special point is called the Flory temperature O; a solvent used at T = O is called a O solvent ' The calculation of a and O and their relation to thermodynamic quantities is treated in Section C. Neutron-scattering experiments (Chapter 8C) show that the above considerations apply as well to chain conformations in the melt and the glassy state
Nonlinear Chains. A branched molecule occupies a smaller volume than a linear one with the same number of segments, that is, the same molecu& weight. It 5 convenient to express this diminution of size as a factor g = s2 (branched)/s2 (linear), which can be calculated statistically for various degrees and types of branching (Zimm 1949). For random branching, five trifunctional branch points per molecule reduce g to about 0 70. The change in size is the basis of a method to measure branching (Chapter 8E) Billmeyer (1972, 1977) has used a "pop-it" bead model to illustrate the dimensions of polymer chains.
GENERAL. REFERENC,ES
Flory 1953, 1969; Volkenstein 1963; Birshtein 1966; Bovey 1969, 1979, Chapter 3 9; Lowry 1970; Hopfinger 1973; Vollmert 1973, Chapter 32; Morawetz 1975, Chapter 3; Elias 1977, Chapter 4; de Gennes 1979, Chapter 1
C.
T'H,ERMODYNAMIC,S OF POLYMER SOLU'TIONSt
The behavior of polymers toward solvents is characteristic and different from that of low-molecular-weight substances The size and conformations of dissolved polymer molecules require special theoretical treatment to explain their solution properties Conversely, it is possible to obtain information about the size and shape of polymer molecules from studies of their solution properties The first group of properties of interest includes those depending upon equilibrium between two phases, one or both of which is a solution of the polymer. In this section we discuss situations in which one phase is pure solvent, and in Section D solubility phenomena in which both phases contain polymer.
1For 1,eviewsof thermodynamics as background to this subject see Hildeb~and (1950) and Lewis (1961)
POLYMER SOLUTIONS
Thermodynamics of Simple Liquid Mixtures

From the condition for equilibrium between two phases may be derived relations such as that fbr the free energy of' dilution of a solution:
where AG is the free energy of' dilution resulting from the transfer of one molecule to of liquid A from the pure liquid state with vapor pressure pAO a large amount of' solution with vapor pressure pA This expression is written in terms of one of the colligative properties of' solutions: vapor-pressure lowering, freezing-point depression, boiling-point elevation, and osmotic pressure. The values of' these properties cannot be related to the composition of the system by pure thermodynamic reasoning It is necessary to know the type of variation of one property with concentration; the variation of the others may then be deduced
ideal Solutions. In the simplest type of mixing, the molecules of' components A and B have roughly the same size and shape and similar force fields. They may then form an ideal solution, defined as one in which Raoult's law is obeyed, This law states that the partial vapor pressure of each component in the mixture is proportional to its mole fraction.. Therefore
where n denotes mole fraction, and Eq 7-1, for example, becomes
The total free energy of mixing is
The conditions for ideal mixing imply that the heat of mixing AH = 0; that is, the components mix without change in enthalpy.. Since AG = AH - T AS, the entropy of' mixing is given by
which is positive for all compositions, so that, by the second law, spontaneous mixing occurs in all proportions
IHERMODYNAMICS OF POLYMER SOLUTIONS
161
Other 'Types of Mixing. In practice, few liquid mixtures obey Raoult's law. Three types of deviations are distinguished:
"Athermal" solutions, in which AH = 0 but AS is no longer given by Eq 7-5 .. b. "Regular" solutions, in which AS has the ideal value but AH is finite. c. "Irregular" solutions, in which both AH and AS deviate from the ideal values..
a.
It is usually found in systems of similar-sized molecules that AS is nearly ideal when AH = 0; therefore athermal solutions are nearly ideal Many mixtures are fbund fbr which AH is finite, however. Such cases arise when the intermolecular force fields around the two types of molecule are different. Expressions for the heat of mixing may be derived in terms of' the cohesive-energy density of these force fields (Section A )
Entropy and Heat of Mixing of Polymer Solutions Deviations From Ideal Behavior. Polymer solutions invariably exhibit large deviations from Raoult's law, except at extreme dilutions, where ideal behavior is approached as an asymptotic limit At concentrations above a few percent, deviations from ideality are so great that the ideal law is of little value for predicting or correlating the thermodynamic properties of polymer solutions Even if the mole fraction is replaced with the volume fraction, in view of the different sizes of the polymer and solvent molecules, there is not a good correlation with the expe~imental results Entropy of Mixing. Deviations from ideality in polymer solutions arise largely from small entropies of mixing These are not abnormal, but are the natural result of the large difference in molecular size between the two components They can be interpreted in terms of a simple molecular model The molecules in the pure liquids and the mixture are assumed to be representable without serious error by a lattice A two-dimensional ~epresentationof such an arrangement for nonpolymer liquids is shown in Fig 7-5a. Whereas the molecules of a pure component can be arranged in only one way on such a lattice, assuming that they cannot be distinguished from one another, the molecules of a mixture of two components can be arranged on a lattice in a large but calculable number of ways, W. By the Boltzmann relation, the entropy of mixing AS = k In W Equation 7-5 results for the case of molecules that can replace one another indiscriminately on the lattice. It is assumed that the polymer molecules consist of a large number x of chain segments of equal length, flexibly joined together Each link occupies one lattice site, giving the arrangement of Fig. 7-5b The solution is assumed to be concentrated enough that the occupied lattice sites are distributed at ando om rather than lying in well separated regions of x occupied sites each. It can now be seen qualitatively
162
POLYMER SOLUTIONS
11_-1-1
-X X X X X X
Y
(a)
(b)
liquids and (b) a polymex molecule located FIG. 7-5. Two-dimensional representation of (a) nonpolyme~ in the liquid lattice
why the entropy of' mixing of' polymer solutions is small compared to that with normal solutes.. There are fewer ways in which the same number of lattice sites can be occupied by polymer segments: Fixing one segment at a lattice point severely limits the number of' sites available for the adjacent segment. The approximate calculation of' W fbr such a model is due separately to Flory (1942) and Huggins (1942a,b,c); their results (which differ only in minor detail) are known as the Flory-Huggins theory of' polymer solutions. The entropy of' mixing is analogous to that given in Eq 7-5 for simple liquids: For polymer solutions
where subscript 1 denotes the solvent and 2 the polymer, and u , and u, are volume fractions defined as
Heat and Free Energy of Mixing. The heat of mixing of' polymer solutions is analogous to that of' ordinary solutions: i'
iThe nomenclature of Flory is uszd throughout Huggins writes p,O instead of expresses AH in terms of the solubity parameter, as in Section A
x,, and
Hildeb~and
THERMODYNAMICS OF P L M R SOL.UTIONS OY E
163
where X, characterizes the interaction energy per solvent molecule divided by kT. Combining Eqs.. 7-6 and 7-8 gives the Flory-Huggins expression for the free energy of mixing of a polymer solution with normal heat of mixing:
From this expression may be derived many useful relations involving experimentally obtainable quantities For example, the partial molar free energy of mixing is
and fiom it is obtained the osmotic pressure
where V, is the molecular volume of the solvent If the logarithmic term in Eq '71I is expanded and only low powers of u2 are retained, the fbllowing equation fbr the osmotic pressure is obtained:
In expansions of this type the coefficient of u,Z is known as the second vzrial coefficient A2 (see also Chapter 8B) Here it is defined as
where U2 is the specific volume of the polymer. This simple treatment does not describe the temperature and molecular-weight dependence of A, in good agreement with experiment
Experimental Results With Polymer Solutions
The first system for which accurate experimental results were compared with the Flory-.Huggins theories was that of rubber in benzene (Gee 1946, 1947). Free energies for the system were calculated from vapor-pressure measurements by Eq '7-,l, and heats of' solution were measured in a calorimeter. The predicted concent~ation dependence of the partial molar heat of dilution
164
P L M R SOLUTIONS OY E
was not observed. Despite this fact, the entropy of mixing, calculated fiom the heat and free energy, was in h i s agreement with theory except for the dilute-solution region The properties of other systems do not in general confbrm as well to the predictions of the theory For example, the anticipated independence of X , of' v, is usually not observed (Fig 7-6) (but see the coxresponding-state theories described below).
Dilute Solutions Flory-Krigbaum T'heory. The lattice model used in the Flory-Huggins treatment neglects the fact that a very dilute polymer solution must be discontinuous in structure, consisting of' domains or clusters of polymer chain segments separated on the average by regions of polymer-free solvent Flory and Krigbaum (1950) assume such a model in which each cloud of segments is approximately spherical, with a density that is a maximum at the center and decreases in an approximately Gaussian function with distance from the center. Within the volume occupied by
FIG.
Experimentally observed variation of X , with concentration: curve 1, poly(dimethy1siloxane) in benzene (Newing 1950); curve 2, polystyrene in benzene (Bawn 1950); curve 3, rubber in benzene (Gee 1946, 1947); curve 4, polystyrene in toluene (Bawn 1950)
THERMODYNAMICS OF POLYMER SOLUTlONS
165
the segments of one molecule, all other molecules tend to be excluded Within such an excluded volume (Section 3 ) occur long-range intramolecular interactions whose thermodynamic functions can be derived The partial molar heat content, entropy, and free energy of these inter actions are, respectively,
where, by comparison with Eq.. 7-12,
It is convenient to define as a parameter the Flory temperature @ such that @ QJ, It follows that
K,T/
and that at the temperature T = O the partial molar free energy due to polymer solvent interactions is zero and deviations from ideal solution behavior vanish The excluded volume becomes smaller as the solvent becomes poorer and vanishes at T = @, where the molecules interpenetrate one another freely with no net interactions At temperatures below @ they attract one another and the excluded volume is negative If the temperature is much below @, precipitation occurs (Section D) The parameter a can be evaluated in terms of thermodynamic quantities:
where C,,, lumps together numerical and molecular constants This equation, which involves the assumption of validity of theories involving intermolecular interactions as well as intramolecular interactions, must be accepted with some reservations.lt predicts that a increases without limit with increasing molecular weight; since roZ is proportional to M, r2/M should increase with M. It also leads to the fact that, at T = 0 , a = 1 and molecular dimensions are unperturbed by intramolecular interactions Since a depends on the entropy parameter $, it is larger in better solvents
Corresponding-State Theories
From what has been said, it can be seen that both the Flory-Huggins and FloryKrigbaum theories have serious shortcomings Both are based on and conserve the important features of the theories of "regular" solutions of small molecules. Only the appropriate combinatorial entropy of mixing has been modified to fit the polymer
166
POLYMER SOLUTIONS
case The most important assumption that is retained is that there is no volume change on mixing The deviations between theory and experiment for the interaction parameter X , led to its reinterpretation as a combined entropy and enthalpy parameter, as in Eqs 7-1 1 and 7-12 The entropy term was considered to be a small negative cor~ection, corresponding to the plausible idea of a small increase in order as polymer-polymer and solvent-solvent interactions were replaced with polymer-solvent interactions Typical values of $, and K, show that this is not the case, for the entropic contribution to X , is positive and much larger than the enthalpic term Still other failures of the traditional theories of the thermodynamics of polymer solutions are discussed in Section D These difficulties have been overcome in cor~esponding-statetheories first developed around 1952 by Prigogine (1957) and his co-workers, and extended and put into practice by Flory (Flory 1965, Eichinger 1968~-d),and others The major new factor in the theories is the recognition of the dissimilarity in the free volumes of the polymer and the solvent as a result of their great difference in size, the usual solvent being much more expanded than the polymer Mixing is not unlike the condensation of a gas (the solvent) into a dense medium (the polymer) The total volume change on mixing is usually negative and is accompanied with a negative (exothermic) AH and a negative contribution to AS Thus the total AH on mixing consists of the usual positive interaction term (Eq 7-8) and the new negative term; both contribute to a new interaction parameter like X , Similarly, the entropy of mixing consists of the Flory-Huggins combinatorial term (Eq 7-6) plus the new negative teIm; only the latter contributes to the new X . The relative magnitude of these contributions is discussed further in Section D Their exact mathematical formulation is beyond the scope of this section It can be said, however, that a basic assumption of the free-volume theories is that all liquids, polymers when noncrystalline, and mixtures of these follow the same reduced equation of state In Flory's formulation (see, for example, Carpenter 1970) this can be written as
which simplifies fur.ther at low pressures, including atmospheric, where ~5 -+ 0 Herep, p, f a r e the ratios of the real pressure, volume, and temperature to reduction parameters that can be evaluated from the thermal coefficient of expansion and the isothermal compressibility It is possible to calculate reduced parameters for mixtures as well as fbr the pure components; from the resulting equations of state we can evaluate the volume change on mixing, the heat and entropy of mixing, and such details as the dependence of the new x on concentration. It has been found that volume changes on mixing can be predicted to within 10--15% of the experimental values, and the correct variation of x with concentration is predicted, in contrast to the result of the Flory-Huggins theory discussed above.
THERMODYNAMICS OF POLYMER SOLUTIONS
Scaling Concepts
From the previous discussion we know the properties of polymer chains in both dilute and concentrated solutions Dilute nonoverlapping coils in a good solvent are swollen, with a size proportional to xO @'. the other hand, in very concentrated On solutions or melts the chains are essentially ideal, with a size proportional to xO and they interpenetrate one another very strongly. What has happened in between? The only available theory for a long time was that of Flory and Huggins, and it is not adequate at low and intermediate concentrations The solution has now been elucidated both experimentally, through neutron-scattering experiments (Chapter SC), and theoretically, through the application of scaling concepts (de Gennes 1979). The two extreme concentration regions described above are separated by a region of t~ansition properties at a concentration c = c* where the chain coils begin to in overlap (Fig. 7-7) The scaling properties of c* are important It is comparable to the local segment concentration within a single coil:
In terms of the volume fraction, uz*

X
x-OaO
and is quite small
for large
When c is small compared to c* (the dilute-solution region), Flory (1953) showed that the equation of state takes the form
which leads by scaling concepts to A, - x - O 20, in good agreement with experiment In the transition region, where u2is in the broad region between u,* and 1, application of scaling concepts to the osmotic pressure leads (des Cloizeaux 1975) to nlkT v2225, in contrast to the Flory-Huggins prediction of u,Z This has been confirmed experimentally (Daoud 1975)
c < c*
= C*
c>cX
(a) (b) (c) FIG. 7-7. Schematic of chain conformations in the (a) dilute, (b) transition, and (c) semi-dilute concentration regions (From de Gennes 1979 Copyright 1979 by Cornell University Used by permission of the publisher, Cornell University Press )
POLYMER SOLUTIONS
FIG. 7-8. Qualitative behavior of the parameter X, according to the corresponding-state theories: curve 1, contribution from enthalpic polymer-solvent interactions (as predicted by the FloryHuggins theory); curve 2, contribution due to free-volume dissimilarity between polymer and solvent; curve 3, total x,, sum of 1 and 2 (Patte~son1969)
GENERAL REFERENCES
Flory 1953; Huggins 1958; Carpenter 1970; Tager 1978; de Gennes 1979, Chapter 3; Kwei 1979, Chapter 4 2
D.
PHASE EQUILIBRIUM IN POLYMER SOLUTIONS
Polymer-Solvent Miscibility
This section is concerned with the equilibrium between two liquid phases, both of' which contain amorphous polymer and one or more solvents. The treatment of cases involving a crystalline polymer phase is more complex and is given in part in Section E and in part in Chapter 10C. When the temperature of' a polymer solution is raised or lowered, the solvent eventually becomes thermodynamically poorer. Finally a temperature is reached beyond which polymer and solvent are no longer miscible in all proportions. At more extreme temperatures the mixture separates into two phases. Such phase separation may also be brought about by adding a nonsolvent liquid to the solution. In either case it takes place when the interaction parameter X, exceeds a critical value near 4 (see below) It was shown in Section C that the Flory-Huggins theory attributes X , to polymersolvent interactions alone, and predicts it to increase monotonically as the temperature decreases (Fig 7-8). Thus phase separation is predicted to take place only on lowering the temperature, the phase diagram looking like that in the lower portion of' Fig. 7-9.. The maximum temperature fbr phase separation is designated the upper critical solution temperature Although it was not frequently observed
P A E EQUILIBRIUM IN POlYMER SOLUTIONS H S
FiCt 7-9. Phase d~agram the system polystyrene In acetone, showlng phase separation at upper and of lower crltlcal solutlon temperatures for polymer of the molecular welghts lndlcated Repnnted wlth peImlsslon from Slow 1972 Copyright 1972 by the Amerlcan Chem~calSoclety )
until recent years (Freeman 1960), it is now recognized that phase separation mvariably occurs also when the temperature is raised until a lower critical solution temperaturef (Fig 7-9, upper area) is reached. This phenomenon is explained by .he corresponding-state theories of polymer solutions described qualitatively in Section C. The contribution to X, from the free-volume dissimilarity between polymer and solvent is an increasing function of the temperature (Fig. 7 8) The total interaction parameter thus goes through a minimum, and two critical values of X , 1Ie accessible. Each has the characte~of a @ temperature (see below); at each the
The perverse names of the upper and lowe1 critical solution temperatures relate more di~ectlyto an liternative case (the common example is nicotine in water) in which the upper critical solution temperature s higher than the lower one and the phase diagram consists of a region of immiscibility completely urrounded by a one-phase region
170
POLYMER SOLUTIONS
second virial coefficient approaches zero and the polymer chain approaches its unperturbed dimensions. For the remainder of' this section reference is made only to phase separation at an upper critical solution temperature. The predictions of'the Flory-Huggins theory remain unchanged except for the sign of the temperature dependence in the case of' a lower critical solution temperature
Binary Polymer-Solvent Systems. The condition for equilibrium between two phases in a binary system is that the partial molar free energy of' each component be equal in each phase. This condition corresponds to the requirement that the first and second derivatives of' AG, (Eq. 7-10) with respect to v, be zero. Application of this condition leads to the critical concentration at which phase separation first appears:
This is a rather small volume fraction; fbr a typical polymer (x = lo4), vZL= 0.01 The critical value of X, is given by
The critical value of X, exceeds d by a small increment, depending on molecular weight, and at infinite molecular weight equals 1. The temperature at which phase separation begins is given by
where C is a constant for the polymer-solvent system Thus 1/T, (K) varies linearly square root of molecular weight. The Flory tempeIature O is temperature in the limit of infinite molecular weight of the theory are in agreement with experiment. The temperature on polymer concentration is shown in Fig and v2, correspond to the maxima in the curves. In the point the polymer concentration in the dilute phase is extremely small The phenomenon of the coexistence of two liquid phases, one of which is a dilute solution and the other nearly pure solvent, is called coacervatzon (Section E) The discrepancy between theory and experiment results (Koningsveld 1968) from the effect of finite molecular-weight distribution breadth in the polymer samples used
HAS EQUILIBRIUM IN POLYMER SOLUTIONS
171
=IC.7-10. Phase diagram showing p~ecipitationtemperature as a function of polyme~concentration fbr f~actions polystyrene in diisobutyl ketone (Shultz 1952) Higher-molecu1a1-.weight of fractions precipitate at higher temperatures (solid lines, experimental; dotted lines, theoretical)
The dependence of T, upon molecular weight is shown in Fig 7-1 1 The curves n e accurately linear, and the Flory temperatures @ obtained from the intercepts igree within experimental erxor (<lo) with those derived from osmotic measurenents, tpkin'g @ to be the temperature whexe A, is zero Precipitation measurements ,n a series of sharp fractions offer perhaps the best method of determining @ Although a thorough discussion of ternary systems is outside he scope of this book, a few cases have particular interest The most commonly 3ncountered system is that of polymer, solvent, and nonsolvent The phase relations Ire conveniently displayed in the usual ternary diagram (Fig. 7-12) The position ~f the bznodal curve, along which two phases are in equilibrium, depends upon molecular weight; the limiting critical point at infinite molecular weight is the analog 3 @ in a two-component system f
rernary Systems. Multicomponent Systems. The theory of phase separation in systems comprising a heterogeneous polymer in a single solvent is developed with the simplifying assumption that the interaction constants x for all values of x are identical; only
POLYMER SOLUTIONS
phase separation on molecular welght f o ~ (0) FIG. 7-1 1. Dependence of cntlcal temperature f o ~ polystyrene In cyclohexane, and (6)polylsobutylene In d~lsobutyl ketone (Shultz 19.52)
the size parameter x itself varies from one molecular species to another In theory, all details ofthe system could be calculated from a knowledge ofthe size distribution of' the polymer and the equilibrium conditions. In practice, this would be an enosmous task; but the principal point of interest, namely, the efficiency with which moleculas species are separated between the two phases, is simply derived. It is given by the equation
where v,' and v, are, respectively, the concentrations of species x in the precipitated or more concen ,ated phase (primed) and the dilute phase (unprimed), and
ere v 2 ' and v, ase the total polymer concentrations in the two phases, and 2, is the number average of'x. The parameter a thus depends in a complicated way on the relative amounts of all the polymer species in each phase. The following conclusions may be drawn from the theory:
a.
A part of any given polymer species is always present in each phase. In fact, every species is actually more soluble in the precipitated phase, that is, v,' > v, for all values of .x.
Polymer Nonsolvent
FIG. 7-1 2. Ternary phase diagram for a polymer-solvent-nonsolvent system, showing phase separation boundaries for the indicated values of x Dotted lines aie tie lines for x = 100, connecting points representing the compositions of pairs of' phases in equilibrium Dashed lines represent constant solvent-nonsolvent ratios ( 8 , critical points; 0, precipitation threshold for x = 100)
b. Since u depends on the concentrations of all species and on the details of

their weight distribution, the result of a fractionation cannot be predicted in advance unless the details of the distribution are known The Flory-Huggins theory predicts that only the number average 2, for each phase appears in cr, while an alternative analysis (Stockmayer 1949) expresses u in terms of higher moments of the distribution for the whole polymer Fractionation is significantly more efficient in very dilute solutions: If R = v ' I v , the ratio of the volumes of the precipitated and dilute phases, the fraction f , of the constituent x in the dilute phase is
f
c.
"
-1 -
1 t.Reux
and that in the precipitated phase is
174
POLYMER SOLUTIONS
If R 4 1, most of' the smaller species remain in the dilute phase because of its greater volume. The larger partition factor v,'lv, for the higher- molecular^.. weight species causes them to appear selectively in the precipitated phase despite its smaller volume The resulting inefficient nature of the fractionation process is illusbated in Figs 7-13 and 7-14 Figure 7-13.shows in the three lower curves the distribution of polymer species left in the dilute phase after precipitation at various values of R The initial distribution is shown in the upper curve The increase in efficiency as R becomes smaller is apparent Figure 7-14 shows the overlapping distribution curves resulting even when R is unusually small It is clear that separation into fractions that are close to monodisperse is a most difficult and time-consuming process It should be noticed that the distribution curves of the fractions in Fig 714 become narrower at lower molecular weights The foregoing discussion refers to fractionation by cooling from a single solvent Similar considerations apply to fractionation at constant temperature in a solventnonsolvent system A difference arises in that the solvent composition is different in the two phases This difference should lead to more eff~cientfractionation in solvent-precipitant mixtures Experimental evidence for such an effect has not been reported
FIG. 7-13.
Calculated separation of a polymer having the molecular-weight distribution shown in the upper curve with R = 0 1, 0 01, and 0 001, f, chosen to be t at x = 200 in each case (Flory 1953)
175
FIG. 7-14.
Calculated distributions of a series of elght fractions separated f ~ o m distribution of the the upper curve (Flory 1953); R = 0 001 In each case Dashed curves represent the polymer remaining in the dilute phase after each successive precipitation
Scaling Concepts. Figure 7-15 shows one important result of the application of
scaling concepts to polymer physics It is a typical phase diagram of a polymer solution, with the coordinates concentration c and reduced tempeIature T = (T - @)/ 6 , showing the regions in which the application of scaling concepts becomes 3 important The dashed curve in the negative region of T (region IV) is a coexistence or phase-separation curve of the type shown in Fig 7-10 Region I' is the O region, of dilute solutions near T = O and T = 0 Region I is the dilute region, limited by a value of T proportional to ( ? ) - ' I 2 , and the line of c* the critical concentration - "*. at which the chains begin to overlap; c* is also propor.tiona1 to (7) Region I1 is the'semidilute region within which chain overlap becomes more and more irnportant The concentration c** is, as seen, proportional to T Region 111 is the semidilute and concent~atedO region, not yet studied in detail In several of these regions the Flory--Huggins theory describes the situation adequately The critical point for phase separation, for example, is located correctly, but the shape of the coexistence cuwe in that region is not properly described by that theory The properties of the polymer-rich phase are adequately described by the Flory-Huggins theory
Polymer-Polymer Miscibility
Because of the commercially interesting properties of compatible polymer blends, interest in phase sepasation in polymer-polymer systems has increased in recent
POLYMER SOLUTIONS
FIG. 7-1 5.
Typical phase diagram of a polymer solution See the text for explanation (Reprinted with permission from Daoud 1976, Figure 3, page 974 )
years In general, polymers mixed in pairs tend to be totally immiscible, but a significant number of exceptions exist and have been listed by Krause (1978) Each of the several theories of polymer solution thermodynamics contributes to an understanding of polymer-polymer miscibility Use of the solubility parameter predicts, for example, that for miscibility over the entire composition range the difference in solubility parameters of the two polymers cannot exceed 0.7 at M = 10,000, 0 2 at M = 100,000, or 0 08 at M = 1,000,000 The Flory-Huggins theory describes all the observed behavio~, provided that interaction parameters are determined experimentally, but it does not provide a derivation of them, an understanding of the origin of the observed behavior, or predictive powers. The conesponding-state theory overcomes many of these limitations and leads to several conclusions at va~iance with the predictions of the Flory-Huggins theory. The thermodynamics of phase separation in block copolymers has been developed by Krause (1970) and Meier (1969)
GENERAL REFERENCES
Flory 1953; Casassa 1977; Elias 1977, Chapter 6.6; Paul 1978; de Gennes 1979, Chapter 4; Kwei 1979, Chapter 4.3; Olabisi 1979
FRACTIONATION OF POLYMERS BY SOLUBILITY
177
E.
FRACT'IONAT'ION OF POLYMERS BY SOLUBlLlTY
In this section are discussed only the most important and widely practiced techniques for fractionating polymers by solubility differences Many variations and less useful techniques exist and are reviewed by Hall (1959), Guzman (1961), and Cantow (1967)
Bulk Fractionation by Nonsolvent Addition

Fractional precipitation in bulk (Kotera 1967) is car~ied by adding the nonsolvent out to a dilute solution of the polymer until a slight turbidity develops at the temperature of fractionation To ensure the establishment of equilibrium the mixture may be warmed until it is homogeneous and allowed to cool slowly back to the required temperature, which should thereafter be carefully maintained The precipitated phase is allowed to settle to a coherent layer, and the supernatant phase is removed A further increment of precipitant is added and the process repeated The polymer is isolated from the precipttated phase, which may still be relatively dilute (perhaps 10%polymer) Refractionation is often used to achieve better separation A more efficient procedure is to restrict the initial experiments to very dilute solutions, ot to refine each initial fraction by a single reprecipitation at higher dilution, returning the unprecipitated polymer from this separation to the main solution before the next fraction is removed The solvent and the precipitant should be chosen so that precipitation occu~s over a relatively wide range of solvent composition yet is complete before too high a ratio of precipitant to solvent is reached Other impor.tant considerations aIe the stability and volatility of the liquids and their ability to form a highly swollen, mobile gel phase Relatively simple equipment, for example, a three-neck flask of several liters capacity (Fig 7-16), is adequate for bulk precipitation fractionation
Column Elution Solvent-Gradient Elution. In elution methods (Elliott 1967), polymer is placed in contact with a series of' liquids of' gradually increasing solvent power. Species of' lowest molecular weight, and thus highest solubility, dissolve in the first liquid, and successively higher molecular-weight fractions in subsequent liquids. To ensure rapid equilibration, the polymer must be present as a very thin film. For convenience, the polymer is often applied to a finely divided substrate such as sand or glass beads, which is subsequently used to pack a column.. By the use of' mixing vessels, a continuous gradient of solvent-nonsolvent composition can be produced and eluted through the column Thermal-Gradient Elution. A modification of' the solvent-gradient method utilizes a small temperature gradient from one end to the other of the column, in addition to the gradient of solvent composition (Baker 1956, Porter. 1967). The
POLYMER SOLUTIONS
'-.
u
FIG. 7-16. An apparatus suitable for bulk fractional precipitation (Hall 1959)
sample is initially confined to a small zone of the packing at the warm end of' the column. Each species undergoes a series of solution and precipitation steps as it progresses down the column
Analytical Precipitation Techniques

Analytical fractionation techniques are those in which fractions are not isolated, in contIast to the methods discussed above. They may be but ase not necessarily carried out on a smaller scale than their preparative counterparts..
Summative Fractionation. In this method (Billmeyer 1950, 1973), a series of polymer solutions is used. In each, part of the polymer is precipitated, the point of fractionation being varied throughout the series. Weight fractions w, and moleculru weights M, of the precipitates are obtained; the product wxMxmay be related to the integral of the cumulative distribution curve. Since the two differentiations required to obtain the differential distribution would introduce intolerably large errors into the resulting differential curve', the data are properly analyzed to provide only information equivalent to the position of the maximum and breadrh of the molecular-weight distribution curve. The method is advantageous for obtaining limited infbrmation about the molecular-weight distribution with a minimum of time, preliminary calibration effort, and equipment expense
FRAC:'I'IONATIONO POLYMERS BY SOLUBILITY F
179
Turbidimetric Titration. In this technique (Giesekus 1967) a precipitant is added slowly to a polymer solution, and the turbidity due to precipitated polymer is measured by the decrease in intensity of a beam of transmitted light or by the increase in intensity of scattered light Selection of an appropriate solvent-nonsolvent system for a given polymer requires much preliminary study, and it is usually not possible to obtain accurate information on molecular weight distribution from the experimental data. Nonetheless, the method is attractive for control purposes once the proper operating conditions are established Effect of Polymer Structure on Solubility-Based Fractionation Effect of Chemical Type. It must be reemphasized (Section A ) that the sblubility of polymers is determined primarily by their chemical composition, and in the usual case molecular weight is only a secondary variable If chemical composition differences exist among the polymer species in a sample, fractionation is likely to occur primarily as a result of these differences and to yield information on them rather than on the distribution of molecular weights (Fuchs 1967). Turbidimetric titration has often been used to explore qualitatively the effects of chemical composition in copolymer systems (Giesekus 1967) Effect of Chain Branching. On the basis of limited information (Schneider 1968), it appears that branching increases the solubility of high polymers Thus at any given stage in a solubility-based fractionation process, the branched polymer being separated will consist of a mixture of species, some with low branching and low molecular weight, others more branched but with compensating higher molecular weight, all having the same solubility Additional information is needed to characterize such a system in terms of either variable separately Effect of Crystallinity. Precipitation of polymer to a crystalline, rather than an amorphous, phase can lead to fractionation by molecular weight (with control of the other vxiables just mentioned) only if the '(depressed) crystalline melting point is a strong function of molecular weight This is the case only for polymers of rather low moleculax weight (say, less than 20,000), but in these cases the technique works rather well fbr the production of narrow fractions in large amounts (Pennings 1968). Treatment of Data
Info~mationabout the distribution of molecular weights in a polymer is often the goal of a fractionation experiment. Such information requires not only that the paragraphs be controlled, but also the measvariables mentioned in the p~evious urement of the weight and molecular weight of each fraction. For convenience, intrinsic viscosity is often used for the molecular-weight measurement, but this must be limited to experiments with linear polymers
180
POLYMER SOLUTIONS
The data are treated (Tung 1967) by plotting against molecular weight the cumulative or integral distribution curve of the combined weight of' all fractions having molecular weight up to and including M. The differential distribution curve is obtained by differentiation. Examples are given in Chapter 8F Experimental erlors, indicated by the scatter of' points in the integral curve, ase magnified in this process Only the gross features of the differential curve, such as the position of the peak and the breadth of the curve, are experimentally significant. Unless the fractionation is performed with extreme care, spurious features, such as double peaks in the differential distribution curve, may appear
Hall 19.59; Guzm6n 1961; Schneider 1965; Cantow 1967; Johnson 1967; Tung 1977
1. List and discuss the variables affecting the solubility of polymers 2. Describe the two stages of the process of dissolving a polymer and suggest how to speed each of them up
3. Calculate the solubility parameter 6 of bis-phenol A polycarbonate (Chapter 15B) from molar attraction constants
4. It is observed that a styrene-butadiene copolymer (6 = 16 5) is insoluble in pentane (6 = 14 5) and ethyl acetate (6 = 18 6 ) ,but soluble in a 1:1 mixture of the two Explain 5. Describe the conformation of' a polymer chain in solution. Define and relate various measures of the size of' the chain, show the distribution of sizes by graphs, and discuss the effect of molecular structure on size
6. For polystyrene of'molecular weight 416,000 dissolved in toluene, the chain expansion factor a is 3 . 2 . Calculate (a) the contour length, (b) the randomflight end-to-end distance, (c) the random-flight radius of' gyration, (d) the unperturbed end-to-end distance, (e) the actual end-to-end distance, (f') the weight concentration of polymer at which spheres with radius 2..5 times the actual end-to-end distance would just touch (this radius is that at which the chance of finding chain segments outside the sphere becomes small; for simplicity, pack the spheres as if they were cubes), and (g) the volume fraction of' polymer at this concentration (the effective density of the polystyrene chain in toluene is 1. 10)
7. Define ideal solutions and show why polymer solutions can never be ideal, even at extreme dilutions
BIBLIOGRAPHY
181
8. Define excluded volume and discuss the difficulty of its evaluation
9. Discuss the basis and contributions to polymer science of (a) the FloryHuggins theory, (b) the Flory-Krigbaurn theory, (c) the cor.respondingstate theories, and (d) scaling concepts
10. Define the Flory temperature O and describe two ways of' determining it from thermodynamic considerations.. 11. Derive Eqs. 7-10 and 7-1 1 from Eq. 7-9
r
12. Calculate the free energy of mixing 100 g of polystyrene, of molecular weight 100,000, with 1000 g of benzene Take X , = 0 23 and assume that the polymer and the solvent have the same density Comment on the sign and magnitude of the answer 13. In the fractional precipitation of polymer in a mixture of' good and poor solvents, how do the compositions of the two phases depend on polymer chain length? 14. Explain why poly(methy1 methacrylate) and polystyrene are immiscible whereas their monomers are miscible in all proportions Under what circumstances would this miscibility carry over to a copolymer of the two monomers?
15. The following data were synthesized to represent the fractionation of a polymer of relatively narrow molecular-weight distribution. Calculate M, and M, and plot the integral and differential distribution curves
Weight Percent in Fraction

3 5 6 13
Molecular Weight
50,000 75,000 95,000 125,000 150,000 185,000 220,000 270,000 325,000 400,000
8 25 16 9 10 5
BIBLIOGRAPHY
Asmussen I962 F~ithjofAsmussen and Kurt Uebeneiter, "Velocity of Dissolution of Polymers Piut 11," J Polym Sci 57, 199-208 (1962) ASTM 03132 Standard Test Method for Solubility Range of Reslns and Polymers, ANSIIASTM
Designation: D3132, American Society for Test~ng Mate~ials, and Ph~ladelphia, Pennsylvania, 1976
182
Baker 1956 C A Baker and R J P Will~ams, New Chromatographic Procedure and Its Appllcation "A to H ~ g h Polymers," I Chem Soc 1956, 2352-2362 (1956) Bawn 1950 C E H Bawn, R F J Freeman, and A R Kamallddln, "High Polymer Solutions Part I Vapour Pressure of Polystyrene Solutions," Trans Faraday Soc 46, 677-684 (1950) Billmeyer 1950 F W Billmeyer, Jr ,and W H Stockmayer, "Method of Measunng Molecular Weight D~stnbution," I Polym Sci 5, 121-137 (1950) Brllmeyer 1969 Fred W Blllmeyer, JI , "Molecular Structure and Polymer Properties," J Paint Technol 41, 3-16, 209 (1969) Brllmeyer 1972 Fred W Billmeyer, Jr , Synthetrc Polymers Butldrng the Grant Molecule, Doubleday, Garden C ~ t y New York, 1972 , Brllmeyer 1973 Fred W Blllmeyer, JI , and Leonard R Siebert, "Appllcation of the Summative* Fractlonatlon Method to the Determlnat~onof M,IM. for Narrow Dlstrlbution Polymers," Chapter 2 In Myer Ezrin, ed , Polymer Molecular Werght Methods, American Chemlcal Soclety, WashIngton, D C , 1973 Brllmeyer 1977 Fred W Brllmeyer, JI , "The Slze and y e ~ g h of Polymer Molecules," Chapter 4 in t Herman S Kaufman and Joseph J Falcetta, eds , Introduction to Polymer Scrence and Technology An SPE Textbook, John Wlley & Sons, New York, 1977 Brrshtern I966 T M Blrshteln and 0 B Ptitsyn, Conformatrons of Macromolecules (translated y Serge N Timasheff and Marina J Timasheff), John Wlley & Sons, New York, 1966 Bovey 1969 Frank A Bovey, Polymer Conformation and Configuratron, Academlc Press, New York, 1969 Bovey 1979 F A Bovey and T K K w e ~ ,"M~c~ostructure Cham Conformation of Macromoleand cules," Chapter 3 In F A Bovey and F H Winslow, eds , Macromolecules An Jntroductron to Polymer Scrence, Academic Press, New York, 1979
Burrell 1955 Hany Burrell, Solubility Parameters for Film Formers," Off Drg Fed Soc Parnt Technol 27, 726-758 (1955) Burrell 1975 H Burrell, "Solubility Parameter Values," pp IV-337-IV-359 In J Brandrup and E H Immergut, eds , with the collaboration of W McDowell, Polymer Handbook, 2nd ed , WlleyIntersc~ence,New York, 1975 Carztow 1967 Manfred J R Cantow, ed , Polymer Fractronation, Academlc Press, New York, 1967 Carpenter 1970 D K Carpenter, "Solut~on Properties," pp 627-678 in Herman F Mark, Norman G Gaylord, and Norbert F B~kales,eds ,Encyclopedra ofPolymer Scrence and Technology, Vol 12, W~ley-Intersclence,New York, 1970 Casarsa 1977 Edward F Casassa, "Phase Equlllbrlum In Polymer Solut~ons," Chapter 1 in L H Tung, ed , Fractronatron of Synthetrc Polymers Prrnclples and Practrces, Marcel Dekker, New York, 1977 Collrns 1973 Edward A Collins, Jan Bares, and Fred W Blllmeyer, Jr , Experrments rn Polymer Science, Wiley-Intersc~ence,New York, 1973 Daoud 1975 M Daoud, J P Cotton, B Famoux, G Jannlnk, G Sarrna, H Benoit, R Duplessix, C Picot, and P G de Gennes, "Solutions of Flexlble Polymers Neutron Experiments and Interpretat~on," Macromolecules 8, 804-818 (1975) of Daoud I976 M Daoud and G Jannink, "Temperature-Concentrat~on Dlag~am Polymer Solutions," J Phys Pans 37, 973-979 (1976) de Gennes 1979 P~ene-Glllesde Gennes, Scahng Concepts m Polymer Physrcs, Cornell University Press, ithaca, New York, 1979 des Clorzeaux 1975 J des Clolzeaux, "The Lagrangian Theory of Polymer Solutions at Intermediate Concentrat~ons," J Phys Purrs 36, 28 1-291 (1975) Erchinger 1968a B E Elchlnger and P J Flory, "Thermodynamlcs of Polymer Solutions Part I Natural Rubber and Benzene," Trans Faraday Soc 64, 2035-2052 (1968)
BIBLIOGRAPHY
183
Elchinger 1968b B E Eichinger and P J Flory, "Thermodynamics of Polymer Solutrons Part 2 Polyisobutylene and Benzene," Trans Faraday Soc 64, 2053-2060 (1968) Elchlnger 1968c B E Eichinger and P J Flory, "Thermodynamics of Polymer Solutions Part 3 Polyisobutylene and Cyclohexane," Trans Faradq Soc 64, 2061-2065 (1968) Eichrnger 1968d B E Eichinger and P J Flory, "Thermodynamics of Polymer Solutions Part 4 Polyisobutylene and n-Pentane," Trans Faraday Snc 64, 2066-2072 (1968) Ellas 1977 Hans-Georg Elias, Macromokcules I Structure and Propertres, (translated by John W Stafford), Plenum Press, New York, 1977 Ell~ott 1967 John H Elliott, "F~actionalSolution," Chapter B 2 in Manfred J R Cantow, ed ,Polymer Fractionation, Academic Press, New York, 1967 Flory 1942 Paul J Flory, "Thermodynamics of High Polymer Solutions," I Chem Phys 10, 51-61 (1942) Flory 1950 P J Flory and W R Krigbaum, "Statistical Mechanics of Dilute Polymer Solutions," J Chem Phys 18, 1086-1094 (1950) Flory 1953 Paul J Flory, Prlncrples of Polymer Chemistry, Cornell University Press, Ithaca, New York, 1953 Flory I965 P 1 Flory, "Statistical Thermodynamics of Liquid Mixtures," J Am Chem Soc 87, 1833-1838 (1965) F lory 1969 Paul J Flory , Statistical Mechanics of Chain Molecules, Wiley-Interscience, New York, 1969 Freeman 1960 P I Freeman and J S Rowlinson, "Lower Critical Points in Polymer Solutions," Polymer 1, 20-26 (1960) Fuchs 1967 0 Fuchs and W Schmieder, "Chemical Inhornogeneity and Its Determination," Chapter Academic Press, New York, 1967 D in Manfred J R Cantow, ed , Polymer Fract~onat~on, Gardon 1965 J L Gadon, "Cohesive-Energy Density," pp 833-862 in Herman F Mark, Norman ofPolymer S~ience Technology, Vol and G Gaylord, and Norbert M Bikales, eds , Encycloped~a 3, Wiley Interscience, New York, 1965 Gee 1946 Geoffrey Gee and W J C Orr, "The Interaction between Rubber and Liquids VIII A Benzene," Trans New Examrnation of the Thermodynamic Roperties of the System Rubber Faraday Soc 42, 507-51 7 (1946) Gee 1917 Geoffrey Gee, Part 5, "Equ~libr~um P~opertres High Polymers and Gels," in "Drscussion of Z on 'Some Aspects of the Chemistry of Macromolecules,'" J Chem Soc 1947, 280-288 (1947) G ~ e s e k 1967 Hanswalter Giesekus, "Turbidimetric Titration," Chapter C 1 in Manfred J R Cantow, ~~s ed , Polymer Fract~onat~on, Academic Press, New York, 1967 Guznrh 1961 G M GuzmPn, "Fractionation of High Polymers," pp 113-183 in J C Robb and F W Peake~, , Progress in H ~ g h eds Polymers, Vol 1, Academic Press, New York, 1961 Hall I959 R W Hall, "The Fractionation of High Polymers," Chapter 2 in P W Allen, ed , Techniques of Polymer Characterrzatron, Butterworths, London, 1959 Hlldebrmui 1950 Joel H Hildebrand and Robert L Scott, The Sobbrhty of Nonelectrolytes, 3rd ed , Reinhold, New York, 1950 (reprinted by Dover Publications, New York, 1964) Hopfinger 1973 A J Hopfinger, Conformational Proper t ~ e s Macromolecules, Academic Press, of New York, 1973 Hoy 1970 K L Hoy, "New Values of the Solubility Parameters from Vapo~ Ressure Data," J Palnt Iechnol 42, 76-118 (1970) Huggim- 1942a Maulice L Huggins, "Thermodynamic Properties of Solutions of Long-Chain Compunds," Ann N Y Acad Scr 42, 1-32 (1942) Huggins 1942b Maurice L Huggins, "Some Properties of Solutions of Long-Chain Compounds," 7 P h ~ sChem 46, 151-158 (1942) Hugpim 1942c Maurice L Huggins, "Theory of Solutions of High Polymers," I Am Chem Soc at, 1712-1719 (1942)
184
Huggins 1958 Maurice L Huggins, Physical Chemistry of'High Polymers, John Wileg & Sons, New York, 1958 Johnson 1967. .Julian F Johnson, Manfred J. R Cantow, and Roger S Porter, "Fractionation," pp 231.-260 in Herman F Mark, Norman G Gaylord, and Norbert M Bikales, eds , Encyclopedia oj Polymer Science and Technology, Vol. '7, Wiley-Interscience, New York, 1967. Koningrveld 1968. R Koningsveld, "Liquid-Liquid Separation of Polymer Solutions," pp 172-213 in Donald McIntyre, ed., Characterization of Macromolecular Structure, Publication No. 1573, National Academy of Sciences, Washington, D C , 1968 Kotera 1967 Akira Kotera, "Fractional Precipitation," Chapter B 1 in Manfred J R Cantow, ed., Polymer Fractionation, Academic Press, New York, 1967 Krause 1970. Sonja Krause, "Microphase Sepaation in Block Copolymers Zeroth Approximation Energies," Macromolecules 3, 84-86 (1970) Including Surface F ~ e e Krause 1978 Sonja Krause, "Polymer-Polymer Compatibility," pp. 16-1 13 in D . R Paul and Seymour Newman, eds , Polymer Blendr, Vol 1, Academic Press, New York, 1978 Kwei 1979. T K Kwei, "Macromolecules in Solution," Chapter 4 in F A . Bovey and F H Winslow, eds , Macromoleculer An Introduction to Polymer Science, Academic Press, New York, 19'79 Lewis 1961 Gilbert N. Lewis and Merle Randall, Thermodynamics, revised by Kenneth S Pitzer and Leo Brewer, 2nd ed., McGraw-Hill, New York, 1961 1 Lowry 1970 George G Lowry, ed., Markov Chains and Monte Carlo Calculationsin Polymer Science, Marcel Dekker, New York, 1970 Meier I969 D . J Meier, "Theory of' Block Copolymers I . Domain Formation in A-B Block Copolymers," 1 Polym Sci C26, 81-98 (1969) Morawetz 1975. Herbert Morawetz, Macromolecules in Solution, 2nd ed , Wiley-Interscience, New York, 1975. Newing 1950 M J Newing, "Thermodynamic Studies of Silicones in Benzene Solution," Trans Faraday SOC 46, 61 3-620 (1950) Olabiri 1979. Olagoke Olabisi, Lloyd M Robeson, and Montgomery T Shaw, Polymer-Polymer Miscibilify, Academic Press, New York, 1979 Patterson 1969. D . Patterson, "Free Volume and Polymer Solubility: A Qualitative View," Macromolecules 2, 672-6'79 (1969) Paul 1978 D R Paul and Seymour Newman, eds., Polymer Blendr, Academic Press, New York, 1978 Pennings 1968 A . J Pennings, "Liquid-Crystal Phase Separation in Polymer Solutions," pp 214244 in Donald McIntyre, ed , Characterization oj Macromolecular Structure, Publication No 1573, National Academy of Sciences, Washington, D C ., 1968. Porter 1967. Roger S Porter and Julian F Johnson, "Chromatographic Fractionation," Chapter B 3 in Manfred J R Cantow, ed., Polymer Fractionation, Academic Press, New York, 1967 Prigogine 19.57. I . Prigogine (with the collaboration of' A Bellemans and V. Mathot), The Molecular Theory oj Solutions, Interscience, New York, 1957 Rayleigh 1919 Lord Rayleigh, "On the Problem of Random Vibrations, and of Random Flights in One, Two, or Three Dimensions," Philos Mag. 37, 321-347 (1919). Schneider 1965. Nathaniel S Schneider, "Review of Solution Methods and Certain Other Methods of Polymer Fractionation," J Polym Sci C8, 179-204 (1965) Schneider 1968 N S Schneider, R T Traskos, and A S . Hoffinan, "Characterization of' Branched Polyethylene Fractions from the Elution Column," .J Appl. Polym Sci 12, 1567-1587 (1968). Shultz 19.52 A . R Shultz and P. J Floly, "Phase Equilibrium in Polymer-Solvent Systems," J . Am Chem Soc 74,4760-4767 (1952). Siow 1972. K . S . Siow, G Delmas, and D Patterson, "Cloud.-Point Curves in Polymer Solutions with Adjacent Upper and Lower Critical Solution Temperatures," Macromolecules 5, 29-34 (19'72) Snyder 1980 Lloyd Snyder, "Solutions to Solution Problems-2," Chemtech 10, 188-193 (1980)
Stockmayer 1949 W H Stockmayer, "Solubility of Heterogeneous Polymers," J Chem Phys 17, 588 (1949) Tager 1978 A Tagex, Physical Chemistry of Polymers (translated by Dav~dSobolev and Nicholas Bobrov), Mir, Moscow, 1978 (Imported Publications, Chicago) Taylor 1948 William J Taylor, "Average Length and Radius of Normal Paraffin Hydrocarbon Moiecules," J Chem Phys 16, 257-267 (1948) 7reloar 1958 L R G Treloar, The Physics of Rubber Elasticity, 2nd ed , Clarendon h e s s , Oxford, 1958 Tung 1967 L H Tung, "Treatment of Data," Chapter E in Manfred J R Cantow, ed , Polymer Fractronatlon, Academic hess, New York, 1967 Tiing 1977, L H Tung, ed , Fractronatron of Synthetic Polymers Principles and Practices, Marcel Dekker, New York, 1977 Ueberreiter 1962 Kurt Ueberreiter and Frithjof Asmussen, "Velocity of D~ssolutionof Polymers Part I," J Polym Sci 57, 187-198 (1962) Volkenstein 1963 M V Volkensteln, Configurational Statistics of Polymeric Chains, translated by Serge N Tlrnasheff and M J T~masheff, Wlley-Intersclence, New York, 1963 Vollmert 1973 Bruno Vollmert, Polymer Chemrstry (translated by Edmund H Immergut), Springer, New York, 1973 Zimm 1949 Bruno H Zimm and Walter H Stockmayer, "The Dimensions of Chain Molecules Containlng Branches and Rlngs," J Chem Phys 17, 1301-1314 (1949)
CHAPTER EIGHT
MEASUREMENT OF MOLECULAR WEIGHT AND SIZE
The molecular weights of polymers can be determined by chemical or physical methods of functional-group analysis, by measurement of the colligative proper.ties, light scattering, or ultracentrifugation, or by measurement of dilute-solution viscosity Ail these methods except the last are, in principle, absolute: Molecular weights can be calculated without reference to calibration by another method Dilutesolution viscosity, however, is not a direct measure of molecular weight. Its value lies in the simplicity of the technique and the fact that it can be related empirically to molecular weight for many systems With the exception of some types of end-group analysis, all molecular-weight methods require solubility of the polymer, and all involve extrapolation to infinite dilution or operation in a O solvent in which ideal-solution behavior is attained The term molar mass 1s preferred by some authors (e g , Billingham 1977) to molecular weight It is regrettable that over the years far more use is being made of empirical rather than absolute methods for determining molecular weight Such methods require the use of calibration samples to give them meaning Absolute methods for molecularweight measurement, providing such samples, are practiced far less than a decade ago, and much special apparatus developed for these techniques is no longer commercially available These methods are vital, and their continuing development and practice must be encouraged (Billmeyer 1976)
A.
END-GROUP ANALYSIS
Molecular-weight determination through group analysis requires that the polymer contain a known number of' determinable groups per molecule.. The long-chain
COLLIGATIVE PROPERTY MEASUREMENT
187
nature of polymers limits such groups to end groups. Thus the method is called end-group analysis Since methods of end-group analysis count the number of molecules in a given weight of sample, they yield the number-average molecular weight (Section B) for polydisperse materials The methods become insensitive at high molecular weight, as the fraction of end groups becomes too small to be measured with precision Loss of precision often occurs at molecular weights above 25,000, the limitation being as much or more due to the inability to purify samples and reagents as to lack of sensitivity in the methods The discussion of end-group methods is conveniently divided to cover condensation and addition polymers because of the difference in the types of end groups usually found
Condensation Polymers. End-group analysis in condensation polymers usually
involves chemical methods of analysis for functional groups Carboxyl groups in polyesters and in polyamides are usually titrated directly with base in an alcoholic or phenolic solvent, while amino groups in polyamides may be titrated with acid under similar conditions Ilydroxyl groups are usually determined by reacting them with a titratable reagent, but infrared spectroscopy has been used The chemical methods are often limited by insolubility of the polymer in solvents suitable for the titrations
Addition Polymers. No general procedures can be given for end-group analysis
in addition polymers because of the vaxiety of type and origin of the end groups When the polymerization kinetics is well known, analysis may be made f o initiator ~ fragments containing identifiable functional groups, elements, or radioactive atoms, or for end groups arising from transfer reactions with solvent or for unsaturated end groups in linear polyethylene and poly-a-olefins, as in the infrared spectroscopic analysis for vinyl groups
GENERAL REFERENCES
Hellman 1962; Collins 1973, Exp 10; Garmon 1975; Billingham 1977, Chapter 9
B.
COL.LIGATIVE PROPERTY MEASUREMENT
The relations between the colligative properties and molecular weight for infinitely dilute solutions rest upon the fact that the activity of the solute in a solution becomes equal to its mole fraction as the solute concentration becomes sufficiently small.. The activity ofthe solvent must equal its mole fraction under these conditions, and it follows that the depression of' the activity of the solvent by a solute is equal to the mole fraction of' the solute. The colligative property methods are based on vapor-pressure lowering, boilingpoint elevation (ebulliometry), freezing-point depression (cryoscopy), and the os-
188
MEASUREMENT OF MOLEC.ULAR WEIGHT AND SIZE
motic pressure (osmometry) The working equations for these methods are derived from Eq 7-10. When Boltzmann's constant is replaced by the gas constant throughout to convert to a molar basis, the equations for the methods of interest for polymer solutions are
where ATb, ATf, and r are the boiling-point elevation, freezing-point depression, and osmotic pressure, respectively; p is the density of the solvent; AH, and AHf are the enthalpies of vaporization and fusion, respectively, of the solvent per gram; and c is the solute concentration in grams per cubic centimeter The number-average molecular wezght M,, defined below, has been inserted to make the equations applicable to polydisperse solutes The relative applicability of the colligative methods to polymer solut~ons1s demonstrated in Table 8-1. It is clear that direct measurement of vapor-pressme lowering in polymer solutions is unrewarding It is possible, however, to utilize vapor-pressure lowering indirectly through the technique of vapor-phaseosmometry, in which one measures a temperature difference relatable to vapor-pressure lowering through the Clapeyron equation This temperature difference is of the same order of magnitude as those observed in cryoscopy and ebulliometry While such differences, of the order of 1 X 10-30C, can currently be measured with considerable precision, the larger effect observed in the osmotic experiment makes this technique more useful for polymer solutions, and it has been more widely used than other colligative techniques for polymer systems To distinguish it from vapor-phase osmometry, we refer hereafter to membrane osmometry as the specific name for the last-named technique in Table 8- 1
Number-Average Molecular Weight
Since typical polymers consist of mixtures of many molecular species, molecularweight methods always yield average values. The measurement f the colligative properties in effect counts the number of moles of solute per unit w k . sample. This number is the sum oveI a11 molecular species of the number of m o b N, of each species present:
COL.l.ICAl'IVE PROPERIY MEASUREMENT
TABLE 8-1. Colligative Properties of a Solution of Polystyrene of M = 20,000 at c. = 0.01 g/crn3 in Benzene
----
Property
Value
4 x mm Hg 1 3 x IO-~OC 2 5 x IO-~"C 15 cm solvent -.
Vapor-pressu~elowering Boiling-point elevation Freezing-point depression Osmotic pressure
The total weight w of the sample is similarly the sum of the weights of each molecular species,
The average m o l e c u l ~weight given by these methods is known as the numberaverage molecular weight Mn B y the definition of molecular weight as weight of sample per mole,
It is easy to show that the colligative methods measure ii?,, Taking membrane osmometry as an example, we may write Eq 8-1 separately for each species i in the mixture present Omitting the limit sign for convenience and transposing, we obtain
Summing over i, we may replace C T , by T ,since each solute molecule contributes independently to the osmotic pressure, and note that c = Cci. The mixture is then described by an unspecified average molecular weight M:
Solving, we have M = clZ(c,lM,) For unit volume the c's may be replaced by w's Noting that w, = M,N, shows that M = M,,
190
MEASUREMENT OF MOLECUL.AR WEIGHT AND SIZE
End-group methods as well as colligative methods give the number-average molecular weight.. The number average is very sensitive to changes in the weight fractions of low-molecular-weight species, and relatively insensitive to similar. changes for high-molecular-weight species
Concentration Dependence of the, Colligative Properties
For convenience, the concent~ation dependence of the colligative properties is developed in terms of the osmotic pressure; the results are directly applicable to substitution in Eq. 8-1, By Eq.. 7-1 1 or cryoscopy and ebulliometry by approp~iate 7-12 a general expression for the osmotic pressure at a finite concentration is
where r = A,IA, and g is a slowly varying function of the polymer-solvent interaction with values near zero fbr poor solvents and values near 0..25 for good solvents (Krigbaum 1952, Stockmayer 1952) In most cases the term in c2 may be neglected; when dependence on c2 is significant, it may be convenient to take g = 0.25 and write
'77
RTc
Mn
In te~ms the polymer-solvent interaction constant X , of' the Flory-Huggins and/ of Flory--Krigbaum theories, the second virial coefficient is given by
where subscript 1 indicates the solvent, and 2 the polymer. As noted in Chapter 7C, this expression does not account for the molecular-weight dependence of A, cor~cctly As Eqs.. 8-5 and 8-7 suggest, it is usual to plot TIC versus c. In general, a " straight line results whose intercept at c = 0 is A, = l l ~ and whose slope is A, and allows evaluation of'the polymer-solvent interaction constant X,. If the solvent is good enough or the concentration high enough so that the c2term is significant, the points may deviate from a straight line In such cases it is useful to plot (TI RTc)'12versus c, as suggested by Eq. 8-6. Typical data are discussed in the section Membrane OsmomeQy.
COLLlGATlVE PROPERTY MEASUREMENT
Vapor-Phase Osrnornetry
An indirect measurement of vapor-pressure lowering by the technique known as of vapor-phase osmometry is the method of choice for measuring M,, samples too low in molecular weight to be measured in a membrane osmometer. In this method the property measured, in a so-called "vapor-pressure osmometer" (Fig 8-I), is different rates of solvent evaporation the small temperature difference resulting f ~ o m from and condensation onto droplets of pure solvent and polymer solution maintained in an atmosphere of solvent vapor The basis of the method is that this temperature difference is proportional to the vapor-pressure lowering of the polymer solution at equilibrium and thus to the numbe~ -average molecular weight: Because of heat losses, the full temperature difference expected from theory is not attained Measurements must be made at several concentrations and extrapolated to c = 0. Like ebulliometry and cryoscopy, the method is calibrated with low-molecularweight standards and is useful for values of M~at least up to 40,000 in favorable cases It is rapid and has the additional advantage that only a few milligrams of that sample are required The lower limit of M,, can be measured is that at which the solute becomes appreciably volatile Since the method does not measure equilibrium vapor-pressure lowering but depends on the development of quasi-steady-
FIG. 8-1. Measurement chamber of a vapor-phase osmometel (Pasternak 1962) Droplets of solvent and polymer solution are placed, with the aid of hypodermic syringes, on the "beads" of two thermistors used as temperatu~e-sensingelements and maintained in equilibration with an atmosphere of solvent vapoI The lower activity of the solvent in the solution droplet leads to an excess of solvent condensation over evaporation there compared to the droplet of pure solvent The excess heat of vaporization thus liberated leads to a rise in temperature, usually somewhat less than that predicted from thermodynamic considerations because of heat losses
192
state phenomena, care must be taken to standardize such variables as time of' measurement and drop size between calibration and sample measurement.
Ebulliometry
In the boiling-point elevation experiment the boiling point of the polyme~ solution is compared directly with that of' the (condensing) pure solvent in a vessel known as an ebulliometer. Sensing devices include differential thermometers and multijunction thermocouples or thermistors arranged in a Wheatstone bridge circuit. It is customary to calibrate the ebulliometer with a substance of' known molecular weight, for example, octacosane (M = 396) or tristearin (M = 892). Although with care accurate molecular weights can be obtained up to = 30,000 or more, the ebulliometric experiment occasionally suffers from a limitation in the tendency of' polymer solutions to foam on boiling Not only may this lead to unstable operation, but also the polymer may concentrate in the foam because of' its greater surface, thus rendering uncertain the actual concentration of' the solution. No equipment suitable for ebulliometry in the high-,polymer range is commercially available, B and this method remains largely in the category of' a referee technique
Cryoscopy
The freezing-point depression or cryoscopic method is similar to the ebulliometric technique in several respects The preferzed temperature-sensing element is the thermistor used in a bridge circuit The freezing points of solvent and solution are often compared sequentially Calibration with a substance of known molecular weight is customary Although the limitations of the method appear to be somewhat
' 1 Semipermeable membrane

of FIG. 8-2. The principle of' ope~ation the membrane osmometer
COL.LIGATIVE PROPERTY MEASUREMENT
193
less severe than those of ebulliometry, care must be taken that supercooling be controlled.. The use of a nucleating agent to provide controlled crystallization of' the solvent is helpful in this respect. Reliable results are obtained fbr molecular weights at least as high as 30,000.. Again, lack of' commescial equipment relegates this method largely to referee status, despite the appeals of' simplicity and ease of' operation.
Membrane Osmometry
The principle of membrane osmometry is illustrated in Fig 8-2. The two compartments of an osmometer are separated by a semzpermeable membrane, through which only solvent molecules can penetrate, and which is closed except for capillary tubes With the polymer solute confined to one side of the osmometer, the activity of the solvent is different in the two compartments The thermodynamic drive toward equilibrium results in a difference in liquid level in the two capillaries The resulting
Cell
assembly
Glass cell
FIG. 8-3. Diagram of the Zimm-Myerson osmometer (Zimm 1946) A typical diameter for the measf uring and reference (solvent cell) capillay is 0 5-1 mm The closure o the filling tube is
a 2-mm metal rod A mercury seal is used at the top to ensure tightness
194
MEASUREMENT O F MOLEC;UI.AR WEIGHT' A N D SIZE
hydrostatic pressure increases the solvent activity on the solution side until, when the applied pressure equals the osmotic pressure, equilibrium is achieved.
Osmometers. Two types of osmometer have been used extensively in the past. One, the block osmometer, is a relatively large and bulky metal instrument. The membrane area is large and the solution volume small-advantages for rapid equilibration. More popular a e relatively small and simple osmometers based on the Zimm-Myerson design (Zimm 1946), in which two membranes are held against a glass solution cell by means of perforated metal plates, as shown in Fig. 8-3. The assembled instrument is suspended in a large tube partly filled with solvent. In the Stabin (1954) modification the membranes are supported on both sides by channeled metal plates to ensure their rigidity.. Advantages of these osmometers are small size and low cost, making multiple installations feasible, allowing the possibility of complete immersion in constant-temperature baths, and providing ease of' filling and adjusting osmotic height Modern membrane osmometers are rapid-equilibrating automatic devices in which the solvent compartment is completely closed and fitted with a sensitive pressure- ; sensing device rather than a capillary (Fig. 8-4). The instruments can be brought to balance before an appreciable amount of solvent has to pass through the membrane. As a result of' this rapid action, the osmometers come to equilibrium in 15 min instead of' 10-20 h, as typically required in conventional instruments Membranes. The usual materials used as osmotic membranes include collodion (nitrocellulose of 11-13.5% nitrogen); regenerated cellulose, made by denitration of collodion; gel cellophane that has never been allowed to dry after manufacture; bacterial cellulose, made by the action of certain strains of bacteria; rubber; poly(viny1
Sample outlet
Suction tube for level control
Diaphragm
Fixed electrodes
COLLlGATlVE PROPERTY MEASUREMENT
195
TABLE 8-2. Apparent Molecular Weights of Branched Polyethylenes by Membrane Osrnornetrya
Apparent M,, Sample of'

-
Method Ave~age cryoscopy, ebulliometry, and of vapor-phase osmometry (Table 8-3) Rapid membrane osrnometer, first observable values (5-7 min after sample introduction) Rapid membrane osmometer, 36 min after sample introduction Conventional membrane osmometer , 1000 min afte~sample introduction
75
76
77
alcohol); polyurethanes; poly(viny1 butyral); and polychlorotr~fluo~oethyleneOf these, gel cellophane is probably most widely used at present
Permeability of Solute Through the Membrane. The success of the osmotic experiment depends on the availability of a membrane through which solvent but not solute molecules can pass freely Existing membranes only approximate ideal semipermeability, and the chief limitation of the osmotic method is the diffusion of low-molecular-weight species through the membrane As currently practiced, the method is reliable and accurate only for experiments in which diffusion is absent, such as measurement of unfractionated polymers with M, > 50,000, or polymers from which low-molecular-weight species have been removed by fractionation or extraction, with M,, 20,000 The upper limit to molecula~weights > that can be measured by the osmotic method is about lo6, set by the precision with which small osmotic heights can be read When diffusion is present, the apparent osmotic pressure is always less than the true osmotic pressure and falls with time Extrapolation back to zero time still gives too low an osmotic pressure and hence too high a molecular weight The magnitude of the error resulting from solute diffusion and its dependence on the type of measurement are illustrated in Table 8-2 Typical Data. Figures 8-5 and 8.6 show typical osmotic data In both cases concentrations were low enough so that dependence of nlRTc on c2 was negligible The parallelism of the lines in Fig 8 -5, representing samples of different molecular weight in the same solvent, illustrates the usual result that the polymer-solvent interactions are slowly varying functions of molecular weight The constant intercept of the lines in Fig. 8-6 obtained for one sample in several different solvents illustrates the independence of (nlRTc),=, = llK of solvent type and the range of A2 or X, that can be obtained with a given polymer and a range of solvents
FIG. 8-5. Plots of nlRTc

1949)
I/&',
c, g/mi x l o 3 versus c for cellulose acetate fractions in acetone solution (Badgley
FIG. 8-6. Plot of d R T c = IIMn versus c for nitrocellulose in (a) acetone, (b) methanol, and (c) nitrobenzene (Gee 1944, data of Dobry 1935)
COLLfCATlVE PROPERTY MEASUREMENT
197
T'ABlE8-3.
Number-Average Molecular Weight of Polyethylene by Various Methodsa
Sample
-Method
Branched Polyethylene
Linear Polyethylene
75
76
77
99
99H
99L
101
End groups by infrared -8,100 37,400 4,800 11,700 Ebulliometry 11,150 18,400 -Cryoscopy 10,700 13,300 19,100 8,100 - 3,320 11,800 Membrane osmometry (30,300) (26,600) (3 1,400) 40,600 -Vapor-phase osmometry 11,000 16,300 18,800 8,600 38,400 3,500 11,900
- - -
"Billmeyer (19646)
The advantage of selecting a poor solvent for osmotic measurements in order to minimize the erIor in exhapolation is evident At the Flory temperature @ the slope of .rrlRTc versus c is zero to concentrations of several percent
Comparison of Data
Although it is conservative to state that most methods for measuring polymer molecular weights have an accuracy of only 5- lo%, significantly closer agreement among methods can be achieved in favorable cases An extensive intercomparison of methods for determining &,, gave the data in Table 8-3 Here, despite additional difficulties imposed by high-temperature operation, several methods gave excellent agreement for samples of polyethylene. The osmotic molecular weights in parentheses show the effects of diffusion of low-molecular-weight species through the membranes They were obtained with conventional osmometers; data on these samples obtained with rapid automatic osmometers are given in Table 8-2
GENERAL. REFERENCES
Bonnar 1958; Mclntyre 1968; Collins 19'73, Chapter 7A; Glover 197.5; Morawetz 197.5, Chapter 4A; Billingham 1977, Chapters 1, 3, and 4; Billmeyer 1977; and specifically the fbllowing: Vapor-Phase Osrnometry.. Collins 19'73, Exp. 12; Billingham 1977, Chapter 4; Burge 1977; ASTM D3.592.. Ebulliometry . Glover 1966, E z ~ i n1968. Membrane Osrnometry. Krigbaum 1967; Coll 1968; Collins 19'73, Exp. 11; Ulrich 1975; Billingham 1977, Chapter 3; Burge 197'7; ASTM D37.50.
198
C.
LIGHT SCATTERING
Light scattering occurs whenever a beam of' light encounters matter. The nuclei and electrons undergo induced vibrations in phase with the incident light wave and act as sources of light that is propagated in all directions, aside from a polarization effect, with the same wavelength as the exciting beam. Light scattering accounts for many natural phenomena, including the colors of' the sky and the rainbow, and of most white materials In 1871 Lord Rayleigh applied classical electromagnetic theory to the problem of the scattering of light by the molecules of a gas.. He showed that the quantity of' light scattered, for particles small compared to the wavelength of the light, is inversely proportional to the number of scattering particles per unit volume and to the fourth power. of' the wavelength; the latter dependence accounts fbr the blue color of the sky In Rayleigh's treatment it was assumed that each particle scattered as a point source independent of' all others.. This is equivalent to assuming that the relative positions of'the particles are random. In liquids this is not the case, and the scattered light intensity is reduced about 50-fold because of destructive interference of the light scattered from different particles. In the calculation of scattered intensity in liquids, due to Einstein and Smoluchowski in the early 1900's, the scattering is considered to be due to local thermal fluctuations in density that make the liquid optically inhomogeneous.
Light Scattering From Solutions of Particles Small Compared to the Wavelength. In solutions and in mixtures of' liquids, additional light scattering arises from irregular changes in density and refractive index due to fluctuations in composition. Debye (1944, 1947) calculated the effect of these fluctuations, relating them to the change in concentration c associated with the osmotic pressure 71. per mole of' solute:
32n3 RTc AT = -- (rz 3A4 No
2i2/($)
where 7 is the turbidity, AT represents the excess turbidity of the solution over that of'thepure solvent, A is the wavelength, and n is the refractive index. In the absence of absorption, T is related to the primary-beam intensities before and after passing through the scattering medium. If the incident intensity I. is reduced to I in a length I of sample, then
LIGHT SCATTERING
199
The turbidity, which is the total scattering integrated over all angles, is often replaced by the Rayleigh ratio Re, which relates the scattered intensity at angle 0 to the ) incident beam intensity FOI particles small compared to A, T = ( 1 6 ~ 1 3R(,=,., Inserting the relation between osmotic pressure and molecula~weight yields the Debye equatzon
where
K =
2n2n2 dn .q ) and ( z
H =
Equation 8-10 forms the basis of the determination of polymer molecular weights by light scattering Beyond the measurement of T or R,, only the refractive index n and the spec@ refractive increment dnldc require experimental determination The latter quantity is a constant for a given polymer, solvent, and temperature and is measured with an interfexometer or a differential refractometer (Brice 1951). Equation 8-10 is corxect only for vertically polarized incident light and for optically isotropic particles The use of unpolarized light requires that T be multiplied by 1 f cos2 8 This term arises from the fact that light can be propagated only at right angles to the direction of oscillation of the electric moment Whether polarized or unpolarized light is used, the scattered intensity is symmetrical about the angle of observation, 90" If the scattering particles are anisotropic, a correction for depolarization is required (Cabannes 1929)
Weight-Average Molecular Weight
I
In the derivation of Eq 8-10 it is shown that the amplitude of the scattered light is proportional to the polarizability and hence to the mass of the scattering particle Thus the intensity of scattering is proportional to the square of the particle mass. If the solute is polydisperse, the heavier molecules contribute more to the scattering than the light ones. The total scattering at zero concentration is
defining the weight.-average molecular weight according to any of' the relationships
200
MEASUREMENT OF MOLEC.ULAR W E I G H T A N D SIZE
where wi is the weight fraction and
as in the corresponding definition of' number-average molecular weight The significance of the weight- and numbe1-average molecular weights is noteworthy. The quantity M, is always greater than M,, except for a monodisperse is system. The ratio M,/R a measure of the polydispersity of the system M, is particularly sensitive to the presence of high-molecular-weight species, whereas is influenced more by species at the lower end ofthe molecular-weight distribution
Light Scattering in Solutions of Larger Particles

When the size of' a scattering particle exceeds about Al20, different parts of the particle are exposed to incident light of' different amplitude and phase. The scattered light is made up of. waves coming from different parts of the particle and interfering with one another. Consequently, the scattered light intensity varies with the angle 8 . Debye and others described this variation with a particle scattering factor P(0) that depends upon the model selected to describe the scattering system For spheres P(e) where
=
'
[($)
(sin u - u cos u )
1;
d is the diameter of' the spheres, and A, = Aln is the wavelength in the solution. For a monodisperse system of' randomly coiling polymers the radial density function is the Gaussian distribution, and the intensity of scattering is given by
where
and 2 is the mean-square radius of gyration of' the molecule Similar functions can be derived fbr other models, such as rods and discs. The functions P(8) are plotted in Fig. 8-7. It can be seen from the curves that the
FIG. 8-7. The variation of the paticle scattering function P ( 0 ) with the size parameter u (spheres), u (monodisperse random coils), x (rods), or y (discs)
scattered intensity is greater in the forward than in the backward direction, falling off smoothly with angle except for rather large spheres It is sometimes convenient to characterize the size of the scattering particles by the dissymmetry of light scattering, z = i4,alt,3,-, measured at two angles symmetrical about 90" The disisymmetry increases rapidly with increasing particle size. If the scattering particles are not small compared to the wavelength, as is usually the case with polymer solutions, Eq. (8-10) (in which M is replaced by M, for application to polydispe~sesystems) 'is modified to insert P(9):
It has been shown (Zimm 1948a,b) that a z-average dimension is obtained from P(9) for polydisperse systems In practice, the conection for dissymmetry can be made in two ways In Debye's o~iginal method, the dissymmetry was measured and compared with the functions of Fig 8-7 to evaluate particle size and thence P(9) In more common curxent use is Zimm's method, in which the left-hand side of Eq. 8-15 is plotted (Zimm plot) against sin2 (812) -t kc, where k is an arbitrary constant. A rectilinear grid results (Fig 8-8), allowing extrapolation to both c = 0 and 8 = 0, where P(8) = 1.
MEASUREMENT' OF MOLECULAR WEIGHT AND SIZE
FIG. 8-8.
Determination of Molecular Weight by Light Scattering Photometers. Modern light-scattering instruments use either a mercury arc or a laser. as a source and detect the scattered light photoelectrically.,Many ase interfaced to computers for contsol, data handling, and computation of results. The essential features of a "classical" light-scattering photometer are described with respect to the widely used Brice-Phoenix instrument (Brice 1950) (Fig. 8-9). Light from a mercury-arc source ( S ) passes through a lens (L), a polarizes (P,), and a monochromatizing filter (F). It then st~ikes either a calibrated reference standard or a glass cell (C) holding the polymer. solution. After passing through the cell, the primary light beam is absorbed in a light-trap tube (T)..Scattered light from cell or standard is viewed by a multiplier phototube (R) after passing through a slit system D,-D, and polarizer (P,). The phototube is powered by high voltage from a regulated electsonic power supply. Its output signal is transmitted either to a strip-chart recorder or to an analogue-to-digital converter for computer use. A typical scattering cell consists of a glass cylinder having flat entrance and exit windows. The scattering cell is centered on the axis of rotation of the receiver phototube
llGHT SCATTERING
FIG. 8-9. Sketch of the essential components of a light-scattering photometer
Sample Preparation. The proper preparation of a sample for light-scattering measurement is of major importance Polymer molecules are usually small compared to particles of dust or other extraneous material The scattering of these extraneous substances, while usually confined to low angles, may easily outweigh the scattering from the polymer solution Consequently, solvents and polymer solutions must be clarified by filtration or by ultracentrifugation Proper clarification, as evidenced by obtaining a rectilinear Zimm plot, is essential for the unambiguous interpretation of light- scattering data Accuracy in light-scattering measurement is favored by the proper choice of index between polymer and solvent should be solvent The difference in ref~active as large as possible In addition, the solvent should itself have relatively low scattering, and the polymer-solvent system should not have too high a second virial coefficient, since the extrapolation to zeIo polymer concentration becomes less certain for high A, Mixed solvents must be avoided unless both components have the same refractive index Calibration. ' Although, in principle, observed scattered intensity can be related to turbidity through knowledge of the geometry of the photometer, calibration with substances of known turbidity is common practice These substances include reflecting standards, colloidal suspensions, and simple liquids Tungstosilicic acid, H4SiW,,0,, M = 2879, has been recommended as a primary calibration standard in p~eferenceto puxe liquids (too sensitive to impurities) or uniform-particle-size latexes (too sensitive to residual polydispersity) (Kratohvil 1966, Billmeyer 1971, Levine 1976) Treatment of Data. The weight-average molecular weight is the inverse of the intercept at c = 0 and 0 = 0 in the Zimm plot (Fig. 8-8). The second virial coefficient is calculated from the slope of the lines at constant angle by Eq. 8-15, and to a first approximation is independent of angle. The radius of gyration is
204
derived from the slope of' the zero-concentration line as a function of angle The appropriate relationship, derived from Eq 8-14, is
sz2
= -X
3X,Z
initial slope intercept
16n2
Range of Applicability. Light scattering from solutions has been used to measure molecular weights as low as that of' sucrose and as high as those of' proteins. In practice, polymer molecular weights on the order of' 10,000-10,000,000 can readily be measured, with the possibility of' extending the range in either direction in favorable cases. Applications include determination of' absolute values of' M ~es, pecially for proteins and other biological materials, and calibration of' intrinsic viscosity-molecular weight relations (Section D). It is no;, however, possible to obtain information on particle shape or polydispersity from light-scattering measurements except under most unusual conditions. The method is applicable to copolymers only with severe restriction, owing to the difference in refiactive indelq that generally exists between the two types of chain repeat units. Branched polymers can be measured without restriction
Other Scattering Methods Dynamic Light Scattering. This term, among others (Brillouin scattering, laser light scattering, laser spectroscopy, light-beating spectroscopy, photon correlation spectroscopy), is applied to a variety of related methods in which the spectrum of' scattered light (secondary peaks, linewidths, etc ) is determined by both static and dynamic means The most valuable additional information obtained is the rotational diffusion coefficient of the polymer (Cummins 1973, Chu 1974, Kinsinger 1975, Berne 1976, Shimoda 1976). Neutron Scattering. In experiments entirely analogous to light scattering, the angular dependence of' scattering at small angles can be observed when a monochromatic collimated beam of coherent "cold" neutrons is directed at a solid polymer sample (Allen 1976).. Advantage is taken of' the much greater scattering power of deuterons compared to protons to provide the equivalent of'the refractiveindex difference between polymer and solvent in light scattering: Fully deuterated polymer samples are prepared and studied in mixtures with normal, protonated samples. The major results are in the area of' polymer conformations in the solid state (Benoit 1974, Richards 1978) Light Scattering From Very Large Particles. The measurement by light sc&ng of the size of particles considerably larger than polymer molecules is discussed by van de Hulst (1957), Billmeyer (1964a), Kerker (1969), and Livesey (1965s)
GENERAL REFERENCES
Billmeyer 1964a; Eskin 1964; McIntyre 1964; Kerker 1969; Huglin 1972; Collins 1973, Chapter 7B, Exps. 13 and 14; Casassa 1975; Morawetz 1979, Chapter 5C; Billingham 1977, Chapter .5; Elias 197'7, Chapters 9.5 and 9 6; Hyde 1978; ASTM D4001
Ultracentrifugation techniques are the most intricate of the existing methods for determining the molecular weights of high polymers They are far more successful in application to compact protein molecules than to random-coil polymers, where extended conformations increase deviations from ideality and the chances of mechanical entanglement of the polymer chains Despite recent theoretical and experimental developments that increase the utility of the techniques for random-coil polymer systems, the methods ale used overwhelmingly for biological materials
Experimental Techniques
The ultracentrifuge consists of an aluminum rotor several inches in diameter that is rotated at high speed in an evacuated chamber The solution being cent~ifuged is held in a small cell within the rotor near its periphery The ~otor may be driven electrically or by an oil or air turbine The cell is equipped with windows, and the concentration of polymer along its length is determined by optical methods based on measurements of refractive index or absorption. Solvents fox ultracentrifugation experiments must be chosen for difference from the polymer in both density (to ensure sedimentation) and refractive index (to allow measurement) In addition, mixed solvents must usually be avoided, and low solvent viscosity is desirable
Sedimentation Equilibrium
In the sedimentation equilibrium experiment, the ultracentrifuge is operated at a low speed of rotation for times up to 1 or 2 weeks under constant conditions. A thermodynamic equilibrium is reached in which the polymer is distributed in the cell solely according to its molecula weight and molecular-weight distribution, the force of sedimentation on each species being just balanced by its tendency to diffuse back against the concentration gradient resulting from its movement in the centrifugal field The force on a particle of mass m at a distance r from the axis of rotation is w2r(l - bp)m, where w is the angular velocity of rotation, b is the partial specific volume of the polymer, and p is the density of the solution Writing the partial
206
molar free energy in terms of this force and applying the conditions fbr equilibrium gives the result, for ideal solutions,
where c, and c, are the concentrations at two points r , and r , in the cell; r , and r, must be taken at the meniscus and the cell bottom to include all the molecular species in the average. If' the data are treated somewhat differently, it can be shown that the z-average molecular weight M;is obtained:
The problem of nonideality is best met by working at the 0 temperature In nonideal solutions the apparent molecular weight is a linear function of concentration at temperatures near 0, the slope depending primarily on the second virial coefficient The major disadvantage to the sedimentation equilibrium experiment lies in the fact that several days may be required to reach equilibrium, even if the time is shortened by the use of short cells or a synthetic boundary cell to reduce the distance the various species must travel as equilibrium is established The distribution of molecular weights can be obtained directly from sedimentation-equilibrium measurements The solution of an integral equation is involved, however, and the analysis remains a difficult one
'
Equilibrium Sedimentation in a Density Gradient. In this approach a mixed solvent is chosen so that the relative sedimentation of the two components gives xise to a density gradient. The solute forms a band centering at the point where its effective density equals that of' the solvent mixture. The band is Gaussian in shape with respect to solute concentration, the half-width being inversely proportional to the solute molecular weight.. A feature that is of major importance is the sensitivity of the method to small differences in effective density among the solute species..
Approach to Equilibrium
Archibald (1947) showed that measurement of c and dcldr at the cell boundaries allows molecular weight to be determined at any stage in the equilibrium process. If' measurements are made early enough, before the molecular species have time to redistribute in the cell, the weight-average molecular weight and second virial coefficient can be evaluated. In practice, measurements can be made about 10-60 min after the start of the experiment
Sedimentation Transport
In the sedimentation transport or sedzmentatlon veloczty experiment, the ultracentrifuge is operated at high speed so that the solute is transported to the bottom of the cell The rate of sedimentation is given by the sedimentation constant
and related to particle mass through a frictional coefficient f . For rigid particles f is well defined (for solid spheres of diameter d, f is given by Stokes' law, J' = 3~77d), but for random-coil polymers the evaluation of f is more difficult. It can be related to the diffusion coefficient D: At infinite dilution D = k T l f , where
However, even for a single solute species the equations have not been extended to take into account simultaneously the effects of diffusion in the ultracentrifuge cell, deviations from ideality, and compressibility of solvent and solute in the high centrifugal fields required For polydisperse solutes additional complications arise in assigning average molecular weights, since D and s must be averaged separately In general, simple averages such as & and M, are not obtained (Elias 1975) The most fruitful result of the sedimentation transport experiment now appears to be examination of the distribution of sedimentation constants By working in a O solvent, the concentration dependence of s and D is reduced enough to avoid extrapolation to c = 0 The observed distribution of s over the cell must still be extrapolated to infinite time to eliminate the effects of diffusion relative to sedifor mentation, and cor~ection compression must be made The resulting distribution of s can be converted to a distribution of molecular weight, either through detailed knowledge of the dependence of s and D on M or from the relation
and which is very nearly true at T = O..The numbe~~of'approximations the complex treatment of the data required in this procedure suggest that its use be limited to qualitative observations. In this respect, however, the experiment can be extremely valuable, for it provides directly qualitative information about the nature of the distribution of species present.
Preparative Separations. The usefulness ofthe ultracentrifuge in the preparation of samples mher than in the production of analytic data should not be overlooked. Preparative ultracentrifuges have utility in fractionating polymer samples and in freeing them from easily sedimented impurities..
208
GENERAL. REFERENCES
Svedberg 1940; Fujita 1962, 1975; Bowen 1970; Williams 1972; McCall 1973; Morawetz 1975, Chapters 4B and 6A; Scholte 1975; Billingham 1977, Chapter 6; Elias 1977, Chapter 9 7
E.
SOLUTION VISCOSITY AND MOLECULAR SlZE
The usefulness of solution viscosity as a measure of' polymer molecular weight has been recognized ever since the early work of' Staudinger (1930). Solution viscosity is basically a measure of the size or extension in space of' polymer molecules. It is empirical1,jrelated to molecular weight fbr linear polymers; the simplicity of the measurement and the usef'ulness of the viscosity-molecular weight correlation ase so great that viscosity measurement constitutes an extremely valuable tool for the molecular characterization of' polymers
Experimental Methods
Measurements of' solution viscosity are usually made by comparing the effux time t required fbr a specified volume of' polymer solution to flow through a capillary tube with the co~responding ef'fux time to of' the solvent. From t, to, and the solute concentration are derived several quantities whose defining equations and names are given in Table 8-4 Two sets of' nomenclature are in use for these quantities; one (Cragg 1946) has had long and widespread application, the other (International Union 1952) was proposed for greater clarity and precision. In this book the common nomenclature is retained. In this system the concentration c is expressed in grams per deciliter (gldl or g/ 100 ml) . The intrinsic viscosity [q] is independent of' concentration by virtue of' extrapolation to c = 0, but is a function of the solvent used. The inherent viscosity at
'TA8L.E 8-4.
Nomenclature of Soiution Viscosity Recommended Name

Viscosity ratio
-
Common Name
Relative viscosity Specific viscosity Reduced viscosity Inhe~ent viscosity
Symbol and Defining Equation

q r = rl/llo " t/t, qSp= q, - 1 = (7-1 qo)/q0 = (f - fo)/to -?red = qrp1c ymnh (In ~ , ) / c =
Intxinsic viscosity
Viscosity number Logarithmic viscosity number Limiting viscosity number
[rll = (rlTp/~)'=o [(in qr)lcl,=,, =
SOLUTION VISCOSITY AND MOL.ECULAR SIZE
Etched lines
FIG.. 8-10. Capilla~yviscometers commonly used fbr measurement of polymer solution viscosities: (a) Ostwald-Fenske and (b) Ubbelohde
a specified concentration, usually 0 5 gldl, is sometimes used as an approximation to [TI Dilute-solution viscosity is usually measured in capillary viscometers of the Ostwald-Fenske or Ubbelohde type (Fig 8-10a,b, respectively) The latter has the advantage that the measurement is independent of the amount of solution in the viscometer; measurements at a series of concentrations can easily be made by successive dilutions For highest precision, the follo&ing precautions are usually observed The viscosity measurement is made in a constant-temperature bath regulated to at least 2 0 02C. The effux time is kept long (preferably more than 100 sec) to minimize the need for applying corrections to the observed data For accuracy in extrapolating to c = 0, the solution concentration is restricted to the range that gives relative viscosities between 1 I and 1.5
Treatment of Data. Viscosity data as a function of' concentration are extrapolated to infinite dilution by means of' the Huggins (1942) equation
21 0
MEASUREMENT OF MOLEC,UL.AR WEIGH'T AND SIZE
where k' is a constant for a series of' polymers of different molecular weights in a given solvent. The alternative definition of' the intrinsic viscosity leads to the equation (Kraemer 1938)
where k' - k" = 4. Typical data are shown in Fig. 8-1 1 . The slopes of' the lines vary as [TI', as required by Eqs. 8-22 and 8-23 for polymer species of' different molecular weight in the same solvent. At intrinsic viscosities above about 2, and even lower in some cases, there may be an appreciable dependence of viscosity on rate of shear in the viscometer. This by dependence is not elimi~ated extrapolation to infinite dilution; measurement as a function of shear rate and extrapolation to zero s h e a rate are also required. Special appaatus and extreme care are required (Zimm 1962)
Empirical Correlations Between Intrinsic Viscosity and Molecular Weight for Linear Polymers
Staudinger's prediction in 1930 that the reduced viscosity of' a polyme~is proportional to its molecular weight has needed only slight modification: The intrinsic viscosity has been substituted fbr the reduced viscosity, and it has been recognized that the p~oportionalityis to a power of' the molecular weight somewhat less than 1. The relation is expressed in the equation
where K' and a are constants determined from a double logarithmic plot of' intrinsic viscosity and molecular weight (Fig. 8-12). Such plots are usually found to be
Concentration, g/dl
FIG. 8-11. Reduced and inherent viscosity-concentration curves for a polystyrene in benzene (Ewart
1946).
SOLUTION VISCOSITY AND MOLECULAR SIZE
30
-
I l l 1
Cyclohexane, 30C
10 171
---- -
03
Di~sobutylene, 20C --
- - u u
lo5 3 lo6 Molecular we~ght Intnnslc v~scos~ty-molecula~ we~ght relat~onships polylsobutylene in dl~sobutylene for and cyclohexane (Flory 1943, Kngbaum 1953)
lo4
straight lines within experimental error over large ranges of the variables For randomly coiled polymers the exponent a varies from 0 5 in a O solvent to a maximum of about 1 0 For many systems a lies between 0 6 and 0 8 Typical values of K' range between 0 5 and 5 x Both K' and a are functions of the solvent as well as of the polyme~type This empirical relation between viscosity and molecular weight is valid only for linear polymers Intrinsic-viscosity measurement leads to the viscosizy-average molecular weight, defined as
Thus M,depends on a as well as on the distribution of molecula species. For many polymers M,is 10--20% below M,.For a = 1, M, = M,.If fractionated polymers, for. which M, = M, = M,, are available, any absolute molecular-weight measurement may be combined with a viscosity measurement to evaluate the constants in Eq. 8-24.. If' the molecular-weight distribution of the samples is known in sufficient detail to calculate M, from a molecular-weight average that has been measured, the constants in the equation
21 2
MEASUREMENT OF M0LEC:ULQR WEIGHT AND SIZE
may be evaluated. Since M, is neaser to M, than to M,,, weight-average molecular the weight is preferred for. this calibration.. It can be shown that polymers with fairly broad but well-known distributions are preferred to fractions for this purpose. A less precise relation,
is sometimes used to relate weight-average molecular weight directly to intrinsic viscosity for a limited series of samples. Extensive tabulations of K t and a are available (Kurata 1975). Equation 8,-27is known to become inaccurate for molecular weights below about 50,000, because deviations from the linear relationship set in. For better results in this region, and with some theoretical justification (Stockmayer 1963), use of one of several expressions of' the fbrm
[q] = KMli2
+ K"M
(8-28)
I
is recommended Here the first term is determined by short-range interactions, as in Eq 8-31, and the second term by long-range interactions The form and value of K" vary among the several theories available (Cowie 1966)
intrinsic Viscosity and Molecular Size
Theories (Flory 1949, 1953) of the frictional properties of polymer molecule5 in solution show that the intrinsic viscosity is proportional to the effective hydrodynamic volume of the molecule in solution divided by its molecular weight. The effective volume is proportional to the cube of a linear dimension of the randomly is the coiling chain If (7)'" dimension chosen,
is a universal constant Replacing (7)' a(;T)112and rxallicg that by r/ :M is a function of chain structure independent of its surroundings 13: mo!szular weight, it folIows that where -
~ where K = Q ( ~ / M )is~a/ constant for a given polymer, independ& of c.ds.ent and molecular weight The dimensions (7)"' and ( 2 ) " ' and the e x p z - ~ ?=tor~ ~ ~z a used here are usually identified with those d~scussed Chapter 7 F-_x=~t;i..Y.~~ in of the Flory-Krigbaum treatment (Chapter 7C) imply that Q, is not a u n l v w d cr,rL9mt, and that the power of a in Eq. 8-30 is somewhat less than 3 (Kurata i'J/1- :'?53)
SOLUTION VISCOSITY AND MOLECULAR SIZE
21 3
It f'ollows from the properties of a that, at T = O, a = 1 and
There is ample experimental evidence for the validity of this equation. Values of K are near 1 X for a number of' polymer systems. An important feature of' Flory's viscosity theory is that it furnishes confirmation (if not evaluation) of the 63 temperature as that at which a = 4, and it allows the determination of' the unperturbed dimensions of' the polymer chain. Even if' a 6 3 solvent is not available, several extrapolation techniques (summarized by Cowie 1966) are available for estimating the unperturbed dimensions fiom viscosity data in good solvents.. The best and simplest of' these techniques appears to be that of Stockmayer (1963) based on Eq. 8-28
Determination of Chain Branching

The size of' a dissolved polymer molecule, which can be estimated by means of' its intrinsic viscosity, depends upon a number of factors. These include the molecular weight, the local chain structure as reflected in the factor ~ I M the perturbations , reflected in a, and finally the gross chain structure, in which the degree of longchain branching is an important variable. [The presence of short branches, such as ethyl or butyl groups in low-density polyethylene (Chapter 13A), has little ef'ect on solution viscosity.] By evaluating or controlling each of' the other factors, vis-, cosity measurements may be used to estimate the degree of' long-chain branching of' a polymer . The relation between molecular size and the number and type of' branch points has been calculated (Zimm 1949), leading to a relation g between the radii of' gyration fbr branched and lineas chains of' the same numbers of' segments:
g =
7(branched) s2(linear)
Several theories have been proposed relating g to [q](branched)/[q] (linear). Only recently has it been possible to decide which of these yields the best fit to experimental data; the situation is well summarized by Graessley (1968) At present it still seems appropriate to consider the ratio [q] (branched)l[q] (linear) evaluated at constant molecular weight as a qualitative indication of chain branching rather than to attempt to assign numerical values to the degree of chain branching in a given polymer sanlple
GENERAL REFERENCES
Kurata 1963; Lyons 1967; Moore 1967; ASTM D2857; Collins 1973, Chapter 7C, Exp.. 1.5; Carpenter 1975; Billingham 1977, Chapter 7; Elias 1977, Chapter 9.9.
214
MEASUREMENT OF M0LEC:ULAR WElCHr AND SIZE
F.
GEL PERMEATION CHROMAIOGRAPHY
Gel pe~meation chromatography, more cor~ectly termed size exclusion chrornatograph,y, is a separation method for high polymers, similar to but advanced in practice over gel filtration as carried out by biochemists, that has become a prominent and since its discovery widely used method for estimating molecular-weightdist~ibutions (Moore 1964) just over two decades ago in 1961..The separation takes place in a chromatographic column filled with beads of' a rigid porous "gel"; highly crosslinked porous polystyrene and porous glass are preferred column-packing materials. The pores in these gels are of the same size as the dimensions of'polymer molecules A sample of a dilute polymer solution is introduced into a solvent stream flowing through the column. As the dissolved polymer.molecules flow past the porous beads
Sample mixture
ty~.q*<c*
b * ; ~ ~ o ~ ~
0 0 0 0 00
ooooooo
000000 00000 0 000 000 oO,O ; O o O 0 000 0 0 0 0 0 0 0 0 0 0000000 00 000 0
,0
Separation begins
* I@~::$$ 8te0
00'0'00 0 0000 oo 0 0000 0 Ooo O o O O,"o%O 0
0"
0 000 :,"o 00 0 0 00 0 0 0 0 0 0000000 0% 900
Partial > P ~ < ~ $ P ' separation .d.,eQb,Q~.

00 a00 0 oOO 0 0 0 0 0 0 0 0 ~OOooOo 0 000
"oooog
ooo00~ 0 000000 0Q O Q aW0 o~Q~o..,
00
sa
Se complete
oQ * : . o$ 0 o*O 0 o o oOOooO 00000 ooOOOo 00 0 0 0 0 0 00 0 0 0 0:-yo 0

-0
Separated\* PI"'"
I;:./>
---.*
...* .00 .. ' 7
leave column
FIG. 8-13. Principle of'the separation of moiecules according to size by gel permeation chromatography (after Cazes 1966 and Billmeyer 1969)
GEL PERMEATION CHROMATOGRAPHY
+ -
Retention volume
=
FIG. 8-14.
A typical get permeation chromatogram: polystyrene in tetrahydrofuran, with M J M , 29
(Fig 8-13), they can diffuse into the internal pore structure of the gel to an extent depending on their size and the pore-size distribution of the gel. Larger molecules can enter only a small fraction of the internal portion of the gel, or are completely excluded; smaller polymer molecules penetrate a larger fraction of the interior of the gel The larger the molecule, therefore, the less time it spends inside the gel, and the sooner it flows through the column The different molecular species are eluted from the column in order of their molecular size (Benoit 1966) as distinguished from their molecular weight, the largest emerging first A complete theory predicting retention times or volumes as a function of molecular size has not been formulated for gel permeation chromatography A specific column or set of columns (with gels of different pore sizes) is calibrated empirically to give such a relationship, by means of which a plot of amount of solute versus retention volume (the chromatogram, Fig 8-14) can be converted into a molecularsize-distribution curve For convenience, commercially available nar~ow-distribution polystyrenes are often used If the calibration is made in terms of a mo1ecula1size parameter, for example, [ q ] M (whose relation to size is given by Eq 8-29), it can be applied to a wide variety of both linear and branched polymers (Fig 8- 15) As in all chromatographic processes, the band of solute emerging from the column is broadened by a number of processes, including contributions from the apparatus, flow of the solution through the packed bed of gel particles, and the permeation process itself (Kelley 1970) Cor~ectionsfor this zone broadening can be made empirically; it usually becomes unimportant when the sample has M,IM,, > 2 Gel permeation chromatography is extremely valuable for both analytic and preparative work with a wide variety of systems ranging from low to very high
MEASUREMENT OF MOLECULAR WEIGHT' AND SIZE
% F
80 FIG. 8-15.
100 120 Retent~onvolume, ml
140
Calrbxat~oncurve fox gel permeallon chromatog~aphybased on hydrodynamic volume as expressed by the product [q]M (Grublslc 1967) Among the polynier types shown are I~near polystyrene, two types o bxanched polystyrene, poly(methy1 methacrylate), poly(vlny1 f chloride), polybutadlene, poly(pheny1 s~loxane),and two types of copolymer
molecular weights. The method can be applied to a wide variety of' solvents and polymers, depending on the type of gel used. With polystyrene gels, relatively nonpolar polymers can be measured in solvents such as tetrahydrofhran, toluene, or (at high temperatures) 0.-dichlorobenzene; with porous glass gels, more polar systems, including aqueous solvents, can be used. A few milligrams of sample suffices for analytic work, and the determination is complete in as short a time as a few minutes using modem high-pressure, high,-speed equipment.. The results of' careful gel permeation chromatography experiments for molecularweight distribution agree so well with results from other techniques that there is serious doubt as to which is correct when residual discrepancies occur. Figure 816 shows the extent of' agreement between this method and a solvent-gradient elution
GEL PERMEATION CHROMATOGRAPHY
100 80
u
L
d
,Y
3
0
5 2 m
s+
40
a,
20
0 lo4
I ~ I I I I L L u
105
lo6
Molecular we~ght
FIG. 8-16
Typical cumulative molecular weight dlstnbution cur.ve for a sample of polypropylene and (Crouzet 1969) gradlent-elution data (0) data from gel permeation chromatography (6) Molecular welght IS plotted logarrthmically because of the very broad dtstrlbution tn this sample
fractionation, while Fig 8- 17 demonstrates the degree of fit between the experiment and a distribution curve calculated from polymerization kinetics
GENERAL REFERENCES
Altgelt 1971; Collins 1973, Chapter 7D, Exp 15; Ouano 1975; ASTM D3535; Billingham 1977, Chapter 8; Tung 1977; Dawkins 1978; Yau 1979; ASTM D3593
x 10-~
FIG. 8-1 7.
Fit of gel permeation chromatography data fbr polystyrene to a molecular-weight distribution curve calculated from polymerization kinetics (May 1968)
21 8
C. POLYELECTROLYTES
Polymers with ionizable groups along the chain, termed polyelectrolytes, no~mally exhibit properties in solution that are quite different from those with nonionizable structures. There are many examples of' polyelectrolytes, including polyacids such as poly(acrylic acid) and hydrolyzed copolymers of' maleic anhydride, polybases such as poly(viny1 amine) and poly(4-vinyl pyridine), polyphosphates, nucleic acids, and proteins When they are soluble in nonionizing solvents-fbr example, poly(acry1ic acid) in dioxane-polyelectrolytes behave in completely normal hshion, but in aqueous solution they are ionized, with three major results First, the mutual repulsion of their charges causes expansions of'the chain f .beyond those resulting from changes I from good to poor solvents with ordinary polymers The size of the polyelectrolyte random coil is, moreover, a function of' the concentrations of' polymer and added salt, if' any, since both influence the degree of' ionization Second, the ionization of' the electrolyte g~oups leads to a variety of unusual effects in the presence of' small amounts of' added salt.. The intensity of' light scattering decreases because of' the ordering of the molecules in solution, while the osmotic pressure and ultracentrifugation behavior are determined predominantly by the total charge on the molecule (Donnan effect) Finally, the ionic charges attached to the chains create regions of' high local charge density, affecting the activity coefficients and properties of small ions in these localities. Although the various effects cannot be separated completely, the results of' chain expansion are of' primary interest for the measurement of' molecular weight and size.. Those properties depending on the size of the chain, such as viscosity and angular dependence of' light scattering, are strongly affected by chain expansion. The viscosity may even increase markedly as polymer concentration decreases, with consequent increase in the degree of ionization of the polymer. When very high chain extensions are reached (up to half' of the fully extended chain length), the effect reverses, but it does not disappear at infinite dilution. On the other hand, the addition of low-molecular-weight electrolyte (salt) to the aqueous solution increases the ionic strength of' the solution outside the polymer coil relative to that inside, and also reduces the thickness of'the layer of' "bound" counterions around the chain.. Both effects cause the chain to contract, and when the concentration of added salt reaches, say, 0.1 M, behavior is again normal. With some special precautions, molecular weights may be measured by light scattering and equilibrium ultracentrifugation, and intrinsic viscosity-molecular weight relations may be established for polyelectrolytes in the presence of' added salt Because of' the preponderance of' small ions, the colligative properties of' polyelectrolytes in ionizing solvents measure counterion activities rather than molecular weight. In the presence of added salt, however, correct molecular weights of' polyelectrolytes can be measured by membrane osmometry, since the small ions can equilibfite across the membrane..The second virial coefficient differs from that previously defined, since it is determined by both ionic and nonionic polymersolvent interactions.
DISCUSSION QUESTIONS AND PROB1.EMS
219
The transport and electrophoretic properties of' polyelectrolytes are beyond the scope of this book
GENERAL. REFERENC;ES
Armstrong 1969; Morawetz 1975, Chapter 7; Anderson 1982
DISCUSSION QUESTlONS A N D PROBLEMS

1. Discuss the value of knowledge of the molecular weight and distribution of a polymer to the plastics engineer 2.
3.
Which methods would you use to obtain this information on a routine basis, as in process control? Why? Which methods would you use to obtain this information for a new polymer type not previously studied? Why?
4. Of what value is the ability to make measurements by such absolute methods as membrane osmometry or light scatte~ing?Review your answers to the previous questions, assuming that these absolute methods were not available, and discuss 5.
You wish to determine M, for a polymer, but find it too high to be measu~ed by vapor-phase osmometry, yet, when using membrane osmometry, there is significant diffusion through the membrane. Outline two different ways of' obtaining the desired information, and discuss their relative merits
6. If light scattering and osmotic pressure are measured in the same solvent, compare the slopes of the plots of 1/M versus c
7.
Over what molecular.-weight range can light scattering be used to measure
M,,,, what limitations set this range? and

8. What types of ultracentrifugation experiment are suitable fbr ando om-coil polymers? I
9. What molecular-weight method would be suitable for. a polymer with an intrinsic viscosity of about 10 dl/g? 10. What kind of information would light scattering give you about the polymer of Question 9?
11. What molecular-weight technique is best for calibrating [ql-.M relations?

12. What kind of' infb~mationdoes one obtain from measuring [q]?
13. How can you make the best guess at M from measurements of [q]on an entirely new kind of' polymer?
-
14. Under what conditions can you use [q]to measure $/M?
220
15. How can ?/M be used to measure chain branching? 16. Calculate the end-to-end distance of a polymer with molecular weight 1,000,000 and intrinsic viscosity of 2 10 dllg; take Q> = 2 1 X 10" What can you say about the light-scattering behavior of such a polymer?
' 17.
Light scattering and osmometry were performed on a solution of a polymer in a O solvent at 27C Concentration was 0 01 glliter Take R = 1 x lo5, K = 0 4 x lo-' for these units, .rr = 3 5 cm of solvent, and Re = 0 01 cm-I Calculate M, and M, and comment on their ratio
18. A polymer with M = 100,000 obeys the Mark-Houwink equation with K = 1 x and a = 0 80 Huggins' constant is 0 33 Calculate the relative viscosity at c = 0 30 gldl
19. A sample of polyethylene was studied by vapor-phase osmometry in o-dichlorobenzene at 130C Data for the polymer and for tristearin, M = 891, a e given below Calculate M, and A, for the polyethylene sample
3
Polyethylene
Tristearin
c (glliter)
Ar
c (glliter)
Ar
20. The following freezing-point data were obtained for a sample of polyethylene in hexamethylbenzene at 16SC:
Chart Divisions From Solvent Freezing Point 60 103 148 189
Polymer Added (g) 0.428 0 739 1 055 1 348
21.
For convenience, a calibration was made with octacosane, M = 395, that gave a fieezing-point depression of 1223 divisions per gram in the same solvent. The density of hexamethylbenzene at the freezing point may be taken as 0.85, to a first approximation. Calculate M,, A*, and The f'ollowing data were obtained by membrane osmometr:y of' solutions of' a polyethylene sample in xylene at 90C. At this temperature the density of'
D1SC:USSION QUESTIONS AND PROBLEMS
221
xylene is 0.8014 and that of polyethylene is 0.8173. Calculate the molecular weight of' the sample and the second virial coefficient.
c (glliter): .rr (cm xylene):
* 22.
2 00 2,585
4.00 5.965
6 .OO 9.660
8 00 13.935
Angular-light,-scattering measurements on solutions of' a nonionic polymer in water at X = 4359 gave the fbllowing results fbr ARe as a function of the angle of observation 8 and concentration c:
ARU x lo4 for c (glliter)
If n
23.
1 340 and dnldc
0 192 mllg, calculate M,, A2, and
Determination of the sedimentation coefficient of a polymer in aqueous solution at a concentration c led to the following data, where r is the distance of the boundary from the axis of rotation and t is the time of centrifugation in minutes:
t (min)
r. (cm)
222
The speed of' centrifugation was 59,780 rpm, the temperature 20C. Calculate the sedimentation coefficient. 24. The following solution-viscosity data were obtained for samples of polystyrene in 2-butanone at 2SC, in which successive increments of a stock solution of polymer were added to solvent in a Ubbelohde viscometer. Finally, the total weight of' polymer in the viscorneter was measured Sample
Efflux time (s) for 20 ml of solvent 20 ml solvent + 5 ml stock 20 ml solvent 10 ml stock 20 ml solvent + 15 ml stock 20 ml solvent 20 ml stock Total grams of' polymer in viscometer at end of' run M, from light scattering
Is there evidence fbr chain branching in any of' the samples? Explain and discuss
25. The fbllowing data were obtained by gel permeation chromatography in tetrahydrofuran at 25C: of a sample of poly(methy1 methacrylate):
Retention Volume (ml) Recorder. Chart Divisions Above Base Line
Samples of narrow-distribution polystyrene chromatogrammed under the same conditions gave a linear calibration curve with M = 98,000 eluting at 130 ml and M = 1800 eluting at 165 ml a Calculate M, and M,, assuming the above calibration applies. b. Comment on possible systematic errors in the calibration and outline in detail a better method Specify what additional information you would need to apply it. c Derive the equation defining the type of molecular-weight average obtained in the "universal" calibration method in gel permeation chromatography
BIBLIOGRAPHY
Allen 1976 G Allen, "A Review of Neutron Scattering with Special Reference to the Measurement of the Unperturbed Dimensions in Macromolecules," Chapter 1 in K J Ivin, ed , Structural Studres of Macromolecules by Specrroscoprc Methods, John Wiley & Sons, New York, 1976 Altgelt I971 Klaus H Altgelt and Leon Segal, eds , Gel Permeation Chromatography, Marcel Dekker, New York, 1971 Anderson 1982 Charles F Anderson and Herbert Morawetz, "Polyelectrolytes," pp 495-530 in Martin Grayson, ed , Kirk-Othmer Encyclopedia of Chemical Technology, 3rd ed , Vol 18, WileyInterscience, New York, 1982 Archrbald 1947 W J Archibald, "A Demonstration of Some New Methods of Determining Molecular Weights from the Data of the Ultracentrifuge," I Phys Collord Chem 51, 1204-1214 (1947) Armstrong 1969 R W Armstrong and U P Strauss, "Polyelectrolytes," pp 781--861 in Herman F Mark, Norman G Gaylord, and Norbert M B~kales, , Encyclopedia of Polymer Science and eds Technology, Vol 10, Wiley-Interscience, New York, 1969 ASTM 02857 Standard Test Method for Dllute Solutton Vrscoszty of Polymers, ASTM-ANSI Designation: D2857, American Society for Testing and Materials, Philadelphia, Pennsylvania, 1970
224
ASTM 03536 Standard Test Method for Molecular WerghtAverages and Molecular Weight D~strrbutzon of Polystyrene by Lrqurd Exclusron Chromatography (Gel Permeation Chromatography-GPC), ASTM-ANSI Des~gnat~on. D3536, Amencan Soc~etyfor Test~ngand Matenals, Ph~ladelph~a, Pennsylvania, 1976 ASTM 03592 Standard Recommended Practice for Determinrng Molecular Werght by Vapor Pressure D3592, Amer~can Soc~ety Testing and Materials, Ph~lfor Osmometry, ASTM-ANSI Des~gnat~on. adelph~a,Pennsylvania, 1977 ASTM 03593 Standard Test Method for Molecular Weight Averages andMolecular Werght Drstrrbutron of Certaln Polymers by Llqurd Srze-E rclusron Chromatography (Gel Permeatron ChromatographyGPC) Uslng Un~ver Calibratron, ASTM Des~gnat~on sal D3593, Amerlcan Soc~ety Test~ng for and Pennsylvan~a,1980 Matenals, Ph~ladelph~a, AS7M 03750 Standard Pract~ce Determrnatron of Number-Average Molecular Weight of Polymers for by Membrane Osmometry, ASTM-ANSI Deslgnat~on D3750, Amer~canSoc~etyfor Test~ng and Pennsylvan~a,1979 Matenals, Ph~ladelph~a, ASTM 04001 Standaxd Pract~ce Determrnatron of Werght-Average Molecular Weight of Polymers for for by Lrght Scatterrng, ASTM-ANSI Designat~onD4001, Amerlcan Soc~ety Testlng and Matenals, Ph~ladelph~a, Pennsylvan~a,1981 Badgley 1949 W J Badgley and H Mark, "Osmometry and V~scometry Polymer Solut~ons," pp of 75-112 m R E Burk and Ol~verGrumm~t,eds , High Molecular Weight Compounds (Frontrers t tn Chemistry), Vol 6, Intersc~ence,New Yolk, 1949 Benolt 1966 Henn Beno~t, Zlatka Grub~sic,Paul Rempp, Danlelie Decker, and Jean-Georges Z~lliox, "Study by L~qu~d-Phase Chromatography of L~nearand Branched Polystryrenes of Known Structure" (m French), J Chem Phys 63, 1507-1514 (1966) Benort 1974 H Beno~t,J P Cotton, D Decker, B Faxnoux, J S H ~ g g ~ nG, Jann~nk,R Ober, C s P~cot, J des Clo~zeaux, and "Conformation of Polymer Chams In the Bulk," Macromolecules 7, 863-871 (1974) Berne 1976 Bruce J Berne and Robert Pecora, Dynarnrc Ltght Scattering, John W~ley Sons, New & York, 1976 Blllrngham 1977 N C B~lllngham, Molar Mass Measurements in Polymer Science, Kogan Page Ltd and Halsted Press, John Wiley & Sons, New York, 1977 Brllmeyer 1955 F W B~llmeyer,JI , and C B de Than, "D~ssymmetryof Molecular Light Scattenng in Polymethyl Methacrylate," I Am Chem Soc 77, 4763-4767 (1955) Brllmeyer 1964a Fred W B~llmeyer,Jr , "Prlnc~plesof L~ght Scattenng," Chapter 56 In I M Kolthoff and Ph111p J Elv~ng, , with the assstance of Ernest B Sandell, Treatise on Analytrcal Chem~stry, eds Part I, Vol 5, W~ley-Intersc~ence, New York, 1964 Blllmeyer 1964b F W B~llmeyer,Jr , and V Kokle, "The Molecular Structure of Polyethylene XV Companson of Number-Average Molecular We~ghtsby Various Methods," J Am Chem Soc 86, 3544-3546 (1964) Brllmeyer 1969 Fred W Blllmeyer, Jr , "Molecular Structure and Polymer Propert~es," J Palnt Techno1 41, 3-16, 209 (1969) Brllmeyer 1971 Fred W B~llmeyer, Jr , Harold I Levlne, and Peter J L~vesey,"The Refract~on Co~rect~on Light Scatter~ng,"J ColloidInterface Scr 35, 204-214 (1971) m B~llmeyer1976 Fred W B~llmeyer,Jr , "Trends In Polymer Charactenzation." J Polym Sci Symp 55, 1-10 (1976) Billmeyer 1977 Fred W B~llmeyer,Jr , "The S ~ z e We~ght Polymer Molecules," Chapter 4 In and of Herman S Kaufrnan and Joseph J Falcetta, eds ,Introdu~tton Polymer Scrence and Technologyto An SPE Textbook, John Wlley & Sons, New York, 1977 Bonnor 1958 R U Bonnar, M D~mbat,and F H Stross, Number-Average Molecular Weights, Intersclence, New York, 1958 Bowen 1970 T J Bowen, An lntroductlon to Ultracenrr$ugation, Wlley-Intersclence, New York, 1970
BIBLIOGRAPHY
225
Brice 1950 B A Brice, M Halwer, and R Speiser, "Photoelectric Light-Scattering Photometer for Determlning High Molecular Weights," I Opt Soc Am 40, 768-778 (1950) Brice 1951 B A Bnce and M Halwer, "A D~fferent~al Refractometer," I Opt Soc Am 41, 10331037 (1951) Bulge 1977 Dav~d Burge, "Molecular Weight Determination by Osmometry," Am Lab 9 (b), 41E 51 (1977) Cabannes 1929 J Cabannes and Y Rocard, La diffusion molPcularre de la lumit?re, Presses Un~versltarres de France, Pans, 1929 Carpenter 1975 Dewey K Carpenter and Lowell Westerman, "Viscometric Methods of Studying Molecular Weight and Molecular Weight Distribution," Chapter 7 in Philip E Slade, JI , ed , Polymer Molecular Werghtr, Part I, Marcel Dekker, New York, 1975 Casassa 1975 Edward F Casassa and Guy C Berry, "L~ght Scattering from Solut~ons Macroof molecules," Chapter 5 in Philip E Slade, 71 , ed Polymer Molecular Weights, Part I, Marcel Dekker, New York, 1975 Cazes 1966 Jack Cazes, "Topics in Chemical Instrumentation XXIX Gel Permeation Chromatography," J Chem Educ 4.3, A567-A582, A625-A632 (1966) Chu 1974 Benjamin Chu, Laser Lrght Scattering, Academic Press, New York, 1974 Coll1968 Hans Coll and F H Stross, "Determination of Molecular Weights by Equilibrtum Osrnot~c Pressure Measurements," pp 10-27 in Donald Mclntyre, ed , Characterrration ofMacromolecular Structure, Publication No 1573, Nat~onalAcademy of Sciences, Washington, D C , 1968 Collins 1973 Edward A Collins, Jan BareS, and Fred W Billmeyer, Jr , Experiments rn Polymer Scrence, W~ley-Interscience,New York, 1973 from Viscosity MeasCowie 1966 J M G Cowie, "Estimation of Unperturbd Polymer Dimens~ons urements in Non-Ideal Solvents," Polymer 7, 487-495 (1966) Cragg 1946 L H Cragg, "The Terminology of Intrinsic Viscosity and Related Functions," J Colloid Scr 1, 261-269 (1946) Crouzet 1969 P Crouzet, F Fine, and P Magnin, "Comparison of the Gel Penneation Chromatographic and Gradient Elution Fractionation Methods for the Determination of the Molecular Weight Dls ti-ibut~on Polypropylene" (in French), J Appl Polym Sci 13, 205-213 (1969) of
Cummins 1973 H Z Cummins and E R Pike, eds , Photon Correlatron and Lrght Beating SpecNew York, 1973 troscopy, Plenum P~ess, Dawkins 1978 J V Dawkins and G Yeadon, "High Pertormance Gel Permeation Chromatography," Chapter 3 In J V Dawkins, ed , Debelopmenrs rn Polymer Characterisation - I , Applied Science, London, 1978 Debye 1944 P Debye, "Light Scattering in Solutions," J Appl Phys 15, 338-342 (1946) Debye 1947 P Debye, "Molecular-Weight Determination by Light Scattering," I Phys Colloid Chem 51, 18-32 (1947) Dobry 1935 A Dobry, "Apparatus for the Measurement of Very Small Osmotic Pressures in Colloidal Solutions" (in French), J Chlm Phys 32, 46-49 (1953) Ehas 1975 Hans-Georg Elias, "Polymolecularity and Polydispersity in Molecular Weight Determinations," Pure Appl Chem 43, 115-147 (1975) Elias 1977 Hans-Georg Ellas, Macromolecules I Strrccture and Properties (translated by John W Stafford), Plenum Press, New York, 1977 Eskin 1964 V E Eskin, "Light Scattering as a Method of Studying Polymers," Sov Phys Usp 7 , 270- 304 (1964) Ewarr 1946 R H Ewart, "Significance of Viscosity Measurements on Dilute Solutions of High Polymers," pp 197-251 in H Mark and G S Whitby, eds , Screntz3c Progress in the Field of Rubber and Synthetrc. Elastomers (Advances m Colloid Science) Vol 2, Interscience, New York, 1946
226
MEASUREMENT OF MOL.ECULAR WEIGHT AND SIZE
Ezrin 1968 Myer Ezrin, "Detemnnation of Molecular Weight by Ebulliometry," pp 3-9 In Donald McIntyre, ed , Characterrzation of Macromolecular Structure, Publication No 1573, National Academy of Sciences, Washington, D C , 1968 Flory 1943 Paul J Flory, "Molecular Weights and Intnnslc V~scosltles P~l~isobutylenes," Am of J Chem Soc 65,372-382 (1943) Flory 1949 Paul J Flory, "The Configuration of Real Polymer Chains," J Chem Phys 17,303310 (1949) Flory 1953 Paul J Flory, Prrnciples of Polymer Chemrstry, Cornell University P r c s ~ ,Ithaca, New York, 1953 Fu~rta1962 Hlroshl Fujita, Mathematrcal Theory of Sedrmentatron Analysrr, Academic Press, New York, 1962 Fujita 1975 Hiroshi Fujita, Foundations of UltracentrifugalAnalysis, Wiley.Interscience, Kew York, 1975 Garmon 1975 Ronald G Garmon, "End Group Detemnatlons," Chapter 3 in Philip E Slade, Jr , ed , Polymer Molecular Werghts, Part I, Marcel Dekker, New York, 1975 Gee 1944 Geoffrey Gee, "The Molecular We~ghts Rubber and Related Materials V ?he Interpreof tatlon of Molecular Weight Measurements on H ~ g h Polymers," Trans F a r h y Soc 40,261-266 (1944) Glover 1966 Clyde A Glover, "Detennlnation of Molecular Weights by Ebull~ornetry," pp 1-67 rn Charles N Rellley and Fred W McLafferty, eds , Advances in Analytrcal Chrmrstq and Instrumentatron, Vol 5, Wiley-Interscience, New York, 1966 Glover I975 Clyde A Glover, "Absolute Colligatrve Property Methods," Chapter 4 In PhiIxp E Slade, Jr , ed , Polymer Molecular Werghts, Part I, Marcel Dekker, New York, 1475 Graessley 1968 William W Graessley, "Detection and Measurement of Branching In PoI:cmers," pp 371-388 in Donald McIntyre, ed , CharacterrzationojMacromolecular Struturr, Pubfzcation NO 1573, National Academy of Sciences, Washington, D C , 1968 Grubrsic 1967 Z Grublslc, P Rempp, and H Beno~t,"A Universal Calibratrrn for Gel Permeation Chromatography," J Polym Scr B5, 753-759 (1967) Hellman 1962 Max Hellman and Leo A Wall, "End-Group Analysis," Chapter 5 ~nfiordor_M Kline, ed ,Analytical Chemrstry oj Polymers, Part 111, Wlley-Intersc~ence,New Ym-ii, 1962. Holleran I968 Peter M Holleran and Fred W B~llrneyer, Jr , "Rap~d Osrnmfi-dry w& D~ffus~ble Polymers," J Polym Sci B6, 137--140 (1968) Huggins 1942 Maunce L Hugglns, "The Viscosity of Dilute Solutions of Longaain .Cfciscules IV Dependence on Concentration," J Am Chem Soc 64, 27 16-271 8 (1942: Huglrn I972 M B Huglin, ed , Lrght Scattering from Polymer Solutions, Aca&.sz kc?;. w Yolk, se 1972 Hyde 1978 A J Hyde, "Light Scattenng in Synthetrc Polymer Systems," Chaprs C ir. I ;Dawklns, ed , Developments in Polymer Characterisatron-I, Applied Science, Loriix. 15'2 Internatronal Union I952 International Union of Pu1.e and Applied Chemistry, ''kqy-,fl<a u~menclature In the Field of Macromolecules," J Polym Scr 8 , 255-277 (1952) Kelley 1970 Richard N Kelley and Fred W Billmeyer, Jr , "A Revtew of Pezz E r c x m g m Gel Chomatography," Sep Scr 5, 291-316 (1970) Kerker 1969 Milton Kerker, The Scatterrng oj Light and Other Electromagnerv Y d ~ r a w Academic Press, New York, 1969 Kinsmger 1975 Jack B Kinsinger, "Macromolecules: Their Masses, Sizes, and i s e 3-=ibutions," ia pp 291-306 in John J Burke and Volker Weiss, eds , Choracterrzatronof 4 f - x d c x Research Ceramics and Polymers, Syracuse University Press, Syracuse, New York, :>7f Kraemer 1938 Elmer 0 Kraemer, "Molecular Weights of Cellulose and Celluk.*.e k ~ a - . v e s , " Ind Eng Chem 30, 1200-1203 (1938)
BIBLIOGRAPHY
227
Kratohvil 1966 J P Kratohv~l,L E Oppenhe~mer,and M Kerker, "Correlat~on of Turb~d~ty and Activtty Data 111 The System Tungstostlicrc Acld-Sodlum Chlonde-Water," I Phys Chem 70, 2834-2839 (1966) Krigbaum 1952 W R Krigbaum and P J Flory, "Treatment of Osmotic Pressure Data," I Polym Sci 9 , 503-588 (1952) Krigbaum 1953 W R Krigbaum and P J Flory, "Stat~sticalMechanics of Dtlute Polymer Solutions IV Variation of the Osmottc Second Coefficient with Molecular We~ght,"J Am Chem Soc 75, 1775-1784 (1953) Krigbaum 1967 W R Kxighaum and R J Roe, "Measurement of Osmotic Pressure," Chapter 79 in I M Kolthoff and Philip J Elving, eds , with the assistance of Ernest B Sandell, Treatise on Analytical Chemistry, Part I , Vol 7, Wiley-Interscience, New York, 1967 Kurata 1960 Michio Kurata, Walter H Stockmayer, and Antonio Roig, "Excluded Volume Effect of Linear Polymer Molecules," J Chem Phys 33, 151- 155 (1960) Kurata 1963 M Kurata and W H stock ma ye^, "Intrinsic Viscosities and Unperturbed Dimensions of Long Chain Molecules," Adv Polym Scr 3 , 196-312 (1963) Kurata 1975 M Kurata, Y Tsunashima, M Iwama, and K Karnada, "Viscosity-Molecular Weight Relationships and Unperturbed Dimensions of Linear Chain Molecules," pp IV-I-IV-60 in J Brandrup and E H Irnmergut, eds ,with the collaboration of W McDowell, Polymer Handbook, 2nd ed , Wiley-Interscience, New York, 1975 Levine 1976 Harold I Levine, Robert J Fiel, and Fred W Billrneyer, Jr , "Very Low-Angle Light Scattering A Characterization Method for High-Molecular- Weight DNA," Bropolymers 15, 12671281 (1976) Lrvesey 1969 P J Livesey and F W Billmeyer, JI , "Particle-Size Determinatton by Low-Angle Light Scattering: New Instrumentation and a Rapid Method of Interpreting Data," I Collozd Interface Sci 30, 447-472 (1969) Lyons 1967 John W Lyons, "Measurement of Viscosity," Chapter 83 in I M Kolthoff and Philip 1 Elving, eds , with the assistance of Ernest B Sandell, Treatrse on Analytzcal Chemistry, Part I, Vol 7, Wiley-lnte~science,New York, 1967 McCall I973 J S McCall and B J Potter, Ultracentrijugation, Macmillan, New Yo~k,1973 McIntyre 1964 D McIntyre and F Gomick, eds , Lzght Scattering from Dilute Polymer Solutions, Gordon and Breach, New York, 1964 McIntyr e 1968 Donald McIntyre, ed , Character izatron of Macromolec ular Structure, Publication No 1573, National Academy of Sciences, Washington, D C , 1968 May I968 James A May, Jr , and William B Smith, "Polymer Studies by Gel Permeation Chromatography 11 The Kinetic Parameters for Styrene Polymerization," J Phys Chem 72, 216221 (1968) Morawetz 1975 Herbert Morawetz, Macromolecules In Solution, 2nd ed , Wiley-Interscience, New York, 1975 M0or.e 1964 J. C Moore, "Gel Permeation Chromatography.. I . A New Method fbr Molecular Weight Distribution of High Polymers," J Polym. Sci A2, 835-843 (1964) Moore 1967 W R Moore, "Viscosities of Dilute Polymer Solutions," Chapter 1 in A D lenkins, ed , Progress m Polymer Sczence, Vol 1, Pergamon Press, New York, 1967 Ouano 1975 Augustus C Ouano, Edward M Banall, II, and Julian F Johnson, "Gel Permeation Chromatography." Chapter 2 in Philip E Slade, 11 , ed , Polymer Molecular Wezghts, Part 11, Marcel Dekker, New York, 1975 Pasternak I962 R A Pasternak, P Brady, and H C Ehrmantraut, "Apparatus f o ~ Rapid Deterthe mination of Molecular Weight," DECHEMA Monogr 44, 205-207 (1962) Rrchards 1978 R W Richards, "Molecula~ Dimensions of Amorphous Polymers by Neutron Scatterlng," Chapte~ in J V Dawkins, ed , Developments in Polymer Characterisatron-I, Applied 5 Science, London, 1978
228
MEASUREMENT OF MOLECUL.AR WEIGHT AND SIZE
Scholte 1975 Th G Scholte, "Sedlmentat~on Techn~ques," Chapter 8 in P h ~ l ~ p Slade, Jr , ed , E Polymer Molecular Weights, Part 11, Marcel Dekker, New York, 1975 Sh~moda1976 K Shlmoda, ed , H~gh-Resolutron Laser Spectroscopy, Springer-Verlag, New York,
1967
Stabin 1954 J V Stabln and E H Immergut, "A H~gh-Speed Glass Osmometer," J Polym Sci 14,
209-212 (1954)
Staudrnger 1930 H Staudlnger and W Heuer, "H~ghlyPolymenzed Compounds XXXIII A Relat~on Between the V~scosity the Molecular We~ght Polystyrenes" (In German), Ber Dtsch Chem and of Ges B 63, 222-234 (1930) Stockmayer 1952 Walter H Stockmaye1 and Edward F Casassa, "The Th~rdVlnal Coeffic~entIn Polymer Solut~ons,"J Chem Phys 20, 1560-1566 (1952) Stockmayer 1963 W H Stockmayer and Marshall F~xman,"On the Estlmat~onof Unperturbed DImenslons from Intnns~c V~scos~t~es,"Polym SCI C1, 137-141 (1963) J Svedberg 1940 The Svedberg and Kal 0 Pedersen, 7he Ultracentrrfuge, Clarendon Press, Oxford, 1940, Johnson Repnnt Corp , New York, 1959 Tung 1977 L H Tung and John C Moore, "Gel Pe~meat~on Chromatography," Chapter 6 In L H New Tung, ed , Fractionairon oj Synthet~cPolymers Princrples and Practices, Marcel Dekke~, York, 1977 d Ulrlch 1975 Robert D Ul~lch,"Membrane Osmometiy," Chapter 2 In Ph~llpE Slade, Jr , ed , Polymer Molecular Weights, Part I, Marcel Dekker, New York, 1975 van de Hulst 1957 H C van de Hulst, Llght Scatterlng by Small Partrcler. John Wiley & Sons, New York, 1957 W~llrams1972 J W Will~ams, Ultracentrlfugarron of Macromolecules Modern Toptcs, Academic Press, New York, 1972 Yau 1979 W W Yau, J J Klrkland, and D D Bly, Modern S ~ z e Exclus~on Lzquid Chromatography & Practrce oj Gel Permeation and Gel Frltration Chromatography, John W~ley Sons, New York,
1979
Zrmm 1946 B H Zlmm and I Myerson, "A Conven~entSmall Osmometer," J Am Chem Soc 68,
911-912 (1946)
Z~mm 1948a Bruno H Zlmm, "The Scattering of L~ght the Radlal D~strlbut~on and Funct~on H ~ g h of Polymer Solut~ons,"J Chem Phys 16, 1093-1099 (1948) Z~mm1948b Bruno H Zlmm, "Apparatus and Methods for Measurement and Interpretatron of the of Scatterlng Prel~m~nary Results on Polystyrene Solut~ons,"J Chem Angular Vanat~on L~ght Phys 16, 1099-1 116 (1948) Z~mm 1949 Bruno H Zlmm and Walter H Stockmayer, "The D~mensrons Cham Molecules Conof tainlng Branches and R~ngs," J Chem Phys 17, 1301-1314 (1949) Z~mm 1962 Bruno H Zimm and Donald M Crothers, "Slmpl~fied Rotat~on Cyl~nderV~scometerfor DNA," Proc Not Acad Sci U S A 48, 905-911 (1962)
CHAPTER
ANALYSIS A N D TESTING OF POLYMERS
The purpose of this chapter is to assemble information on the applicability to polymers of various standard methods of physical and chemical analysis. A background is thus established for the uninterrupted discussion in succeeding chapters of of the ~esults such analyses and tests In no case is the treatment in this chapter exhaustive; it is intended to provide, with minimum detail, a broad picture of the applicability of each method to polymer systems and of the type of information obtained
A.
CHEMICAL ANALYSIS OF POLYMERS
The chemical analysis of polymers is not basically different from the analysis of low-molecular weight organic compounds, provided that appropriate modification is made to ensure solubility or the availability of sites for reaction (e g , insoluble specimens should be ground to expose a large surface a ~ e a ) The usual methods for functional group and elemental analysis are generally applicable, as are many other standard analytic techniques Sevexal books on the identification and analysis of plastics have appeared (Haslam 1965, Saunders 1966, Lever 1968, Wake 1969, Braun 1982, Kxause 1983); the analytical chemistry of the more common plastics is treated by Kline (1959). Chemical reactions of polymers provide additional means of chemical analysis, as do their reactions of degradation (Chapter 6 0 ) Collins (1973) discusses various analyses as means of following the course of polymerization Brief descriptions follow of two powerful techniques for analyzing lowmolecular.weight products of reactions (usually degradation) of polymers
230
ANALYSIS A N D TESTING O F POLYMERS
Mass Spectrometry (Wall 1962, Sedgwick 1978). In the most common applications of mass spectrometry to polymer systems, the polymer is allowed to react (as in thermal degradation, Chapter 6 0 ) to form low-molecular-weight fragments that ase condensed at liquid-air temperature. They ase then volatilized, ionized, and separated according to mass and charge by the action of' electric and magnetic fields in a typical mass spectrometer analysis. From the abundance of' the various ionic species found, the structures of the low-molecular-weight species can be inferred. Gas Chromatography (Cassel 1962, Stevens 1969) Gas chromatography is a method of separation in which gaseous or vaporized components are distributed between a moving gas phase and a fixed liquid phase or solid adsorbent By a continuous succession of adsorption or elution steps, taking place at a specific rate for each component, separation is achieved The components are detected by one of several methods as they emerge successively from the chromatographic column From the detector signal, proportional to the instantaneous concentration of t4e dilute component in the gas stream, is obtained information about the number, nature, and amounts of the components present
GENERAL REFERENCES
Kline 1959; Brauer 1965; Feinland 1965; B~aun1982; Krause 1983
B.
SPECTROSCOPlC METHODS
Infrared Spectroscopy
Emission or absorption spectra arise when molecules undergo transitions between quantum states corresponding to two different internal energies The energy difference AE between the states is related to the frequency of' the radiation emitted or absorbed by the quantum relation AE = h v . Infrared frequencies in the wavelength range 1-50 pm'i' are associated with molecular vibration and vibrationrotation spectra A molecule containing N atoms has 3N normal vibration modes, including rotational and translational motions of the entire molecule. For highly symmetrical molecules with very few atoms, the entire infiar.ed and Raman$ spectrum can be
'i Infrared absorption wavelengths are often expressed in wave numbe~s (wavelength-') in centimeter- ':
1 pm = 10,000 cm-I, 50 p m = 200 cm-I, and so on, where pm = micrometer, 10 'meter, formerly (and still widely) called the micron
$The Raman effect results when the frequency of' visible light is altered in the scattering process by the absorption or emission of energy produced by changes in molecuiar vibration and vibration-lotation quantum states
SPECTROSCOPIC METHODS
231
correlated to and explained by the vibrational modes, but even fbr most lowmolecular-,weight substances, N is too large fbr such analysis. Useful information can still be obtained, however, because some vibrational modes involve localized motions of small groups of' atoms and give rise to absorption bands at frequencies characteristic of these groups and the type of motions they under.go In polymers the infixed absorption spectrum is often surprisingly simple, if one considers the large number of' atoms involved. This simplicity results first from the fact that many of the normal vibrations have almost the same frequency and therefore appear in the spectrum as one absorption band and, second, from the strict selection rules that prevent many of the vibrations from causing absorptions.
Experimental Methods. General features of infrared spectroscopy are by now so well known that it suffices to comment briefly on special problems of sample preparation for polymers One of the greatest experimental difficulties in work with substances that absorb heavily in the infrared, including many polymers, is obtaining sufficiently thin samples. Common methods of sample preparation include the following: compression molding, by far the most widely used preparation technique; dissolving the polymer in a solvent, such as carbon disulfide or tetrachloroethylene, whose spectrum is relatively free of intense absorption bands; preparing a thin film by microtoming ox milling; casting a thin film from solution; and pressing a finely ground mixture of the sample with KBr to form a disc or wafer Detection of Chemical Groups in Typical Spectra. Figure 9-1 gives a few of the many chemical linkages or groups that can be detected in polymer spectra, together with the approximate wavelengths at which they occur Among the infrared absorption bands of interest in common polymers are those in polyethylene corresponding to C-H stretching (3.4 pm), C-H bending of CH, groups (6 8 pm) and CH, groups (shoulder at 7 25 pm on an amorphous band at 7 30 pm), and CH, ~ocking sequences of methylene groups in paraffin structures (13 9 pm) of (The structure of polyethylene is discussed in more detail in Chapter 13 ) Other absorption bands of interest are those due to C=C at 6 1 p n in natural rubber; carbony1 at 5 8 pm and ether at 8 9 pm in poly(methy1 methacrylate); aromatic structures at 6 2, 6 7, 13.3, and 14 4 pm in polystyrene; C-CI at 14 5 pm in poly(viny1 chloride); peptide groups at 3 0, 6.1, and 6 5 pm in nylon; and CF2 at 8 2-8 3 pm in polytetrafluoroethylene Related Methods. It is desirable to supplement as much as possible the observations of infrared absorption usually obtained in the 2-15 pm wavelength range. For this purpose observations in the far-inpared region (say up to 200 ~ m become ) important. Despite serious experimental difficulties, valuable information is available in this region: For some polymers, such as polytetrafluoroethylene, most of the absorption bands occur above 15 pm. The application of'Fourier-transform infrared spectroscopy to polymers has been reviewed (D'Esposito 1978). Since the requirements for activity of a vibration for causing absorption in the infrared and fbr causing Raman ~catteringare often

I
Assignments
X-H
Strething CEX
I
X-H Deformation and C - X
Stretching
C = X Stretching
Stretching
f--f
Functional group absorption regions
-1
I
Rocking
C-H(alkane)
I
C= N
I
C-N
C-H(arornatfc)
I
S -H
I
N-H
CH2, CH3
I
C-H
N- H Rocking
C-F
I I
13
14
15
5000
3000
2000
1000 1500 Wavenumbers, cm-I
900
800
700
FIG. 9-1.
Infzared absorption bands of interest in polymers a~rangedby approxlmate wavelength and frequency
different, information fiom Raman experiments in general supplements that obtainable from infrared absorption.. Utilization of' lasers as light sources has greatly decreased experimental difficulties in this technique, and the method has now become one of the standard tools fbr polymer analysis (Hendra 1969, Koenig 1971)
Dichroism. For any molecular vibration leading to infrared absorption, there is a periodic change in electric dipole moment If the direction of this change is parallel to a component of the electric vector of the infrared radiation, absorption occurs; otherwise it does not. In oriented bulk polymers, the dipole-momentchange can be confined to specified directions The use of polarized infrared radiation then leads to absorption that is a function of the orientation of the plane of polarization This phenomenon is called dichroism and is usually measured as the dichroic ratio, the ratio of the optical densities of an absorption band measured with radiation polarized parallel and perpendicular, respectively, to a specified direction in the sample. Dichroic ratios depend upon both the degree of orientation and the angle between the direction of the transition moment and the selected direction in the sample (for example, the axial direction in a fiber) They usually range between 0.1 and 1 0 (see also Chapter 10E) Crystallinity. The infrared absorption spectra of' the same polymer in the crystalline and amorphous states can differ fbr at least two reasons. First, specific intermolecular interactions may exist in the crystalline polymer that lead to sharpening or splitting of' certain bands; and second, some specific confbrmations may
233
exist in one but not the other phase, leading to bands character.istic exclusively of' either c~ystalline amorphous material. An example of the latter effect is poly(ethy1ene or terephthalate), in which the -OCH,CH,O-portion of each repeat unit is restricted to the all-trans conformation in the crystal, but can exist in part in the gauche fbrm in the melt. Several bands characteristic of each confbrmation have been identified (Ward 1957). In favorable cases, such as 66-nylon (Starkweather 1956), percent crystallinity can be determined in absolute terms from infrared absorption data
Geometric Isomerism. The determination of' the various types of' geometric isomers associated with unsaturation in polymer chains is of great importance, for example, in the study of' the structure of modern synthetic rubbers (Chapter 1.3) Table 9-1 lists some of' the important infrared absorption bands resulting from olefinic groups In synthetic "natu~al" rubber, cis-1,4-polyisoprene, relatively small amounts of 1,2.-and 3,4-addition can easily be detected, though it is more difficult to distinguish between the cis and trans configurations. Nuclear magnetic resonance spectroscopy (see below) is also useful fbr this analysis Nuclear Magnetic Resonance Spectroscopy
Since about 1960 nuclear magnetic resonance (NMR) spectroscopy has become a major tool for the study of chain configu~ation, sequence distribution, and microstructure in polymers Its use has evolved from early broad-line studies of the onset of molecular motion in solid polymers, through the widely practiced solution studies of proton NMR, to the application of the more difficult but more powerful carbon 13 NMR methods to both liquids and solids Despite the widespread use of NMR, a brief summary of its origins and experimental methods is warranted
Experimental Methods. The NMR technique utilizes the property of spin (angular momentum and its associated magnetic moment) possessed by nuclei whose atomic number and mass number are not both even Such nuclei include the isotopes of hydrogen ?nd I3C, I5N, I7O, and 19F Application of a strong magnetic field to materiai containing such nuclei splits the energy level into two, representing states
TABLE 9-1. Absorption Wavelengths of Olefinic Groupsa - -
Group Containing C=C

Vinyl, R,CH=CH2 trans-R,CH=CHRZ Vinylidene, R,R2C=CH, R,R,C=CHR, cis-R,CH=CHR2
Wavelength (ym)
10 1 and 1 1 0 10 4 11 3 12 0 14 2 (variable)
"CIOSS (1950)
234
with spin parallel and antiparallel to the field. Transitions between the states lead to absorption or. emission of' an energy
where the frequency vo is in the microwave region for. fields of' strength Hoof' the order of 10,000 gauss and up, and p, is the magnetic moment of' the nucleus. The energy change is observed as a resonance peak or line in experiments where either H or vo is varied, with the other. held constant With an assembly of nuclei, the field on any one is modified by the presence of' the others:
where H, is the local field with a strength of 5-10 gauss A distribution of local fields usually exists so that the resonance line becomes broadened This broadening was once studied as a function of temperature to indicate the temperature of the onset of molecular motion, but it was soon found that with ample molecular motion present, as with liquids and solutions, quite narrow lines can be observed The positions of these lines on the scale of frequency or magnetic field depend on the local fields, which in turn result from the nature and location of the atomic groups in the vicinity of the protons Ths displacements in the resonances, called chemical shifts, are measured in parts per million in frequency (or the equivalent field strength) on a scale labeled 6. The zero of the 6 scale is a reference point provided by the single resonance of the equivalent protons in a substance showing minimum chemical shift, such as tetramethyl silane. Tables of chemical shifts resulting from groups in the neighborhood of the nucleus being studied are now common and are not repeated here In proton NMR, additional complexity and additional information result from coupling of the resonances of protons on adjacent carbon atoms, resulting in the 1 peaks, where n is the number of equivalent splitting of their resonances into n neighboring protons To aid in interpretation, two experimental modifications a e useful The first of these is the use of high magnetic field strengths, m the range 60,000-220,000 gauss compared to the few tens of thousand gauss used in early broad-line work Superconducting magnets are used to obtain the highest field strengths. As Fig 9-2 shows, the increased resolution at the higher field strength aids in the unambiguous interpretation of the spectra. The second development, double resonance or spin decoupling, effects great simplifications in the spectra. A second radiofrequency field is used that has the effect of removing the coupling and collapsing multiplet spectra to much simpler ones (Fig 9-3). Mention should also be made of the use of 2H NMR to provide spectra simplified by elimination of proton-proton coupling due to the lower frequency of occurlence of the deuteron A great advantage of I3C NMR over proton NMR is the much greater range of chemical shifts exhibited by 13C(and most other nuclei), some 200 ppm in contrast
rnD
__ A7 20
W
169
6 0 MHz CDCI, SCCVENT r -H+

0
rn
CDCI3 SOLVENT
- . - - . 4- h -.
_I___-I___ L -H-720 6 0
C C C
220 MHz C D C l j SOLVENT

-J. - -
J b
L
1 69
-H+
0
-I
720
El
la
FIG. 9-2. High-resolution NMR spectra of bisphenol-A polycarbonate in deuterochlo~oformsolution showing the effect of frequency on resolution (Sudol 1969)
to 10 ppm for 'H. In addition, there is no carbon-carbon spin coupling due to the low chemical abundance of the l3C nucleus, 1 . I % relative to I2C. Finally, the resonances of carbon nuclei are sensitive to configurational as well as chemical environments. Offsetting these advantages are the lower abundance ofthe 13Cnuclei and their lower nuclear magnetic moment, about one-quarter that of' the proton; both effects reduce the intensity of the NMR signal This lower sensitivity is overcome by the use of pulsed Fourier-transform NMR, in which a high-power microsecond pulse of radiofrequency energy sets all the carbon nuclei into resonance at once, eliminating the need to sweep the frequency or magnetic field. The data are recorded as the subsequent decay of' the resonance with time, which is the Fourier transform of the desired spectrum. Records from
Normal spectrum
11
Double resonance spectrum
10 0
95
25
20
6 FIG. 9-3. NMR spectra of propylene oxide in carbon tetrachlonde solution: (top) conventional spectrum and (bottom) double-resonance spectrum showing s~mplificat~on resulting from collapse of the splitting due to coupling (Bovey 1965)
repeat pulses are summed by computer, which transforms the data to the spectrum desired when the signal-to-noise ratio has reached a satisfactory level Interest in studying molecular motion in solid polymers has returned with the development of the theory of line broadening It can be shown that the local field seen by a pair of like nuclei in a solid sample is given by an isotropic component (the origin of the chemical shift obserlred in solution) plus a broadening component proportional to 3 cos28 - 1, where 8 is the angle between the line connecting the nuclei and the direction of the magnetic field If the structure of the polymer solid is random, this component broadens the resonance to a width greater than the total I3C spectrum; but if the specimen can be oriented and spun around an axis at 8 = 54.7", for which 3 cos2 8 - 1 = 0-the so-called "magic angleM--the line width is reduced enough so that all the carbon resonances can be resolved.
Stereochemical Configuration. A major application of NMR to polymer systems has been the elucidation of the stereochemical configurations of' polymer chains; this topic is discussed fhrther in Chapter 10A. Poly(methy1 methacrylate) was first studied over 20 years ago (Bovey 1960). It is now possible to analyze for the statistical frequency of occurrence of all possible combinations of' up to four successive pairs of units (dyads) that can occur with either the same (meso) or opposite (racemic) configurations
Geometric Isomerism. Both pr.oton and 13C NMR have proven very useful in supplementing infrared spectroscopy as a means of' determining isomerism around the carbon-carbon bond in polymer chains (Duch 19'70)
The principles of analyzing short sequences of monomers in a copolymer are not different from those of determining stereochemical configuration NMR results have contributed extensively in this field
Electron Paramagnetic Resonance Spectroscopy
Copoiymer Sequences.
Although electron paramagnetic resonance (EPR) and NMR spectroscopy are similar in basic principles and in experimental techniques, they detect different phenomena and therefore yield different information The major use of EPR spectroscopy lies in the detection of free radicals (Chapter 3A) These species are uniquely characterized by their magnetic moment, arising from the presence of an unpaired electron Measurement of a magnetic property of a material containing free radicals, such as its magnetic susceptibility, gives the concentration of free radicals, but the method lacks sensitivity and cannot reveal the structure of the radicals Electron paramagnetic resonance spectroscopy is essentially free from these limitations
Experimental Methods. As in NMR spectroscopy, the action of a strong magnetic field on a material containing free radicals removes the degeneracy of their groundstate energy levels For low radical concentrations the new energy levels are given by two terms The first is
where g is a tensor relating the field direction and the symmetry directions in the radical, 6 is the magnetic moment of the electron spin, and pois the magnetic permeability of a vacuum The second term represents coupling of the electron spin with the nuclear spins in the molecule. This coupling results in the splitting of the resonance line into a symmetrical group of lines whose positions and amplitudes depend on, and therefore give information about, the structure of the radical
Applications to Polymers. The investigation of free radicals produced by highenergy irradiation (Chapter 6 0 ) of polytetrafluoroethylene (Rexroad 1958, Lebedev 1960) serves as an example of an early application of EPR to polymer radicals The EPR spectrum of irradiated polytetrafluoroethylene is interpreted as arising from radicals of the type
remaining trapped in the polymer after a C--F bond has been broken and the fluorine atom has diffused away The spectrum shows fine structure due to coupling between the unpaired electron and neighboring 19Fnuclei The fine structure is lost
238
below 270 K as the motion of the I9Fnuclei is slowed down. The spectrum does not change at higher temperatures, indicating that the radical is stable to at least 550 K. Secondary products, resulting f ~ o m reaction of' the primary radical with the such substances as 0, and NO, were also studied It may be noted that, as discussed in Chapters 1C and 6 0 , the strength of' the c-F bond and the tendency of' polytetrafluoroethylene to degrade rather than crosslink on irradiation would suggest the breaking of a C-C rather than a C-F bond as a likely source of' radicals in this polymer.
GENERAL REFERENCES
Infrared spectroscopy. Zbinden 1964; Hummel 1966; Kossler 1967; Elliott 1969; Boerio 1972; Collins 1973, Exp 32; Siesler 1980. NMR spectroscopy: Bovey 1968, 1969, 1972, 1976; Schaefer 1974, 19'76, 1979; Cunliffe 1978.. EPR spectroscopy: Bresler 1966; Bullock 1976; RBnby 1977; Boye~1980.
C.
X-RAY DIFFRACTION ANALYSIS
The .x-ray dzffracrion method is a powerful tool for investigating orderly arrangements of atoms or molecules though the interaction of' electromagnetic radiation to give interference effects with structures comparable in size to the wavelength of' the radiation. If' the structures are arranged in an orderly array or lattice, the interferences are sharpened so that the radiation is scattered or diffracted only under specific experimental conditions.. Knowledge of these conditions gives information regasding the geometry of the scattering structures. The wavelengths of' x-rays are cornpasable to interatomic distances in crystals; the infbrmation obtained from scattering at wide angles describes the spatial arrangements of the atoms. Low-angle x-ray scattering is useful in detecting larger periodicities, which may arise from lamellar crystallites (Chapter 10C) or from voids. The results of the application of' the x-ray method in determining the crystal structures of' polymers ase described in Chapter 10B. Collected crystallographic data for polymers have been tabulated (Miller 1975)
X-rays are usually produced by bombasding a metal target with a beam of high.voltage electrons. This is done inside a vacuum tube, the x-rays passing out through a beryllium or polyester film window in the tube in a well-defined beam. Choice of'the target metal and the applied voltage determines the wavelength or wavelengths of x-rays produced. Experiments in which nearly monochromatic x-rays are used are of' the greatest interest
X-RAY DIFFRACTION ANALYSIS
The diffiacted x-rays may be detected by their action on photographic films or plates, or by means of a radiation counter and electronic equipment feeding data to a computer. Each method has its advantages. Qualitative examination of the diffraction pattern and accurate measurement of angles and distances are best made from a photographic record, whereas for precise measurement of the intensity of' the diffiacted beam the counting technique is preferred X-rays of a given wavelength ase diffiacted only for certain specific orientations of' the sample If the sample is a single crystal, it must be placed in all possible orientations during the experiment, usually by rotating or oscillating the sample about one of its axes to achieve the desir.ed orientations with respect to the x-ray beam. Alternatively, a sample made up of' a powder of very small crystals may be used. If the minute powdered particles are randomly oriented, all orientations will be included within the sample.. The powder method is more convenient, but gives less information than the single-crystal method, since the latter allows orientations about one crystal axis at a time to be investigated
Application to Polymers
Since polymer single crystals as now prepared (Chapter 10C) are too small for xray diffraction experiments, the crystal structure of a polymer is usually determined from x-ray patterns of a fiber drawn from the polymer. Because of the alignment of the crystalline regions with the long axes of the molecules parallel to the fiber axis (Chapter IOE), the pattern is essentially identical to a rotation pattern from a single crystal. In such a pattern diffraction maxima occur in rows perpendiculas to the fiber axis, called layer lines (seen, for example, in Fig.. 10-226).
Chain Conformations. As the periodicity of molecular structure of a polymer is characterized by the existence of a repeat unlt, so the periodicity of its crystal is characterized by a repeat dzstance The repeat distance 1s directly determined by measuring the distance between the layer lines: The greater the repeat distance, the clo~er together are the layer lines Determination of the repeat unit is considerably moIe difficult Classically, it proceeds through derivation, from the positions of the diffracted x-ray beams, of the dimensions of the unzt cell (usually the simplest geometric volume unit that, by repetition, builds up the three-dimensional crystalline ar~ay)The positions of the atoms in the unit cell are then derived from the relative intensities of the diffracted beams The diffraction patterns of polymers, however, do not provide sufficient information to allow such analyses to be carried to completion Additional structural information is utilized from other sources, such as normal bond lengths and angles and atomic arrangements along the chain suggested by the chemical structure of the sample For example, the repeat distance of 2 55 A in the crystals of polyethylene is readily identified with a single repeat unit in the planar zigzag conformation Many polymer conformations can be described in terms of atoms regularly spaced along helices (Chapter 10B) Methods for analyzing the diffraction patterns
240
from helical structures provide a highly versatile technique for the determination of the repeat unit in such polymers.
Chain Packing. The packing of the chains is described most completely in terms of the unit cell and its contents (shown for polyethylene in Fig. 10-3). The volume of the unit cell, and hence the volume occupied in the crystal by a single repeat unit, can be obtained f'inm the repeat distance and the positions of the diffraction spots on the layer lines. This volume measures the density of the crystals, which is useful in determining the degree of'crystallinity, as outlined in Chapter 10D. The atomic arrangement within the unit cell is much more difficult to determine than the cell dimensions. Trial structures can often be deduced from these dimensions and a knowledge of'the chain conformation, and can be tested in terms of'calculated and observed diffraction intensities, using well-established crystallographictechniques Disorder in the Crystal Structure. In the discussion so f i r it has been assumed that polymer crystals consist of' perfect geometrical arrays of' atoms. HowevCr, defects and distortions are present in crystals of' all materials, and, together with grosser types of disorder, they play an important role in the crystal structures of' polymers. X-ray diffiaction data give a great deal of' information about the qualitative aspects of disorder in the crystal structure, since disordering results in the broadening ofthe diffraction maxima.. Since different classes of' diffraction maxima are sensitive in different ways to particular types of' disorder, a systematic study of'the broadening yields detailed infbrrnation on the types of' disturbances in the crystals (Wunderlich 1973) Quantitative measurement of the disorder is difficult because broadening of the diffraction maxima is caused both by small crystallite size and by distortions within larger, crystals.. As discussed in Chapter 10C, the fbrmer explanation was traditionally adopted in interpreting the diffraction pattern from polymers. However, the distortion of perfect crystals into paracrystals (Hosemann 1962) is now considered important in many polymers. Orientation. Unless the crystalline regions in polymers are oriented, as by drawing a fiber, the diffraction maxima merge into rings, made up of the maxima from a large number ofcrystallites in many different orientations (Chapter IOE, especially Fig. 10-22).. If orientation is intermediate between these extremes, the rings split into arcs, and a quantitative evaluation of the crystallite orientation is made by measuring the angular spread and intensity of these arcs
GENERAL REFERENCES
Alexander 1969; Kakudo 1972; Collins 1973, Chapter 8B, Exp. 18; Wunderlich 1973; Herglotz 1975; Brown 1978
D. MICROSCOPY
While the techniques of' light and electron microscopy are well described in the literature, their application to polymer analysis involves sufficient extension beyond the ordinary techniques to warrant brief' comments
Light Microscopy. Reflected-light microscopy is valuable for examining the texture of solid opaque polymers For materials that can be prepared as thin films (often, for example, by casting on the microscope slide), examination by transmitted light is usual, but little detail can be seen without some type of enhancement Two common techniques are used One is polarized-light microscopy, in which advantage is taken of the ability of crystalline materials to rotate the plane of polarized light Thus the structure of spherulites (Chapter IOC) is studied with the sample between crossed polarizers, and the crystalline melting point is taken as the temperature of disappearance of the last traces of crystallinity when using a hot-stage polarizing microscope A second useful technique is phase-contrast microscopy, which allows observations of structural features involving differences in refractive index rather than absorption of light as in the conventional case Interference microscopy, allowing measurement of thicknesses as low as a few angstrom units, has proved valuable in the study of polymer single crystals (Chapter IOC) Resolution, of course, is limited to object sizes about half the wavelength; practically, about 2000 '4 Electron Microscopy and Electron Diffraction. Resolution of smaller objects can be achieved in electron microscopy, though the theoretical resolving power mentioned above is scarcely approached, the practical limit of resolution being a few angstrom units Electron microscopy has been a powerful tool in the study of the morphology of crystalline polymers (Chapter 10C) The usual techniques of replication, heavy-metal shadowing, and solvent etching are widely used The direct observation of thin specimens, such as polymer single crystals, is also poss~ble and allows the observation of the electron-diffraction pattern of the same speclmen area, invaluable for determining crystallographic directions and relating them to morphology A severe problem, however, is damage of the specimen by the electron beam Typically, polymer single crystals are severely damaged in times of a few seconds to a minute This problem can be alleviated by maintaining the sample well below room temperature with a cold stage, by the use of accelerating voltages several times higher than the usual 50,000- 100,000 V, or by the use of an image intensifier providing a cathode-ray tube image with far less beam current than that ordinarily required Scanning Electron Microscopy. In scanning electron microscopy a fine beam of electrons is scanned across the surface of an opaque specimen to which a light conducting film has been applied by evaporation Secondary electrons, backscattered electrons, or (in the electron mzcroprobe) x-ray photons emitted when the beam hits the specimen are collected to provide a signal used to modulate the
242
intensity of' the electron beam in a television tube, scanning in synchronism with the microscope beam. Because the latter maintains its small size over large distances relative to the specimen, the resulting images have great depth of' field and a remarkable three-dimensional appearance (conventional stereoscopic pairs of' photographs can be made). Resolution is cur~entlylimited to the order of 100 A .
GENERAL REFERENCES
Birbeck 1961; Hartshorne 1964; Fischer 1966; Geil 1966; Sjostrand 1967; Thornton 1968; Collins 1973, Chapter 8C,D, Exps. 19 and 20; Hemsley 1978
E.
THERMAL ANALYSIS
The field of the thermal analysis of polymers has expanded greatly since the introduction of' simple, inexpensive instruments (compared to classicaI instrumentatibn) for several types of thermal measurements just over 20 years ago In addition to the traditional calorimetric and differential thermal analysis, the field now includes equipment fbr thermogravimetric analysis, thermomechanical analysis, electrical thermal analysis, and effluent gas analysis. Not only can one study the enthalpy ' changes associated with heating, annealing, crystallizing, or otherwise thermally treating polymers, but one can now study a wide variety of' responses ofthe system to temperature, including polymerization, degradation, or other chemical changes
Differential Scanning Calorimetry Experimental Methods. In contrast to earlier use of' a large, expensive adiabatic calorimeter for measurements of' specific heat and enthalpies of transition, these measurements are now usually carried out on quite small samples in a differential scanning calorimeter (DSC). The term is applied to two different modes of analysis, of which the one more closely related to traditional calorimetry is described here. In DSC an average.-temperaturecircuit measures and controls the temperature of' sample and reference holders to conform to a predetermined time-temperature program. This temperature is plotted on one axis of' an .x-y recorder At the same time, a temperature-difference circuit compares the temperatures of the sample and reference holders and proportions power to the heater in each holder so that the temperatures remain equal. When the sample undergoes a thermal transition, the power to the two heaters is adjusted to maintain their temperatures, and a signal proportional to the power difference is plotted on the second axis of'the recorder. The area under the resulting curve is a direct measure of' the heat of' transition. Although the DSC is less accurate than a good adiabatic calorimeter (1-2% versus 0. I%), its accuracy is adequate for most uses and its advantages of' speed and low cost make it the outstanding instrument of' choice for most modern calorimetry
THERMAL ANALYSIS
243
Application to Polymers. As a typical result, specific heat--temperature curves obtained (by adiabatic calorimetry) (Smith 1956) on heating quenched (amorphous) specimens of'poly(ethy1ene terephthalate) are shown in Fig. 9-4.. Each curve rises linearly with temperature at low temperatures and then rises more steeply at the glass transition, 60-80C. With the onset of' mobility of' the molecular chains above this transition, crystallization takes place, as indicated by the sharp drop in the specific heat curve. At still higher temperatures, 220-27O0C, the crystals melt with a cor~esponding rise in the specific heat curve
Temperature, "C
FIG. 9-4. Curves of specific heat as a function of increasing tempelahre for quenched (amorphous) poiy(ethy1ene terephthalate) (Smith 1956)
Differential 'Thermal Analysis Experimental Methods. In differential thermal analysis (DTA) the sample and an inert reference substance, undergoing no thermal transition in the temperature range of interest, are heated at the same rate. The temperature difftrence between sample and reference is measured and plotted as a function of sample temperature The temperature difference is finite only when heat is being evolved or absorbed because of exothermic or endothermic activity in the sample, or when the heat capacity of the sample is changing abruptly Since the temperature difference is directly proportional to the heat capacity, the curves resemble specific heat curves, but are inverted because, by convention, heat evolution is registered as an upward peak and heat absorption as a downward peak Application to Polymers. A typical DTA result (Ke 1960) is the differential thermal analysis curve for poly(ethy1ene terephthalate) shown in Fig. 9.-5.Features of' the curve may readily be identified by comparison with Fig 9-4. The lower crystalline melting range in the specimen of Fig. 9.-5 is attributed to impuritiek in the polymer. Other Thermal Methods Thermogravimetric Analysis. In thermogravimetric analysis (TGA) a sensitive balance is used to follow the weight change of' the sample as a function of tern.perature. Typical applications include the assessment of' thermal stability and decomposition temperature (Fig 9-6), extent of cure in condensation polymers, com-
Temperature, "C
FIG. 9-5. Differential thermal analysis curve for amorphous poly(ethy1ene terephthalate) (Ke 1960); cornpare with Fig 9-4
PHYSICAL. TESTING
245
Temperature, "C
FIG. 9-6.
Relative thermal stability of' polymers as determined by weight loss on heating at S0C/min in nitrogen in thermogravimetric analysis (Chiu 1966) Poly(viny1chloride) (PVC) first loses HCI; late1 the mixture of unsatu~atedcarbon-carbon backbone and unchanged poly(viny1 chloride) partly chars and partly degrades to small fragments Poly(methy1 methacrylate) (PMMA), branched polyethylene (HPPE), and polytetrafluoroethylene (PTFE) degrade completely to volatile fragments (Chapter 6D), while a polyimide (PI) partially decomposes, forming a char above 800C
position and some information on sequence distribution in copolymers, and composition of filled polymers, among many others.
Thermomechanical Analysis. Thermomechanical analysis (TMA) measures the mechanical response of a polymer system as the temperature is changed Typical measurements include dilatometry, penetration or heat deflection, torsion modulus, and stress-strain behavior. GENERAL REFERENCES
Ke 1966; Slade 1966-1970; Smothers 1966; Miller 1969; Schwenker 1969; Mackenzie 1970; Wunderlich 19'70, 1971; Reich 1971; Collins 1973, Chapter 9C,E, Exps. 21, 22, and 24; Richardson 1978; Turi 1981.
F .
PHYSICAL TES'TING
The purpose of' this section is to provide brief' descriptions of' the more important test methods for measuring the physical properties of polymers. The discussion is not intended to be comprehensive or detailed. Test methods related to rheology and viscoelasticity are discussed in Chapter 11.
246
ANALYSIS AND TESTING OF POLYMERS
The literature related to the physical testing of polymers is extensive Several compilations are useful: the series edited by Schmitz (1965, 1966, 1968) and Brown (1969) and those volumes of methods of test and recommended practices dealing with plastics issued by the American Society for Testing and Materials (ASTM) There are also many pertinent articles in the Encyclopedza oj Polymer Science and Technology (Mark 1964-1970) Beyond this listing, specific references have largely been omitted from this section
Mechanical Properties Stress-Strain Properties in Tension. One of the most informative mechanical experiments for any material is the determination of its stress-strazn curve In tenslon This is usually done by measuring continuously the force developed as the sample is elongated at constant rate of extension The generalized s'tress-straln curve for plastics shown in Fig 9-7 serves to define several useful quantities, including modulus or stiffness (the slope of the curve), yleld stress, and strength and elongation at break This type of curve is typicallof a plastic such as polyethylene Figure 9-8 shows stress--strain curves typical of some other classes of polymeric materials The properties of these polymer types are related to the characteristics of their stress-strain curves in Table 9-2 Tensile properties are usually measured at rates of strain of 1 -100%/min At highel rates of strain-up to 106%/min-tensile strength and modulus usually increase severalfold, while elongation decreases The interpretation of these results is complicated by large temperature rises in the test specimen In addition to tensile measurements, tests may also be performed in shear, flexure, compression, or torslon For materials in film form, flexural tests are often
Strain-
FIG. 9-7.
Generalized tensile stress-strain curve for plastics (Winding 1961)
PHYSICAL TESTING
Hard and strong
Hard and tough
r '
FIG. 9-8. Tensile stress--strain curves for several types of polymeric materials (Winding 1961)
used These may include (for stiffer materials) measurement of flexural modulus or (for less stiff materials) flexural or foldzng endurance tests (see next pargraph)
Fatigue Tests. When subjected to cyclic mechanical stresses, most materials fail at a stress considerably lower than that required to cause rupture in a single stress cycle This phenomenon is called fatigue Various modes of fatigue testing in common use include alternating tensile and compressive stress and cyclic flexural stress Results are reported as plots of stress versus number of cycles to fail Many materials show a fatigue endurance limit, or a maximum stress below which fatigue failure never takes place
TABLE 9-2. Characteristic Features of Stress-Strain Curves as Related to Polymer Propertiesa
Characteristics of Stress-Strain Curve Description of Polymer

Soft, weak Soft, tough H a d , brittle Hard, strong H a d , tough
"Winding (1961)
Modulus
Low Low High High High
Yield Stress
Low Low None High High
Ultimate Strength
Low Yield stress Moderate High High
Elongation at Break
Moderate High Low Moderate High
248
Fatigue failure may arise from the absorption of energy in a material that is not perfectly elastic. This energy is manifested as heat, leading to a temperature rise, a lower modulus, and rapid failure. Energy absorption is accompanied and measured by a difference in phase between stress and strain in the cyclic test
Impact Tests. In most cases ruptuIe in polymer samples may be divided into two classes: brittle and ductile Brittle rupture occurs if the material behaves elastically up to the point of failure, that is, does not yield or draw, whereas in ductile rupture the specimen is permanently distorted near the point of failure Brittle failure is characterized by lack of distortion of the broken parts Two aspects of brittle rupture a e of interest: the temperature below which brittle failure occurs under a given set of experimental conditions, and measures of the toughness of materials at ambient temperature The test methods used to obtain both types of information are highly empirical Since the specimen is destroyed, multiple testing is usually employed to allow statistical evaluation of the results The brittlepoznt, or temperature at the onset of brittleness, is usually determiyd by subjecting a specimen to impact in a standardized but empirical way. The temperature of the test is varied until that temperature is found where, statistically, half the specimens fail by brittle rupture The brittle point is roughly related to the glass transition temperature (Chapter 11D) Impact strength of plastics is commonly measured by tests in which a pendulum with a massive striking edge is allowed to hit the specimen From the travel of the pendulum after breaking the specimen can be calculated the energy required to cause the break The specimen is often notched in an effort to improve the reproducibility of the mode of failure Other forms of impact test include a large variety in which falling objects (balls, darts, etc ) strike the specimen. Usually the height from which the missile must be dropped to cause failure in half the specimens is taken as a measure of the toughness of the material 'Tear Resistance. When plastics are used as films, particularly in packaging applications, their resistance to tearing is an important property. In one test of tear. strength a specimen is torn apart at a cut made by a sharp blade. Energy is provided by a falling pendulum, and the work done is measured by the residual energy of the pendulum.. Tear strength and tensile strength are closely related Hardness. As usually conceived, hardness is a composite property combining concepts of resistance topenetrarzon, scratching, marring, and so on. Most hardness tests for plastics are based on resistance to penetration by an indentor pressed into the plastic under a constant load. Abrasion Resistance. Abrasion resistance in plastics usually takes the fbrm of a
scratch test, in which the material is subjected to many scratches, usually fiom contact with an abrasive wheel or a stream of falling abrasive material The degree of abrasion can be determined by loss of' weight for severe damage, but is more
PHYSICAL TESTING
249
usually measured by evidence of' surface marring, such as loss of' gloss or development of haze in transparent specimens Friction, hardness, and abrasion resistance are related closely to the viscoelastic properties of' polymers (Chapter 11C).
Thermal Properties Softening Temperature. In addition to the rheological tests discussed in Chapter
1 IA, a thermal property of' great interest is the softening temperature of a plastic Various ways of measuring this p1,operty include obsewation of' the temperatures at which (a) an indentor under fixed load penetrates a specified distance into the material (Vicat test; see Chapter. 120); (b) a bar, held in flexure under constant load, deforms a specified amount (dejlection temperature or heat distortion test); (c) a polymer sample becomes molten and leaves a trail when moved across a hot metal surface with moderate pressure (polymer melt or stick temperature test); and (d) a polymer specimen fails in tension under its own weight (zero-strength temperature test)
Flammability. The flammability of plastics is usually tested as the burning rate of a specified sample The self-extinguishing tendency of the material on the removal of an external flame is also important Optical Properties Transmittance and Reflectance. A major determinant of' the appearance of' a transparent material (one that does not scatter light) is its transmittance, the ratio of the intensities of' light passing through and light incident on the specimen. Similarly, the appearance of an opaque material (one that may reflect light but does not transmit it) is characterized by its reflectance, the ratio ofthe intensities of' the reflected and the incident light. A translucent substance is one that transmits part and reflects part of'the light incident on it. Transmittance and reflectance may be measured as a function of' the wavelength of light in a spectsophotometer. When adjusted to cor~espond visual perception, to by weighting the effects at various wavelengths according to the power incident from a specified light source (often daylight) and the response of' the human eye to light flux, these quantities are called luminous transmittance and luminous reflectance. Color. Color is the subjective sensation in the brain resulting from the perception of those aspects of the appearance of objects that result from the spectral composition of the light ~eaching eye Other aspects of appearance (gloss, haze, transparency; the see below) are properly not part of the phenomenon of color Since color is subjective, it cannot be desc~ibedcompletely in physical terms, though to a first app~oximationcolor depends largely on the spectral power distribution of a light source, the spectral reflectance of the illuminated object, and the spectral response
250
curves of the eye. In terms of' visually perceived quantities, the description of'color requires specification of' three variables, a common set of'which are hue, the attribute that determines whether the color is red, green, blue, and so on; lightness, the attribute that permits a color to be classified equivalent to some member of' a scale of' grays ranging from white to black; and saturation, the attribute of' any color possessing a hue that determines the degree of' difference ofthe color from the gray of the same lightness. The instrumental "measurement" of' color consists in determining sets of numbers, correlating approximately with visually perceived quantities fbr sufficiently restrictive conditions, that allow one to judge whether two colors are alike or different. Despite the many limitations involved, the ability to make such determinations quantitatively and objectively is of' great commercial importance. The principles of color technology, including visual perception, the measurement of color and color difference, color matching, and the coloring of' plastics ancl other materials, have been discussed by Billmeyer (1981)
Gloss. Gloss is the geometrically selective reflectance of' a surhce responsible for its shiny or lustrous appearance. Surface reflectance is commonly at a maximuh in or near the specula direction, that is, the direction at which a mirror would reflect light. Photoelectric instruments are available fbr measuring gloss at a variety of' angles of incidence and reflection Haze. For transparent materials, haze is that percentage of transmitted light that in passing through the specimen deviates fiom the incident beam by forward scatte~ing. commercial hazemeters only light deviating more than 2.5" from the In transmitted beam direction is considered haze.. The effect of haze is to impart a cloudy or milky appearance to the sample, but its transparency (defined in the next paragraph) need not be reduced.. Transparency. For a transparent plastic material transparency is defined as the state permitting perception of objects through or beyond the speclmen A sample of low transparency may not exhibit haze, but objects seen through it will appear blurled or distorted Transpaency can be measured as that fraction of the normally without deviation from the primary beam direction incident light that is transm~tted of more than 0 1" (Webber 1957) Electrical Properties Dielectric Constant and Loss Factor. The dielectric constant of' an (insulating) material is the ratio of' the capacities of a parallel plate condenser measured with and without the dielectric material placed between the plates. The difference is, of course, due to the polarization of the dielectric. If'the field applied to the condenser is time dependent (as in alternating current), the polarization is time dependent also.. However, because of' the resistance to motion of' the atoms in the dielectric, there is a delay between changes in the field and changes in the polarization. This delay s is often expressed as a phase difference or l o ~ angle 6 . The power factor is then
PHYSICAL. TESTING
251
defined as sin 6, and the dissipation,factor as tan 6 The product of the dielectric constant and the power factor is called the 103s factor. It is proportional to the energy absorbed per cycle by the dielectric from the field Dielectric constant and loss factor are usually measured over a frequency range from 60 cps to thousands of megacycles per second
Resistivity. The resistance of most polymers to the flow of direct current is very high, and conductivity probably results from the presence of' ionic impurities whose mobility is limited by the high viscosity of' the medium.. Both susf'ace and volume resistivity ase important properties for applications of polymers as insulating materials.. Dielectric Strength. Insulators will not sustain an indefinitely high voltage: As the applied voltage is increased, a point is reached where a catastrophic decrease in resistance takes place, accompanied by a physical breakdown of the dielectric As in mechanical fatigue experiments, application of lower voltages causes eventual breakdown: Curves of time to fail versus voltage can be plotted Such curves show two distinct regions of failure At short times failure is presumed to occur as a result of the inability of the electrons conducting the curTent to dissipate rapidly enough the energy they receive from the field Breakdown at longer times appears to be due to corona attack, which is sensitive to the atmosphere surrounding the dielect~icand to the presence of mechanical strains Arc Resistance. The surfaces of some polymers may become carbonized and conduct curxent readily when exposed to an electrical discharge This property is important for such applications as insulation for gasoline-engine ignition systems Electronic Properties. Recognition of the possibility of producing unusual electronic properties in polymers has led to considerable research activity and several new products in recent years The phenomena involved include electrical conductivity, charge storage, energy transfer and contact electrification or t~iboelectricity Both real and polarization charges can be stored for long periods of time by polyethylene, poly(ethy1ene terephthalate), and a variety of fluorine-containing polymers, the most important of which is poly(viny1idene fluoride). In the latter polymer, thermal or sonic energy can be converted to electrical energy and vice versa by pyroelectric and piezoelectric phenomena, leading to electret microphones and loudspeake~s with remarkable fidelity among other new products The electrical conductivity of such polymers as polyacetylene, -(CH=CH),-, and poly-p-phenylene can be no less than 20 orders of magnitude greater than that of the usual polymer insulators Chemical Properties
The following properties may not involve chemical treatment or attack exclusively, but are conveniently grouped together.as different from those considered in previous sections
252
ANALYSIS A N D 'TESTING OF POLYMERS
Resistance to Solvents. The effect of' solvents (broadly defined as liquids in general) on polyme~smay take several forms: solubility (Chapter 2A); swelling, including the absorption of' water; environmental stress cracking, in which the specimen fails by breaking when exposed to mechanical stress in the presence of an organic liquid or an aqueous solution of' a soap or other wetting agent; and crazing, in which a specimen fails by the development of'a multitude of' very small cracks in the presence of' an organic liquid or its vapor, with or without the presence of mechanical stress Vapor Permeability. The permeabi'lity of a polymer to a gas or vapor is the product of the solubility of' the gas or vapor in the polymer and its dzffusion coejficient. Permeability is directly measured as the rate of transfer of vapor through unit thickness of the polymer in film form, per unit area and pressure difference across the film Weathering. For convenience and reproducibility, the behavior of materials on long exposure to weather is often simulated by exposure to artificial weatherihg sources, such as filtered carbon-arc lamps in controlled atmospheres. Such tests are usually not appreciably accelerated over natural weathering, except that they may be applied continuously
GENERAL REFERENCES
Baer 1964; Mark 1964-1970; Ritchie 1965; Schmitz 1965,1966,1968; Lever 1968; Bmwn 1969; Collins 1973, Chapter 10B-D, Exps. 27-29; Allcock 1981, Chapter 21; Duke 1982; Horowitz 1982; Mort 1982; Seanor 1982; ASTM.
Which of'the techniques described in this chapter could contribute to the solution of each of the following problems? Discuss the advantages and disadvantages of each. (a) Locate a crystalline melting temperature; (b) determine degree of orientation (see also Chapter. 10E); (c) ascertain the arrangement of' molecular chains in a polymer crystal; (d) locate a glass-transition temperature; (e) chasacterize the double bonds in a diene polymer; (f') measure an enthalpy of fusion; (g) investigate the mechanism of oxidation of' a polymer; (h) study molecular motion in polymer chains; (i) estimate degree of c~ystallinity; measure the (j) amide content of' an ester-amide copolymer; (k) determine the mechanism of' polymerization of'
2. Explain how infrared spectroscopy can be used to determine copolymer reactivity ratios. Give an illustrative example and discuss limitations to the method
BIBLIOGRAPHY
253
3.
Show how NMR can be used to (a) distinguish between head-to-head and headto-tail polymerization in polymers and (b) distinguish between a random co-. polymer and a mixture of homopolymers
4. How may x-ray diffraction data be used to estimate the sizes of' polymer crystallites, and what alternative interpretations of' the data exist? (See also Chapter IOC..)
5. What microscopic methods would be suitable for studying (a) the morphology of' crystalline polymers (Chapter 10C), (b) surface oxidation in a hydrocarbon polymer, (c) the structure of the beads used in a GPC column (Chapter 8F), (d) suspected gel particles in branched polyethylene (Chapter 13A)?
6. With the aid of sketches of the instruments, describe the difference between DTA and DSC, and discuss the determinations for which each method is best suited
7. Draw typical DSC and DTA thexmograms fbr a crystalline polymer, showing the glass transition, crystallization, crystalline melting, and thermal degradation
8.
Which of' the stress-strain curves in Fig. 9-8 would you expect to find for a polymer useful for (a) a gear in a machine, (b) the housing fbr an automatic pencil, (c) a garden hose, (d) the packing material to cushion delicate instruments, (e) the outer covering of a basketball, (f) a deco~ative paperweight, (g) a fiber useful fbr rope
Alexander 1969 Leroy E Alexander, X-Ray Diffractzon Methods rn Polymer Sczence. John Wiley & Sons, New York, 1969 Allcock 1981 Harry R Allcock and Frederick W Lampe, Contemporary Polymer Chemistry. PrenticeHall, Englewood Cliffs, New Jersey, 1981 ASTM Annual Book of ASTM Standards, Section 8-Plastics (Vols 08 01--08 03), American Society for Testing and Materials, Philadelphia, Pennsylvania, 1983, and annually thereafter Baer 1964 Eric Baer, ed ,Engineering Design for Plastrcs, Reinhold, New York, 1964 Billmeyer 1981 Fred W. Billmeyer, Jr , and Max Saltzman, Principles oj'Color Technology, 2nd ed , John Wiley & Sons, New York, 1981 Bzrbeck 1961 M S C Birbeck, "Techniques for the ElecBon Microscopy of Proteins," Chapter 1 in P Alexander and R J Block, eds , Analytrcal Methods of Protein Chemistry, Vol 3, Pergamon Press, New York, 1961 Boerio 1972 F J Boerio and J L Koenig, "Vibrational Spectroscopy of Polymers," J Macromol Sci Rev Macromol Chem C7, 209-249 (1972) Bovey 1960 F A Bovey and G V D Tiers, "Polymer NSR Spechoscopy I1 The High Resolution Spectra of Methyl Methacrylate Polymers Prepared with Free Radical and Anionic Initiators," J Polym Scr 44, 173-182 (1960) Bovey 1965 Frank A Bovey, "Nuclear Magnetic Resonance," Chem Eng News 13 (35), 98-121 (1965) Bovey 1968 F A Bovey, "Nuclea~ Magnetic Resonance," pp 356-396 in Herman F Mark, Norman
254
ANALYSIS AND TESTING OF POLYMERS
G Gaylord, and Norbert M Blkales, eds , Encyclopedra of Polymer Science and Technology, Vol 9, Wiley-Inte~science,New York, 1968 Bovey 1969 Frank A Bovey, Nuclear Magnetic Resonance Spectroscopy, Academlc Press, New York, 1969 Bovey 1972 F A Bovey, Hlgh Resolutron NMR of Macrom~lecules,Academlc Press, New York, 1972 Bovey 1976 F A Bovey, "High-Resolution Carbon-13 Studres of Polymer Structure," Chapter 10 in K J Ivln, ed , Structural Studres of Macromolecules by Spectroscoprc Methods, John Wlley & Sons, New York, 1976 Boyer 1980 R F Boyer and S E Kelnath, eds , Moleculur Motlon rn Polymers by ESR, Harwood Academic, New York, 1980 Brauer 1965 G M Brauer and G M Kline, Chemical Analys~s,"pp 632-665 In Herman F Mark, Norman G Gaylord, and Norbert M Bikales, eds , Encyclopedra of Polymer Science and Technology, Vol 3, W~ley-inter science, New York, 1965 Brau~r1982 Dretrlch Braun, Simple Methodr for Identrfication of Plastrcs, rev~seded , Macmillan, New York, 1982 Bresler I966 S E Bresler and E N Kazbekow, "Electron-Spin Resonance," pp 669-692 In Hcrman F Mark, Norman G Gaylord, and Norbert M Blkafes, eds , Encyclopedia of Polymer Sc~ence and Technology, Vol 5, Wtley-Interscience, New York, 1966 Brown 1969 W E Brown, ed , Testing of Polymers, Vol 4, Wiley-Interscience, New York, 1969 Brown 1978 D S Brown and R E Wetton, "Recent Advances in the Study of Polymers by Small Angle X-Ray Scattering," Chapter 6 in J V Dawklns, ed , Developments in Polymer Character iration-I, Applled Sclence, London, 1978 Bullock 1976 A T Bullock and G G Cameron, "E s I Studies of Spin-Labeled Synthetic Polymers," Chapter 15 m K J Ivin, ed , Structural Studles of Macromolecules by Spectroscopic Methods, John Wiley & Sons, New York, 1976 Cassel 1962 James M Cassel, "Chromatography," Chapter 10 in Gordon M Kllne, ed , Analytical Chemrstry of Polymers, Part 11, Wlley-Interscience, New York, 1962 Chiu 1966 Jen Chiu, "Appl~cations of Thermog~avimetryto the Study of Hlgh Polymers," Appl Polym Symp 2, 25-42 (1966) Collrns 1973 Edward A Colhns, Jan Bare:, and Fred W Blllmeyer, Ir , Experiments rn Polymer Scrence, Wlley-Interscience, New York, 1973 Cross 1950 L H Cross, R B Richards, and H A W~lhs,'''The Infrared Spectrum of Ethylene Polymers," Disc Faraday Soc 9, 235-256 (1950) Cunizffe 1978 A V Cunliffe, "L3CNMR Spectroscopy of Polymers," Chapter 1 m J V Dawkins, ed , Developments In Polymer Characterisat~on--1, Applied Science, London, 1978 D'Esposiro 1978 L D'Esposito and 3 L Koenig, ''Apphcatlons of Founer Tiansfom IR to Synthetic Polymers and Blologlcal Macromolecules," Chapter 2 in John R Ferraro and Louis J Besile, eds , Fourrer Transform IR Spectroscopy, Appkcatrons to Chemical Systems, Vol 1, Acadcmlc Press, New York, 1978 Drtch 1970 Michael W Duch and David M Grant, "Carbon-13 Chem~calShlft Studles of thc 1,4Polybutadlenes and the 1,4-Polylsoprenes," Macromolecules 3, 165-174 (1970) Duke 1982 Charles B Duke and Harry W Gibson, "Polymers, Conductive," pp 755-793 in Martin Grayson, ed , Kirk-Othmer Encyclopedia of Chemical Technology, 3rd ed , Vol 18, WifeYInterscience, New York, 1982 Elliott 1969 Arthur Elliott, Infrared Spectra and Structure of Organrc Long-Churn Polymerc. St Mart~n'sPress, New York, 1969 Feinland 1965 R Felnland, "Chromatography," pp 731-762 in Herman F Mark, Norman G Gay lord, and Norbert M Blkales, eds , Encyclopedra of Polymer Science and Technology, Vol 3, WileY Intersclence, New York, 1965
BIBLIOGRAPHY
255
Frscher 1966 E W Flscher and H Goddar, "Electron-Diffraction Analysis," pp 641-661 tn Herman F Mark, Norman G Gaylord, and Norbert M Bikales, eds , Encyclopedra of Polymer Science and Technology, Vol 5, Wiley-Intersclence, New York, 1966 Geil 1966 Phillip H Geil, "Electron Microscopy," pp 662-669 in Herman F Mark, Norman G Gaylord, and Norbert M Bikales, eds , Encyclopedra of Polymer Scrence and Technology, Vol 5 , W~ley Interscience, New York, 1966 Hartshorne 1964 N H Hartshorne and A Stuart, Practrcal Optical Crystallography, 2nd ed , American Elsevier, New York, 1964 Haslam 1965 John Haslam and Harry A Willis, Identificatron andAnalysis of Plastrcs, D Van Nostrand, Princeton, New Jersey, 1965 Hemsley 1978 D Hemsley, "M~croscopy of Polymer Surfaces," Chapter 8 In J V Dawkins, ed , Developments rn Polymer Characterisatron-I, Applled Science, London, 1978 Hendra I969 P J Hendra, "Laser-Raman Spectra of Polymers," Adv Polym Sci 6 , 151-169 (1969) Herglotz 1975 H K Herglotz, "Characterlzat~onof Polymers by Unconvent~onal X-Ray Techniques," Chapter 5 in John J Burke and Volker Wexss, eds , Characterrzatlon of Materrals rn Research Ceramic r and Polymers, Syracuse University Press, Syracuse, New York, 1975 Horowrtz 1982 Emanuel Horow~tz,"Plastics Testing," pp 207-228 In Martin Grayson, ed , KrrkOthmer Encyclopedia of Chemical Technology. 3rd ed , Vol 18, Wiley-Interscience, New York, 1982 Hosemann 1962 R Hosemann and S N Bagchi, Drrecr Analyszs of Drfiactzon by Matter, NorthHolland, Amstexdam, 1962 Humme11966 D~eter0 Hummel, Infrared Spectra of Polymers In the Medium and Long Wavelength Region, John Wlley & Sons, New York, 1966 Kakiido 1972 M Kakudo and N Kasa~, X-Ray Drffractron by Polymers, Elsevier, New York, 1972 Ke 1960 Bacon Ke, "Application of Differential Thermal Analysls to High Polymers," pp 361--392 In John Mitchell, JI ,I M Kolthoff, E S Proskauer, and A Weusberger, eds , Organrc Analysts, Vol 4, Intersc~ence, New York, 1960 Ke 1966 Bacon Ke, "D~fferential Thermal Analysis," pp 37-65 In Herman F Mark, Norman G Gaylord, and Norbert M Blkales, eds , Encyclopedra of Polymer Scrence and Technology, Vol 5, W~ley-Interscience,New York, 1966 Kl~ne 1959 Gordon M Kllne, ed , Analytrcal Chemrstry of Polymers, Part I, Intersclence, New York, 1959 Koenrg 1971 J L Koenig, "Raman Scattering of Synthet~c Polymers-A 4, 233--306 (1971) Review," Appl Spec Rev
Kossler 1967 Ivo Kossler, "Infrared-Absorption Spectroscopy," pp 620-642 in Herman F Mark, Norman G Gaylord, and Norbert M Blkales, eds , Encyclopedia of Polymer Scrence and Technology, Vol 7, Wiley-Interscience, New York, 1967 Krause 1983 Anneliese Krause, Anfrled Lange, and Myer Ezrin, Plastrcs Analysrs Guide, Macmillan, New York, 1983 Lebedev 1960 Ya S Lebedev, Iu D Tsvetkov, and V V Voevodskli, "Paramagnetlc Resonance Spectra of Fluoroalkyl and Nltrose-Fluoroalkyl Radicals in Irradiated Teflon," Opt Spectrosc 8, 426-428 (1960) Lever 1968 A E Lever and J A Rhys, Propertres and Testing of Plastrcs Materrals, 3rd ed , Chemical Rubber Co , New York, 1968 Mackenzre 1970 R C Mackenzie, Drrerential Thermal Analysrs Fundamental Aspects. Vol 1 , Academlc Press, New York, 1970 Mark 1964-1970 Herman F Mark, Norman G Gaylord, and Norbert M Bikales, eds , Encyclopedra of Polymer Science and Technology, Wiley-Intersc~ence,New York, 14 volumes + suppl , 19641970
256
ANALYSIS A N D TESTING OF POLYMERS :
Mlller 1969 Gerald W Mrller, "The Thermal Character~zationof Polymers," Appl Polym Symp 10, 35-72 (1969) Mtller I975 Robert L M~ller,"Crystallographic Data f o ~ Va~lousPolymers," pp 111-1-111-137 In 1 Brandrup and E H Immexgut, eds , with the collaboratron of W McDowell, Polymer Handhook, 2nd ed , Wiley-Interscience, New York, 1975 Mort 1982 1 Mort and G Pfister, eds , EIectronrc Propertres of Polymers, W~ley-Intersclence,New York, 1982 Rdnby I977 B Rinby and J F Rabek, ESR Spectroscopy m Polymer Research, Springer, New York, 1977 Rerch 1971 Leo Re~chand Davtd W lev^, "Thermogravimetr~c Analys~s," pp 1-41 In Herman F Mark, Norman G Gaylord, and Norbert M Btkales, eds , Encyclopedra of Polymer Scrente and Technology, Vol 14, Wrley-Interscience, New Yolk, 1971 Rexroad 1958 Harvey N Rexroad and Walter Gordy, "Electron Sptn Resonance Stud~es Irrddldted of Teflon Effects of Vanous Gases," J Chem Phys 30, 399-403 (1958) R~chardson1978 M J Richardson, "Quantit,ativc Drfferent~alScannlng Calonmetry," Chapter 7 In J V Dawkins, ed , Developments m Polymer Characterrsatron-I, Applred Sctence, London, 1978 Rrtchre 1965 P D Rltchre, ed , Phys~csoJ Plastrcr, D Van Nostrand, Pnnceton, New Jersey, 1965 Saunders 1966 K J Saunders, The Identlficatlon ojPlastrc r andRuhhers, Chapman and Hall, London, 1966 Schaefer 1974 Jacob Schaefer , "The Carbon-1 3 NMR Analysrs of Synthettc H ~ g h Polymers," Chapter 4 in George C Levy, ed , Topics rn Carbon-I3 NMR Spectroscopy, Vol 1 , Wiley-Interccrcnce, New York, 1974 Schaefer 1976 J Schaefer, "The Analysis of "C n m r Relaxation Experiments on Polymers," Chapter 11 In K J Ivin, ed , Structural Studies oj Macromolecules by Spectroscoprc Methods, John Wiley & Sons, New York, 1976 Schaefer 1979 Jacob Schaefer and E 0 Stelskal, "H~gh-Resolut~on NMR of Sol~dPolymers," I3C Chapter 4 m George C Levy, ed , Toprcs m Carbon-I3 NMR Spectroscopy, Vol 3, W ~ l e Interscience, New York, 1979 Schmrtz 1965 John V Schmltz, ed , Testmg of Polymers, Vol 1, Wiley-Interscience, New York, 1965 Schmrtz 1966 John V Schmitz, ed , Testrng of Polymers, Vol 2, W~ley-lntersc~ence, York, 1966 New Schmltz I968 John V Schmttz, ed , Testtng of Polymers, Vol 3, W~ley-Intersc~ence, York, 1968 New Schwenker 1969 Robert F Schwenker, Ir and Paul D Garn, eds , Thermal Analysrs. Instrumenfuliofl. organ^ Materrals and Polymers, Vol 1, Academrc Press, New York, 1969 Seanor 1982 Donald A Seanox, ed , Electrrcal Propertres of Polymers, Academic Press, Xew York, 1982 Sedgwrck 1978 R D Sedgw~ck, "Mass Spectrometry," Chapter 2 In I V Dawklns, ed ,Developmenls rn Polymer Characterrsatron--1, Applted Scrence, London, 1978 Sresler 1980 H W Slesler and K Holland-Montz, IR and Raman Spectroscopy of Polymers, Marcel Dekker, New York, 1980 S~ostrand1967 F r ~ t ~ o f Sjostrand, Electron Mrcroscopy of Cells and 7rssues. Instrumentairon and S Techniques, Vol 1, Academic Press, New York, 1967 Slade 1966-1970 Phllip E Slade, Jr and Lloyd T Jenkms, eds , Technzques and Methods (4Pc1lymer Evaluatron, Vol 1, Thermal Analyrrs. 1966; Vol 2, Thermal Characterization Techniquer, 1970, Marcel Dekker, New York Smrth 1956 Carl W Sm~thand Malcolm Dole, "Spec~fic Heat of Synthet~cHigh P o l y m < VIr Polyethylene Terephthalate," J Polvm Scr 20, 37-56 (1956) Smothers 1966 W J Smothers and Y Ch~ang, Handbook of Dflerentral Thermal Analysrr, ~ k r n ~ c a l Publishing Co , New York, 1966
BIBLIOGRAPHY
257
Starkweather 1956 Howard W Starkweather and Robert E Moyn~han,"Density, Infrared Absorption, and Crystallinity in 66 and 610 Nylons," / Polvm SCI 22, 363-368 (1956) Stevens 1969 Malcolm P Stevens, Characterization andAnalysis ojPolymers by Gas Chromatographv, Marcel Dekker, New York, 1969 Sudol 1969 Robert S Sudoi, "Determination of Polymer Structure by High-Resolution Nuclear Mag netic Resonance Spectroscopy," Anal Chrm Acta 46, 231--237 (1969) 7lrornton I968 P R Thornton, Scanning Electron M~croscopj, Applrcatrons to Mater~alrand Device Science, Chapman and Hall, London, 1968 Turz 1981 Edith A Turi, ed , Thermal Characterlst~csof Polymeric Materials, Academic Press, New York, 1981 Wake 1969 William C Wake, The Analysis of Rubber and Rubber-Llke Polymers, 2nd ed , John Wiley & Sons, New York, 1969 Wall 1962 Leo A Wall, "Mass Spectrometry," Chapter 6 in Gordon M Kline, ed , Analytical Chemistry of Polymers, Part 11, Wiley-Interscience, New York, 1962 Ward 1957 I M Ward, "Configurational Changes in Polyethylene Terephthalate," Chem 2nd 1957, 1102 (1957) Webber I957 Alfred C Webber, "Method for the Measurement of Transparency of Sheet Mate. rials" J Opt Soc Am 47, 785-789 (1957) Wrnding 1961 Charles C Winding and Gordon D Hiatt, Polymeric Materrals, McGraw-Hill, New York, 1961 Wunderlrch 1970 B Wunderlich and H Baur, "Heat Capacities of Linear High Polymers," Adv Polym Scr 7 , 151-368 (1970) Wunderhch 1971 Bernhad Wunderlich, "Differential Thermal Analysis," Chapter 8 in Arnold Weissberger and Bryant W Rossiter, eds , Physical Methods of Chemrstry, Vol 1 of A Weissberger, ed , Techniques oj Chemistry, Part 5 , Wiley-Interscience, New York, 1971 Wunderlich 1973 Bernhard Wunderlich, Macromolecular Phys~cs,Cnstals, Structure. Morphology, New York, 1973 and Defects, Vol 1, Academlc P~ess, Zbrnden 1964 Rudolf Zbinden, Infrared Spectroscopy of Hrgh Polymers, Academic Pxess, New York, 1964
PART FOUR
STRUCTURE AND PROPERTIES
CHAPTER TEN
MORPHOLOGY A N D ORDER IN CRYSTALLINE POLYMERS
A.
CONFIGURATIONS OF POLYMER CHAINS
In this book the word configuratzon is used to describe those arrangements of atoms that cannot be altered except by breaking and reforming primary chemical bonds In contrast, as discussed in Chapter 78, arrangements that can be altered by rotating groups of atoms around single bonds are called conformations. Examples involving conformations of polymer chains include trans versus gauche anangements of consecutive carbon-carbon single bonds and the helical arrangements found in some polymer c~ystal structures (Section B) Examples involving configurations include head-to-head, tail-to-tail, and head-to-tail arrangements in vinyl polymers (Chapter 3A) and the several stereoregular arrangements described in this section, including 1,2- and 1,4-addition, cis and trans isomers, and arrangements around asymmetric carbon atoms.
Configurations Involving an Asymmetric Carbon Atom
Staudinger (1932) recognized early that polymers of monosubstituted olef'ins should contain a series of asymmetric carbon atoms along the chain, and much speculation followed over the possibility of' synthesizing such polymers in stereo~egularforms As discussed in Chapter 4 0 , such syntheses have been achieved by several routes. The regular structure of the resulting polymers, notably poly(a-olefins), was recognized by Natta (1955, 1959a), who devised the nomenclature now accepted (Huggins 1962) to describe stereoregular polymers of this type. As indicated in Fig.
FIG 10-1.
Diagrams (a) and models (b) showing the Irregular atactic (top) and the stereoregular isotactic (center) and syndiotactic (bottom) configurations in a vinyl polymer (Natta 1959b) The maln carbonxarbon chain is depicted In the fully extended (all trans) planar zlgzag conformation
CONFIGURATIONS OF POLYMER CHAINS
26 3
10-1, where the polymer chain is depicted in the fhlly extended (all trans) planar zigzag conformation, the configuration resulting when all the substituent groups R on the vinyl polymer lie above (or below) the plane of' the main chain is called jsotuctic If' substituent groups lie alternately above and below the plane, the configuration is called syndiotactic, whereas a random sequence of positions is said to lead to the atacti'c configuration. More complicated arrangements and their nomenclature are beyond the scope of this book.. The synthesis of isotactic and syndiotactic polymers has now become almost commonplace, and many examples of' each type are known. In addition to poly(aolef'ins), polystyrene and poly(methy1 methacrylate) are examples of polymers that can be made in the isotactic configuration, while poly(methy1 methacrylate) and polybutadiene in which 1,2-.addition occurs exclusively, giving the structure
are examples of polymers that can be prepared in the syndiotactic conf'iguration. It has been shown that some stereoregular polymers had been made earlier but were not recognized at the time, a notable example being poly(viny1 isobutyl ether) (Schildknecht 1948), which is isotactic As shown in Section B. whether a polymer is isotactic or syndiotactic usually determines its crystal structure, and the assignment of' an all-,isotactic or all-qndiotactic structure to a polymer can be made from its crystal structure. The use of NMR spectroscopy (Chapter 9B) is a more powerful technique, however, since it allows determination of' the stereoregular configuration of' successive monomers in sequences up to about 5 units long. For example, examination of' the resonance absorption of the methylene protons in poly(methy1 methacrylate) allows one to distinguish between and determine the relative numbers of' sequences of' two monomers (dyads) that have syndiotactic (racemic, r) and isotactic (meso, m) symmetry The a-methyl proton resonance allows estimation of the numbers of -monomer sequences (triads) with configurations mm, mr, and rr. Use of 13C-NMR allows the configurations of'tetrads and pentads to be determined The data can be interpreted to indicate whether the probability P,,, adding the next monomer with the same of (meso) configuration as the last one added is independent of'the configuration of the growing chain, or not, and if' the former, to determine this probability. Typical of' the results are that fbr predominantly syndiotactic poly(methy1 methacrylate), P, = 0.24; for free-radical poly(viny1 chloride), P , = 0.43, cor~espondingto a slightly syndiotactic but nearly atactic structure (fbr which P, = 0.50); and fbr predominantly isotactic poly(methy1 methacrylate) and polypropylene, the probability of' stereoregularity of propagation is more complex (Bovey 1979). For most polymers tacticity changes very little with the temperature of polymerization
3r
264
M O R P H O L O G Y AND ORDER IN CRYSTALLINE POLYMERS
Head-to-Head, Tail-to-Tail Configurations

Nuclear magnetic resonance spectroscopy (Chapter 9B) has provided the first detailed information on the frequency of occurrence of' a monomer adding in the reverse fishion, to provide an occasional head-to-head, tail-to-tail instead of the normal head-to-tail sequence.. An example is the use of I9F-NMR to study triad sequences in poly(viny1idine fluoride), in which it has been fbund that a monomer adds in reverse fashion 5-6% of the time, but the "mistake" is immediately rectified with the next addition
Configurations involving a Carbon-Carbon Double Bond

As discussed in Chapter 6 0 , polymers of 1,3-dienes, containing one residual double bond per repeat unit after' polymerization, can contain sequences with several different configurations FOI a monosubstituted butadiene such as isoprene, the following structures are possible:
Stereoregular polymers with several of' these configurations are known (Chapter 13D,E). Their structures are confirmed by x-ray diffraction and infrared spectroscopy.
Other Stereoregular Configurations

There is no doubt that stereoregularity plays an important role in the structures of proteins, nucleic acids, and other substances of biological importance. Deoxyribonucleic acid (DNA), for example, has a highly stereoregular double helix structure, now famous (Watson 1953) (Fig. 10-2). Other stereoregular structures in nonbiological polymers are also well known (Chapter 15).
Optically Active Polymers

Many polymers, both biological and synthetic, are capable of rotating the plane of polarization of light and are said to be optically active. While in simple lowmolecular-weight compounds optical activity is associated with the presence of
CONFIGURATIONS OF POIYMER CHAINS
-.
FIG 'O-2.
The famous double helix of DNA (Kombcrg 1962, aftel Feughelman 1955)
asymmetric carbon atoms, this is not universally true in polymers. Staudinger (1932) pointed out that every second chain carbon atom in vinyl polymers is formally asymmetric Yet such polymers are not usually optically active, even when isotactic or syndiotactic, because of intramolecular compensation Optically active polymers can be prepared, however, in several ways (Schulz 1968): (1) polymerization of optically active monomers in such a way that their asymmetric centers are preserved; (2) introduction of activity into optically inactive polymers by reactions that place asymmetric centers on side chains; (3) stereospecific
266
M O R P H O L O G Y A N D ORDER IN CRYSTAL.LINE POLYMERS
polymerization of optically inactive monomers with optically active catalysts in such a manner that the repeat units have one or more asymmetric carbon atoms of' a given configuration; or (4) polymerization of racemic monomer mixtures to yield polymer from only one of' the monomers (as by using a selective initiator) or from both polymerizing simultaneously and independently to yield a mixture of dextsorotatory and levorotatory polymer which can be sepasated.. An example of' an optically active polymer is that made from L-propyleneoxide (Price 1956):
GENERAL REFERENCES
Natta 1967; Farina 1968; Schulz 1968; Morawetz 1975, Chapter 3; Tadokoro 1979
B.
CRYSTAL. STRUCTURES O F POLYMERS
Structural Requirements for Crystallinity
Many polymers, including most fibers, are partially crystalline The most direct evidence of this fact is provided by x-ray diffraction studies The x-ray patterns of crystalline polymers show both sharp features associated with regions of threedimensional order, and more diffuse features characteristic of molecularly disordered substances like liquids The occurrence of both types of feature is evidence that ordered and disordered regions coexist in most crystalline polymers Additional evidence to this effect comes from other polymer properties, such as densities, which are intermediate between those calculated for completely crystalline and amorphous species. The close relation between regularity of molecular structure and crystallizability has long been recognized Typical crystalline polymers are those whose molecules are chemically and geometrically regular in structure. Occasional ir~egularities, such as chain branching in polyethylene or copolymerization of a minor amount of isoprene with polyisobutylene as in butyl rubber, limit the extent of crystallization but do not prevent its occurrence Typical noncrystalline polymers, on the other hand, include those in which ir~egularityof structure occurs: copolymers with significant amounts of two or more quite different monomer constituents, or atactic polymers Before the recognition of the importance of stereoregularity to crystallinity in polymers, it was thought that the bulkiness of the side groups in some atactic polymers, such as polystyrene and poly(methy1 methacrylate), alone prevented their
CRYSTAL STRUCTURES OF POLYMERS
267
crystallization The argument was based in part on the fact that poly(viny1 alcohol) can be made to crystallize, although it is derived by hydrolysis from poly(viny1 acetate), which has never been crystallized The crystallinity of poly(viny1 alcohol) was once taken as evidence that ~t was isotactic. Therefore poly(viny1 acetate) should have been similarly isotactic and would be expected to crystallize-perhaps not in a completely extended conformation-despite the bulkiness of the acetate groups Poly(viny1 alcohol) is in fact atactic. Other irregular polymers, such as copolymers of ethylene and vinyl alcohol, ethylene and tetrafluoroethylene, and ethylene and carbon monoxide, are also crystalline All these polymers have very similar structures It is now clear that the CH,, CHOH, CF,, and C=O groups are near enough to the same size to fit into similar crystal lattices despite the stereochemical irregularity of the polymers On the other hand, the CI-ICl group is apparently too large, since chlorinated polyethylene is noncrystalline The absence of crystallinity in poly(vinyl acetate) and similar polymers is due to the combination of their atactic structure and the size of their substituent groups When the disordering features of atactic configuration and bulky side groups are combined with strong interchain forces, the polymer may have a structure that is intermediate in kind between crystalline and amorphous An example is polyacrylonitrile, in which steric and intramolecular dipole repulsions lead to a stiff, irregularly twisted, backbone chain conformation The stiff molecular chains pack like rigid rods in a lattice array; lateral order between chains is present, but longitudinal order along chains is absent
Structures Based on Extended Chains
There is little doubt that the fully extended planar zigzag is the conformation of minimum energy for an isolated section of a hydrocarbon chain, the energy of the trans conformation being some 3 3 W/mole less than that of the gauche It is to be expected, therefore, that fully extended chain conformations will be favored in polymer crystal structures unless substituents on the chains cause steric hindrance relieved by the assumption of other forms Fully extended chains are found in the crystal structures of polyethylene, poly(viny1 alcohol), syndiotactic polymers including poly(viny1 chloride) and poly(l,2-butadiene), most polyarnides, and cellulose.
Polyethylene. Except for end groups, the arrangement of chains in the usual orthorhombic structure of polyethylene is essentially the same as that in crystals of linear paraffin hydrocarbons containing 20-40 carbon atoms The unit cell of polyethylene is rectangular with dimensions a = 7 41 A,b = 4 94 A, and c = 2 55 A The latter is the chain repeat distance, which is identical with the fully extended zigzag repeat distance The arrangement of the chains with respect to one another is shown in Fig 10-3. The positions of the hydrogen atoms are not detectable by the x-ray method (since the scattering power of an atom is proportional to its atomic weight) but have been calculated assuming tetrahedral bonds and a C-H distance of 1 10 A
M O R P H O L O G Y A N D ORDER IN CRYSTALLINE POLYMERS
F I G . 10-3.
Arrangement of chains in the orthorhombic unit cell of polyethylene (Geil 1963)
Polyethylene, like many other polymers, exhibits polymorphism, the ability to exist in more than one crystal structure, depending on conditions When crystalline polyethylene is stressed, as in cold drawing (Section E), the rectangular face of' the unit cell perpendicular to the chain direction is distorted into a parallelogram with one included angle, designated P, becoming 107.9". The lengths of' the axes are altered to a = 8..09A and b = 4..79 A . The chain fbrmerly in the center of the unit cell now lies in the center of the a axis This unit cell is monoclinic. At high pressures another, hexagonal, structure is formed in which the order associated with the all-trans conformation is lost and the molecules pack like rods (Bassett 1976). Similar structures exist in normal paraffins at temperatures close to their melting points
Poly(viny1 alcohol). The crystal structure of'poly(viny1 alcohol) is similar to that of' polyethylene, as might be expected, since the CHOH group is small enough to fit into the polyethylene structure in place of' a CH, group with little difficulty. The unit cell is monoclinic with a = 7.81 A, b = 2.52 A (chain axis), c. = 5.51 A; p = 9142'. Pairs of' chains are linked together by hydrogen bonds, and similarly linked into sheets, insofar. as the stereochemical irregularity allows Poly(viny1 chloride). Crystallinity in free-radical poly(viny1 chloride) is not well developed; the polymer. is primarily syndiotactic but with considerable ir~egularity Its crystal structure appears to be typical of' syndiotactic polymers, with a repeat
C'RYST'AL. STRUCTURES OF POLYMERS
269
distance corresponding to fbur chain carbon atoms Poly(l,2-butadiene) has a similar structure, but the carbon-.carbon chains are slightly distorted from the trans con-, fbrrnation to relieve repulsion between side-chain and neighboring-chain atoms
polyamides. The structures of poly(hexamethy1ene adlpamide) (66-nylon), poly(hexamethy1ene sebacamide) (6 10-nylon), and polycaprolactam (6-nylon) are made up of fully extended chains linked together by hydrogen bonds to form sheets that may be packed together in two different ways, giving two crystal modifications Oxygen atoms of one molecule are always found opposite NH groups of a neighboring molecule, with the N--H---0 distance of 2 8 A smaller than norrnai because of hydrogen bond formation Other nylons (99-, 106-, 1010-, and 11-nylon) contain chains slightly distorted from the planar zigzag form Cellulose. The crystal structure of cellulose is of historic interest, since it represents the first instance in which the concepts of a crystal unit cell and a chain structure were reconciled by the concept of the chain extending through successive unit cells. The basic feature of the crystal structure of cellulose is the repeating segment of two cellobiose units, joined in I,4-linkages, with the --CH,OH side chains in successive units occurring on opposite sides of the chain Many forms and derivatives of cellulose have crystal structures in which these chains are packed in slightly different ways Distortions From Fully Extended Chains Polyesters. In most aliphatic polyesters and in poly(ethy1ene terephthalate) the polymer chains are shortened by rotation about the C--0 bonds to allow close packing As a result, the main chains are no longer plawr The group
in poly(ethy1ene terephthalate) is planar, however, as required by resonance.
Polyisoprenes and Polychloroprene. Polymers of isoprene and chloroprene in which 1,Caddition predominates can occur in all-cis or all-trans configurations; natural rubber and gutta-percha are naturally occurring cis-1,4-polyisoprene and trans-l,4-polyisoprene, respectively All these polymers have similar crystal structures except that the repeat distance of the trans polymers corresponds to one monomer unit and that of the cis polymers to two The repeat distance of trans-1,4...polyisoprene (gutta-percha) ( 4 . 7 2 4 7 7 A) is smaller than that corresponding to a fully extended chain (5.04 A).. The structure of' natural rubber (cis-polyisoprene) is similar to that of guttapercha, except as required by the longer repeat distance due to the cis configuration
2 70
MORPHOL.OCY AND ORDER IN CRYSTALLINE POLYMERS
The chain is slightly out of plane at the single bonds and in plane around the double bond, the observed repeat distance of 8 1 A being less than the extended distance of 10.08 A. The structure of polychloroprene is similar to that of gutta-percha except that the molecules are oriented differently with respect to one another in the crystal because of differences in polarity between the chlorine and methyl groups..
Polypeptides. Proteins, including the natural fibers wool and silk, are made up of a-amino acids joined into long polypeptide chains with the planar structure
where R represents the amino acid chain These polymer chains can exist in two quite different types of crystal structure One of these forms, the P-keratin structure, is made up of nearly extended into sheets through hydrogen bonding The sheets are polypeptide chains ar~anged not quite planar and have been described as pleated or rippled A number of possible of structures have been proposed, differing in the detailed ar~angement the chains into sheets One of these, the antiparallel-chain, pleated sheet structure, is now considered the most likely structure for silk fibroin. The helical a-keratin structure of polypeptides is described below
Helical Structures
Polymers with bulky substituents closely spaced along the chain often take on a helical conformation in the crystalline phase, since this allows the substituents to pack closely without appreciable distortion of' chain bonds. Most isotactic polymers, as well as polymers of some 1,l-disubstituted ethylenes such as isobutylene, fall in this class. Other helical structures of interest are those of'polytetrafluoroethylene and the a-keratin structure
P
isotactic Polymers. Many isotactic polymers crystallize with a helical confbrmation in which alternate chain bonds take trans and gauche positions. For the gauche positions the rotation is always in that direction that relieves steric hindrance by placing R and H groups in juxtaposition, generating either a left-hand or a righthand helix. If' the side group is not too bulky, the helix has exactly three units per turn and the arrangement is similar to that in Fig. 10-4a. This arrangement is fbund in isotactic polypropylene, one form of' poly(1-butene), polystyrene, and others.. More bulky side groups require more space, resulting in the formation of looser helices as shown in Fig. 1 0 - 4 M Isotactic poly(methy1 methacrylate) forms a helix with five units in two turns, while polyisobutylene fbrms a helix with eight units in five turns
CRYSTAL STRUC:TURES OF POLYMERS
271
FIG. 10-4. Helical conformations of' isotactic vinyl polymers (Gaylord 1959)
Polypeptides. Helical structures are of' great importance in biological materials In addition to helical conformations proposed for DNA and other polymers of biological interest, the a,-keratinstructure assumed by many polypeptides is a helix, in which about 3 6--3.7 polypeptide residues occur per turn (Fig. 10-5)..Adjacent turns are held together by intramolecular hydrogen bonding. Several variations of' this structure have been proposed.
Polytetrafluoroethylene. This polymer exists in two helical conformations that may be described as twisted ribbons in which the fully extended plan= fbrm is distorted to have a 180" twist in 13 CF, units in the more stable (low-temperature)
MORPHOLOGY A N D ORDER IN CRYSTAl
POLYMERS
FIG. 10-5. Helical structure of a-keratin proteins (Pauling
form. Above 19C this form is replaced by a slightly untwisted conformation with 15 CF, units per half-twist. Above 19C the x-ray diffraction pattern shows diffuse streaks that are interpreted as resulting from small angular displacements of molecular segments about their long axes. Above 30C more diffuseness occurs in the x-ray pattern, increasing as the temperature is raised to and above the melting point, 327C. This additional diffuseness is attributed to random angula displacement of the molecules about their long axes. In terms of' molecular motion, it is considered that the molecules undergo slight torsional motion-that is, twisting and untwisting of' the helices-that causes the lattice to expand laterally, and at 19C is strong enough to cause the untwisting
,qORPHOL.OGY O F CRYSTALLINE POLYMERS
273
from 13 to 15 CF, groups in each 180" twist Between 19 and 30C the molecules are "locked" into the threefold helix cor~esponding the 15 CF, repeat distance; to their motion is restricted to small-angle oscillations about their "rest positions." Above 30C the molecules lose their prefer~edangular orientation, becoming randomly misaligned about their axes Nealy cylindrical in shape, they pack like rods Above 30C torsional motion increases, with the twisting and untwisting becoming more pronounced, until at T,,, all crystalline order is lost In this respect, polytetrafluoroethylene is an example of a newly recognized type of polymer mesophase, termed a condis (for. conformationally disordered) crystal In such a mesophase cooperative motion between various conformational isomers is permitted (Wunderlich 1983)
Helix-Coil Transitions in Polypeptides (Poland 1970) The a-keratin helix structure that has been described as chaacteristic of polypeptides and many polymers of biological origin can undergo sevsral phase transitions, including melting to random coils at T, or in the presence of a solvent A transition of particular interest results from the fact that the helix can exist as a stable conformation in solution, so that the helix-coil transition can occur without involving a soIid phase. The stability of the helix in solution results from intramolecula hydxogen bonding Since disruption of the helix requires breaking about three such bonds per turn, some 20-40 k of enthalpy must be expended to realize the greater entropy of the J random-coil state Once the transition is initiated, by a change in temperature or composition, the prefer~edconformation is perpetuated over many bonds In the simplest case the helical conformation occurs either not at all or in long sequences with infrequent interruption if not throughout the chain. The change is fairly abrupt and resembles a phase transition
. GENERAL REFERENCES
Geil 1963; Miller 1966; Corradini 1968; Bryan 1969; Wunderlich 1973; Miller 1975; Tadokoro 1979; Fava 1980
C.
MORPHOLOGY OF CRYSTALLINE POLYMERS
Polymer Single Crystals
Although the foxmation of single crystals of polymers was observed during polymerization many years ago, it was long believed that such crystals could not be produced from polymer solutions because of molecular entanglement In 1953 (Schlesinger 1953) and in several laboratories in 1957 (Fischer 1957, Keller 1957, Till 1957), the growth of such crystals was reported The phenomenon has been reported for so many polymers, including gutta-percha, polyethylene, polypropylene and other poly(a-olefins), polyoxymethylene, polyamides, and celluose and its
2 74
desivatives, that it appeass to be quite general and universal (Geil 1963, Blackadder
Lamellae. All the stIuctures described as polymer single crystals have the same general appearance, being composed of thin, flat platelets (lamellae) about 100 8, thick and often many micrometers in lateral dimensions. They ase usually thickened by the spiral g~owthof additional lamellae from screw dislocations. A typical lamellar crystal is shown in Fig.. 10-6..The size, shape, and regularity ofthe crystals depend on their growth conditions, such f'actors as solvent, temperature, and growth rate being important. The thickness of' the lamellae depends on crystallization temperature and any subsequent annealing treatment (Hendra 1976) Electron-diffraction measurements, performed in almost all the researches cited above, indicate that the polymer chains are oriented normal or very neasly normal to the plane of the Iamellae. Since the molecules in the polymers are at least 1000 A long and the lamellae are only about 100 thick, the only plausible explanation is that the chains ase folded (Keller 1957) This arIangement has been shown to be sterically possible In polyethylene, for example, the molecules can fbld in such a way that only about five chain carbon atoms ase involved in the fold itself. Many single crystals of essentially lineas polyethylene show secondary structural features, including corrugations (Fig. 10-7) and pleats (Fig.. 10-8). Both these
FIG. 10-6. Electron micrograph of a single crystal of 6-nylon grown by precipitation from dilute glycerine solution (Geil 1960) The lamellae are about 60 8, thick
MORPHOL.OGY O F CRYS7ALLINE POLYMERS
275
FIG 10-7. Optlcal mlcrog~aphshowlng coIrugatlons In slngle crystals of 11neal polyethylene grown from a solutlon In pe~chlorethylene(Reneke~1960)
features result from the fact that many crystals of polyethylene, and perhaps other polymers too, grow in the form of hollow pyramids. When solvent is removed during preparation of the crystals for microscopy, surface-tension forces cause the pyramids to collapse The hollow pyramidal structure is related to the packing of the folded chains , If successive planes of folded molecules are displaced from their neighbors by an Integral (or, with rotation of 180, by a half-integral) number of repeat distances, a pyramidal structure with a slope of 32" (or 17") results
More Complex Structures. The growth of crystal structures from polymer solutions can ~esultin more complex structures sometimes reminiscent of features found in polymers crystallized from the melt Typical of these are sheaflike arrays corresponding to the nuclei and initial growth habit of spherulites. Each ribbon in the array is composed of a number of lathlike lamellae with the usual 50-100 A thickness. Twinned crystals, dendritic growths, clusters of hollow pyramids, spiral growths, steps, Bragg extinction lines, dislocation networks, moir6 patterns, and epitaxial growths have been described, among others
276
MORPHOLOGY A N D ORDER IN CRYSTALLINE POLYMERS
FIG. 19-8. Electron rn~crographshowing pleats in a crystal like that of Fig 10-7 (Reneker 1960)
Disorder and Nature of the Fold Surfaces. The major unsolved problem (Keller 1969) in polymer single-crystal studies has for some time centered on the nature of the fold surfaces Although there is strong structural and kinetic evidence favoring regular folding with immediate adjacent reentry of the chain into the crystal, there is even more convincing evidence that considerable molecular disorder exists in polymer single crystals The extent of this disorder appears too great to be compatible with intercrystalline defects and to require some sort of amorphous region or layer at the fold surface. Such a region might involve loose or irregular folds or a more random "switchboard" model as suggested by Flory (1962), and as indicated schematically in Fig 10-9 Although there appears to be agreement that the fold surfaces in polyme~single crystals are rather regular and tight, the controversy remains unresolved into the 1980s Structure of Polymers Crystallized From the Melt The Fringed Micelle Concept. Although x-ray diffraction studies show recognizable crystalline features in some high polymers, the Bragg reflections appear b~oadand diffuse compared to those from well-developed simple crystals.. Diffraction theory indicates that this broadening can arise horn eithe~small crystallite size or the presence of lattice defects (Bunn 1961), but the diffraction patterns from
MORPHOLOGY OF CRYSTAL.LINE POLYMERS
(4
FIG. 10-9
(d)
Schematic two-dimensional representations of models of the fold surface in polymer larnellae (Ingram 1968): (a) sharp folds, (b) "switchboard" model, (c) loose loops with adjacent reentry, (d) a combination of several features
polymers are usually too weak to permit discrimination between these possibilities IIistorically, the hypothesis of small crystallite size was selected as more probable With this assumption, rough estimates from the width of diffraction rings indicated that crystallite size rarely exceeds a few hundred angstroms. A substantial background of diffuse scattering suggested the coexistence of an appreciable amorphous fraction It was suggested that the polymer chains are precisely aligned over distances cor~espondingto the dimensions of the crystallites, but include more disordered segments that do not crystallize, and hence are included in amorphous regions Since the chains are very long, they were visualized as contributing segments to several crystalline and amorphous regions, leading to a composzte singlephase structure known as thefrznged micelle or fringed crystallite model (Fig 10lo), the fringes representing t~ansitionmaterial between the crystalline and amorphous phases.
FIG. 10-10.
Fringed micelle model of the crystalline-amorphous structure of polymers (Bryant 1947)
278
M O R P H O L O G Y A N D ORDER IN CRYSTALLINE POLYh4EI;S
The fringed micelle concept enjoyed popularity for many years, largely b s c a u s ~ it led to simple and appealing consequences such as the strong bonding of crystalline. and amorphous regions into a composite structure of good mechanical properties, and the simple interpretation of degree of crystallinity in terms of thc percentages of well.defined crystalline and amorphous regions However, it tends to dm\\ attention away from the details of fine structule and gives little insight inlo thc structures of larger entities such as spherulites Almost simultaneously with the discovery of polymer single crystals nith ,t lamellar structure involving chain folding, it was found that spherulitc~ complr\ are structures of lamellae and that individual lamellae in polymers crystallized fiv.7rn the melt are almost macroscopic in size These features, incompatible with thc' interpretation of x-ray diffraction-pattern broadening in terms of small crystdlire~. suggested that reexamination of the interpretation of the x-ray data was n e d d
The Defect Structure of Crystalline Polymers. It is rewarding to recog* te h existence of defects in crystalline polymers just as in any other crystalline rn=ri.~l The pIesence of seve~al kinds of such defects is now well recognized, Among IIIPXII are the following:
a.
b.
Point defects, such as vacant lattice sites and interstitial atoms Chain sndb. which themselves must be considered defects because they differ cheri:all\ from the rest of the chain, are usually accompanied by vacancies. Intsxiti-i! atoms or groups may be foreign material or may be associated ~5 thr' chain, for example, as certain kinds of side chains Dislocations, primarily screw dislocations and edge dislocations S:R\~ dislocations resulting in growth spirals have been observed r i n y t b s in hs\z both polymer single crystals and bulk polymers Edge d i s f c ~ a t i o r s been discovered more recently through the study of moirt pmrns
Two-dimensional imperfections, those of particular interest k i n g t 5 fold ~ surfaces (Fig 10-9) d. Chain-disorder defects, including folds, changes of alignment. and l&e e. Amorphous defects, defined as disorders large enough to d f c ~ ~the hxiw pt surrounding their own immediate area by forcing other atom. o u t of 33rnl.d lattice positions The net effect of all these defects, in particular the fold surface5 arh J U I - I C ~ ~ ~ I ~ S regions, is to provide localized amorphous regions contributing u, diffu~e X.-ri\ scatte~ing,and deformed or distorted lattices, termed paracrystallr:~ ( H G % ~ ~ G 1962), which contribute in part to x-ray line broadening In s u m m q - the h ~ d micelle concept has largely been supplanted, especially for highly c f . s d b ymers, by a defect-crystal concept accounting equally well for the o k m a k k ?:.. 3; r and utilizing, in addition, experimentally verified defect structurs M* 1 X! X! remalns to be learned, however, about the structure of polymers ,c/:% I f l x ~ 1 termediate crystallinity; here the fringed micelle concept may give 3/22 :.u, *=n.: folding model with partial adjacent and partlal random reentry (Pa%;:n ig1 57
c.
M O R P H O L O G Y O F C:RYST'AL.LINE POLYMERS
279
Extended-Chain Crystals. In addition to producing folded-chain structures, polymers can also crystallize from the melt in the form of extended-chain crystals (Fig 10-11) (Geil 1964, Wunderlich 1968) This morphology is more easily obtained with low-molecular-weight polymers, but this depends on crystallization conditions, including applied pressure and supercooling The extended-chain morphology, in which the size of the crystal in the chain direction is essentially equal to the extended chain length, is now considered the equilibrium morphology for polyethylene and probably many other crystalline polymers (Bassett 1976b) Structure of Spherulites. The most prominent organization in polymers on a scale larger than lamellae is the spherulite, (ideally) a spherical aggregate ranging from submicroscopic in size to millimeters in diameter in extreme cases Spherulites are recognized (Fig 10-12) by their characteristic appearance in the polarizing microscope, where they are seen as (ideally) circular birefringent areas possessing a dark Maltese cross pattern The birefringence effects are associated with molecular
FIG. 10-11. Extended-chain crystals of linear polyethylene (Prime 1969) Fracture surface of a sample crystallized at 5000 atm
MORPHOL.OCY AND ORDER IN CRYSTALLINE POLYMERS
FIG. 10-12. Ringed spherulites of poly(trirnethy1ene glutarate) between crossed polarizers (Keller 1959)
in the optical microscope
orientation resulting from the characteristic in spherulites
now recognized
Relation of Spherulites to Crystallites. Although there has not been general agreement on whether spherulites ase formed by rearrangement of previously crystallized material or as products of' primary crystallization, the latter interpretation is considered more probable. It has been shown by examination of polymers that are partially spherulitic that in the usual case the spherulites are crystalline, whereas the intermediate nonspherulitic material is amorphous. Spherulites thus appear to represent the crystalline portions of the sample, growing at the expense of' the noncrystalline melt. The point of initiation of spherulite growth, its nucleus, may be a foreign particle (heterogeneous nucleation) or may arise spontaneously in the melt (homogeneous nucleation). Morphology of Spherulites. Electron-microscopic evidence indicates that spher-
ulites have a lamellar structure for almost all polymers. Crystallization spreads by the growth of individual lamellae. Growth may begin fiam screw dislocations at the nucleus or at spiral growths originating at crystal defects hrther out in the spherulite. When two spherulites meet during crystallization, lamellae from both extend across the boundary into any uncrystallized material available. This results in interleaving, which holds the material together. Evidence that the lamellar structure persists throughout the body of spherulites, and that lamellae thus are basic structural elements of most solid polymers, is furnished by electron microscopic examination of' fracture surfaces
CRYSTALLIZATION AND MELTING
281
FIG. 10-1 3.
Electron micrograph of a portion of a ringed spherullte in linear polyethylene (photograph by E W Fischer, from Geil 1963)
The morphology of spherulites in some polymers is complicated by the development of a ring structure (Fig 10-12) In the electron microscope it is seen (Fig 10-13) that the ring structure is associated with a periodic twisting of the lamellae; the cause of the twisting is unknown
GENERAL REFERENCES
Geil 1963; Ingram 1968; Wunderlich 1973; Schultz 1974, Chapters 2 and 3; Mandelkern 1975; Bassett 1976a, Chapters 2-4; Khoury 1976; Fava 1980; Peterlin 1980; Hobbs 1982
D. CRYSTALLIZATION A N D MELTING
Crystallization Kinetics
Experimental observations of' the development of crystallinity in polymers are of' two types In one case a property varying with total amount of'crystallinity, such as specific volume, is observed as a function of' time at constant temperature. Alternatively the rate of formation and growth of' spherulites is observed directly with the microscope. The two techniques lead to essentially the same results
MORPHOLOGY AND ORDER IN CRYSTXLL.INE POLYMERS
Time, hr
FIG. 10-14. Volume-time relations during the crystallization of natural rubber (Wood 1946) Arrows indicate points where crystallization is half complete
The development of crystallinity in polymers 1s not instantaneous Curves of specific volume as a function of time at temperatures below the crystalline melting point (Fig 10-14) show that crystallization sometimes cannot be considered complete for long periods Since the time for complete crystallization is somewhat indefinite, it is customary to define the rate of crystallization at a given temperature as the inverse of the time needed to attain one-half of the total volume change The rate so defined is a characteristic function of temperature (Fig 10-15) As the temperature is lowered, the rate increases, goes through a maximum, and then decreases as the mobility of the molecules decreases and crystallization becomes diffusion controlled At temperatures at which the rate is very low, the polymer may be supercooled and maintained in the amorphous state. If data for specific volume as a function of time during crystallization, such as those for natural rubber in Fig 10-14, are plotted against log time, all the curves have the same shape and can be superposed by a shift along the time axis This behavior > y c a l of the crystallization of low-molecular-weight substances, and curves of t 's type can be fitted with an equation due to Avrarni (1939):
where V,, V,, and Vo are specific volumes at the times indicated in the subscripts, w, is the weight fraction of material crystallized, k is the constant describing the rate of' crystallization, and n is an exponent varying with the type of nucleation and growth process Although this analysis provides a satisfying analogy between crystallization in polymers and that in low-molecular-weight substances, quantitative aspects of the agreement with theory are not completely satishctory . The shape of' the curves, as in Fig. 10-14, is not very sensitive to values of' the exponent n in the expected range of' 2-4, and it is not always possible to determine n or indeed to ascertain
CRYSTAL.LIZATION AND MELTING
Temperature, " C
FIG. 10-15.
Rate of e~ystallization rubber as a function of temperature (Wood 1946) of
whether n is constant throughout the crystallization process. It is generally found that crystallization continues in polymer systems for much longer times than Eq 10-1 predicts This results from the fact that crystallization in polymers involves the steps of (primary) nucleation and relatively rapid sphemlite growth, followed by a slow, kinetically difficult improvement in crystal pe~fection "Amorphous" chains remaining between the spherulite lamellae after the first stage also crystallize in the slow second step Extension of these kinetics to the case of polymers was carried out by Lauritzen (1960) and Hoffman (1961), who calculated the rate at which a nucleus of critical size to sustain growth into a crystallite is formed by the addition of strands of polymer chain They evaluated the driving force for crystallization as the difference in free energy between the mass of the nucleus and a similax mass in the melt This expression involved terms for the free-energy change due to supe~cooling and for sur.face free energy Their final equation, though complicated, predicts the correct order of magnitude of the thickness of the resulting lamellar crystal, and the temperature dependence of crystallization rate of the form shown in Fig 10-15 An alternative thermodynamic approach (Peterlin 1960) has received less general acceptance The concept of the lamellar structure of polymers implies large-scale molecular motion in both the melt and the solid crystalline polymer Either prefolding and some local orientation exists in the crystallizing polymer, or the molecules must undergo a considerable degree of motion during crystallization Extensive melting and crystallization with an extremely rapid, severalfold increase in fold period occur during annealing of polymer single crystals and melt-crystallized lamellae; the extent to which the crystal morphology is disrupted, even at temperatures below T,,,, is illustrated in Fig 10-16 The development of spherulites in polymers quenched from the melt and then held at temperatures between T, and T, is additional evidence of high chain mobility below T,,, The existence of mobile lattice defects provides a plausible mechanism for this mobility in crystalline material It has been found, as might be expected, that the ease with which these molecular motions take place decreases as molecular weight increases
284
FIG. 10-16.
A single crystal of lineaz polyethylene crystallized from perchlorethylene solutlon and 30 min at 125"C, at least 10C below 7 , (Statton 1960) The fold from about 100 8, to almost 200 8, during annealing Compare Fig
Determin ron of 7,
The disappearance of'a polymer crystalline phase at the melting point is accompanied by changes in physical properties: The material becomes a (viscous) liquid, with discontinuous changes in density, refractive index, heat capacity: transparency, and other. properties. Measurement of' any of' these prope~ties may be used to detect the crystalline melting point T,,. Since melting occurs over a temperature range, the melting point determined by a particular method may depend to some extent on the sensitivity of the method. For example, crystalline melting points may be determined by noting the temperature of disappearance of the last traces of' crystallinity as evidenced by birefringence observed between crossed polarizers on a
CRYSTALLIZATION A N D MELTING
285
hot-stage microscope. This melting point is usually several degrees higher than changes in properties such as those obtained from methods depending on g~.oss specific volume and specific heat. Because of retention of' order above T , and the existence of' a melting range in polymers, properties sensitive to the amount of' crystalline material present undergo change over a relatively wide range in temperature, as evidenced by experimental data for specific volume or specific heat (see, for example, Fig. 9-4).. Today almost all measurements of' T , are made by differential thermal analysis (Chapter 9E).. Other methods that could be used to detect T, are x-ray diffraction and infrared and NMR spectroscopy, each of these being suited to the detection of' crystalline material as described in Chapter 9. As the experiments are usually performed, they are not adapted to detecting T, but, rather, are carried out at constant temperature.
Thermodynamics of Crystalline Melting
For a polymer homogeneous in molecular weight, a statistical thermodynamic analysis based on a lattice model yields the relation 1 T
I T, AH,
(- + '
xw,
x -- 5
where wa is the weight fraction of material that is amorphous at temperature T (expressed in Kelvins as required in all thermodynamic equations) The equilibrium melting point of the pure polymer at infinite chain length is T, = AH,,,IAS, The degree of polymerization is x , and 5 is a parameter characterizing the crystailite size Equation 10-2 indicates that even for a polymer of uniform chain length, melting occurs over a finite temperature range The physical reason for this behavior is the higher surface free energy of small crystallites Equation 10-2 also shows that the temperature interval for fusion decreases with increasing molecular weight, and further analysis indicates that it broadens with increasing breadth of molecularweight distribution On the other hand, as wa approaches unity, its derivative dwal dT does not vanish; in consequence, the last traces of crystallinity disappear at a well-defined temperature, as confirmed by experiment The extended-chain crystals described in Section C are now considered to be the equilibrium state for high polymers, and thus the only state whose melting is properly described by Eq 10-2 The more common folded-chain crystals are not in their most stable state, and as T, is approached, they can undergo several possible changes in structure, exemplified by the reorganization and recrystallization seen during annealing The melting point in systems of this kind occurs at the point where the free-energy change AG is zero, even though the system is not at equilibrium The process is termed zero-entropyproductzon melting The melting point is a measure of the degree of impedection of the metastable system, and is greatly dependent on the thermal history of the sample and in particular the heating rate (Jaffe 1967)
286
M O R P H O L O G Y A N D ORDER IN CRYSTAL.LINE POLYMERS
For polymer-diluent systems, use of the Flory-Huggins expression fbr the free energy of mixing (Chapter 7C) leads to the expression
1
T,
1 - R vz (u1 - xlu12) TmO A H , V ,
where T , and TmO the melting points of the poly~er-diluent mixture and the are pure polymer, respectively, V is molar volume, u is volume fraction, and xI is the polymer-diluent interaction parameter defined in Chapter 7C. Thus the meltingpoint depression depends on the volume fraction of diluent added and its interaction with the polymer. For thermodynamically good solvents, large depressions (4050C) are sometimes observed in dilute solutions Since the melting temperature is only slightly dependent on molecular weight, crystallization does not provide an efficient method of' molecular-weight fractionation..
Heats and Entropies of Fusion

I
Measurement of the heat and entropy of fusion yields important experimental data for assessing the efftct of' polymer structure on Tm, as discussed in Chapter 12A Heats of fusion are readily derived from several experimental sources, at present most often differential scanning calorimetry (Chapter 9E). The observed heat of fusion AH* calculable from specific-heat measurements is not AH,n, the heat of fusion per mole of' repeating unit, but is related to it by the degree of crystallinity w,: AH* = w,AH,,, It is AHm that appears directly in Eq. 10-3, however, and in the related equation for the melting points of copolymers,
s the mole fraction of crystallizing units Thus AHm can be determined where from melti -point data for these two types of systems even if w,is not known In some c ses AHm can be inferred from the heats of fusion of low-molecularweight homo1 gues It is found, for example, that the increment per CH, group in the heat of sion of n-paraffins becomes constant for chain lengths greate1 than six carbon atoms With certain assumptions, this increment may be taken as the heat of fusion of crystalline polyethylene The method can be extended to the calculation of the lattice or sublimation energy at 0 K , a measure of the intermolecular forces in the crystal A comparison is made in Table 10-1 of the values of A H , for polyethylene determined from specific-heat measurements, the melting-point depressions of polymer-diluent mixtures, and extrapolation of data on paraffins Heats of fusion for several other polymers are included The total entropy change on fusion can be calculated from T,, and the heat of fusion, since AG, = AH, - T AS,,, = 0 at the melting point The total change
CRYSTALLIZATION A N D MELTING
287
TABLE 10-1. Heat and Entropy of Fusion of Crystalline Polymers
Polymer Polyethylene From specific heat From diluents From extrapolation Polyoxymethylene Polytetrafluoroethylene Natural rubber Gutta percha
"Wunderlich 1957 hQuinn 1958
' Bilirneyer 1957 dSta~kweathe~ 1960
'Robexts 1955 'Mandelkern 1956
in entropy is made up of the change in entropy on fusion at constant volume (AS,"), plus the change in entropy due to expansion to the volume of the liquid phase The latter can be evaluated by a simple thermodynamic relation to yield the equation
where V, and V, are the volumes of the solid and liquid phases, respectively Values of (AS,), so calculated, as well as of AS,, are included in Table 10-1
Degree of Crystallinity
The determination of the degree of crystallinity of a polymer is affected by the conceptual difficulties d~scussed Section C The interpretation of measurements in of crystallinity depends on the model used in designing the experiments: Each of the methods to be described involves a simplifying assumption, usually concerned with the degree to which defects and disordered regions in paracrystalline material are weighted in the determination of the amorphous fraction Fortunately, independent techniques lead to unexpectedly good experimental agreement, implying that differences in their resolution, specifically the difference in the lateral extent of lattice order detected, are relatively minor (Richardson 1969) The three major methods of determining crystallinity, discussed in the following paragraphs, are based on specific volume, x-ray diffraction, and infrared spectroscopy Polyethylene is used throughout as an example, although each method is applicable, at least in principle, to other polymers as well Another well-established method, based on measurement of heat content as a function of temperature through the fusion range, is now easily carlied out using differential scanning calorimetric measurements Other methods are based on inc~eased swelling or increased chemical reactivity of the amorphous regions Nuclear
MORPHOLOGY AND ORDER IN CRYSTALLINE POLYMERS
magnetic resonance spectroscopy (Chapter 9B) has been correlated with crystallinity, but this technique places more emphasis on other properties than on differences between crystalline and amorphous character..
The crystallinity of' a material is given in terms of'the specific volumes of'the specimen (V), the pure crystals (V,), and the completely amorphous material (V,), as
Specific Volume.
This relation assumes additivity of the specific volumes, but this requirement is fulfilled for polymers of homogeneous structure, as far as is known It is implicit that the sample be free of voids Since specific volume can be determined to 1 or 2 parts in lo4, the method can attain high precision Its accuracy depends on the uncertainty in Va and V,, which are known to about 1 part in lo3 for polyethylene The crystalline specific volume is determined from x-ray unit cell dimensions, while Va is usually obtained by extrapolation of the specific volumes of polymer melts
X-Ray Diffraction. The x-ray method allows calculation of the relative amounts of crystalline and amorphous material in a sample if it is possible to resolve the contributions of the two types of structure to the x-ray diffraction pattern A favorable situation is depicted in Fig 10-17, where the scattering envelope of a polyethylene sample (outer line) has been resolved into contributions of two crystalline peaks (with indices 110 and 200) and a broad amorphous peak The estimation
Dtffractton angle
FIG. 10-17. Resolution of the x-ray scattering curve of a polyethylene Into contribut~ons from two crystalline peaks (~ndices110 and 200), the amorphous peak, and background
STRAIN-,INDUCEDMORPHOLOGY
289
of amount of crystallinity is usually based on comparison of the areas under the peaks, but other measures, such as peak height, may be used With proper attention to experimental detail, this method provides one of the fundamental measures of crystallinity in polymers
Infrared Absorption. Suitable infrared absorption bands can be used as primary measures of crystallinity in polymers The application requires that the band bear a simple and unambiguous relationship to the crystalline or amorphous character of the polymer, and that absorption data can be obtained or inferred for the pure crystalline and amorphous polymers These conditions are satisfied, for example, for an infrared band at 7.67 pm in polyethylene identified with a methylene wagging motion in the amorphous regions, and a band at 10.6 p n in 610-nylon whose specific absorbance varies linearly with specific volume
GENERAL. REFERENCES
Mandelkern 1964; Sharples 1966; Price 1968; Sanchez 1974; Schultz 1974, Chapter 9; Bassett 1976a, Chapters 5-8; Hoffman 1976; Wunderlich 1976, 1980; Hobbs 1982
E. STRAIN-INDUCED MORPHOLOGY
When a polymer is crystallized from solution in the absence of' external forces, the in single crystals desc~ibed Section C fbrm. If the solution is subjected to an external stress such as stirring, an entirely different morphology can result. When a polymer mass is crystallized in the absence of' external fbrces, there is no preferred direction in the specimen along which the polymer chains lie. If such an unoriented crystalline polymer is subjected to an external stress, it undergoes a rearrangement of' the crystalline material. Changes in the x-ray diffraction pattern suggest that the polymer chains align in the direction of' the applied stress At the same time the physical properties of the sample change markedly.
Fibrillar Crystallization
If' a dilute (-1%) polymer solution is stirred during crystallization, c~ystalline aggregates consisting of a central rod or ribbon with lamellar overgrowths can be f'o~med (Fig. 10.-18).The morphology has, for obvious reasons, been called the "shish kebab" morphology.. The backbone, or "shish," consists of' relatively but not fully extended chains, with a significant number of defects, probably chain folds. The "kebabs" form by epitaxial growth using the backbone as a nucleus This is perhaps the most spectacular of' a number of' morphologies produced by crystallization under unusual conditions
MORPHOLOGY A N D ORDER I N CRYSTALLINE POLYMERS
FIG. 10-18. Schematic model of "shish kebab" morphology formed by crystallization of s st~rred polymer solution (Reprinted with permission from Pennings 1977 Copyright 1977 hj John Wiley & Sons )
Cold Drawing
When the orientation process is carried out below T, but above the glass-transitioa temperature T,, as when an unoriented fiber is stretched rapidly, the sample d ~ 8 not become gradually thinner, but suddenly becomes thinner at one point. in a process known as "necking down" (Fig. 10-19). As the stretching is continued. the thin or drawn section increases in length at the expense of' the undrawn porrion of' the sample. The diameters of' the drawn and undrawn portions remain about t k same throughout the process. The draw ratio, or ratio of the length of' the drawn fiber to that of'the undrawn, is abeut 4 or 5 to 1 for a number of polymers, including branched polyethylene, polyesters, and polyamides, but is much higher (10 to I ( 6 more) in linear polyethylene In general, the degree of' crystallinity in the specimen does not change greatf? during drawing if' crystallinity was previously well developed. If' the undrauffi polymer was amorphous or only partially crystallized, crystallinity is likely increase during cold drawing
ffj
STRAIN-INDUCED MORPHOLOGY
FIG 10-19
(a) Cold-drawn fibe~ polyethylene, showing sectlons of drawn and undrawn matenal of (Bunn 1953) (b) Cold-drawn flber of polyethylene observed between crossed polailzers, showlng evldence of random onentation m the undrawn sectlon and or~entatlon parallel to the dlrectlon of stress In the drawn section
Morphology Changes During Orientation

Observations in the electron microscope indicate that when polymer single crystals are stressed, fibrils 50-100 A in diameter are drawn across the break (Fig 10-20) These fibrils must contain many molecules; smaller fibrils have not been resolved in the microscope. The fibrils appear to come from individual lamellae, and the sharp discontinuity between the drawn and the undrawn material suggests that the molecules are unfolding In some cases the fibrils appear to have a periodicity of 100-400 A along their length; a similar period is observed by low-angle x-ray diffraction Examination of melt or solution-cast specimens shows that the lamellae tend to rotate toward the draw direction and on further elongation break up into microlamellae and finally into submicroscopic units Since the elongation at break in colddrawn specimens is less than could be obtained by complete unfolding, it has been suggested that the smallest units are microcrystallites in which the chains remain folded Periodic spacing of these crystallites within the fibrils would account for their observed periodicity
292
M O R P H O L O G Y A N D ORDER IN CRYSTALLINE POL.YMERS
FIG. 10-20. Electron micrograph of a film of linear polyethylene crystallized by evaporation of the solvent (Keller 1957) Flbnls drawn across the breaks are 50-100 A in dlameter and appear to have a 100-300 A perlodlclty along the11 length
Spherulites tend to remain intact during the first stages of' drawing, often elongating to markedly ellipsoidal shapes. Rupture of' the sample usually occurs at spherulite boundaries The structure of' fibers, whet he^. fbrmed by drawing or crystallized from an oriented melt, is still unsettled. X-ray evidence indicates that the molecules are aligned and shows that a long periodicity is present. Microfibrils are undoubtedly important structural elements. Among the models proposed are fringed fibrils, a development of the fringed micelle model (Section C) in which the crystalline regions are aligned in the direction of the fiber axis; a "string" model (Statton 1959), in which the microfibrils have alternating crystalline and amorphous regions, but there are fewer ties molecules between them than in the fringed fibril concept; a "folded fibril" model in which the microfibrils consist of small stacked sections of' lamellae; and a paracrystalline model in which the imperfect crystalline regions are thought to be much larger than required by the other models Perhaps the final stages in the fbrmation of an oriented fiber by drawing can be imagined (Peterlin 1965, 1967) as shown in Fig. 10-21. A multilayer lamellar crystal is destroyed with tilting and slipping of the chains. At points of concentration of defects, blocks of folded chains break off' and are incorporated along with unfolded chains into the new fiber structure. In consequence of' these processes, the number of' chain folds decreases and the number of' tie molecules incseases
STRAIN-,INDUCED MORPHOLOGY
293
Increasing deformation of single crystals
FIG. 10-21. Suggested model for fiber formation by chain tilting and slipping followed by break~ng off of blocks of lamellae (Peterlin 1967)
The latter and the partial interpenetration of lamellae increase the mechanical stability, tensile strength, and stiffness of the fiber
Degree of Orientation
As in the measurement of degree of crystallinity (Section D), the determination of degree of orientation is complicated both by the complexity of the types of orientation that can occur and by the conceptual difficulties involved in the current picture of the stxucture of an oriented specimen The interpretation of the measurements depends on the model and simplifying assumptions in each method, but these differences are relatively minor, and the agreement among results by different methods is in general satisfactory.
X-Ray Diffraction. As indicated in Chapter 4C, x-ray diffraction patterns of unoriented polymers resemble small-molecule powder photographs, characterized by rings rather than diffraction spots As the specimen is oriented, these rings break into arcs, and at high degrees of orientation approach the relatively sharp patterns characteristic of paracrystals (Fig 10-22). From the distribution of scattered intensity in the arcs or spots, the degree of orientation can be measured in terms of distribution functions. These functions describe the amount of material occurring in crystalline regions having angular coordinates describing their orientation that fall within certain limits. For practical purposes, only average values of the functions are important
?t
M O R P H O L O G Y A N D ORDER I N CRYSTALLINE POLYMERS

i(
X-ray diffraction patterns for unoriented (a) and oriented (b) polyoxymethylene (courtesy of E S . Clark)
Birefringence. The birefringence, or change in index of refraction with direction, of a crystalline polymer is made up of' contributions from the crystalline and amor. phous regions plus a cont~ibution, form birefringence, resulting from the shape of' the crystals or the presence of' voids. For completely unoriented material, the contributions of both the crystalline and the amorphous regions are zero. The increase in birefringence on orientation is due primarily to the crystalline regions
295
and is proportional to the degree of'crystallinity, the intrinsic birefringence of' the crystals (which can be calculated), and an o~ientationfactor that can be identified with that determined by the x-ray technique..
infrared Dichroism. Infrared absorption is dependent upon change in dipole moment. Such a change is a vector quantity, that is, one that is confined to definite directions for certain groups..In oriented samples the amount of' absorption of planepolzized infrared radiation may vary markedly with change in direction of' the plane of polarization. Thus the stretching vibrations of' the C=O and N-H groups in polyamides involve changes in dipole moment that are almost exactly perpendicular.to the chain axis; the corresponding absorption bands are weak f b polarized ~ radiation vibrating along the chain axis and strong for that vibrating perpendicular to the axis. When separate absorptions can be found for both the crystalline and the amorphous ~.egions, dichroism of these bands gives information about orithe entation in both regions of' the specimen. Orientation in Noncrystalline Polymers
The presence of crystalline material is not essential for the development of molecular orientation. Even polymers such as polystyrene, which do not crystallize, undergo considerable molecular orientation when subjected to stress at temperatures above T, This orientation (which is often seen in injection-molded pieces) is stable below T, and is accompanied by the development of birefringence and of enhanced strength properties in the direction of the applied stress This type of orientation can be measured by birefringence, by a property change associated with the orientation, or by a change on subsequent heating above T, The force with which the specimen tends to retract on heating above T, is a suitable and readily measured property characterizing orientation in amorphous polymers
GENERAL REFERENCES
Peterlin 1967; Szabolcs 1967; Ingram 1968; Wilchinsky 1968; Desper 1975.
1. Draw structural formulas indicating the stereoregular chain configuration in (a) atactic polystyrene, (b) isotactic polypropylene, and (c) syndiotactic poly(viny1 chloride)
2. Name two examples of crystallizable atactic polymers
3.
Polyisobutylene has unit cell dimensions of approximately 18 x 12 x 7 A and contains two chains of eight repeat units each per unit cell. What would
MORPHO1.OGY AND ORDER IN CRYSTALL.INE P L M R OY E S
be the density of' 100% crystalline polyisobutylene? Why is your answer different from the observed density of' 0.915?
4.
Polytetrafluoroethylene has a first-order phase transition at 19OC. The crystal structures above and below the transition are hexagonal, with the fbllowing parameters: Interchain Distance (A) Repeat Length (A) CF, per Repeat Length
Temperature Low High
5.62 .5.66
Calculate the density of the polymer above and below the transition and discuss the phase change in t&ms of' molecular motion and possible effects on the mechanical properties of polytetrafluoroethylene
5.
Contrast the fringed micelle and lamellar models of'the structure of crystalline polymers with respect to (a) description and sketch of the basic units, (b) observation basis, (c) interpretation of x-ray evidence, (d) sphetulite structure, (e) correlation between density and crystallinity, and (f) changes on cold drawing.
6 . Discuss, with sketches, the structure of solution-grown polyethylene single crystals.

7.
Discuss the evidence for chain fblding in polymers and the conditions under which it occurs
8. Describe the growth and structure, including moleculat. arrangement, ofspherulites in polymers.
9.
Sketch graphs of' the fbllowing: (a) specific volume versus time during crystallization at several temperatures below T,, indicating the definition of rate of crystallization; (b) rate of crystallization versus temperature, showing the locations of T, and T,; and (c) equilibrium specific volume versus temperature over a range including both T, and T,
10. On the basis of the data in Table 10-1, compare the fbllowing on the basis of intermolecular forces, energy bar.r.iersto rotation about bonds, and other structural features: (a) (AS,), for polyethylene and polyoxymethylene, (b) for (AS,), for polyethylene and polytetrafluoroethylene, and (c) AH,,, natural rubber and gutta-percha (see also Chapter 1.30). 1 1 . Using the data of Table 10-1, calculate T, for natural rubber and gutta-percha Predict the properties of' these polymers at room temperature. Do your predictions agree with the actual properties of' the polymers? Can you reconcile any differences?
12. From the data plotted in Figs. 10-14 and 10-15, predict the behavior of' natural
rubber subjected to the following thermal treatments: cool rapidly to -- 40C, hold fbr 24 hr; raise rapidly to - 5'C, hold fbr 24 hr; raise rapidly to -t.2O0C, hold for 24 hr
13.
Draw a sketch of a fiber. during the process of cold drawing. Indicate the orientation of' the chains in the drawn and undrawn regions in terms of' the probable structural elements present..
14. Discuss (a) extended-chain crystals, (b) "shish kebab" morphology, and (c) structural changes in single crystals on annealing.
BIBLIOGRAPHY
Avrami 1939 Melvin Avrami, "Kinetics of Phase Change I General Theory," I Chem Phys 7 , 1103-1 112 (1939) Bossett 1976a D C Bassett, Prtnciples of Polymer M o q d ~ ~ l o gCambridge University Press, Camy, bridge, 1976 Bassert 1976b D C Bassett, "Chain-Extended Polyethylene in Context: A Review," Polymer 17, 460--470 (1976) Billmeyer I957 Fred W Billmeyer, Ir , "Lattice Energy of Crystalline Polyethylene," I Appl Phys 28, 1114-1118 (1957) Blackodder 1967 D A Blackadder, "Ten Years of Polymer Single Crystals," I Macromol Sci Rev Macromol Chem C1 (2), 297-326 (1967) Bovey 1979 F A Bovey and T K Kwei, "Microstructure and Chain Conformation of Macromole cules," Chapter 3 In F A Bovey and F H Wlnslow. eds , Macromolecules An Introduction to Polymer Science, Academic P~ess,New York, 1979 Bryan 1969 W P B~yan, E Hein, and M F Perutz, "Proteins," pp 620-677 in Herman F Mark, G Norman G Gaylord, and Norbert M B~kales,eds , Encyclopedra of Polymer Science and Technology, Vol 11, Wiley-Inte~science,New Yo~k,1969 Bryant 1947 W M D Bryant, "Polythene Fine Structure." I Polym Scr 2, 547-564 (1947) Bunn 1953 C W Bunn, "Polymer Texture," Chapter 10 in Rowland Hill, ed , Fibres from Synthertc Polymers, Elsevier, New York, 1953 Bunn 1961 Charles W Bunn, Chemical Crystallography. 2nd ed , Oxford Unive~sityP~ess, London, 1961 Corradini 1968 P Co~radlni,''Cham Conformation and Crystallin~ty," Chapter 1 in A D Ketley, ed , The Stereochemlstry oj Macromolecriles, Vol 1 , Marcel Dekker, New York, 1968 Desper 1975 C Richard Desper, "Characterization of Molecular and Crystalline Orientation of Anisotropic Solid Polymers," Chapter 16 in John J Burke and Volker Weiss, eds , Characterizatron of Materrals rn Research Ceramics atzdPolymers, Syracuse University Press, Syracuse, New York, 1975 Farrna 1968 Mario Farina and Giancarlo Bressan, "Optically Active Stereoregular Polymers," Chapter 4 in A D Ketley, ed , The Stereochemrstry of Macromolecules, Vol 3 , Marcel Dekker, New York, 1968 Fava 1980 R A Fava, ed , Polymers, Part B Crystal Structure and Morphology, Vol 16 of L Marton and C Marton, eds , Methods ojExperrmenra/Physics, Academic Press, New York, 1980 Ferighelman I955 M Feughelman, R Langridge, W E Seeds, A R Stokes, H R Wilson, C W Hooper, M H F Wilklns, R K Barclay, and L D Hamilton, "Molecular Structure of Deoxynbose Nucleic Acid and Nucleop~ote~n," Natu~e175, 834-838 (1955)
298
Fischer I957 E W Fischer, "Step and Spiral Crystal Growth of High Polymers" (in German), Z Naturforsch 12a, 753-754 (1957) Flory 1962 P J Flory, "On the Morphology of the Crystalline State in Polymers," J Am Chem SOL 84, 2857-2867 (1962) Gaylord I959 Norman G Gaylord and Herman F Mark, Linear and Stereoregular Addition Polymers, Interscience, New York, 1959 Geil I960 P H Geil, "Nylon Single Crystals," J Polym Sci 44, 449-458 (1960) Gerl I963 Philllp H Geil, Polymer Srngle Crystals, Wiley-Interscience, New York, 1963 Gel1 I964 Philllp H Gell, Franklin R Anderson, Bernhard Wunderlich, and Tamio Arakawa, "Morphology of Polyethylene Crystallized from the Melt Under Pressure," I Polym Sci A2, 37073720 (1964) Hendra I976 P J Hendra, "The Measurement of Lamellar Thickness by Raman Methods," Chapter 6 in K J Ivin, ed , Structural Studies of Macromolecules by Spectroscopic Methods, John WlIey & Sons, New York, 1976 Hobbs I982 S Y Hobbs, "P?lymer Morphology," Chapter 7 in Mahendra D Baijal, ed , Plastics Polymer Science and Technology, John Wlley & Sons, New York, 1982 with Hoffman I961 John D Hoffman and John I Lauritzen, JI , "Crystallization of Bulk Polyme~s Chain Foldlng: Theory of Growth of Lamella Spherul~tes," J Res Nat Bur Stand 65A, 297336 (1961) Hoffman 1976 John D Hoffman, G Thomas Davis, and John I Lauritzen, 11 , "The Rate of Crystallization of Linear Polymers with Chain Folding," Chapter 7 in N B Hannay, ed , Treatrse on Solid Stare Chemrstry, Crystalline and Noncrystalline Solrds, Vol 3 , Plenum Press, New York, 1976 Hosemann I962 R Hosemann and S N Bagchi, Direct Analysrs oJ Drffraction by Matter, NorthHolland, Amsterdam, 1962 Hugg~ns1962 M L Huggins, G Nana, V Desreux, and H Mark, "Report on Nomenclature Dealing with Steric Regularity in High Polymers," J Polym Sci 56, 152-161 (1962) Ingram 1968 P Ingram and A Peterlin, "Morphology," pp 204-274 in Herman F Mark, Norman G Gaylord, and Norbert M Blkales, eds , Encyclopedia ojPolymer Sczence and Technology, Vol 9, Wiley-Intersclence, New York, 1968 Jaffe 1967 M Jaffe and B Wunderlich, "Melting of Polyoxymethylene," KolloidZ Z Polym 216217, 203-216 (1967) Keller 1957 A Kelier, "A Note on Single Crystals in Polymers: Evidence for a Folded Chain Configuration," Philor Mag [8] 2, 1171-1 175 (1957) Keller I959 A Keller, "Investigations on Banded Spherulites," I Polym Sci 39, 151-173 (1959) Keller 1969. A Keller, "Solution Grown Polymer Crystals--A Survey of Some Problematic Issues," Kolloid Z Z . Polym. 231, 386-421 (1969) Khoury 1976 F Khoury and E Passaglia, "The Morphology of Crystalline Synthetic Polymers," Chapter 6 in N B Hannay, ed , Treatise on Solid State Chernlstry, Crystalline and Noncrystalline Solrds, Vol 3, Plenum kess, New York, 1976 Kornberg 1962 Arthur Kornberg, Enlymatic Synthesis of DNA, John Wiley & Sons, New York, 1962 Laurrtzen 1960 John I Lauritzen, Jr ,and John D Hoffman, "Theory of Formation of Polymer Crystals with Folded Chains in Dilute Solution," I Res Nut Bur Stand 64A, 73-102 (1960) Mandelkern I956 L Mandelkern, F A Quinn, Jr , and D E Roberts, "Thermodynamics of Crystallization in High Polymers: Gutta-Percha," I Am Chem Soc 78, 926-932 (1956) Mandelkern 1964 Leo Mandelkern, Crystallrzatron of Polymers, McGraw-H111, New York, 1964 Mandelkern I975 L Mandelkern, "Morphology of Semlcrystalllne Polymers," Chapter 13 In John J Burke and Volker Welss, eds , Characterization oJMaterials In Research Ceramrcs and Polymers. Syracuse Un~versltyPress, Syracuse, New York, 1975
BIBLIOGRAPHY
299
Mrller 1966 R0ber.t L Miller, "Crystallinlty," pp 449-528 In Herman F Mark, Norman G Gaylord, and Norbert M Bikales, eds , Eruyclopedia of Polymer Scrence and Technology, Vol 4, WileyInterscience, New York, 1966 Mtller 1975 Robert L Miller, "Crystallographic Data for Va~lousPolymers," pp 111-1-111-137 in J Brandrup and E H Immergut, eds , with the collaboration of W McDowell, Polymer Handbook, 2nd ed , Wiley-Intersclence, New Yolk, 1975 m Morawetz I975 Herbert Morawetz, ~Vacromolecules Solution, 2nd ed , Wiley-Interscience, New York, 1975, Chapter 3 Nalta 1955 G Natta, Piero Pmo, Paolo Corradtn~.Ferdinand0 Danusso, Ennco Mant~ca,G ~ o r g ~ o Mazzanti, and G~ovanniMorangllo, "Crystalline Hlgh Polymers of a-Olefias," J Am Chem Soc 77, 1708-1710 (1955) Natta 1959a G Natta and F Danusso, "Nomenclature Relatlng to Polymers Having Stencally Ordered Structure," J Polym Set 34, 3-1 1 (1959) Natta 19596 G m l ~ o Natta and Paolo con ad in^, "Conformatron of Linear Chains and Their Mode of Packing in the Crystal State," J Polym Scr 39, 29-46 (1959) Natta 1967 Giulio Natta and Fe~dinandoDanusso, eds , Stereoregular Polymers Stereoregular Polymers and Stereospecific Polymerrzatlon, Pergamon Press, New York, 1967 Paullng 1951 Linus Pauling, Robert B Corey, and H R Branson, "The Structure of Proteins: Two Hydrogen-Bonded Helical Configurat~ons the Polypeptide Chain," Proc Nat Acad SLI U S A of 37, 205-21 1 (1951) Pennrngs 1977 A I Penn~ngs, "Bundle-like Nucleation and Longitudinal Growth of Fibrillar Polymer Crystals from Flowing Solutions," I Polym Scr Polym Symp 59, 55-86 (1977) Peterltn 1960 Anton Peterl~n, "Chain Fold~ng Free Energy Density in Polymer Crystals," J Appl and Phys 31, 1934-1938 (1960) Peterlin 1965 A Peterl~n,"Crystalline Character In Polymers," I Polym Scr C9, 61-89 (1965) Peterltn 1967 A Peterlin, "The Role of Chain Folding m Flbers," pp 283--340 in H F Mark, S M Atlas, and E Cemia, eds , hfan-Made Fibers, Sclence and Technology, Wiley-Interscience, New York, 1967 Peterlrn 1980 A Peterlin, "Chain Folding in Lamellar Crystals," Macromolecules 13, 777-782 (1980) Poland 1970 Douglas Poland and Harold A Scheraga, Theory ojHelrx-Cot1 Translrtonr tn Btopolymers, Academic Press, New York, 1970 Prrce 1956 Charles C Price and Maseh Osgan, "The Polymer~zationof I-Propylene Oxide," J Am Chem Soc 78, 4787-4792 (1956) Prrce 1968 Fraser P Price, "Kinetics of Crystallization," pp 63-83 in Herman F Mark, Norman G Gaylord, and Norbert M Bikales, eds , Encyclopedra of Polymer Sctence and Technology, Vol 8, Wiley-Interscience, New York, 1968 Prlme 1969 R Bruce Prime, Be~nhardWunderlich, and Louis Melillo, "Extended-Chain Crystals V Thermal Analysis and Electron M~croscopyof the Melting Process in Polyethylene," J Polym Sci A-2 7, 2091-2097 (1969) Qurnn 1958 F A Quinn, 11 , and L Mandelkern, "Thermodynamics of Crystallization In High Polyme~s:Poly-(ethylene)," J Am Chem Soc 80, 3178-3182 (1958) Reneker 1960 D H Reneker and P H Geii, "Morphology of Polymer Single Crystals," I Appl Phys 31, 1916-1925 (1960) Rzchardson 1969 M J Richardson, "Crystallinity Determination in Polymers and a Quantitative Comparison for Polyethylene," Br Polym J 1, 132-137 (1969) Roberts 1955 Donald E Roberts and Leo Mandelkern, "Thermodynamics of Crystallization in High Polyme~s:Natural Rubber," J Am Chem Soc 77, 781-786 (1955) Sanchez 1974 Isaac Sanchez, "Modern Theories of Polymer Crystallization," J Macromol Sct Rev Macromol Chem C10, 113-148 (1974)
300
MORPHOLOGY A N D ORDER I N CRYSTALL.INE POLYMERS
Schlldknecht 1948 C E Schildknecht, S T Gross, H R Davidson, J M Lambert, and A 0 Zoss, "Polyvinyl Isobutyl Ethers-Pxoperties and Structures," Ind Eng Chem 40, 2104-21 15 (1948) Schleslnger I953 Walter Schlesinger and H M Leeper, "Gutta I Single Crystals of Alpha-Gutta," I Polym Scc 11, 203-213 (1953) Schultz 1974 Jerold Schultz, Polymer Marerlals Science, Prentice-Hall, New York, 1974 Schulz I968 Rolf C Schulz, "Optically Actrve Polymers," pp 507-524 in Herman F Mark, Norman G Gaylord, and Norbert M Bikales, eds , Encyclopedia of Polymer Science and Technology, Vol 9, Wiley-Interscience, New York, 1968 Sharples I966 Allan Sharples, introduction to Polymer Crystaflizahon, St Martin's Press, New York, 1966 Starkweather 1960 Howard W Starkweather, Jr , and Richard H. Boyd, "The Entropy of Melting of Some Linear Polymers," I Phys Chem 64, 410-414 (1960) Statton 1959 W 0 Statton, "Polymer Texture: The Anangement of Crystallrtes," I Polym Sci 41, 143-155 (1959) Statton 1960 W 0 Statton and P H Geil, "Rec~ystallizationof Polyethylene during Annealing," J Appl Polym Scc 3, 357-361 (1960) Staudinger I932 Hermann Staudinger, Dee Hochmolecularen Organcschen Verbindungen, (High Molecular Welght Organcc Compounds) (in German), Springer-Verlag. Berlin, 1932 f Szabolcs 1967 0 Szabolcs and 1 Szabolcs, "Current Ideas on the Morphology of Synthetic Fibers," pp 341-374 in H F Mark, S M Atlas, and E Cernia, eds , Man-Made Flbers, Science and Technology, Vol 1, Wiley-Intersc~ence,New York, 1967 radokoro 1979 Hiroyuki Tadokoro, Structure oj Crystallcne Polymers, John Wiley & Sons, New York, 1979 Tzll 1957 P H Till, Jr , "The Growth of Single Crystals of Linear Polyethylene," 1 Polym Sci 24, 301-306 (1957) Watson 1953 J D Watson and F H C Crick, "A Structure of Deoxyribose Nucleic Ac~d,"Nature 171, 737-738 (1953) W~lrhtnsky 1968 Zigmond W Wrlchinsky, "Orientation," pp 624-648 in Herman F Mark, Norman G Gaylord, and Norbert M Bikales, eds , Encyclopedia of Polymer Science and Technology, Vol 9, Wiley-Interscience, New York, 1968 Wood 1946 Lawrence A Wood, "Crystallization Phenomena in Natural and Synthetic Rubbers," pp 57-95 in H Mark and G S Whitby, eds , Sclentcfic Progress in the Field oj Rubber and Synthetcc Elastomers (Advances in Colloid Science), Vol 2, Interscience, New York, 1946 Wunderlrch 1957 Bernhard Wunderlich and Malcolm Dole, "Specrfic Heat of Synthetic High Polymers VII Low Pressure Polyethylene," J Polym Sci 24, 201-213 (1957) Wunderlcch 1968 Bernhard Wunderlich and Louis Melillo, "Morphology and Gxowth of Extended Chain Crystals of Polyethylene," Makromol Chem 118, 250-264 (1968) Wunderlich 1973 Bernhard Wunderlich, Macromolecular Physics, Volume 1 Crystal Structure, Mor phology, Defects, Academic Press, New York, 1973 Wunderlich 1976 Bexnhard Wunderlich, Macromolecular Physics, Volume 2 Crystal Nucleation, Growth, Annealing, Academic Press, New York, 1976 Wunderlich 1980 Bernhard Wunderlrch, Mncromo!ecular Physics, Volume 3 Crystal Meltmg, Academic Press, New York, 1980 Wunderllch 1983 Bernhard Wunderlich and Janusz Grebowicz, "Thermotropic Mesophases and Mesophase Transitions of Linear, Flexible Mac~omolecules," Adv Polym Sci 60-61, 1-59, 1983
CHAPTER ELEVEN
RHEOLOGY AND THE MECHANICAL PROPERTIES OF POLYMERS
Rheology is, by definition, the science of deformation and flow of matter The rheological behavior of polymers involves several widely different phenomena, which can be related to some extent to different molecular mechanisms These phenomena and their associated major mechanisms are as follows:
Vzscous flow, the ir~eversiblebulk deformation of polymeric material, associated with ir~eversible slippage of molecular chams past one another b. Rubberlike elastzczty, where the local freedom of motion associated with small-scale movement of chain segments is retained, but large-scale movement (flow) is prevented by the restraint of a diffuse network structure c. Viscoelasticity, where the defbrmation of the polymer specimen is reversible but time dependent and associated (as in rubber elasticity) with the distortion of' polymer chains fiom their equilibr.ium conformations through activated segment motion involving rotation about chemical bonds. d. Hookean elasticity, where the motion of chain segments is drastically re-. stricted and probably involves only bond stretching and bond angle deformation: The material behaves like a glass
a.
These four phenomena are discussed in Sections A-D, respectively. Together they form the basis fbr a description of the mechanical properties of amorphous polymers. The mechanical properties of semicrystalline polymers, however, depend intimately on the restraining nature of their crystalline regions and can be inferred
302
R H E O L O G Y A N D THE MEC:HANIC:AL PROPERTIES O F POLYMERS
only in par.t from the rheological behavior of' amorphous polymers. The mechanical properties of crystalline polymers are therefbre discussed separately in Section E.
A.
VISCOUS FLOW
Phenomena of Viscous Flow
If' a force per unit area s causes a layer of' liquid at a distance x from a fixed boundary wall to move with a velocity u , the viscosity q is defined as the ratio between the shear stress s and the velocity gradient d u l d x or rate of shear j:
If q is independent of the rate of shear, the liquid is said to be Newtonian or to exh~bit ideal flow behavior (Fig 11-la). Two types of deviation from Newtonian flow are commonly observed in polymer solutions and melts (Bauer 1967) One is shear thinnzng or pseudoplastic behavior, a reversible decrease in viscosity with increasing shear rate (Fig 1I-lb) Shear thinning results from the tendency of the applied force to disturb the long chains from their favored equilibrium conformation (Chapter 7B), causing elongation in the direction of shear An opposite effect, shear
Newtonian
Region of increasing
0,o
FIG. 11-1.
0,o
So
(c)
(d )
Dependence of shear rate y on shear stress s for (a) Newtonian, (b) pseudoplastic, and (c) dilatant behavior, and (d) the presence of a yield stress so followed by Newtonian behavior
VISCOUS FLOW
303
rtlickening or dilatant behavior (Fig. 11-lc), in which viscosity increases with increasing shear rate, is not observed in polymers A second deviation from Newtonian flow is the exhibition of' a yield value, a critical stress below which no flow occurs. Above the yield value, flow may be either Newtonian (as indicated in Fig. 11-16) or non-Newtonian. For most polymer mclts, only an apparent yield value is observed The above effects are shear dependent but time independent. Some fluids also exhibit reversible time-dependent changes in viscosity when sheared at constant stress. Viscosity decreases with time in a thixotropicfluid, and increases with time in a rheopectic fluid, under constant shear stress.. For low-molecular-weight liquids, the temperature dependence of' viscosity is found to fbllow the simple exponential relationship
where E is an actrvation energy for vrscous j b w and A is a constant These features of the flow of liquids can be explained in terms of several molecular theories That of Eyring (Glasstone 1941) is based upon a lattice structure for the liquid, containing some unoccupied sites or holes These sites move at random throughout the liquid as they are filled and created anew by molecules jumping from one site to another Under an applied stress the probability of such jumps is higher in the direction that relieves the stress If each jump is made by ove~coming energy bar~ierof height E, the theory leads to Eq 11-2 The energy an of activation E is expected to be related to the latent heat of vaporization of the liquid, since the removal of a molecule from the surxoundings of its neighbors forms a part of both processes Such a relation is indeed found and is taken as evidence that the particle that moves from site to site is probably a single molecule As molecular weight is increased in a homologous series of liquids up to the polymer range, the activation energy of flow E does not increase proportionally with the heat of vaporization but levels off at a value independent of molecular weight This is taken to mean that in long chains the unit of flow is conside~ably smaller than the complete molecule It is rather a segment of the molecule whose size is of the order of 5-50 carbon atoms Viscous flow takes place by successive jumps of segments (with, of course, some degree of coordination) until the whole chain has shifted
Dynamics of Polymer Melts. It is now accepted that polymer chains are strongly intertwined and entangled in the melt; the dynamic behavior of' such a system has been reviewed (Graessley 1974), but is only poorly understood.. Thermodynamically, the chains are essentially ideal, as was first realized by Flory (1949). Their freedom of motion results from the presence of a correlation hole around each flow unit, within which the concentration of' similar units from other chains is reduced The presence of these correlation holes, and of the ideal but entangled nature of chains in the melt, has been confirmed by neutron-,scattering experiments (Cotton 1974)
304
RHEOLOCY AND THE MECHANKAL PROPERTIES OF POLYMERS
Flow Measurement
Methods commonly used for measuring the viscosity of polymer solutions and melts are listed in Table 11-1. The most important of these methods involve rotational and capillary devices (Van Wazes 1963, Whorlow 1979, Dealy 1981, 1983).
Rotational Viscometry. Rotational viscometers are available with several different geometries, including concentric cylinders, two cones of different angles, a cone and a plate, or combinations of these Measurements with rotational devices become difficult to interpret at very high shear stresses, owing to the generation of heat in the specimen because of dissipation of energy, and to the tendency of the specimen to migrate out of the region of high shear This phenomenon, the Weissenberg effect, arises because the stress in any material can always be analyzed into the components of a 3 x 3 stress tensor, in wh~ch off-diagonal elements, the called normal stresses because they act perpendicular to the surface of the specimen, are not negligible in viscoelastic fluids A simple rotational instrument used in the rubber industry is the Mooney viscosimeter This empirical instrument measures the torque required to revolve a rotor at constant speed in a sample of the polymer at constant temperature It is used to study changes in the flow characteristics of rubber during milling or mastication (Chapter 19) The Brabender Plastograph is a similar device Capillary Viscometry. Capillary rheometers, usually made of metal and operated either by dead weight or by gas pressure, or at constant displacement rate, have advantages of good precision, ruggedness, and ease of operation They may be built to cover the range of shear stresses found in commercial fabrication operations However, they have the disadvantage that the shear stress in the capillary varies from zero at the center to a maximum at the wall An elementary capillary rheometer (extrusion plastometer) is used to determine the flow rate of polyethylene in terms of melt index, defined as the mass rate of
T'ABLE 11-1. Viscosity
Summary of Methods for Measuring
Method
Capillary pipette Falling sphere Capillary extrusion Parallel plate Falling coaxial cylinder Stress relaxation Rotating cylinder Tensile creep
Approximate Useful Viscosity Range (Poise)

~O-~-IO~ 1-loS 1-108 104-lo9 l0~-10" 103--10'0 1-10'2 105-> 1012
VISCOUS FLOW
305
flow of polymer through a specified capillary under controlled conditions of' tern-, perature and pressure
Experimental Results Molecular Weight and Shear Dependence. As discussed further in Chapter 12, the most important structural variable determining the flow properties of polymers is molecular weight or, alternatively, chain length, Z (the number of' atoms in the chain). Although eady data (Flory 1940) suggested that log q was proportional to Z"', it is now well established (Fox 19.56) for essentially all polymers studied that, for values of Z above a critical value Z,,
FIG. 11-2. Dependence of melt viscosity q on chain length 2 for low shea~ , rate, showing the regions below Z, = 600, where q is approximately proportional to z:", and above Z,, where q 2 (Fox 1956) : '
RHEOLOCY A N D THE MECHANICAL PROPERTIES OF POLYMERS
Log shear rate (sec-l)
FIG. 11-3. Generalized melt viscosity-shea~ rate curves for polymers with broad and narrow distributions of molecula~weight
where k is temperature dependent This equation is valid only for shear stress sufficiently low (10'-lo3 dynes/cm2) that the viscosity is Newtonian The weightaverage chain length 2 is usually assumed to be the appropriate average for the , above conditions. de Gennes (1979) discusses the unusual exponent 3 4 in telms of scaling concepts, considering it a majo~unsolved problem Theories to explain it, including his own reptation model (de Gennes 1971), based on the wriggling motion of a chain inside a "tube" formed by its neighbors, fail to predict the exponent closer than 3 0 with no obvious reason for the discrepancy For chain lengths below Z , , which is about 600 for many polymers, the viscosity is found to depend upon a power of 2 (and hence M,) in the range 1 75-2 0 In , this range, shear rate has little effect on viscosity Typical experimental data for the two regions described above are shown in Fig 11-2 While the Newtonian melt viscosity is determined by M, as described in Eq 11-3, the dependence of viscosity on shear rate also depends upon the molecularweight disuibution As depicted schematically in Fig. 11-3, the drop in melt viscosity below its Newtonian value begins at a lower shear rate and continues over a broader range of shear rates for polymers with broader distributions of molecular weight At sufficiently high shear rates, the melt viscosity appears to depend prirather than &.Qualitative information, at least, about the molecularmarily on M,, weight distribution can be obtained from melt viscosity-shear rate studies
Flow Instabilities. At shear stresses in the neighborhood of' 2 x lo6 dynes/cm2 for many polymers, instabilities in the flow appear, with the result that the upper Newtonian region is rarely realized in bulk polymers. These instabilities are manifested as a striking and abrupt change in the shape of'the polymer stream emerging from the capillary of the rheometer. At and above a critical stress, the shape of' the emerging stream changes ffom that of' a regular cylinder to a rough or distorted one
KINETIC THEORY O F RUBBER ELASTlC:lTY
307
Temperature Dependence of Viscosity. On close examination of' polymer systems, the activation energy fbr viscous flow of Eq. 11-2 is found to be constant only for small ranges of temperature For pure liquids, it was fbund many years ago that most of the change in viscosity with temperature is associated with the concurrent change in volume. This observation led to theories of' viscosity based on the concept of free volume, whose application to polymers is discussed in Section D A major result of these theories is the WLF equation (Eqs. 11-13 and 11-22), in which the temperature dependence of melt viscosity is expressed in terms of'the glass transition temperature T, (or another reference temperature) and universal constants. Since the terms describing the variation of' melt viscosity with temperature and with molecular weight are independent, the WLF equation can be combined with Eq. 11-3 to yield the relation, for low shear rates,
log "(1 = 3 4 log Z, -
17 44(T - T,) 5 1 6 t T - T,
+ k'
where kt is a constant depending only on polymer type. This equation holds over the temperature range from T, to about T, + 100 K
GENERAL REFERENCES
Eirich 1956-1969; Van Wazer 1963; Carley 1975; Walters 1975; Nielsen 1977; White 1978; Whorlow 1979; Dealy 1981; Hull 1981; Pierce 1982; Rosen 1982, Chapters 15 and 17
B.
KINETIC THEORY OF RUBBER ELASTICITY
Rubberlike elasticity is in many respects a unique phenomenon, involving properties markedly different from those of low-molecular-weight solids, liquids, or gases. The properties of typical elastomers are defined by the following requirements:
a. They must stretch rapidly and considerably under tension, reaching high elongations (500-1000%) with low damping, that is, little loss of' energy as heat. b. They must exhibit high tensile strength and high modulus (stiffness) when fully stretched c. They must retract rapidly, exhibiting the phenomenon of snap or rebound.. d. They must recover their original dimensions fully on the release of' stress, exhibiting the phenomena of' resilience and low permanent set.
Although the thermodynamics associated with rubber elasticity was developed in the middle of the nineteenth century, the molecular requirements for the exhibition
308
RHEOLOGY A N D THE MECHANICAL PROPERTIES O F POLYMERS
of rubbery behavior were not recognized until 1932. Theories of' the mechanism relating these molecular.-structure requirements to the phenomena of rubber elasticity were developed soon after. As discussed further in Chapter 12E, the molecular requirements of elastomers may be summarized as follows: The material must be a high polymer.. b. It must be above its glass transition temperature T, to obtain high local segment mobility c. It must be amorphous in its stable (unstressed) state for the same reason d. It must contain a network of crosslinks to restrain gross mobility of its chains
a. Thermodynamics of Rubber Elasticity The Ideal Elastomer. In discussions of the mechanical properties of a polymer, parameters related to its distortion must be included, as well as the variables defining the state of the system, such as pressure and temperature In particular, when a sample of lubber is stretched, work is done on it and its free energy is changed. By restricting the development to stretching in one direction, for simplicity, the elastic work We!equals f dl, where f is the retractive force and dl the change in length If the change in free energy is G,
In analogy with an ideal gas, where
an ideal ela~tomeris defined by the condition that
The negative sign results since work is done on the specimen to increase its length
Entropy Elasticity. Equation 11-7 shows that the r.etractive force in an ideal elastomer is due to its decrease in entropy on extension. The molecular origin of' this entr.op,yelasticity is the distortion ofthe polymer chains from their. most probable conf'o~mations the unstretched sample. As described in Chapter. 7 8 , the distriin bution of these confbrmations is Gaussian, the probability of' finding a chain end
KINETIC THEORY OF RUBBER ELASTICITY
309
in a unit volume of the space coordinates x, y, z at a distance r from the other end being
where b2 = x12, x being the number of links, with length I. Since the entropy of' the system is proportional to the logarithm of' the number of configurations it can have,
where k is Boltzmann's constant It follows that for a single polymer chain the retractive fbrce f" for an extension of' magnitude dr is
It is customarily assumed that the retractive force f (Eq 11-7) for a bulk polymer sample can be identified as the sum of the forces f ' for all the chains in the specimen This assumption, which is inaccurate in detail but justified in most cases, implies that individual chains contribute additively and without interaction to the elasticity of the macroscopic sample, and that the distribution of end-to-end distances undergoes transformation identically with the sample dimensions (affine transformation)
Stress-Strain Behavior of Elastomers

The stress-strain curves of typical elastomers (Fig 11-4) show marked deviations from the straight line required by Hooke's law The relatively low slope of the curve (defined near the origin as the modulus) decreases to about one-third its original value over the first hundred percent elongation, and later increases, often to quite high values at high elongations To explain this behavior, it is convenient first to examine the stress-strain properties of a simple model, and then to consider the relation of the model to actual elastomers
A Simple Model of an Elastomeric Network (Guth 1946) It is assumed that the actual tangled mass of polymer chains may be represented by an idealized network of flexible chains, irregular in detail but homogeneous and isotropic, extending throughout the sample The network consists of m chains of average length (p)"2 per unit volume directed along each of three perpendicular axes. To obtain the proper space-filling properties of the model it is assumed that the space between the chains is filled with an incompressible fluid exerting a hydrostatic pressure p outward against the elastic tension of the chains By considering the equilibration of inward and outward forces when a cube of this material is stretched
RHEOL.OCY A N D I H E MECHANICAL PROPERTIES OF POLYMERS
Elongation, % FIG. 11-4. Stress-strain curve for a typical elastomer (Guth 1946)
into a parallelepiped, using Eq. 11-10, it can be shown that the stress s is related to the strain y by
The modulus i' is given by
y As the strain increases fiam zero ( = 1) to large values of' y, the slope decreases to one-third its initial value as required by the experimental facts. Equation 11-12 predicts the stress-strain curve of actual elastomers very well up to elongations of 300% or more (Fig. 11-5).
t Note that the symbol G is conventionally used to represent both the Gibbs free energy (Eq 11-5) and, throughout the remainder of this chapter, the modulus of elasticity
Relat~velength
FIG. 11-5.
Experimental and theoretical stressstrain curves (Guth 1946): experimental (-), corn. puted from Eq 11-12 ( )and asymptote of Eq 11 12 for high elongations (- - - ) O,
Behavior at High Extensions. Figure 11-5 suggests that at elongations greater than about 300%, the stress-strain curves of actual elastomers have a higher slope than that predicted by Eq 11-12 This is far below the ultimate elongation of 1000% for a good elastomer One reason for the failure of the theory is that the Gaussian distribution of chain lengths that holds at low elongations fails in the region of interest for elastomers, as the actual lengths approach those of fully extended chains A better approximation than the Gaussian is available With its use the theory predicts the actual stress-strain curves rather well for elastomers that do not crystallize on stretching In the cases where crystallinity does develop, the slope still increases faster than predicted As is true in general for the problem of the mechanical behavior of crystalline polymers (Section c), a quantitative theory of the effect of crystallinity on the modulus of elastomers has not been developed
GENERAL REFERENCES
Treloar 1975; Smith 1977; Brydson 1978, Chapte~3; Gent 1978; Shen 19'78
C. VISCOELAS'TICIT'Y
Sections A and B have dealt with the equilibrium r.esponse of linear and network polymer structures to external stress. In this section are considered the time-dependent mechanical properties of' amorphous polymers. As in the studies of' rate
312
RHEOL.OCY A N D THE MEC'HANIC:AL PROPERTIES OF POLYMERS
phenomena, theories of dynamic mechanical phenomena in polymers are less thoroughly developed than those referring to the equilibrium states. They are also more dependent on the details of models. The present problem is, first, to find suitable models or mathematical functions with which to describe the several types of molecular motion postulated earlier to be associated with the processes of' elasticity, viscoelasticity, and viscous flow in polymers; and, second, to utilize these models in relating the behavior of the polymers to their molecular structure The description of the viscoelastic response of amorphous polymers to small stresses is greatly simplified by the application of'the fbllowing two general principles that ase widely applicable to these systems..
The Boltzmann Superposition Principle. This principle states that strain is a linear function of stress, so that the total effect of applying several stresses is the sum of' the effects of applying each one separately Application of the superposition principle makes it possible to predict the mechanical response of an amorphous polymer to a wide range of loading conditions from a limited amount of' experimental data.. The principle applies to both static and time-dependent stresses
Time-Temperature Equivalence. An increase in temperature accelerates molecula~and segmental motion, bringing the system mole rapidly to equilibrium or apparent equilibrium and accelerating all types of viscoelastic processes A convenient way of formulating this effect of temperature is in terms of the ratio a 7 of the time constant (relaxation time)of a particular response T at temperature T to its value TO at a convenient reference temperature P For many cases, including most nonpolar amorphous polymers, a, does not vary with T , so that changes in temperature shift the distribution of relaxation times, representing all possible molecular Iesponses of the system, to smaller or greater values of 7 but do not otherwise alter it Time and temperature affect viscoelasticity only through the product of a, and actual time, and a, is called a shift factor Its application to experimental data is described in the following section. Despite the marked dependence on molecular structure of the relation between a, and absolute temperature, nearly general empirical relations have been derived by expressing the temperature for each material in terms of its glass-transition temperature T, or some nearly equivalent reference temperature Among the most successful of these relations is the Williams-Landel-Ferry (WLF) equation (Williams 1955):
- 17.44(T
log a, =
- T,)
51.6 +. T - T ,
This equation holds over the temperature range from T, to about T, constants are related to the free volume as described in Section D
100 K The
If elongation is stopped during the determination ofthe stress-strain curve of a polymer (Chapter 9 F ) , the force or stress decreases with time as the specimen approaches equilibrium or quasi-equilibrium under the imposed strain. The direct observation and measurement of this phenomenon constitutes the stress relaxation experiment.. Usually the sample is deformed rapidly to a specified strain, and stress at this strain is observed fbr periods ranging from several minutes to several days or longer. It has been shown both experimentally and theoretically that the stress-relaxation behavior of' rubbery polymers can be factored into independent functions of strain and time. At small strains, the stress-strain function is almost linear and can be represented by a time-dependent modulus of elasticity G(t). Typical stress-,relaxation data for a well-studied sample of polyisobutylene are shown in Fig. 11-6, w h e ~ e G(t) is plotted against time for experiments performed at several temperatures By means of time-temperature equivalence these data may be shifted to produce a "maste~curve," as shown in Fig. 11-7. The curve indicates that any specified modulus can be observed after a period that depends, through
Stress Relaxation.
Time, h r
FIG. 1 1.6. Stress relaxation of polyisobutylene (Tobolsky 1956)
1 RHEOLOGY A N D THE MECHANICAL PROPERTIES O F POLYMERS
Log aT t, h r
FIG. 11 -7. Sbess relaxation rnasteI Arve for polyisobutylene (Tobolsky 1956)
the factor a,, on temperature As temperature is increased, a given modulus is observed at shorter times Master curves such as that of Fig 11-7 show the different types of viscoelastic behavior usually observed with amorphous materials At small times the high modulus and low slope are characteristic of glassy behavior The following region of rapidly decreasing modulus represents the glass transition, described further in Section D This is followed by a flat region of rubbery behavior produced either through entanglements among relatively long molecules or through permanent crosslinks The ultimate slope at long times represents the region of viscous flow in uncrosslinked polymers
Creep. Creep is studied by subjecting a sample rapidly to a constant stress and observing the resulting time-dependent strain for relatively long periods of' time, frequently for a week or more or even for a year or more. Creep and stress relaxation are complementary aspects of plastics behavior and in many cases may provide equivalent information for studies of both fundamental viscoelastic properties and performance in practical applications. Creep experiments are usually easier, more economical, and more feasible for long periods of time
Dynamic Methods. The delayed reaction of a polymer to stress and strain also affects its dynamic properties If a simple harmonic stress of angular frequency o is applied to the sample, the strain lags behind the stress by a phase angle whose tangent measures the internal frlcfion AEIE, where AE is the energy dissipated in taking the sample through a stress cycle and E is the energy stored in the sample when the strain is a maximum The internal friction is a maximum when the dynamic modulus of the material (the ratio of the stress to that part of the strain which is in phase with the stress) is in a region of relaxation In terms of frequency the relaxation behavior of the dynamic modulus and the internal friction is illustrated in Fig. 11-8
WT
FIG. 11-,8. Variation of dynamic modulus and internal friction with frequency
At low frequencies (about 1 sec-') it is convenient to measure the internal friction of polymers by observing the free decay of torsional oscillations of a thin fiber or film loaded with a large moment of inertia (torslon pendulum) The inte~nal friction is the logarithm of the ratio of the amplitudes of successive free oscillations, while the modulus is calculated from the frequency of the oscillations A natural extension of this technique is to force oscillation with an external driving f o ~ c eIn such techniques the specimen is drlven in flexural, torsional, or longitudinal oscillations by an oscillating force of constant amplitude but variable frequency, and the displacement amplitude of the resulting oscillation is observed These techniques are particularly useful in the kilocycle frequency xange At very high frequencies (above 1 megacycle), the dimensions of the sample become awkwardly small for mechanical techniques, and it is convenient to measure the attenuation of sound waves in the sample These methods are discussed fully by Ferry (1980) Spelling (1982) describes a simple classroom experiment illustrating the above principles
Models of Viscoelastic Behavior

Before attempting to devise a model to duplicate the viscoelastic behavior of an actual polymer, it is well to examine the response y of ideal systems to a stress s. An ideal elastic element is represented by a spring which obeys Hooke's law, with
316
RHEOLOGY A N D THE MECHANICAL PROPERTIES OF POLYMERS
FIG. 11-9. Strain-time relationships at constant stress for simple models (Alfrey 1948): (a) ideal elastic spring. (b) Newtonian fluid (dashpot), (c) Maxwell element, (d) Voigt element
a modulus of elasticity G The elastic deformation is instantaneous and independent of time: y = (1lG)s (Fig 11-9a) A completely viscous response is that of a Newtonian fluid, whose deformation is linear with time while the stress is applied and is completely i~recoverable: dyldt = ( l / q ) s , where q is the viscosity of the fluid (Fig 1 1 -9b) A simple mechanical analogy of a Newtonian fluid is a dashpot The two elements of spring and dashpot can be combined in two ways If they are placed in series, the resulting Manvell element (Fig 11-10a) exhibits flow plus elasticity on the application of stress (Fig 11-9c); when the stress is applied, t$e sp~ing elongates while the dashpot slowly yields On the removal of the stress the spring recovers but the dashpot does not The strain is given by the equation
The relation of creep to stress relaxation may be seen by considering the experiment in which a strain is obtained and then held by fixing the ends of the system: dyl dt = 0 in Eq.. 11-14. The equation can then be solved:
(a)
FIG. 11-10.
(b)
(a) A Maxwell element and (b) a Voigt element
where the stress s relaxes from its initial value so exponentially as a function of time. The time q / G after which the stress reaches l l e of' its initial value is the relaxation time T. The response of a tangled mass of polymer chains to a stress is better represented by a parallel combination of' spring and dashpot in a Kelvin or Voigt element (Fig 1 1- 106). This element shows a retarded elastic or viscoelastic response (Fig 1 19 4 . The dashpot acts as a damping resistance to the establishment of the equilibrium of the spring.. The equation for the strain is
If a stress is applied and after a time removed (Fig.. 11.-9d),the deformation curve is given by
where T is a retardation time. When the stress is removed the sample returns to its original shape along the exponential curve:
General Mechanical Models for an Amorphous Polymer. The three components that make up the simplest behavior of an actual polymer sample in creep can be represented by a mechanical model which combines a Maxwell and a Voigt element in series. If a stress is suddenly applied to this model, the strain changes behavior of the model is shown with time as shown in Fig 11-1 1; the cor~esponding inFig 11-12. The model departs from the initial conditions (a) at time t , by an elastic deformation s/G, (6) A viscoelastic response approaching s/G2 as an equilibrium value
FIG. 1 1 -1 1. Strain-time relationship for a generalized mechanical model for creep combining elasticity, viscoelasticity, and flow (Alfrey 1948)
318
RHEOLOCY A N D THE MECIHANIOAL. PROPERTIES O F POLYMERS
and a viscous flow at the rate sly3 follow (c). On the removal of' the stress at time (4 and the viscoelastic one slowly ( e ) , but the viscous flow is never recovered. Alternatively, the creep and stress.-relaxation experiments can be described with a generalized model consisting of a Maxwell and a Voigt element arranged in parallel. The two generalized models ase entirely equivalent.. Although these models exhibit the chief' characteristics of the viscoelastic behavior of polymers, they are nevertheless very much oversimplified. The flow of' the polymer is probably not Newtonian, and its elastic response may not be Hookean Moreover, the behavior of a real polymer cannot be characterized by a single relaxation time, but requires a spectrum of'relaxation times to account for all phases of' its behavior..
t,, the elastic element relaxes immediately
Treatment of Experimental Data Distribution of Relaxation Times. Equation 11-17 indicates that most of the relaxation associated with a single element takes place within one cycle of log time In contrast, experiments indicate (as in Fig 11-7) that relaxation phenomena in polymers extend over much wider ranges of time Thus the actual behavior of polymers can be described only in terms of a distribution of model elements and an associated distribution of relaxation or retardation times: ~(t) s =
Im
- rn
j (log r ) ( l - e-"')d log
(1 1-19)
where J = 1/G is the elastic compliance and j(1og 7) the distribution of' retardation times.. In principle, knowledge of y (t)over the entire range of time allows 5(10g 7) to be evaluated explicitly. This is difficult in practice, however, and it is convenient for mathematical simplicity to adopt simple empirical forms of the distribution
(a)
FIG. 11-1 2.
(b)
(c)
Interpretation of strain-time curve of Fig 11 model fbr creep
(d) (el terms of the generalized mechanical
function ?(log 7)and evaluate their parameters from the more limited experimental data at hand Alternatively, the experimental data may be fitted graphically by summing the theoretical curves corresponding to a small number (thee or four) of discrete relaxation times The usefulness of the distribution of relaxation times arises from its identification with certain molecular parameters of the specimen, as described below.
Molecular Structure and Viscoelasticity

The phenomena of relaxation processes can be thought of in terms of the effect of thermal motion on the orientation of polymer molecules When a mechanical stress is applied to a polymer, introducing deformations of the chains, the entropy of the system decreases as less probable conformations are taken up The free energy cor~espondinglyincreases If the sample is kept in the deformed state, stress relaxation takes place as a result of the thermal motions of the chains, the molecular deformations are obliterated, and the excess free energy is dissipated as heat. The details of the stress-relaxation process depend upon the multiplicity of ways in which the polymer molecules can regain their most probable conformations through thermal motion These complex motions of a polymer molecule can be expressed as a se~ies characteristic modes requiring various degrees of long-range coopof eration among the segments of the chain Thus the first mode corresponds to translation of the entire molecule, requiring maximum coope~ation,the second corresponds to motion of the ends of the chain in opposite directions, requiring somewhat less cooperation, and so on With each of these modes is associated a characteristic relaxation time; there are so many modes that over most of the time scale the discrete spectrum of el ax at ion times can be approximated by a continuous distribution. Theories based on the above considerations, derived for dilute solutions of polymer molecules, have been combined (Ferry 1980) with the concepts of entanglement of long molecules successful in predicting the molecular-weight dependence of VISCOUS flow (Section A) Although exact numerical agreement is not achieved, for example, in calculating the dynamic modulus of polyisobutylene over a wide range of frequency and molecular weight, the qualitative situation seems extremely satisfactory In the rubbery region, the maximum relaxation time is strongly dependent on molecular weight In this region the motions of the molecules are long range m nature, involving motions of units of the order of the length of the molecule itself This is the region where entanglements are important In the region around the glass transition, however, only vibrations of the parts of the molecule between entanglements are important So long as the molecules are long enough for entanglements to exist, this portion of the curve is expected, and found, to be essentially independent of the molecular weight and to depend primarily upon the local structure of the polymer
320
GENERAL REFERENCES
Alfrey 1948, 1967; Eirich 1956-1969; Tobolsky 1960; Van Wazer. 1963; McCrum 1967; Ward 1971; Nielsen 1974; Sauer 1975; Sternstein 197'7; Kramer 1978; Tager 1978, Chapter 7; F e r ~ y1980; Young 1981, Chapter 5; Rosen 1982, Chapter 18..
D. THE GLASSY STATE A N D THE CL.ASS TRANSITION As anticipated in the previous sections, and in particular in Fig 11-7, all amorphous polymers assume at sufficiently low temperatures the characteristics of glasses, including hardness, stiffness, and brittleness Many aspects of the glassy state and the glass transition were mentioned in Section C; what remains is to discuss measurement of the transition temperature and its molecular interpretation in more detail One property associated with the glassy state is a low volume coefficient pf expansion This low coefficient occurs as the result of a change in slope of the curve of volume versus temperature at the point called the glass-transition temperature T, This behavior is shown for natural rubber in Fig. 11-13 In the hightemperatme region, the slope of the curve (expansion coefficient) is characteristic of a lubber; below T, at about -70C, it is characteristic of a glass Figure 11-13 illustrates another general phenomenon: The amorphous regions in partially crystalline polymers also assume a glassy state, T, being independent of degree of crystallinity to a first approximation The magnitude of the phenomena associated with T, decreases with decreasing amorphous content, however As a result, T, is sometimes difficult to detect in highly crystalline polymers In terms of the lamellar model (Chapter 10C) the glass transition is considered to involve defect regions within or at the boundaries of the lamellae In contrast to crystalline melting at a temperature T,,, (about + 10C in Fig 1113), there is not an abrupt change in volume at T,, but only a change in the slope
Temperature. "C
FIG. 11-1 3. Volume behavior of rubber near the glass-transition temperature (Bekkedahl 1934)
THE GLASSY STATE A N D THE GLASS TRANSITION
321
of' the volume-temperature curve. In analogy to thermodynamic first- and secondorder transitions, T, is sometimes sef'er~ed as a second-order or apparent secondto order transition. This nomenclature is considered poor, however, since it implies more thermodynamic significance than the nature of the transition warrants,
Measurement of T,
The glass-transition temperature can be detected in a variety of experiments, which can be roughly classified into those dealing with bulk properties of the polymer, and those measuring the nature and extent of molecular motion The classification is to some extent arbitrary, since, as indicated below, T, is in fact the temperature of onset of extensive molecular motion
Bulk Properties. Perhaps the most common way of estimating T, is by means of the volume expansion coefficient, as indicated above Other bulk properties whose temperature coefficients undergo marked changes at T, , and which can therefore be used for its determination, include heat content (Chapter 9E), refractive index, stiffness, and hardness
Molecular Motion. Experiments that are sensitive to the onset of molecular motion in polymer chains may be used to detect the glass transition Such methods include the measurement of internal friction (Section C), dielectric loss in polar polymers, and NMR spectroscopy (Chapter 9B) Phenomena Related to T,. The onset of brittleness, as measured in impact tests, and the softening of amorphous polymers, as measured in thermal tests of various sorts, take place at temperatures near T, (see Chapter 9 F ) Time Effects Near T,. If a polymer sample is cooled rapidly to a temper.atuse just below T,, its volume continues to decrease for many hours. In consequence, the value observed as T, in a volume-temperature experiment depends on the time scale of' the measurements. It is conveni~nt define T, in terms of an arbitrary 'to convenient (but not highly critical) time interval, such as 10 min to 1 hr.. For similar reasons, other tests for T, give results somewhat dependent on the time scale of' the experiments, with tests requiring shorter times yielding higher values of T,. The brittle temperature as determined in an impact test is, fbr example, normally somewhat higher than T, as otherwise measured. Molecular interpretation of
1''
In the glassy state, large-scale molecular motion does not take place, rather, atoms and small groups of atoms move against the local restraints of' secondary bond forces, much as atoms vibrate around their equilibriun~ positions in a crystal lattice, except that the glassy state does not have the regularity of'the crystalline state The glass transition corresponds to the onset of' liquidlike motion of much longer seg-
322
RHEOLOCY A N D 'IHE MECHANKAL PROPERTIES O F PO1.YMERS
ments of molecules, characteristic of the rubbery state. This motion requires more free volume than the short-range excursions of atoms in the glassy state The rise in the relative free volume with increasing temperature above T, leads to the higher observed volume expansion coefficient in this region. Since the fully extended chain is the conformation of minimum energy (Chapter lOB), it tends to be assumed more frequently as the tempelature is lowered. As the molecules thus straighten out, the free volume decreases. In consequence, flow becomes more difficult The glass transition (observed at infinite time)--or, alternatively, the onset of crystallization where possible-is taken as the point where the number of possible conformations of the amorphous phase decreases sharply toward one The fraction f of "free" volume may be defined as
Thus f is constant at the value f, for all temperatures below T, Here the volume expansion coefficient a is that resulting from the increase in amplitude of molecular vibrations with temperature. Above T, new free volume is created as the result of an increase Aa in the expansion coefficient. Williams, Landel, and Ferry (Williams 1955) proposed that log viscosity varies linearly with l / f above T,, so that
Substitution into Eq . 11-20 leads to log
);(
= -
a(T - T,) b+T-T,
which is the WLF equation presented in Section C (Eq 11-13), the numerical constants for a and b given there being determined by fitting literature data on the viscosity-temperature behavior of many glass-forming substances. The shift factor a, is seen to be just the ratio of the viscosity at T relative to that at T, The latter is about lOI3 poise for many substances Equation 11-22 also implies that both the viscosity of the polymer and the activation energy for viscous flow AE = 2 3R d(log q)ld(llT) should become infinite at T = T, - b = T, - 51 6. Thus by extrapolating downward from behavior well above T,, one would predict that all molecular motion should become completely frozen at T < T, - 51 6. What happens, of course, is that new mechanisms of deformation take over more or less sharply as this critical range is approached, in fact at T,. The above discussion embodies elements of several theories of the glass transition, based on free-volume concepts, kinetics, and statistical thermodynamics
THE MECHANICAL PROPERTIES O F CRYSTALLINE POLYMERS
323
These theories ase summarized by Ward (1971) and Ferry (1980); important contributions were made by Doolittle (1951), Bueche (1953), Gibbs (19581, DiMarzio (1958), Kovacs (1958), Cohen (1959), and Adam (1965).
Molecular Motion Below T,
The foregoing discussion is concerned with the transition involving the motion of' long segments of the polymer chain. At lower temperatures, other transitions may occur, pxoduced by the motion of' short sections of the main chain or of' side chains. Although some characteristics of the glassy state, such as brittleness, may occasionally occur only below one of' these lower transitions, it is proposed that the transition of' highest temperature be called T, . Alternatively, the transitions may be denoted a,P, y,and so on, in order of descending temperature Transitions due to the motion of short segments of' the main polymer chain occur most prominently in crystalline polymers such as polyethylene, polypropylene, and polytetrafluoroethylene. Such polymers also typically exhibit an a-.transition. Side-chain transitions occur in methacrylate polymers as a result of the relaxation of the carbomethoxy side chain at about 20C (torsion pendulum) and the relaxation of' the aliphatic ester group below .- 150C.
GENERAL REFERENCES
Ward 1971; Nielsen 1974; Armeniades 1975; Tager 19'78, Chapters 8 and 9; Fer~y 1980; Young 1981, Chapter 5.
E. THE MECHANICAL PROPERTIES O F CRYSTALLINE POLYMERS
The models developed in the preceding sections represent well the rheological and mechanical properties of' amorphous polymers. The viscoelastic properties of crys... talline polymers are much more complex, however, and are not amenable to ade-, quate theoretical explanation fbr three reasons.. First, an amorphous polymer is isotropic. This means that models suitable for descxibing shear. stress, for example, are adequate to describe tensile stress or other types. Since crystalline polymers are not isotropic, this universality does not hold and the range of application of any model is severely limited.. Second, the homogeneous nature of amorphous polymers ensues that an applied stress is distributed uniformly throughout the system, at least down to very small dimensions. In crystalline polymers the relatively large crystallites are bound together in such a way that Iasge stress concentrations inevitably develop Finally, a crystalline polymer is a mixture of' regions of different degrees of order ranging all the way from completely ordered crystallites to completely amorphous regions. As the stress on the sample changes, the amounts of these regions change continuously as the crystallites melt or grow. This change of composition with respect to ordering is the most difficult obstacle to overcome in fbrmulating
B
324
RHEOLOCY A N D THE MECHANICAL PROPERTIES O F POLYMERS i
a theory of the mechanical behavior of crystalline polymers Even in the simplest cases the necessity of having the mechanical model change continuously with the 1 3 applied stress has led to serious difficulties In consequence, neither the Boltzmann superposition principle nor time-tern perature equivalence apply to crystalline polymers Without these simplifying prin- f ciples, attempts to explain the viscoelastic response of crystalline polymers in terms 5 f of models become complex and are only qualitative
Typical Behavior
r
L
An orderly, if' qualitative, discussion of' the mechanical properties of' crystalline polymers requires their classification into several categories, as indicated in Table 11-2.. It is in the range of' intermediate degree of' crystallinity that the properties, unique to polymers, of' most importance for mechanical and engineering appIications are found. Further, these properties are found fbr the most part in the temperature range between T, and T,,,, and at temperatures not fBr below T, Well below the glass transition, molecular motion is essentially absent and the material behaves as a hard, glassy solid with the presence or absence of' crystallinity making little difference: fbr example, the properties of atactic and isotactic polystyrene are quite similar at room temperature. Above T,,,, of' course, crystallinity plays no part in the properties of' the amorphous viscoelastic melt Polymers with low crystallinity include plasticized poly(viny1 chloride) and elastic polyamides. These materials behave like lightly crosslinked amorphous poly mers, their crystalline regions acting like crosslinks which are stable with respect to time but unstable with respect to temperature. The viscoelastic properties of these polymers are much like those of' amorphous polymers except that the transition region between glassy and rubbery or liquid behavior is very much broadened on the temperature scale. At very low extensions (< I%), at temperatures well below T,, and at not too long times, polymers with intermediate degrees of'crystallinity (such as low&nsit;; polyethylene) behave much like those with very low crystallinity. Time--tempzature equivalence is not applicable to such polymers. The transition regions of these polymers in modulus and the cor~espondingdistributions of' relaxation tirncs are exceptionally broad, as indicated by the master curves of' Fig 11-14. (It should be
'TABLE 11-2. Classification of Crystalline Polymers
Predominant Properties in Temperature Range

Above T, Below T,
Degree of Crystallinity
-
Low (5-1 0%)

Rubbery Glassy, br~ttle
Intermediate (20-60%)
Leathery, tough Hornlike, tough
High
(7fj-Y)%)
Stiff. f 2 ~ d
f Lr>-~k) Strfl djlrd b; e r : L -
THE MECHANICYAL PROPERTIES O F CRYSTALLlNE POLYMERS
FIG. 11-14.
Stress-relaxation master curves for several crystalline and amorphous polymers (Tobolsky 1960)
pointed out that these curves were themselves derived by application of timetemperature equivalence and must be considered only as idealizations ) At higher extensions, these polymers exhibit the phenomena of a yield stress and cold drawing, with the accompanying changes in crystalline morphology described in Chapter 10E. No adequate theory relating rheological and mechanical behavior in this region to structure has yet been formulated Polymers of intermediate degree of crystallinity are characteristically leathery or horny in texture, and exhibit good impact resistance which in many cases is retained even below T,. The exact structural features responsible for this toughness have not been well defined The major effects of a further increase in crystallinity to very high values include (a) a further increase in modulus, as the high modulus characteristic of the crystalline regions is approached, and (b) the onset of a tendency toward brittleness, in the sense of failure at low strains. Such polymers, of which high-density polyethylene is typical, can be cold drawn only with difficulty Tensile failure usually occurs at or slightly beyond the yield stress, accompanied by distortion or deformation that appears to occur at slip boundaries or dislocations, reminiscent of the viscoelastic behavior of metals
Crystallization on Stressing. The application of a mechanical stress to a noncrystalline but crystallizable polymer can cause crystallinity to develop, either by raising T, or. by increasing the rate of crystallization. An example of the former
326
RHEOL.OCY AND THE MECHANICAL PROPERTIES OF POLYMERS
effect is the crystallization of' natural rubber on stretching described in Section B At room temperature, unstretched natural rubber is above its crystalline melting point. As a tensile stress is applied, T, is raised and the rubber crystallizes to an oriented structure.. When the stress is released, T,,, is reduced and the rubber melts as it retracts An example of the effect of' stress on crystallization rate is seen in polymers such as poly(ethy1ene terephthalate) which can be quenched to a metastable amorphous state at temperatures well below T m .Without an applied stress, the rate of nucleation and crystal growth is very low When a tensile stress is applied, Tmis raised and the rate of' crystallization is greatly increased so that crystallization takes place during stretching. These crystals do not melt when the stress is removed, of' course
GENERAL. R F R N , S E E E CE
Alfrey 1948, 1967; Tobolsky 1960; McCrum 1967; Ward 1971; Nielsen 19f4; Schultz 1974, Chapter 11; McCullough 1977; Tager 1978, Chapter 9; F e r ~ y 1980; Young 1981, Chapter 5
DISCUSSiON QUESTIONS AND PROBLEMS

1. Define and distinguish among the following terms: (a) shear thickening, (b) shear thinning, (c) thixotropic, (d) rheopectic, (e) Newtonian behavior..
2. State the equations relating the melt viscosity of a polymer to (a) molecular weight and (b) temperature
3. List four physical properties characteristic of' typical elastomers

4. List fbur molecular-structure characteristics necessary for the development of' typical elastomer physical properties..
5. Describe two quantitative aspects of the elastomeric state that are explained by the kinetic theory of rubber elasticity
6 . Why can the kinetic theory of rubber elasticity not predict the entire stressstrain curve for a typical elastomer?
7 Derive the stress-strain equation for the simple stretching of' an elastomer, .
Eq. 11-11 8. Derive an equation similar to Eq. 11-11 for the case of' an elastomer with each dimension increased by a factor u due to swelling with an organic solvent
9. List t h ~ e etypes of' response of' a typical polymer to stress and suggest a molecular mechanism for each type.
10. Sketch a strain-time curve for each of the mechanisms described in Question 9
1 1. Describe briefly (a) creep and (b) stress relaxation.. 12. Why cannot the theories of' the viscoelasticity of' amorphous polymers be extended to crystalline polymers?
BIBLIOGRAPHY
Adam 1965 Gerold Adam and Julian H Gibbs, "On the Temperature Dependence of Cooperative Relaxation hoperties in Glass-Forming Liquids," J Chem Phys 43, 139-146 (1965) Alfrey 1948 Tulner Alfrey, Jr ,MechanlcalBehavlor ofHigh Polymers, Interscience, New York, 1948 Alfiey 1967 Turner Alfrey, Jr , and Edward F Gumee, Organic Polymers, Prentice-Hall, Englewood Cliffs, New Jersey, 1967 Armenlades 1975 C D Armeniades and Eric Baer, "Txansit~onsand Relaxations in Polymers," Chapte~ 6 in Herman S Kaufman and Joseph J Falcetta, eds , Introductron to Polymer Scrence and Technology An SPE Textbook, Wiley-Intersclence, New Yor k, 1975 Bauer 1967 Walter H Bauer and Edward A Collins, "Thixotropy and Dilatancy," Chapter 8 in Vol 4 , Academlc Press, New York, Frederick R Eirich, ed ,Rheology-Theory and Applz~atzons, 1967 Bekkedahl 1934 Norman Bekkedahl, "Forms of Rubber as Indicated by Temperatuxe-Volume Relationship," I Res Nut Bur Stand 13, 411-431 (1934) Brydson 1978 J A Brydson, Rubber Chemzstry, Applied Science, London, 1978 Bueche 1953 F Bueche, "Segmental Mobility of Polymers Near Their Glass Temperatures," J Chem Phys 21, 1850-1855 (1953) Carley 1975 James F Carley, "Rheology," Chapter 8 in Herman S Kaufman and Joseph I Falcetta, eds ,Introductron to Polymer Science and Technology An SPE Textbook. Wiley-Interscience, New York, 1975 Cohen 1959 Morrel H Cohen and David Turnbull, "Molecular Transport in Liquids and Glasses," J Chem Phys 31, 1164--1169 (1959) Cotton 1974 J P Cotton, D Decker, H Benoit, B Farnoux, J Higgins, G Janniuk, R Ober, C Picot, and J des Cloizeaux, "Conformation of Polymer Chains in the Bulk," Macromolecules 7 , 863-872 (1974) Dealy 1981 John M Dealy, Rheometers for Molten Plastrcs A Practical Guide f o Testzng & Property Measurement, Van Nostrand Reinhold, New York, 1981 Dealy 1983 J M Dealy, "Melt Rheometer Update," Plastrcs Eng 34, 57-61 (1983) de Gennes 1971 P G de Gennes, "Reptatlon of a Polymer Cham In the Presence of Fixed Obstacles," J Chem Phys 55, 572-579 (1971) de Gennes 1979 Pierre-Gilles de Gemes, Scaling Concepts in Polymer Physzcs, Co~nellUniversity Press, Ithaca, New York, 1979 DrMarzio 1958 E A DiMarzio and J EI Gibbs, "Chain Stiffness and the Lattice Theory of Polymer Phases," J Chem Phys 28, 807-813 (1958) Doohttle 1951 Arthur K Doolittle, "Studies in Newtonian Flow I1 The Dependence of the Viscosity of Liquids on Free-Space," J Appl Phys 22, 1471-1475 (1951) Elrich 1956-1969 Frederick R Eirich, ed ,Rheology-Theory and Applzcation~, Academlc Press, New York, Vol 1, 1956; Vol 2, 1958; Vol 3, 1960; Vol 4, 1967; Vol 5, 1969
328
Ferry 1980 John D Ferry, Viscoelastrc Properties of Polymers, 3rd ed , John Wiley & Sons, New York, 1980 Flory 1940 Paul J Flory, "Viscosities of Linear Polyesters An Exact Relationship between Viscosity and Chain Length," J Am Chem Soc 62, 1057-1070 (1940) Flory I949 Paul J Flory, "The Configuration of Real Polymer Chains," J Chem Phys 17, 303310 (1949) Fox 1956 T G Fox, Serge Gratch, and S Loshaek, "Viscos~tyRelationships for Polymers In Bulk and In Concentrated Solutron," Chapter 12 In Frederick R Elrich, ed , Rheology-Theory and Applicatrons, Vol 1, Academic Press, New York, 1956 Gent 1978 A N Gent, "Rubber Elasticity: Basic Concepts and Behavior," Chapter 1 in Frederick R Eirich, ed , Scrence and Technology of Rubber, Academic Press, New York, 1978 Gibbs 1958 Julian H Gibbs and Edmund A DiMarzio, "Nature of the Glass Transition and the Glassy State," J Chem Phys 28, 373-383 (1958) Glasstone 1941 Samuel Glasstone, Keith J Laidler, and Henry Eyring, The Theory of Rate Processes, McGraw-Hill, New York, 1941 Graessley 1974 William W Graessley, "The Entanglement Concept in Polymer Rheology," Adv Polym Sci 16, 1-179 (1974) Guth 1946 E Guth, H M Iames, and H Mark, "The Klnetic Theory of Rubber Elasticity," fip 253-299 in H Mark and G S Whitby, eds , Scientific Progress in the Field of Rubber and Synrhetrc Elastomers (Advances in Colloid Science), Vol 2, Interscience, New York, 1946 Hull 1981 Harry H Hull, An Approach to Rheology Through Multr-Varrable Thermodynamics Or, Jnsrde the Thermodynamic Black Box, Har~y Hull, Sun City, Florida, 1981 A Kovacs I958 A J Kovacs, "The Isothermal Volume Contraction of Amorphous Polymers" (in French), J Polym Sci 30, 131-147 (1958) "Dynamic Mechanical Properties," Chapter 5 in Frederick Kramer 1978 Ole Kramer and John D Fer~y, and Technology of Rubber, Academic Press, New York, 1978 R Eirich, ed , Sc~ence MtCrum 1967 N G McCrum, B Read, and G Williams, Anelastrc andDielectrrc Effects in Polymerrc Sohds, John Wiley & Sons, New York, 1967 McCullough 1977 R L McCullough, "Anisotrop~c Elastrc Behavlor of Crystalline Polyme~s," In J M Schultz, ed , Treatise onMaterials Screnceand Technology, Vol 10, Part B (Herbert Herman, series ed ), Academic Press, New York, 1977 Nielsen 1974 Lawrence E Nielsen, MechanicalPropertres of Polymers and Composites, Marcel Dekker, New Yolk, 1974 Nielsen 1977 Lawrence E Nielsen, Polymer Rheology, Marcel Dekker, New Yolk, 1977 Pierce 1982 Percy E Pierce and Clifford K Schoff, "Rheological Measurements," pp 259-319 in Martln Grayson, ed , Kirk-OthmerEncyclopedia of Chemrcal Technology, 3rd ed , Vol 20, WileyInterscience, New York, 1982 Rosen 1982 Stephen L Rosen, Fundamental Prrncrples of Polymerrc Materials, Wlley-Interscience, New York, 1982 Sauer 1975 J A Sauer and K D Pae, Mechanical Propert~esof High Polymers," Chapter 7 in to Herman S Kaufman and Joseph J Falcetta, eds ,Inrroductro~t Polymer Scrence and Technology An SPE Textbook, Wiley-Interscience, New York, 1975 Schultz 1974 Jerold Schultz, Polymer Materrals Scrence, Prentice-Hall, New York, 1974 Shen 1978 Mitchel Shen, "The Molecular and Phenomenological Basis of Rubberlike Elasticity," Chapter 4 in Frederick R Eirich, ed , Scrence and Technology of Rubber, Academic Press, New York, 1978 Smrth 1977 Thor L Smith, "Molecular Aspects of Rubber Elasticity," in J M Schultz, ed , Treanse on Materials Science and Technology, Vol 10, Part A (Herbert Herman, senes ed ), Academ~c Press, New York, 1977
Sperlrng 1982 L H Sperling, "Molecular Motion in Polymels," 3 Chem Educ 59, 942-943 (1982) Sternstern 1977 S S Sternstein, "Mechanical Properties of Glassy Polymers," in 3 M Schultz, ed , Treatrse on Materials Scrence and Technology, Vol 10, Part B (Herbert Herman, series ed ), Academic Press, New York, 1977 Tager 1978 A Tager, Physrcal Chemtrtry of Polymers (translated by Davrd Sobolev and N~cholas Bobrov), Mlr, Moscow, 1978 (Imported Publrcatlons, Chrcago) Tobol~ky 1956 Arthur V Tobolsky and Ephnam Catslff, "Elastov~scous Properties of Polylsobutylene (and Other Amorphous Polymers) from Stress Relaxatron Studles IX A Summary of Results," I Polym Scr 19, 111-121 (1956) Tobolsky 1960 Arthur V Tobolsky, Properties and Structure of Polymers, John Wiley & Sons, New York, 1960 Treloar 1975 L R G Treloar, The Physrcs of Rubber Elasrtcrty, 3rd ed , Clarendon Press, Oxford, 1975 Van Wazer 1963 J R Van Wazer, J W Lyons, K Y Kim, and R E Colwell, Kscosity and Flow Measurement. A Laboratory Handbook of Rheology, Wiley-Inte~sc~ence, New York, 1963 Walters 1975 K Walters, Rheometry, John Wiley & Sons, New Yolk, 1975 Wold 1971 I M Ward, Mechanical Properties of Sol~d Polymers, Wlley-Interscience, New York, 1971 Whrte 1978 James Lindsay White, "Rheological Behavior of Unvulcanized rub be^," Chapter 6 in Press, New York, 1978 Frederick R Eirich, ed , Science and Technology of Rubber, Academ~c Whorlow 1979 R W Whorlow, Rheologrcal Terhnrquer, John Wiley & Sons, New York, 1979 Wrlliams I955 Malcolm L Williams, Robert F Landel, and John D Ferry, "The Temperature Dependence of Relaxation Mechanisms in Amorphous Polymers and Other Glass-Forming Liquids," J Am Chem Soc 77, 3701-3707 (1955) Young 1981 Robert J Young, Introductton lo Polymers, Chapman and Hall, New York, 1981
CHAPTER TWELVE
POLYMER STRUCTURE A N D PHYSICAL PROPERTIES
Although inter~elations between the molecular structure of polymers and their properties are mentioned throughout this book, they are emphasized in this chapter First the structural features of polymers most directly responsible for determining their properties are classified, and in succeeding sections we describe how these structures influence various classes of physical properties Finally, the property requirements associated with the uses of various polymers are reviewed Since the acceptance of the macromolecular hypothesis in the 1920's, it has been recognized that the unique properties of polymers--for example, the elasticity and abrasion resistance of rubbers, the strength and toughness of fibers, and the flexibility and clarity of films--must be attributed to their long-chain structure In the examination of structure-property relationships it is advantageous to classify properties into those involving large and small deformations The former class includes such properties as tensile strength and phenomena observed in the melt, while properties involving only small deformations include electrical and optical behavior, such mechanical properties as stiffness and yield point, and the glass and crystalline melting transitions Properties involving large deformations depend primarily on the long-chain nature of polymers and the gross configuration of their chains Important factors for this group of proper.ties include molecular weight and its distribution, chain branching and the related category of side-chain substitution, and c~osslinking Physical properties associated with small deformations are influenced most by factors determining the manner in which chain atoms interact at small distances The ability of polymers to crystallize, set by considerations of symmetry and stedc effects, has major importance here, as do the flexibility of the chain bonds and the
THE CRYSTALLINE MELTING P O I N T
331
number, nature, and spacing of polar groups. To the extent that they influence the achievement of' local order, gross configurational properties are also important. Similar considerations apply to amorphous polymers below the glass transition In crystalline polymers, the nature of the crystalline state introduces another set of' variables influencing mechanical properties. These variables include nature of the crystal structure, degree of' crystallinity, size and number of spherulites, and orientation. Some of' these phenomena are in turn influenced by the conditions of' fabrication of' the polymer Finally, the properties of'polymers can be varied importantly by the addition of' other materials, such as plasticizers or reinforcing fillers. Properties involving both large and small deformations may be influenced in this way As described in Chapter 1, one of the most important determinants of' polymer properties is the location in temperature of the major transitions, the glass transition and the crystalline melting point. It is appropriate therefbre first to discuss the relations between molecular structure and these transition temperatures
A.
THE CRYSTALLINE MELTING POINT
As has been pointed out in Chapter IOD, crystalline melting in polymers is at least a pseudo-equilibrium process, and it is convenient here to describe it in thermodynamic terms, realizing that in a given situation an observed melting point T, may not be the equilibrium value Melting takes place when the free energy of the process is zero: AG = AH,,, -- TmAS, = 0 Thus T, is set by the ratio AH,/ASm, and it is necessary to explore the effect of molecular structure upon both of these quantities to gain insight into the melting process The crystalline melting point is usually taken to be independent of molecular weight in the polymer range It is assumed that AH, and AS, are made up'of molecular-weight-independent terms H , and So plus increments H,and S, for each chain unit Thus, for degree of polymerization x ,
"
LW,,, -= H , + x H ,
AS, So
+ xS,
- -ff1
SI
as .x
(12-1)
Melting Points of Homologous Series

The melting points of homologous series of various types of' polymers are plotted in Fig. 12-.I.The effect of' the spacing of the polar. groups on T, is much as predicted by considering the polymers as derived from polyethylene by replacing methylenes with polar groups, but the overall level of'T, for each series remains to be considered.
-14
16
18 20 22 24 Chain atoms in repeating unit
26
FIG. 12-1. Trend of crystalline meltlng points in homologous series of aliphatic polymers (after H~ll 1948)
Polyesters. Although the low melting points of' linear aliphatic polyesters were once attributed to unusual flexibility of'the C-0 chain bond (Bunn 1955), implying high entropies of'fusion, both AS, and AH, ase significantly lower in these polyesters than in polyethylene (Wunderlich 19.58) Only recently have the reasons for these low values been elucidated in detail, as described later in this section Polyamides. For polyamides it is observed that AH, is lower than for polyeth ylene; therefore molar cohesion cannot account for high values of T, These result from low liquid-state entropies, leading to low values of AS,, which arise from partial retention of hydrogen bonding in the melt and from chain stiffening due to the tendency for resonance of the type
within the amide group Similar considerations undoubtedly apply to the melting points of polyuretbariCC and polyureas, but values of AH, and AS,,, are not available As the spacing between the polar groups is increased, the melting points approach that of polyethylene When the homologous series are examined in more deta~l, thr
THE CRYSTAL1.ING MELTING POINT
333
melting points are fbund to vary in a more complex way with the spacing of' the polar groups than is suggested in Fig 12-1. An alternation of' T,,, with spacing is typical (Fig. 12-2) It results from differences in the crystal structure, which alter-, nates in type f'or chain repeat units with odd and even numbers of carbon atoms.
Effect of Chain Flexibility and Other Steric Factors Chain Flexibility. The flexibility of chain molecules arises from rotation around hindering this ~otationIange saturated chain bonds The potential energy b a r ~ i e ~ s from 0 2 to 1 2 kJ/mole, the same order of magnitude as molecula~ cohesion forces It is not surprising therefore that the flexibility of polymer chains is an important factor in determining theu melting points Thus polytetrafluoroethylene (Tm = 327C) melts much higher than polyethylene because of its low entropy of fusion (Starkweather 1960), which results from the high stiffness of the polymer chains Similarly, the high melting point of isotactic polypropylene (Tm = 165C) is attributed to low entropy of fusion arising from stiffening of the chain in the melt because of the higher energy barrier for rotation about C--C bonds than in polyethylene In neithe~case can a high heat of fusion account for the high value of T,,,, since for both polytetrafluoroethylene and polypropylene AHm is well below that of polyethylene as indicated in Table 10-1 (Dole 1959)
Number of carbon atoms FIG. 12-2. Dependence of crystalline melting point on spacing of polar groups Number of' carbon atoms refers to (a) acid for polyesters made with decamethylene glycol, (b) diamine for polyarnides made with sebacic acid, (c) diamine f'or polyur.ethanes made with tetramethylene glycol, and (d) w-.amino acid polymers (Bannerman 1956)
334
POLYMER STRUCTURE A N D PHYSlC,AL PROPERTIES
The substitution of' an inflexible g ~ o u p the p-phenylene group, like
for six chain CH, groups causes a marked rise in the melting point of the polymer (Edgar 1952). Some examples of this change are shown in Table 12-1 The rise is considerable when the p-phenylene group is connected to CH, groups, but greater still when it is connected to carbonyl,
since the group can resonate as a unit Among other chain-stiffening groups are p,pl-diphenyl, 1,5- or 2,6-naphthyl, diketopiperazine,
TABLE 12-1. Effect of a p-Phenylene Group on the Melting Point of Condensation Polymersa
----
Repeating Unit
-O(CH2)20CO(CH2)6CO--O(CH2)20CO O C O -
Tm
("C)
45
265
235
350h
-NH(CH2)6NHCO(CH2)6CO--NH(CH2)6NHCO O C O -O(CH2)80CO(CH2)8CO--OCH2(/?: -CH2CH2-CH2oCH2CH20COCH2 0 C H : C O -
75'
146
135 380
"Edgar (1 952) *Decomposes 'Estimated
THE CRYSTALLINE MELTING POINT
triazole,
-c\
acetal, and thioketal
/NH\
N-N
/c.-
Side-Chain Substitution. In most cases, the substitution of' nonpolar groups for hydrogens of' a polymer chain leads to a reduction in T,,, or possibly complete loss of crystallinity. If' the substitution is random, as in branched polyethylene, the primary effect is a reduction in the size and perfection of' the crystalline regions, usually accompanied by a decrease in the degree of' crystallinity. The crystalline melting point of polyethylene is lowered 20-25C on going from the linear to the branched material.. Replacement of' an amide hydrogen with an alkyl group has a much larger effect, since hydrogen bonding is destroyed. In general, N-methyl nylons melt at least 100C lower than their unsubstituted counterparts.. When an alkyl group is regularly substituted into a methylene chain, with the retention of ste~eoregularityso that crystallization is possible, two effects compete in setting T,,. As indicated for isotactic poly(a-olefins) in Table 12-2 (with polyethylene omitted because of its widely different crystal structure), an increase in the length of'the side chain results in a looser crystal structure with an increasingly lower melting point On the other hand, an increase in the bulkiness of' the side chain increases T,, since rotation in the side chain is hindered in the liquid state, with consequent decrease in AS,
TABLE 12-2. Effect of Side-Chain Structure on the Crystalline Melting Point of lsotactic P o l y ( ~ o i e f i n s ) ~
Side Chain
-CH3 --CH2CH3 -CH2CHICH3 --CH2CH2CH2CH3 -CH2CHCH2CH3
I
Tm ("(3
165 125 75 - 55 196
"Campbell (1959), Bawn (1960)
3 36
POLYMER STRUCTURE AND PHYSICAL PROPERTIES
Entropy and Heat of Fusion Entropy of Fusion. The effect of structure on the entropy of' fusion can now be calculated, in good agreement with experiment, for many polymers. This had led, fbr example, to a satisfactory explanation (Hobbs 1970) ofthe low entropy of'fusion of the linear aliphatic polyesters whose melting points are depicted in Fig. 12-1 Following Starkweather (1960), the entropy of fusion is considered to consist of independent contributions from volume change on melting and the change in the number of confbrmations that the chain can assume, from one in the crystal to a number determined by the type of chain bonds in the melt. The latter number can be calculated by an enumeration scheme, and is combined with knowledge of the potential barrier to rotation to compute the desired confbrmational entropy. It was concluded that, in comparison to polyethylene, the stiffening effect of the cigid ester bonds outweighs to a small degree the increased flexibility of the ester chains Heat of Fusion. The low melting points of the linear aliphatic polyesters must thus be attributed to their low heats of fusion These are less easily explained in detail, but a very approximate analysis (Hobbs 1970) suggests that the cohesive forces to be overcome in fusion result almost entirely from methylene interactions (dispersion forces) between neighboring chains Since there are fewer methylene groups per unit chain length in the polyesters, their heats of fusion are lower Dipole interactions from the ester carbonyl groups appear not to contribute to the heats of fusion of these polymers, supporting the inference that the dipole bonds in the crystal are almost entirely reformed in the melt A similar effect was postulated for the polyamides by Dole (1959) Effect of C.opolymerization
'
When copolymers are made from monomers that form crystalline homopolymers, degree of crystallinity and crystalline melting point decrease as the second constituent is added to either homopolymer The melting point depends on the mole fraction n of the crystallizing constituent by the relation of Eq 10-4 (Flory 1949):
is where TmO the melting point of the homopolymer and AH, is its heat of fusion A typical case is the copolymer of hexamethylene terephthalamide and hexamethylene sebacamide (Fig 12-3) If, however, the comonomers are isomorphous, that is, capable of replacing each other in the crystals, the melting point may vary smoothly over the composition range An example is the copolymer of hexamethylene terephthalamide and hexamethylene adipamide, also shown in Fig 12-3 Other variations may occur, including the formation of an alternating copolymer with a crystal structure and
THE GLASS TRANSITION
300
$'
280
u -
260'-d
x
H 240
+ m
2
Adipamide Sebacam~de
200
10
20
30
40
50
60
FIG. 12-3
Mole % hexamethylene terephthalarnide Melting points of copolymers of hexamethylene adipamide and terephthalamide, and of hexamethylene sebacarnide and terephthalamide (Edgar 1952)
melting point f8r different fiom those of either homopolymer.. Block and graft copolymers may exhibit two crystalline melting points, one fbr each type of chain segment
GENERAL REFERENCES
Alfrey 1967; Flory 1969; Sweeney 1969; Elias 1977, Chapter 10; Wunderlich 1980
B. THE GLASS TRANSITION

Relation Between T, and 1 ,
With ftw exceptions, polymer structure affects the glass transition T, and the crystalline melting point T,,, similarly This is not unexpected, since similar conf siderations o cohesive energy and molecular packing apply to the amorphous and crystalline or pasacrystalline regions, respectively, in accounting for the temperature levels at which the transitions occur. In consequence, T, and T, ase rather simply related fbr many polymers (Beaman 1952; Boyer 1954, 1963): Depending on symmetry, T, K is approximately one-half to two-thirds T, K (Fig. 12.-4).There are, however, some exceptions to this general rule. The most important appear (Pearce 1969) to be associated with crystallizability or hydrogen bonding in the amorphous
*
POLYMER STRUCTURE A N D PtiYSICAL PROPERTIES
! j
0 Poly(viny1 cyclohexane)
600
Polytetrafluoroethylene
Poly(3.-methyl-1.-butene) O
Cellulose triacetate
500
1
-Linear
Polycarbonate Poly(4-methyl-1-pentene). Poly(methy1 ihpropenyl ketone) a Poly(methy1 isopropenyl

O Polystyrene
ketone)
Poly(allyl cyclohexane)
0 Poly(allyl benzene)
Polypropylene e
0 Poly(4-cyclohexyl-1-butene)
\
Poly(butene-1) 0
lsoiactic poly(methy1 methacrylate)
'PoIy(4-phenyl-1-butene) tmns-l,4-palybutadiene
300
Poly(pentene-4) Poly(ethylene oxide) trans-Polyisoprene Poly(propylene oxide) 1,s-Pohisoprene
cis-Polybutadiene
Poly(dimethy1 siloxane)
200 200
I
300
I 400 500
T, K ,
FIG. 12-4.
Relation between T, and I , for various polymers (Boyer 1963)
regions of polymers.. Within series of' polymers where crystallizability is changed by control of' tacticity (Reding 1962) or structure (Zimmerman 1968), ease of crystallization is associated with a low ratio of I, to T,, and difficulty with a high ratio In polyamides, where T varies with ester group spacing as indicated in , Section A , T, is nearly constant at a level postulated (Woodward 1960, Komoto 1967) to be set by the energy required to break hydrogen bonds in the amorphous regions Copolymers also have ratios of' T to T, difr'ering from those in , Fig. 12-4, as described below
THE GL.ASS TRANSITION

Effects of Molecular Weight and Diluents
339
In the polymer range, T, is more dependent on molecular weight than is T, (Section A), the relation having the form
derived from temperature-volume considerations (Fox 1955), where Tgm the glassis transition temperature at infinite molecular weight, and k is about 2 x lo5 for polystyrene (Fox 1950) and poly(methy1 methacrylate) (Beevers 1960) and 3 5 x lo5 for atactic poly(a-methyl styrene) (Cowie 1968). From the considerations of Chapter 10, it is probable that chain ends, which lead to the terms H, and So for low-molecular-weight substances, are usually associated with defects in the crystalline regions of polymers Hence they are unlikely to make a significant contribution to T,, which is the melting point of the most perfect crystalline regions In amorphous polymers, however, the effect of chain ends on free volume (Chapter 1 ID)should retain importance in proportion to the concentration of ends; thus it is not surprising that T, is found to vary with Mnin the polymer range While the effect of molecular-weight distribution on Tg is accounted for by the appearance of Mnin Eq 12-3, the effect of low-molecular-weight diluents is worth noting The primary example of this effect is plasticization, widely used to improve the flexibility of certain polymers and allow them to remain flexible well below T, of the unplasticized resin (Chapters 14 and 17) Based on the weight of added constituent, plasticization is considerably more effective than copolymerization in lowering T,
Effect of Chemical Structure
The effects of' the nature of' the chain repeat units on T, are closely related to chain stiffness, and symmetry. Probably the most important intermolecular fo~ces, factor among these is hindrance to free rotation along the polymer chain resulting from the presence of stiff' bonds or bulky side groups: Compare polybutadiene, T, = - 85C; styrene-.butadiene copolymer (25/75), - 55C; polysty~ene, 100C; poly(cx-methyl styrene), f 150C; polyacenaphthalene, +285"C. The effect of' intermolecular forces is quantified by the cohesive energy density or solubility parameter (Chapter 7A): Compare polypropylene, 6 = 16 0, T, = -20C; polyacrylonitrile, 6 = 31.5, T, = 90C. The effect of symmetry of the repeat unit is illustrated in Fig.. 12-4: Compare poly(viny1 chloride), T, = 85C; poly(viny1idene chloride), T, = - 17C.
Effect of Chain Topology Copolymerization. The glass-transition temperatures of random copolymers usually fall between those of the corresponding homopolymers, T, for the copolymer
340
1
I
often being a weighted average given by
i
I
where T,, and T,, refer to the homopolymers, w, and w2 ase the weight fractions of monomers 1 and 2 in the copolymer, and a , and a, depend on monomer type (Wood 19.58) There are numerous deviations, both positive and negative, from this lineas relationship, however.. The contsast between this behavior and the common depression of' Tm by co-. polymerization is not surprising, since the changes at T, do not require fitting a structure into a crystal lattice, and in consequence structural irregularity does not affect T, as it does T,. While the above considerations for random copolymers lead to low ratios of' Tm to T,, it is sometimes fbund that block and graft copolymers may have long enough homogeneous chain segments to exhibit the properties of' both homopolymers, rather than intermediate values. Thus a block copolymer in which one homopolymer has high, and the other low, softening and brittle temperatures may exhibit both a high softening point and a low brittleness temperature. Hence it may have values of T,IT, higher than those found for other polymer types
Branching and Crosslinking. The effects of chain branching and crosslinking on T, can be explained in terms of free volume The higher concentration of chain ends in a branched polymer increases the free volume and lowers T,, whereas crosslinking lowers free volume and raises T, Roughly, the latter change may be accounted for in terms of the average molecula weight of the segment between crosslinks by an equation like Eq 12-3 More complete treatments are cited in Nielsen (1969)
GENERAL REFERENCES
Alfrey 1967; Elias 1977, Chapter 10
C.
PROPERTIES INVOLVING LARGE DEFORMATIONS
Melt Properties Melt Viscosity. As discussed in Chapter 1 IA, the viscosity of a polymer melt is a strong function of weight-average molecular weight Melt viscosity is also influenced by chain branching: In polyethylene and in silicone polymers melt viscosity decreases with increasing degree of long-chain branching at constant weight-average molecular weight, whereas in poly(viny1 acetate) melt viscosity increases under the same circumstances The reason for this difference appears to lie in branch length: In experiments with poly(viny1 acetate) branched by graft polymerization (Long
ROPERT TIES INVOLVING LARGE DEFORMATIONS
341
1964), melt viscosity decreased at constant molecular weight when the added branches were shorter than the critical chain length Z, at which the melt viscosity power law changes (Chapter 11A) Only when branches longer than Z, were added did the v~scosityincrease Crosslinking has a pronounced effect on melt viscosity in that the latter becomes essentially infinite along with M, at the onset of gelation However, small, tightly crosslinked network particles may behave like rigid spheres and have little effect on melt viscosity The addition of low-molecular-weight species, as in plasticization, reduces melt viscosity by lowering average molecular weight Bulky side groups may have a 5imilar effect For many polymers an upper limit to M, is set by fabrication requirements on the melt viscosity At the same time, a lower limit on M,, be set by requirements may involving tensile strength, brittleness, or other mechanical properties In such cases, which include polypropylene and probably other polyolefins, the best balance of properties is achieved when the distribution of molecular weights is made as nanow as possible However, the anticipated gain in ease of fabrication on decreasing M, at constant M,, may not be fully realized, since fabricability depends upon melt viscosity at hlgh shear stress In contrast to low shear (Newtonian) viscosity, which depends upon M,, high-shear viscosity depends on a molecular-weight average between M , and M,, (Rudd 1960)
Other Melt Properties. For polymers with very high melt viscosity such as polytetrafluoroethylene, the tensile strength of the melt becomes a property of importance Like melt viscosity, it increases with increasing molecular weight The viscoelastic or elastic properties of polymer melts decrease in magnitude with increasing molecular weight and with increasing chain branching. Tensile Strength and Related Properties
Many polymer properties, including tensile stsength, can be described by an equation of' the type Property = a
-. 7
M n
This is the type of' relation predicted fbr properties depending on the number of ends of' polymer chains. Many such properties, including density and refractive index, attain constant values at molecula weights well below the polymer range. Tensile strength, however, varies significantly with molecular weight in the range of' interest fbr polymers, although the variation may in fBct be with an average between M n and M , (McCormick 1959) It has been shown (Flory 1945) that
342
dependence on M, implies that the tensile strength with tensile strengths ( B ) , is the weight average
-
of a mixture of components
TS = ~ ' w l ( ~ ) ,
I
(12-6)
If a polymer exhibits a yield point and then undergoes extensive elongation before tensile failure, its ultimate tensile strength increases with increasing molecular weight Typical data for branched polyethylene are shown in Fig 12-5 Important structure variables in this polymer were found to be degree of crystallinity and level of molecular weight; in Fig 12-5 crystallinity is replaced by the equivalent property of density, and molecular weight by the logarithm of melt viscosity Although Fig. 12-5 shows that tensile strength is independent of degree of crystallinity for branched polyethylene, this independence does not c a ~ over to y linear polyethylenes, with densities in the range 0 94-0 96: the latter polymers have higher tensile strengths than branched polyethylenes of the same molecular weight (see also Chapter 11E) The difference is attributed to changes in the crystal morphology of the high density polymers during tensile elongation (Chapter 10E) Morphology is indeed important in determining the mechanical properties of crystalline polymers Both tensile strength and the mechanism of failure a e influenced by such factors as spherulite size and structure and the nature of interlamellar ties Polymers with smaller, finer-textured spherulites tend to fail at high elongations after drawing, while those with large, coarse spherulites often fail by brittle fracture between spherulites at low elongations (Collier 1969)
Toughness in Rubber-Modified Classy Plastics

One of the major. developments leading to plastics with outstanding toughness has been the pr.oduction of rubber-modified glassy plastics such as the ABS resins
Ultimate strength, psi
I
I
2000 psi 1000 psi 1500 psi 19:s 1100
0.
0 95
19451975
8;0
1220
I I I 3 4 . 5 6 7 Log melt viscos~ty,polses
FIG. 12-5. Dependence of ultimate tensile strength of polyethylene on molecular structure variables (Sperati 1953)
PROPERTIES INVOLVING S A L DEFORMATIONS M L
34 3
(Chapter 14), made preferably by polymerizing a continuous glassy matrix in the presence of small rubber particles. It is known that for optimum toughness the twophase structure is essential, only a small amount (5--15%) of rubber is needed, the optimum size of the rubber particles is 1-10 pm, and the rubber-matrix interface should be well grafted The mechanism (Bucknall 1967) by which toughness is developed in these materials, in contrast to the brittle failure characteristic of the unmodified glassy polymer, is intimately related to the formation of crazes, regions of low-density material fbrmed as a precursor to cracking in glassy polymers (Sternstein 1977). At low strains, the stress in the sample is largely borne by the matrix, and is concentrated at the rubber particles. As straining continues crazes are initiated here and grow with the absorption of energy as the matrix deforms. Ultimately the applied stress is distributed between the rubber and the crazed matrix. The rubber, now under tension, strengthens the crazed matter and fracture is thus delayed in favor of craze initiation elsewhere..
GENERAL REFERENCES
Alfrey 1967; Kargin 1968; Ward 1971; Boenig 1973, Chapter 7; Nielsen 1974; Elias 1977, Chapter 11
D. PROPERTIES INVOLVING SMALL DEFORMATIONS
This section is concerned with structural determinants of a variety of properties involving small local deformations of polymers in contrast to the gross deformations discussed in Section C Among these are such mechanical p~opertiesas stiffness, yield stress, elongation, and impact strength Related to these properties are hardness, abrasion resistance, and flexural fatigue life, among others. A second group of properties discussed includes solubility and related phenomena, such as swelling, cloud points, sorption of liquids, permeability to gases, and compatibility of plasticizers The final class under discussion includes the effects of electromagnetic radiation, in such optical properties as refractive index and transparency and such electrical properties as dielectric constant, dielectric loss, and dielectric strength.
Effect of Crystallinity Mechanical Properties. The prope~tiesof crystalline polymers are emphasized in this section for two reasons: It is these polymers that are most widely utilized because of their mechanical properties, and structural features related to crystallinity may have profbund effects on these properties The degree of crystallinity alone is effective in determining the stiffness and yield point for most crystalline plastics.. As indicated for branched polyethylene in Fig. 12-6, these properties are independent of molecular weight. As a result, they
POLYMER STRUCTURE A N D PHYSICAL. PROPERTIES

Stiffness x psi
0 89
0
FIG. 12-6.
3 4 5 6 L.og melt viscosity, poises
Dependence of stiffness of branched polyethylene on moleculdr structure var~ables (Sperat~ 1953) (See d~scuss~on Fig 12-5 ) for
can be expressed as single-valued functions of' degree of crystallinity. As crystallinity decreases, both stiffness and yield stress decrease. As a result of'the latter change, the chance of brittle failure is seduced
Solubility and Related Properties. As pointed out in Chapter 7, crystallinity decreases the solubility of' polymers markedly, since the process of' solution involves overcoming the heat and entropy factors associated with crystallization as well as those of the intermolecular interactions in the amorphous regions. Properties related to solubility, such as the cloud point of' dilute solutions, are often functions of' crystallinity relatively independent of' molecular weight The solubility of liquids and gases in polymers is also strongly dependent on crystallinity, since solubility is usually confined to the amorphous regions Permeability, the product of solubility and diffusivity, behaves similarly.. Plasticization is closely related to solubility, and the selection of' an efficient and compatible plasticizer (see Chapter 17) involves considerations similar to those fbr the selection of solvents. Plasticization usually results in loss of crystallinity; however, if c~ystallinityis well developed it may not be possible to find a plasticizer sufficiently compatible with (soluble in) the polymer to have a significant effect on its properties Electrical and Optical Properties. The primary effect of crystallinity on the electrical and optical properties is associated with the changes in dielectric constant
PROPERTIES I N V O l V l h C SMALL DEFORMATlONS
34 5
and refractive index zising from the difference in density between the crystalline and amorphous region5 In the case of visible light, this difference leads to scattering, may be large if the regions responsible (crystallites or lamellae, and spher"lites) are significant in size compared to the wavelength of the light Thus crystalline plastics usually appear translucent or opaque except in thin films, their transparency increasing with decreasing sphemlite size.
Effect of Molecular Weight solubility. Where the presence of' a crystalline phase is not involved, solubility and related phenomena are inverse functions of molecular weight. This fact is reflected in the equations for the thermodynamic properties of polymer solutions, and fbrms the basis of' fractionation methods, as discussed in Chapter 7 , ~lectrical and Optical Properties. The interactions of' electromagnetic radiation with polymers involve, at the most, the cooperative movement of' small groups of' atoms. The molecular. weight dependence of these properties, obeying relations of the type of Eq.. 12-.I, vanishes at moleculw weights far. below the polymer range Except as molecular weight influences some more direct structural determinant of' these properties, they are independent of' this variable Combined Effects of C,rystallinity and Molecular Weight Mechanical Properties. A number of mechanical properties, including hardness, flexural fatigue resistance or flex life, softening temperature, elongation at tensile break (where plastic flow occurs), and sometimes impact strength, are influenced by both degree of crystallinity and molecular weight Typical examples are the softening temperature of branched polyethylene as measured in the Vicat test (Chap" ter 9F), which increases with increasing molecular weight and increasing crystallinity (Fig 12-7), and the flex life of polytetrafluoroethylene, which increases with increasing molecular weight and decreasing crystallinity (Fig 12-8; "standard specific gravity" is a measure df molecular weight, as discussed in Chapter 14) Flex life is sometimes associated with brittleness, samples having low resistance to flexing being brittle The chance of brittle failure is decreased by raising molecular weight, which increases brittle strength, and by reducing crystallinity as indicated previously For amorphous polymers, impact strength is found to depend on the weight-average molecular weight As the degree of crystallinity decreases with temperature during the approach to T,,, (Chapter IOD), stiffness and yield stress decrease correspondingly These factors often set limits on the temperature at which a plastic is useful for mechanical purposes X major determinant of the behavior of a polymer on impact is the relation between the yield stress and the tensile strength in brittle failure, which may be designated the brittle strength If the yield stress or strength is lower than the brittle Strength, plastic flow begins (at the yield point in a tensile experiment) and the
0.89 ~ 0
-ULILL~
1 2 3 4 5 6 7 8 9
Log viscosity at 190C, poises

FIG. 12-7.
Dependence of softening temperature, as measured by the Vicat test, of branched polyethylene on molecular structure parameters (Sperati 1953) (See discussion for Fig
12-5 )
polymer is tough. If' the brittle strength is lower, brittle failure takes place on impact..
Effect of Polar Groups Solubility and Related Properties. In general, the introduction of polar groups into polymers tends to decrease solubility, since strong polymer-polymer. bonds usually develop. The situation is complicated, however, by factors such as the arrangement and bulkiness of the groups, which in turn influence crystallinity
230
226
222
218
214
210
Standard specific grav~ty

FIG. 12-8. Lines of equal flex l ~ f e a clystallin~ty-molecular we~ght"map" for polytetrafluoroon ethylene (Thomas 1956)
PROPERTIES INVOLVING SMALL DEFORMATIONS
347
m e solubility behavior of cellulose and its derivatives provides examples of the of polarity and substitution on solubility Cellulose itself is insoluble because of strong hydrogen bonding and stiff chains which prevent hydrating molecules from penetrating its crystalline regions Substitution of less polar g~oups the for hydroxyls on cellulose leads first to solubility in alkalis, which only swell the parent polymer, and then to water solubility when 0 5--1 hydroxyl per glucose unit is replaced. Crystallinity has usually disappeared at this stage. At higher degrees of substitution, water solubility is replaced by alcohol solubility, and then by solubility in organic solvents whose type depends on the nature of the substituent group When all the substitutents are alike, complete substitution leads to crystallization and more limited solubility The permeability of polymers to gases and liquids decreases with increasing polarity, since a more polar polymer has a higher activation energy for diffusion Even methyl groups contribute sufficiently to high cohesion and high activation energy in rubbers to impart low gas permeability Thus butyl, nitrile, and chloroprene rubbers all have lower permeability than natural or butadiene-based rubbers
Electrical Properties. The electrical properties of polymers depend more on an unbalance or asymmetry of dipoles than on the presence of polar groups per se Thus both polyethylene and polytetrafluorethylene have low dielectric constant and dielectric loss (dissipation factor), but these quantities are much larger for polymers containing both hydrogen and fluorine or chlorine Similarly, the dielectric loss of poly(2,5-dichlorostyrene) is much less than that of poly(3,4-dichlorostyxene) because of cancellation of the dipoles of the para chlorines Effect of Copolymerization Mechanical Properties. The addition of' a comonomer to a crystalline polymer usually causes a marked loss in crystallinity, unless the second monomer crystallizes isomo~phouswith the first (see Section A ) . Crystallinity typically decreases very rapidly, accompanied by reductions in stiffness, hardness, and softening point, as relatively small amounts (10-20 mol.%) of' the second monomer are added. In many cases, a rigid, fiber-forming polymer is converted to a highly elastic, rubbery product by such minor modification The dependence of mechanical properties on copolymer composition in systems which do not crystallize results primarily from changes in intermoleculas forces as measured by cohesive energy. Higher cohesive energy results in higher stiffness and hardness and generally improved mechanical properties The additional var.iable introduced by block or graft copolymerization can be used to alter the properties of such copolymers by changing their method of preparation. A block or graft copolymer consisting of' long-chain segments of widely differing polarity can exist in solution with one or the other type of segment extended, the second relatively contracted, depending on solvent type. When isolated from solution, the copolymer has properties resembling those of' the homopolymer cor~espondingto the extended segments (Merrett 1957). Thus natural
348
rubber with chains of poly(methy1 methacrylate) grafted to it was hard and stiff with a nontacky surface when isolated from a solution where the rubber chains were collapsed and the poly(methy1 methacrylate) chains were extended When prepared under the opposite conditions, it was limp, flabby, and self-adherent like rubber A third form with intermediate properties was isolated from solvents in which both chain segments were relatively extended
Solubility. In essentially random copolymers of' monomers whose homopolymers are noncrystalline, a property such as solubility varies more or less regularly from that of one homopolymer to that of' the other as the relative proportions of' the components are varied. The solubility of copolymers of this type is frequently low in solvents for either homopolymer, but high in mixtures of these solvents The solubility of graft and block copolymers is often unusually high, especially if' the two components have widely different polarities: Block copolymers of' polystyrene with poly(viny1 alcohol) or poly(acry1ic acid), for example, are soluble in benzene, acetone, and water. In water the hydrophilic blocks are solubilized an extended, holding the tightly coiled hydrocarbon segments in solution much as a detergent solubilizes a hydrocarbon by micelle formation In benzene the opposite situation occurs, but in an intermediate solvent such as acetone both blocks are relatively extended, as indicated by higher solution viscosity in intermediate solvents.. These polymers act as efficient detergents and emulsifying and compatibilizing agents, but their virtually universal solubility makes it difficult to isolate or purify them
Effect of Plasticization, Reinforcement, or Crosslinking Plasticization. The addition of' a plasticizer usually reduces stiffness, hardness, and brittleness, and has a similar effect on other mechanical properties, since interchain fbrces are effectively reduced. These changes are accompanied by a reduction in T,, as noted in Section B . As indicated in Chapter 17, plasticization is usually restricted to amorphous polymers or polymers with a low degree of' crystallinity because of' the limited compatibility of' plasticizers with highly crystalline polyme~s Reinforcement and Crosslinking. Whether carried out by chemical crosslinking in an unmodif~edamorphous polymer system or by the addition of a reinforcing filler such as carbon black in rubber (a process involving chemical bonding between polymer and filler), the addition of crosslinks leads to stiffer, stronger, tougher products, usually (in the case of rubbers) with enhanced tear and abrasion resistance as well However, extensive crosslinking in a crystalline polymer may cause loss of crystallinity, with attendant deterioration of the mechanical properties depending on this factor. When this occurs, the initial trend of properties may be toward either enhancement or deterioration, depending on the degree of crystallinity of the unmodified polymer and the method of formation and location (crystalline or amorphous regions) of the crosslinks
PROPERTY REQUIREMENTS AND POLYMER U T I L I Z A T I O N
349
GENERAL REFERENCES
Kargin 1968; Ward 1971; Boenig 1973, Chapter '7; Nielsen 1974; Elias 1977, Chapters 11, 14, and 15
E.
PROPERTY REQUIREMENTS A N D POLYMER UTILIZATION
The variables necessary to define the mechanical and physical properties of'polymers have now been discussed (Fig.. 12-9). The increase of' T, with molecular weight, leveling off as polymer molecular weights are reached, the related approximate behavior of' T,, and the continual increase of' viscosit,~ with molecular weight serve to define, in terms of the variables molecular weight and temperature, regions in which the properties of typical plastics, mbbers, viscous liquids, and so on, may be found. Combinations of properties unique to polymers are evidenced in each of' the major uses, including elastomers, fibers, and plastics, to which macromolecules are put. In this section the property requirements of these end uses ase described and related to polymer structure..
Elastomers
It is a matter of experience that all substances exhibiting a high degree of rubberlike elasticity contain long-chain structures The resto~ing force leading to elastic behavior results directly from the decrease in entropy associated with the distortion of a chain macromolecule from its most probable conformation Two additional property requirements are imposed by the condition that there be sufficient freedom of molecula~ motion to allow the distortions to take place rapidly: First, the polymer must at its use temperature be above T, and, second, it must be amorphous, at least in the undistorted state
Rubber Mobile liquid Viscous
Tm
Tough plastic
Tg
T0C
Crystalline solid
10 FIG. 12-9.
Partially crystalline plastic
I 1000 100.000 Molecular weight 10,000,000
App~oximaterelations among molecular weight, I,, T,,,, and polymer properties
350
POLYMER STRUC:TURE A N D PHYSICAL PROPERTIES
In contrast to the high local mobility of chain segments implied by these requirements, the gross mobility of chains in elastomers must be low.. The motions of chains past one another must be restricted in order that the material can regain its original shape when the stress is released. This restriction of' gross mobility is usually obtained by the introduction of a network of primary bond crosslinks in the material (It cannot be obtained through secondary bond forces; these must be kept low in order to gain local segment mobility .) The crosslinks must be relatively ftw and widely separated, however, so that stretching to large extensions can take place without rupture of primary bonds. The requirement of' low cohesive energy limits the family of' elastomers to polymers that are largely hydrocarbon (or fluorocarbon or silicone) in nature, with polar groups distributed at random and in not too great number, so that crystallinity is absent and T, is sufficiently low, ideally in the range - 50 - -80C. The requirement of' sites for crosslinking often leads to the choice of a diene as a monomer or comonomer for an elastomer In contrast to its equilibrium properties, a stretched elastomer should have the high tensile strength and modulus usually associated with crystalline plastics. Thds rubbers in which crystallinity can develop on stretching, such as natural rubber and its stereoregular synthetic counterparts, usually have more desirable properties than those with less regular structures. However, a reinforcing filler can sometimes impart to a rubber properties similar to those obtained on the development of crystallinity. Thus reinforced styrene-butadiene rubber has p~pperties nearly equivalent to those of'reinfbrced natural rubber, whereas its properties when crosslinked but not reinfbrced are much poorer..
'
Fibers
In contrast to elastomers, the requirements of high tensile strength and modulus characteristic of fibers are almost always obtained by utilizing the combination of molecular symmetry and high cohesive energy associated with a high degree of crystallinity. Usually the fiber is oriented to provide optimum properties in the direction of the fiber axis. The end use requirements of fibers, particularly those involving textiles, lead to restrictions on several properties The crystalline melting point T, must be above a certain minimum, say 200"C, if the resulting fabric is to be subjected to ironing On the other hand, spinning the polymer into a fiber requires either that T,, be below, say, 300C and well below the temperature of decomposition of the polymer, 01 that the polymer be soluble in a solvent from which it can be spun Other requirements limit solubility, for example, in solvents useful for dry cleaning The requirement of orientation in the fiber usually implies that T, be not too high (since orientation by cold drawing and ironing are typically carlied out at or above this temperature) or too low (since orientation and related characteristics, such as crease retention after ironing, must be maintained at room temperature) Thus the selection of a polymer for use as a fiber involves a number of compromises, usually met by choosing a linear polymer wlth high symmetry and high
PROPERTY REQUIREMENT'S AND POLYMER UTILIZATION
351
i n t e r m o l e c ~forces resulting from the presence of' polar groups, high enough in l~ molecular weight so that tensile strength and related properties are fully developed. Branching in the polymer chain is in general detrimental to fiber properties because branch points disrupt the crystalline lattice, lower the crystalline melting point, and decrease stiffness Crosslinking, on the other hand, offers the possibility of obtaining strong interchain bonding. If crosslinks are fbrmed after the polymer is spun into fiber,,and ase relatively few in number, improvement in fiber properties may result.. Thus poly(viny1 alcohol), polyurethanes, and protein fibers may be crosslinked with formaldehyde to give higher melting point, lower solubility and moisture regain, and improved hand, while wool, a natural protein fiber, is crosslinked with cystine links.
General-Purpose and Specialty Plastics
The wide range of end uses of plastics requires a variety of property combinations; correspondingly, a wider variety of structures is important In general, the properties of plastics are intermediate between those of fibers and elastomers, with much overlapping on either end Thus typical plastics may have cohesive energies higher than those of elastomers but lower than those of fibers However, a polymer useful as a fiber when oriented in this form may also be useful as a plastic, where orientation is not readily achieved in the massive pieces used; an example is nylon
Optical Applications. The requirement of good optical properties, especially in massive pieces, imposes severe limitations on the structure of polymers In general, c~ystallinitymust be absent, and most amorphous polymers exhibit softness and brittleness that exclude them from many applications In thin films, structural requirements are not as severe, since crystallinity can often be tolerated if spherulite growth is inhibited and the material can be processed to give sufficiently smooth surfaces Thus the normally incompatible requirements of clarity and toughness can both be met Electrical Applications. To achieve low dielectric loss over a wide frequency range, the structure of a polymer must be selected on the basis of low polarity. All other requirements ase of considerably less importance. It fbllows that polyethylene and polytetrafluoroethylene are the best materials for low-loss applications, particularly at high frequencies. At low frequencies, however, other plastics, such as poly(viny1 chloride), are useful.. Mechanical Applications. Perhaps the most important property requirement for the use of'polymers in mechanical applications is toughness This property is usually achieved by selection of a polymer with a moderate, but not too high, degree of' crystallinity (see the discussion ofthe mechanical properties of crystalline polymers, Chapter 11E). Often, a delicate balance of structural features is needed to achieve the desired combination of properties. Composite structures (as in glass-reinforced or rubber-modified plastics) often provide unique psoperty combinations.
352
POLYMER STRUCT'URE AND PHYSICAL PROPERTIES
Economic Aspects. Though the preceding discussion could be extended considerably, it should be pointed out that a major factor in the selection of the appropriate plastic for a given use is economic. In many (some observers would say far too many) cases, a mass market is achieved by sacrificing properties fbr price. The plastics manufacturer must, therefore, consider what structures give the optimum combination of melt and solid-state properties. He would like, for example, to lower molecular weight to achieve low melt viscosity and more rapid fabrication; but he must maintain levels of' molecular weight consistent with the development of' good mechanical properties.. The selection of raw materials is also important: The rise in prominence of' olefin polymers was closely related to the low cost of' the monomers. The cost of' polymerization itself' appears less important in most cases, but it is clear that monomers that are expensive to polymerize have less opportunity to yield large-volume plastics A ''?rofile" ofthe important properties of' common general-purpose and specialty plastics is given in Fig. 12-10.. Engineering Plastics
!
In contrast to general-purpose and specialty plastics, the term engineering polymers or plastics is applied to those materials that command a premium price, usually
Price Processabilit) Tensile strength Stiffness Impact strength Hardness Useful temperature range Resistance to chemicals Resistance to weather Resistance to water Flammability
FIG. 12-10.
A "profile" of the properties of some general-purpose and specialty plastics (Billmeyer outstanding in the property indicated, among the best performers available; 1968): O,acceptable performance in this property, still suitable in most cases; - , not recommended if' this property is important to the intended use
+,
PROPERTY REQUIREMENTS AND POLYMER UTILIZATION
35 3
associated with relatively low production volume, because of their outstanding balance of properties which allows them to compete successfully with other materials (metals, ceramics) in engineering applications They are strong, stiff, tough, abrasion-resistant materials capable of withstanding wide ranges of temperatures, and resistant to attack by weather, chemicals, and other hostile conditions The value they contribute to the end product justifies their higher price per pound. Figure 1211 compares engineering with general-purpose plastics for price and volume The outstanding properties of engineering plastics come primarily from their crystalline nature and strong intermolecular forces Most of them have quite high melting points, ensuring retention of good physical properties to high temperatwes, and good toughness over wide temperature ranges The necessary high melting points can be obtained in several ways, as discussed elsewhere. These include combining high degree of crystallinity with stiff polymer chains, or finding structural features substituting for crystallinity in imparting rigidlty to the total material structure This can be done in two ways: crosslinking and utilizing composite structures with an extremely rigid material, such as glass fibers So far, these alternatives have been utilized one at a time in almost all cases That is, glass fibers are usually used to reinforce glassy plastics such as polyester-styrene copolymers, crosslinked polymers are not crystallizable, and so on We are now beginning to see instances where two or more of these property-enhancing approaches are combined, with outstanding results-for example, the development of ladder polymers combining the features of crosslinking with crystallizability, and the use of high-performance inorganic fibers (such as boron) to reinforce crystalline plastics Clearly, much effort is being expended toward progress in these directions. A "profile" of' the important properties of engineering plastics is given in Fig 12-12
- -
-.
. n -
10
100 Sales (million ib)
1000
3000
FIG. 12-11. Relationship berween prices and production volumes of major engineering and generalpurpobe plastics (Gutoff 1969)
354
Price Processability Tensile strength Stiffness Impact strength Hardness Useful temperature range Resistance to chemicals Resistance to weather Resistance to water Flammability
lo\-/o i
i - i + . i + r m I
FIG. 12-12. "Profile" of the properties of some engineering plastlcs (Billmeyer 1968) (same key a$ forFig 12-10)
'The Role of Plastics Among Materials
In the last two decades, the role of plastics in the overall materials market has changed drastically. Until the recession of 1980 put an end to the growth in the production of all materials, plastics and particularly engineering plastics had grown at a significantly higher annual rate than either metals or ceramics.. By 1970 the annual production of plastics exeeded that of aluminum or copper on a weight basis, and it had been predicted (Gutoff 1969) that on a volume basis (consistent with the higher strength-to-weightratios of'the engineering plastics) production would exceed that of steel by the early 1980s..Until the drastic increase in the price of' petroleum in the mid-1970s caused prices of' plastics to rise disproportionally in comparison had for many years been to those of' other materials, the trend in plastics p~ices downward in contrast to the trends fbr most metals. It is anticipated that when the world economy regains a semblance of normalcy, the expansion in the use of plastics seen for so many years in the past will continue
GENERAL REFERENCES
Riley 1968; Bakker 1980; McQuiston 1980; National Research Council 1981; Platzer 1981; Thommes 1981; Kossoff' 1982
1. Compare the crystalline melting points of the following classes of polymers at comparable numbers of chain atoms in the repeat unit, and explain in terms
BIBLIOGRAPHY
355
of thermodynamic properties and their molecular origins: polyethylene, polyamides, polyesters, polyureas, polyurethanes
2.
Predict and explain the difference in crystalline melting point between the and polymers with the repeat units --(CH,)$JHCO(CH,),CONH-. -(CH2)6N(CH3)CO(CH2)4CON(CH,)-.
3. Explain why 6-nylon is soluble in some solvents at room temperature, whereas linear polyethylene is not..
4. Predict and explain which of'the fbllowing have similar, and which different, properties at room temperature: polystyrene, poly(methy1 methacrylate), poly(viny1 acetate), poly(viny1 alcohol).
5. What sort of comonomer could be used to raise the glass-transition temperature of poly(viny1 acetate)?
6 . Why does polypropylene undergo a greater change in physical properties near T, than does linear polyethylene?
7. Predict and explain the effect, if any, of varying molecular weight and degree
of short-chain branching on each of the following properties of polyethylene: ultimate tensile strength, stiffness, T,,,, sorption of organic liquids
8 Explain the effect of molecular weight on the specific gravity of poly. tetrafluoroethylene
BIBLIOGRAPHY
Alfrey 1967 Turner Alfrey, Jr , and Edward F Gumee, Organ~c Polymers, Prentice-Hall, Englewood Cliffs, New Jersey, 1967 Bakker 1980 Marilyn Bakker, "Engineenng Plastics," pp 118-137 in Martin Grayson, ed , KirkOthmer Encyclopedta of Chemical Technology, 3rd ed , Vol 9, Wiiey-Intersclence, New Yolk, 1980 Bannerman 1956 D G Bannerman and E E Magat, "Polyamides and Polyesters," Chapter 7 in C E Schildknecht, ed , Polymer Procesrer, Wiley-Interscience, New York, 1956 Bawn 1960 C E H Bawn, "High Polymers and Molecular Architecture," Chem Ind 1960, 388391 (1960) Beaman 1952 Ralph G Beaman, "Relation between (Apparent) Second.Order Transition Temperature and Melting Point," J Polym Sct 9, 470-472 (1952) Beevers 1960 R B Beevers and E F T White, "Physical P~oper.tles Vinyl Polymers Part I of Dependence of the Glass-TransitionTernperatureof Polymethylmethacrylate onMolecular Weight," Trans Faraday Soc 56, 744-752 (1960) Brllmeyer 1968 Fred W Billmeyer, h , and RenCe Ford, "The Anatomy of Plastics," Sct Techno1 73, 22-37, 81-82 (1968) Boenig 1923. Herman V Boenig, Strucrr4r.e and Propertres of'PoIymers, John Wiley & Sons, New York, 19'73 Boyer 1954 R F Boyer, "Relationship of First-to-Second-Order Transition Temperatures for Crystalline High Polymers," I Appi Phys 25, 825--829 (1954)
356
Boyer 1963. Raymond F. Boyer, "The Relation of Transition Temperatures to Chemical Structure in High Polymers," Rubber Chem Techno1 36, 1303-1421 (1963) Bucknall 1967 C B Bucknall, "The Relationship between the Structure and Mechanical Propertles of Rubber-Modified Thermoplastics, Part One," Br Plastics 40 ( l l ) , 118-122 (1967); "Part Two," 40 (12), 84-86 (1967) Bunn 1955 C W Bunn, "The Meltlng Polnts of Chain Polymers," J Polym Scr 16, 323-343 (1955) Campbell 1959 Tod W Campbell and A C Haven, JI , "The Relationship Between Structure and Polyhydrocarbons," J Appl Polpm Sci 1, 73-83 (1959) Properties of Crystalline, H~gh.Melting Coflrer 1969 John R Collier, "Polymer Crystallization History and Resultant Properties," Rubber Chem Techno1 42, 769-779 (1969) Cowle 1968 J M G Cowie and P M Toporowskl, "The Dependence of Glass Temperature on Molecular Weight for Poly(a-methyl styrene)," Eur Polym 1 4 , 621-625 (1968) Dole 1959 M Dole and B Wunderlich, "Melting Points and Heats of Fusion of Polymers and Copolymers," Makromol Chem 34, 29-49 (1959) Edgar 1952 Owen B Edgar and Rowland Hill, "Thep-Phenylene Linkage in Linear High Polymers: Some Structure-Property Relationships," J Polym Sci 8, 1-22 (1952) Elias 1977 Hans-Georg Elias, Macromolecules I Structure and Propertres (translated by John W Stafford), Plenum Press, New York, 1977 d FIory 1945 Paul I Flory, "Tensile Strength in Relation to Molecular Weights of High Polymers," J Am Chem Soc 67, 2048-2050 (1945) Flory 1949 Paul J Flory, "Thermodynamics of Crystallization in High Polymers IV A Theory of Crystalline States and Fusion in Polymers, Copolymers, and Their Mixtures with Diluents," J Chem Phys 17, 223-240 (1949) FIory 1969 Paul J Flory, Stat~stical Mechanics o Chain Molecules, Wiley-Interscience, New York, f 1969 Fox 1950 Thomas G FOX,JI , and Paul J Flory, "Second-Order Trans~tionTemperatures and Related Properties of Polystyrene I Influence of Molecular Weight," I Appl Phvs 21, 581-591 (1950) Fox 1955 T G Fox and S Loshaek, "Influence of Molecular We~ghtand Degree of Crossllnklng on the Speclfic Volume and Glass Temperature of Polymers," J Polym Scr 15, 371-390 (1955) Gutoff 1969 Reuben Gutoff, "Engineering Polymers: Risks and Rewards in the Decade Ahead," talk before the Commercial Chemical Development Association, New York, March 1969 Hrll 1948 R Hill and E E Walker, "Polymer Constitution and Fiber Properties," J Polym Scr 3, 609-630 (1948) Hobbs 1970 Stanley Y Hobbs and Fred W Billmeyer, Jr , "Heats and Entropies of Fusion of Linear Aliphatic Polyesters I1 Molecular Origins," J Polym Scr A-2 8, 1395-1409 (1970) Karg~n1968 V A Kargin and G L Slonlmsky, "Mechanical Propertles," pp 445-516 In Herman F Mark, Norman G Gaylord, and Norbert M Blkales, eds , Encyclopedra of Polymer Scrence and Technology, Vol 8, Wlley-Interscience, New York, 1968 Komoto 1967 H Komoto, "Physico-Chemical Studies of Polyamides I Polyamides having Long Methylene Chain Units," Rev Phys Chem Jpn 37, 105-1 11 (1967) Kossoff 1982 Richard M Kossoff, "The Engineering Plastics Business," Chemtech 12, 552-555 (1982) Long 1964 V C Long, G C Berry, and L M Hobbs, "Solution and Bulk Properties of Branched Poly(viny1 Acetates) IV Melt Viscosity," Polymer 5, 517-524 (1964) McCormrch 1959 Herbert W McCormlck, Frank M Brower and Leo Kln, "The Effect of Molecular We~ght Dlstnbutlon on the Physlcal Propert~es Polystyrene," J Polyrn Scr 39, 87- 100 (1959) of McQuiston 1980 Henry McQuiston, "Designing with Engineering Plastics," Plastrcs Eng 36 (6). 1825 (1980) Preset Molecular Configurations," J Polym Merrett 1957 F M Menett, "Graft Copolymers w ~ t h Sci 24, 467-477 (1957)
BIBLIOGRAPHY
357
National Research Councll 1981 National Research Council, ad hoc Panel on Polymer Science and Engineering, Polymer Scrence and Engineerrng Challenges, Needs, and Opportunities, National Academy Press, Washington, D C , 1981 Nielsen 1969 Lawrence E Nielsen, "Cross-Linking--Effect on Physrcal Properties of Polymers," I Macromol Scl Rev Macromol Chem C3, 69--103 (1969) Nielren 1974 Lawrence E Nielsen, Mechanlcal Propertres of Polymers and Composites. Marcel Dekker, New York, 1974 Pearce 1969 Eli M Pearce, "Polymer Synthesis: Philosophy and Approaches," Trans N Y Acad Scl 31, 629-636 (1969) Platzer 1981 Norbert Platzer, "Commodity and Engineering Plastics," Chemtech 11, 90--94 (1981) Reding 1962 F P Reding, E R Walter, and F J Welch, "Glass Transrtlon and Melting Point of Poly(Viny1 Chloride)," J Polym Sci 56, 225-231 (1962) Riley 1968 Malcolm W Rlley, "MaterlaIs, Selection," pp 419-440 In Herman F Mark, Norman G Gaylord, and Norbert M Bikales, eds , Encyclopedia of Polymer Sclence and Technology, Vol 8, Wrley-Interscrence, New York, 1968 Rudd 1960 John F Rudd, "The Effect of Molecular Weight Distribution on the Rheological Propertles of Polystyrene," J Polym Sci 44, 459-474 (1960) Sperati 1953 C A Speratt, W A Franta, and H W Starkweather, Jr , "The Molecular Structure of Polyethylene V The Effect of Chain Branching and Molecular Weight on Physical Properties," I Am Chem Soc 75, 6127-6133 (1953) Starhweather 1960 Howard W Starkweather, Jr , and Richard H Boyd, "The Entropy of Melting of Some Linear Polymers," J Phys Chem 64, 410-414 (1 960) Sternsteln 1977 S S Sternstein, "Mechanical Propertles of Glassy Polymers," in J M Schultz, ed , Treatlse on Materrals Sclence and Technology, Vol 10, Part B (Herbert Herman, serles ed ), Academic Press, New York, 1977 Sweeney 1969 W Sweeney and J Zimmerman, "Polyamrdes," pp 483-597 in Herman F Mark, Norman G Gaylord, and Norbert M Btkales, eds , Encyclopedia of Polymer Scrence and Technology, Vol 10, Wiley-Intersclence, New York, 1969 Thomas 1956 P E Thomas, J F Lontz, C A Sperati, and J L McPherson, "Effects of Fabrtcation on the Properties of Teflon Resin," Soc Plast Eng 1 12 (6), 89-96 (1956) Thommes 1981 G Thommes, "Making THINGS out of Chemicals," Chemtech 11, 285--287 (1981) Ward 1971 1 M Ward, Mechanrcal P~opertresof Solld Polymers, Wiley-Interscience, New York, 1971 Wood 1958 Lawxence A Wood, "Glass Transition Temperatures of Copolymers," J Polym Sci 28, 319-330 (1958) Woodward 1960 A E Woodward, T M Crissman, and J A Sauer, "Investigations of the Dynamic Mechanical Properties of Some Polyamides," I Polym Scr 44, 23- 34 (1960) Wunderlich 1958 Bemhad Wunderlich and Malcolm Dole, "Specific Heat of Synthetic High Polymers, IX Poly(ethy1ene Sebacate)," 1 Polym Scr 32, 125-130 (1958) Wunderlich 1980 Bernhad Wunderlich, Macromolecular Phystcs, Volume 3 CryrtalMeltrng, Academic Press, New York, 1980 Zimmermari 1968 Joseph Zimmerman, "Melt Blend of Polyamides," U S Patent 3,393,252 (to E I du Pont de Nemours and Co ), 16 July 1968
PART FIVE
PROPERTIES OF COMMERCIAL POLYMERS
CHAPTER THIRTEEN
HYDROCARBON PLASTICS AND ELASTOMERS
In this and the following chapters of Part 5, we discuss the synthesis, structure, properties, and applications of the major commercial polymers in the early 1980's Some material of historical interest is included, and there is implied a familiarity with the material covered in Parts 1-4 and 6 In Parts 5 and 6, three reference series are extremely valuable and are cited extensively (and hence abbreviated) in the bibliography: The Modern Plasfzcs Encyclopedia (MPE), cited with the date 1982 but published annually in October (readers should consult the latest volume for cor~esponding articles); the Encyclopedza o Polymer Sclence and Technology (EPST); and the Kzrk-Othmer Encyclof pedia o Chemzcal Technology (ECT) f
A.
POLYETHYLENE
Low-Density (Branched) Polyethylene

The first commercial ethylene polymer was branched polyethylene, commonly designated as low-density or high-pressure material to distinguish it from the essentially linear material described below After a period of relatively slow growth in the 19407s, the production of branched polyethylene expanded r apidly; this was the first plastic with annual production exceeding 1 billion Ib (in 1959) Reco~d production volume of 7.9 billion lblyr came in 1979, but by 1982 volume had dropped beaause of economic recession to around 6 3 billion Ib (excluding an additional billion Ib of LLDPE; see Section C), at prices around $0 30 per Ib, about triple those of 1969
361
362
H Y D R O C A R B O N PLASTICS AND ELASTOMERS
Polyethylene was first produced in the laboratories of' Imperial Chemical Industries, Ltd. (ICI), England, in a f'ortuitous experiment in which ethylene (and other chemicals that remained inert) was subjected to 1400 atm of' pressure at 170C Traces of' oxygen caused polymerization to take place. The phenomenon was first described by E . W. Fawcett in Staudinger 1936.
Polymerization. Ethylene (b .p. , .- 104C) is made f'rom the thermal (steam) and catalytic cracking of a variety of hydrocarbons, ranging from ethane derived from natural gas to fuel oil. About 25 billion Ib was produced in 1982, almost half' of' it used to produce ethylene polymers and copolymers. High po1ymer.s of' ethylene are made commercially at pressures between 1000 and 3000 atm (15,000-45,000 psi) or possibly higher, and temperatures as high as 250C. Traces of' oxygen initipte ;he polymerization of ethylene readily. Rapid exothermic reactions can occur, and violent explosions have taken place. Many other possible impurities in the monomer, such as hydrogen and acetylene, act as chain transfer agents and must be carefully removed if' high-molecular-weight products are to be obtained Besides oxygen, peroxides (benzoyl, diethyl), hydroperoxides, and azo compounds have been used as initiators Ethylene polymerization can be car~iedout with benzene or chlorobenzene as solvent. At the temperatures and pressures used, both polymer and monomer dissolve in these compounds so that the reactions are true solution polymerizations Water or other liquids may be added to dissipate the heat of reaction.. Batch polymerizations of ethylene cannot be car~ied rapidly with reproduout cibility and good control. Long reaction times, consistent with good control, are not economical. In addition, chain branching becomes excessive at high conversion and results in poor physical properties of' the product.. As a result, balanced, conEmulsion polymerization has had tinuous polymerization systems are prefer~ed. little success One continuous process utilizes tubular reactors, which may have diameters of' less than 1 in, and lengths up to 100 ft. The stainless steel tube may be filled with water, and ethylene containing initiator and possibly benzene is introduced Additional initiator and water or benzene can be injected at one or more points along the tube to keep the initiator concentration more nearly constant throughout the reactor. Ten or more percent of' the ethylene is polymerized at the far end of' the reactor.. Here the gas and liquid phases are taken off' continuously, the polymer is separated, and the ethylene is recycled after purification.. Another process utilizes bulk polymerization in a tower-type reacto r.. Ethylene containing trace amounts of oxygen is charged to the reactor at 1,500atm and 190C.. The reaction is kept essentially isothermal and carried to 10-15% conversion. The effluent f'rom the reactor passes to a separatory vessel in which unconverted ethylene is removed for recycling.. The molten polyethylene is chilled below its crystalline melting point and passed through the usual finishing steps Structure. Low-density polyethylene is a partially (50-60%) crystalline solid melting at about 115"C, with density,in the range 0 91-0 94 It is soluble in many
solvents at temperatures above 100C, but only a few solvent mixtures provide borderline solubility at or near room temperature. In 1940 infrared spectroscopy (Chapter 9B) revealed that low-density polyethylene contains branched chains. These branches are of' two distinct types. Branching due to intermolecular chain transfer, arising from reactions of' the type R I C H ~ C Hf.R z C H z C H A ~
propagating chain dead polymer molecule
~nrermolecular hydrogen transfer
RlCH2CH3 -I-. ~ C H C H ~ R ~ R
dead polymer molecule propagating chain
leads to branches which are, on the average, as long as the main polymer chain This sort of branching has an observable effect on the solution viscosity of the polymer (Chapter 8 E ) and can be detected by comparing the viscosity of a branched polyethylene with that of a linear polymer of the same molecular weight The second branching mechanism in polyethylene is postulated to produce sho~tchain branching by intramolecular chain transfer:
transient six-membered
R C H ~ C H ~ C H ~ C H Z C H ~- formation C- HZ ring
The transient ring mechanism suggests four carbon atoms as the most p~obable length of the short branches Infrared absorption studies and studies of degradation unde~bombardment with high-energy radiation of polyethylene and substituted polyethylenes of known branch structure suggest that both ethyl and butyl branches are present. A mechanism accounting f o ~ ethyl b~anchesassumes a transfer the reaction of the type
after addition of one monomer unit to the radical resulting from a short-chain branching step by the previous mechanism. confo~mations relatively probable dur;ng propagation, are Since the transient ~ i n g the short-chain branching mechanism accounts for the large majority of the chain ends observed in the infrxed A typical low-density polyethylene molecule may contain 50 s h o ~branches and less than one long branch on a number -aveIage basis. t The molecular-weight distribution of typical polyethylenes has been found experimentally and theoretically to be very much broadened by the long-chdin branch-
,364
HYDROCARBON PLASTICS A N D EL.AST'OMERS
ing mechanism. Weight- to number-average molecular weight ratios M,~I@,, of 2050 are considered typical. The distribution of long branches among the molecules is also very broad: Even for highly branched polymer many molecules contain no long branches, while most of' the branches are concentrated on a few very large molecules. Infrared spectroscopy (Chapter 9B) provides much information about the chemical and physical structure of polyethylene. Structural features associated with the crystallinity of polyethylene are described in Chapter 10
Properties. The physical properties of' low-density polyethylene are functions of three independent structural variables: molecular weight, molecular-weight distribution or long-chain branching, and short-chain branching. Short-chain branching has a predominant effect on the degree of' crystallinity and therefbre on the 'density of' polyethylene. (Actually these properties are influenced by total chain branching, but the number of' long-chain branch points per molecule in typical polyethylenes is so much less than the number of' short-chain branch points that the former can be neglected ) Therefbre, as discussed in chapter 12, properties dependent on crystallinity, such as stiffness, tear strength, hardness, chemical resistance, softening temperature, and yield point, increase with increasing density or decreasing amount of short-chain branching in the polymer, whereas permeability to liquids and gases, toughness, and flex life decrease under the same conditions The effect of' molecular weight is largely evidenced (Chapter 12C) in properties of the melt and properties involving large deformations of the solid. As molecular weight increases, so do tensile strength, tear strength, low-temperature toughness, softening temperature, impact strength, and resistance to environmental stress cracking, while melt fluidity, melt "drawability," and coefficient of' friction (film) decrease. These properties are commonly compared on the basis of changes in melt index (Chapter 1 l A ) , which varies inversely with molecular weight The effect of' long-chain branching on the properties of polyethylene is often evaluated in terms of the breadth ofthe molecular-weight distribution M,~M,. With other structural parameters held constant, a decrease in M J M , causes a decrease in ease of' processing but an increase in tensile strength, toughness and impact strength, softening temperature, and resistance to environmental stress cracking.. The intenelation of density, melt index, and M,,IM, in producing physical properties desirable for specific end uses has been described (McGrew 1958). For film uses polymers producing tough and flexible films are needed. Injection-molding applications require polymers characterized by ~igidityand good flow; pipe, by strength; and wire insulation, by good processing characteristics and resistance to stress cracking. The mechanical properties of low-density polyethylene are between those of rigid materials like polystyrene and limp plasticized polymers like the vinyls. Polyethylene has good toughness and pliability over a wide temperature range. Its density falls off f'airly rapidly above room temperature, and the resulting large dimensional changes cause difficulty in some fabrication methods. The relatively low crystalline
'
POLYETHYLENE
365
melting point (about 115C for typical materials) limits the temperature range of good mechanical properties The electrical properties of polyethylene are outstandingly good, probably ranklng next to those of polytetrafluoroethylene for high-frequency uses In thick sections polyethylene is translucent because of its crystallinity, but high transparency is obtained in thin films. Polyethylene is very inert chemically It does not dissolve in any solvent at room temperature, but is slightly swelled by liquids such as benzene and carbon tetrachloride which are solvents at higher temperatures It has good resistance to acids and alkalis At 100C it is unaffected in 24 hr by sulfuric or hydrochloric acid but charred by concentrated nitric acid It is often used in containers for acids, including hydrofluoric Polyethylene ages on exposure to light and oxygen, with loss of strength, elongation, and tear resistance The probable point of attack is the tertiary hydrogens on the chain at branch points Stabilizers retard the deterioration, but few are compatible enough with the polymer to do much good The weathering of carbonblack-pigmented material is quite good. The polymer also undergoes some crosslinking when heated or worked at elevated temperatures Few plasticizers or other additives are compatible with polyethylene in amounts larger than 1% or so.
Applications. Almost two-thirds of the low- and medium-density branched polyethylene produced has gone into film and sheeting uses for many years Few competitive film materials have polyethylene's desirable combination of low density, flexibility without a plasticizer, resilience, high tear strength, moisture and chemical resistance, and little tendency for nicks or cuts to propagate Blown film, produced by extruding a tube of polymer and expanding it by means of internal pressure of inert gas, thus drawing the polymer, is most widely produced The tube may be slit to produce flat film or left as a seamless tube Film thicknesses are usually 0 001-0 005 in. Over three-fourths of the polyethylene film produced goes into packaging applications, including bags and pouches and wrappings for produce, textile products, merchandise, frozen and perishable foods, and many other products Other film uses include drapes and tablecloths, and extensive application in agriculture (greenhouses, ground cover, tank, pond, and canal liners, etc ) and construction (moisture barriers and utility covering material) Extrusion coating for packaging materials is the second-largest market for lowdensity polyethylene, accounting for almost 10% of the use in the United States The constructions, often laminates of foil, paper, and polyethylene, are used in milk-type cartons for a wide variety of foods and drinks Polyethylene has filled a long-standing need for a material that would effectively insulate electrical cables without introducing electrical losses at high frequencies. The nonpolar nature of the polymer makes it ideal for this purpose Television, radar, and multicircuit long distance telephone might well have been impossible without such insulating materials Weathering introduces polar impurities such as carbonyl groups into the polymer and must be carefully guarded against In addition to the high-frequency uses, polyethylene is being more generally utilized for me-
366
HYDROCARBON PLASTICS A N D ELASTOMERS
chanical protection of wire and cables, where its chemical inertness and light weight are advantageous. About 5% of the branched polyethylene psoduced is used fbr wire and cable insulation.. Other important uses for branched polyethylene are housewases, toys, and containers, lids, and closures produced by injection molding (about 10% of'the production), rotational molding, powder coating, and pipe extrusion.
High-Density (Linear) Polyethylene

Linear polyethylene can be produced in several ways, including radical polymerization of' ethylene at extremely high pressures, coordination polymerization of ethylene, and polymerization of' ethylene with supported metal-oxide catalysts. Commercial production of lineas polyethylene, using the second and third routes named, began in 1957 and reached a volume of' 5 billion Iblyr in 1979; its 1982 volume was about 4..8 billion Ib, at list prices of' $0.45-$0.50 per Ib, heavily discounted.
1
Polymerization. The coordination polymerization of' ethylene (Chapter 4 0 ) utilizes a catalyst prepared as a colloidal dispersion by reacting, typically, an aluminum alkyl and TiCI, in a solvent such as heptane.. Ethylene is added to the reaction vessel under slight pressure, at a temperature of' 50-75C. Heat of' polymerization is removed by cooling. Polymer forms as a powder or granules, insoluble in the reaction mixture. At the completion of' the reaction, the catalyst is destroyed by the admission of' water or alcohol, and the polymer is filtered or centrifuged off, washed, and dried. Supported metal-oxide catalysts can be used in a variety of operating modes, including fixed-bed, moving-bed, fluid-bed, or slurry processes.. Ethylene is fed with a paraffin or cycloparaff'in diluent, at 60-200C and around 500 psi pressure. The polymer is recovered by cooling or by solvent evaporation. Structure. Typical linear polyethylenes ase highly (over 90%) crystalline polymers, containing less than one side chain per 200 carbon atoms in the main chain. Melting point is above 127C (typically about 135"C), and density is in the range of 0.95-097. Infrared spectroscopy (Chapter 9B) gives detailed information on the chemical and physical structure of' the polymer. Structural features associated with the crystallinity of lineas polyethylene are discussed in Chapter 10. Properties. Most of' the differences in properties between branched and linear polyethylenes can be attributed to the higher crystallinity of' the latter polymers Linear polyethylenes ase decidedly stif'f'er than the branched material (modulus of' 100,000 versus 20,000 psi), and have a higher crystalline melting point and greater tensile stsength and hardness. The good chemical resistance of' branched polyethylene is retained or enhanced, and such properties as low-temperature brittleness and low permeability to gases and vapors ase improved in the linear material
Applications. The production of' bottles and other containers by blow molding accounts for about 40% of'the linear polyethylene made The adjustment of'structure variables to obtain high resistance to environmental stress cracking, allowing the material to be used in detergent bottles, produced a large expansion in this field.. About 25% of the linear polyethylene produced is used in the injection molding of crates, pails, tubs, caps and closures, and housewases The higher stiffness and heat resistance of the lineas material have led to its replacement of' branched polyethylene in applications where these properties are important. Other major uses of linear polyethylene include film and sheet, rotational molding, wire and cable insulation, extrusion coating, and pipe and conduit High and Ultrahigh Molecular-Weight Polyethylenes. Although most linear polyethylenes have weight-averagemolecularweights in the range 100,000 -200,000, two higher-molecular-weight grades are widely used commercially: "High-molecular-weight" (HMW) polyethylene has M, between 300,000 and 500,000 Still processable by the usual techniques, it has improved environmental stress-crack resistance, impact and tensile strength, and long-term strength retention HMW polyethylene is used for plpe, film, and large blow-molded containers where these properties are important "Ultrahigh molecular-weight" (UI-IMW) polyethylene has i? between 3,000,000 t, and 6,000,000 It has exceptional abrasion and impact resistance compared to other polyethylenes Its uses depend on these properties coupled with those usually found for ethylene polymers, such as low coefficient of friction, and include bearings, sprockets, gaskets, valve seats, conveyor-belt parts, and other high-wear -resistance applications Because of its extremely high molecular weight, UHMW polyethylene does not melt or flow in the normal thermoplastic manner, and most of its fabrication is based on modifications of the compression-molding technique. Fast molding methods combine screw ram injection molding and compression or transfer molding (Chapter 17) Crosslinked Polyethylene
There has been considerable inte~est converting polyethylene to a thermosetting in material, in order to combine its low cost, easy processing, and good mechanical properties with the enhanced form stability at elevated temperatures, resistance to stress crack, and tensile strength expected in a crosslinked polymer.. Incorporation of relatively stable pe~oxides, such as dicumyl peroxide and di-t-butyl peroxide, provides a chemical means of' crosslinking polyethylene. The peroxides ase stable at normal processing temperatures but de.compose to provide free radicals for crosslinking at higher temperatures in a postprocessing vulcanization or curing reaction.. Chemically crosslinked polyethylene is used in the wire and cable industry and is of interest for pipe, hose, and molded articles. Ethylene-propylene rubbers (Section C) are also vulcanized by peroxide curing systems
Chemical Crosslinking.
368
HYDROC.ARBON PLASTICS AND ELASTOMERS
Radiation Crosslinking (Chapter 6 0 ) . The crosslinking of polyethylene by irradiation with high-energy electrons has been used in the commercial production of films combining the properties typical of polyethylene with form stability up to 200C and a significant increase in tensile strength. The film can be made heat shrinkable by biaxial stretching. It is used for insulating (by wrapping) electrical power cables, coils, transformers, and motors and generators
GENERAL REFERENCES
Raff 196.5-1966, 1967; Buckley 1968; Kresser 1969; Boysen 1981; Buchanan 1981; Hogan 1981; Paschke 1981; Short 1981b; Henkel 1982; Hug 1982; Snell 1982; Carol 1983
B.
POLYPROPYLENE
d
With the commercial utilization of coordination polymerization (Chapter 4 0 ) in 1957, the production of polypropylene became possible In the intervening decades this has become one of the world's major plastics, with U S production in 1982 4 of about 3 6 billion Ib, down some 10% from a ~ecord 0 billion lb in 1981, at a price of about $0 40 per Ib
Polymerization. The polymerization of propylene, recovered from cracked gas streams in olefin plants and oil refineries, is carried out with coordination catalysts essentially as described in Section A for linear polyethylene. Ethylene, propylene, and other a-olefins can be polymerized in the same equipment with very little modification, leading to highly flexible operation.. Catalysts and operating conditions must be selected with care to ensure that isotactic polypropylene (see the next paragraph) is produced Structure. Polypropylene can be made in isotactic, syndiotactic, or atactic form (Chapter 10A) The crystallizability of isotactic polypropylene makes it the sole form with properties of commercial interest Isotactic polypropylene is an essentially linear, highly crystalline polymer, with a melting point of 165C Its crystal structure is described in Chapte~IOB. Properties. Polypropylene is the lightest major plastic, with a density of 0.905 Its high crystallinity imparts to it high tensile strength, stiffness, and hardness The resulting high strength-to-weight ratio is an advantage in many applications Finished articles usually have good gloss and high resistance to marring. The high melting point of polypropylene allows well-molded parts to be sterilized, and the polymer retains high tensile strength at elevated temperatures The low-temperature impact strength of polypropylene is somewhat sensitive to fabrication and test conditions This sensitivity results from the presence of a
OTHER OLEFIN-BASED POLYMERS AND COPOLYMERS
369
dominating a-transition (Chapter 1lD) in polypropylene at about OC, resulting in a marked loss in stiffness near this temperature In high-density polyethylene, the dominant transition is the lower-temperature @-transition Thus the restriction of n l o l e ~ ~ lmotion leading to brittle behavior takes place not far below room temar prature in polypropylene, but at a much lower temperature in polyethylene. To overcome brittleness, wide use is made of both random and block copolymers of propylene with ethylene The block copolymers are the most impact resistant, and are used in injection molding applications To retain transparency, random copolymers are used for film applications, while the homopolymer is used almost exclusively for filaments Polypropylene has excellent electrical properties and the chemical inertness and rnoirture resistance typical of hydrocarbon polymers It is completely free from less stable than polyethenvironmental stress cracking However, it is inhe~ently ylene to heat, light, and oxidative attack (presumably because of the presence of tertiary hydrogens) and must be stabilized with antioxidants and ultraviolet light ab5orbers for satisfactory processing and weathering The resulting fo~mulations are quite satisfactory, even for such applications as indoor-outdoor carpeting, but are more expensive
Applications. Injection molding uses, including wide application in the automotive and appliance fields, account for almost half of the production of polypropylene Another third is used as filament (rope, cordage, and webbing) and filament and staple for carpeting Film uses run well behind
GENERAL REFERENCES
Raff 1965-1966, 196'7; Rebenfeld 1967; Frank 1968; Jezl 1969; Kresser 1969; Short 1980, 1981a; Buchanan 1981; Crespi 198 1; Holtgrewe 1982.
C. OTHER OLEFIN-BASED POLYMERS A N D COPOLYMERS
Olefin-Based Plastics
Polymers of a-olefins, including propylene, I-butene, and higher homologues, and copolymers of these monomers with ethylene, can be prepared by coordination polymerization (Chapter 4 0 ) Homopolymers of olefins higher in chain length than propylene that have been offered commercially include those of 1-butene and 4methyl pentene-1. In addition, a number of ethylene copolymers are sold, including those with propylene, I-butene, and isobutylene. Copolymers of ethylene with vinyl acetate, maleic anhydride, and ethyl acrylate are made commercially by radical polymerization Copolymers with carboxyl-containing monomers are crosslinkable by metal-salt bridges Some of these compositions are rubbery in nature, but they are listed here because they are primarily utilized as plastics
370
Polymers of Higher Olefins. Isotactic poly(butene-,I), known in the trade as polybutylene, is known in fbur crystalline modifications A metastable tetragonal fbrm results from crystallization from the melt, and changes i r ~ e ~ e r s i b l y the to stable twinned hexagonal fbrm in 5-7 days at room temperature. The remaining fbrms, orthorhombic and untwinned hexagonal, are formed only after crystallization from solution. Polybutylene's major use is as a pipe resin, based on its flexibility and resistance to creep, environmental stress cracking, chemicals, and abrasion.. It has the highest hydrostatic design stress rating fbr any flexible thermoplastic.. Blownfilm applications are also important.. Copolymers of4-methyl pentene- 1 with undisclosed monomers, presunlably used in small amounts to reduce degree of crystallinity, are known to the trade as p l y methylpentene. It is about 40% crystalline, with the extremely low density o f 0 . 8 3 With veIy high transparency and optical clarity and a thermal expansion coefficient ve~y near that of water, it is used in a variety of applications such as labor at or;^ and medical ware
"Linear Low-Density" (L.LD) Polyethylene. LLD polyethylene (LLDPE). a copolymer of ethylene with an alpha-olefin such as butene, hexene, or octene, is designed to simulate the short-chain branching and density of conventional branched polyethylene without the occuxrence of long-chain branching With unusually good melt-flow and physical properties, LLDPE has been one ofthe fastest growing new plastics in recent years, with about 1 billion Ib produced in 1982. A major use is plastic trash bags
'
Other Ethylene Copolymers. Copolymers of ethylene with ethyl acrylate, methyl

acrylate, and vinyl acetate, are made by high-pressure radical polymerization The main functions of the comonorner are to reduce crystallinity and introduce polarity, leading to more flexible and tougher products The major uses of these materials are as films.
lonomers. The word ionomer (Kinsey 1969) was coined as a generic term fbr a class of'thermoplastics containing ionizable carboxyl groups which can create ionic crosslinks between chains. These substances are produced as copolymers of' aolefins with carboxylic acid monomers, such as methacrylic acid, fbllowed by partial neutralization with a metal cation Crosslinking thus occurs through metal ''bridges ." These crosslinks are labile at processing temperatures, allowing the ionomers to be extruded or molded in conventional equipment. The upper use temperature of' the ionomers is limited, as might be expected, because the crosslinks begin to "melt out." The primary uses of this class of' materials are centered &,ound thei~. cornbination of properties such as high transparency, toughness, flexibility, adhesion, and oil resistance. Food packaging, skin and blister packaging, and several shoe uses are examples
OTtiER OLEFIN-BASED POLYMERS A N D COPOLYMERS
371
Olefin-Based Rubbers Chlorosulfonated Polyethylene. When polyethylene is treated with a mixture of chlorine and sulfur dioxide, some chlorine atoms ale substituted on the chains and some sulfonyl chloride groups (--SO,Cl) are formed. The chlorosulfonation can be car~ied either on the solid material or in solution. There are two results out of these modifications: (a) The chlorine atoms break up the regularity of the polyethylene chain structure so that crystallization is no longer possible, thus imparting an elastomeric character to the polymer, and (b) the sulfonyl chloride groups provide sites for crosslinking A typical polymer contains 25-30% chlorine (one chlorine for every seven carbon atoms) and about 1 5% sulfur (one -SO,CI for every 90 carbon atoms) The elastomer can be crosslinked by a large variety of compounds, including many rubber accelerators Metallic oxides are recommended for commercial cures Fillers are not needed to obtain optimum strength properties Chlorosulfonated polyethylene is resistant to ozone, being better than neoprene and butyl rubber in this respect Oxidative resistance and heat resistance are good Chemical resistance is better than that of the common elastomers The material is poor in "snap" and rebound and has low elongation and some permanent set Its abrasion resistance, flex life, low-temperature brittleness, and resistance to crack growth are good Ethylene-Propylene Rubbers. Copolymerization of propylene with ethylene yields noncrystalline products that have rubbery behavior and are chemically inert because of their saturation They must, however, be crosslinked by use of peroxides or radiation To gain sites for crosslinking, a diene monomer is often added; the resulting terpolymers are known as ethylene-propylene-dienemonomer (EDPM) elastomers The major comonomers are 1,4-hexadiene, dicyclopentadiene,
and ethylidene norbo~nene,
(In the latter two, it is the double bond on the far left that is most favorably opened during polymerization ) These rubbers have several desirable properties, including resistance to ozone, oxygen, and heat However, the additional cost resulting from the use of the expensive diene monomer and other sources has limited them to the special-purpose field to date, with 1982 production about 0 3 billion lb
HYDROCARBON PLASTIC:S A N D ELASTOMERS
GENERAL REFERENCES
Raff 1965-1966; Buckley 1968; Rubin 1968; Kresser 1969; Brydson 1978, Chapter 12; Borg 1979; Johnson 1979a; Ofstead 1979; Holohan 1980; Kochhar 1981; Baker 1982; Bonotto 1982; Chatterjee 1982; Duncan 1982; Ohi 1982; Person 1982
D.
NATURAL RUBBER AND OTHER POLYISOPRENES
Natural Rubber
Natural rubber is a high-molecular-weight polymer of'isoprene, in which essentially all the isoprenes have the cis-1,4 configuration (Chapter 10A). The natural polymer has a number-average degree of polymerization of' about 5000 and a br.oad distri.bution of molecular weights..
Source. Natural rubber can be obtained from nearly five hundred different species of plants The outstanding source is the tree Hevea brasrliensis, from which comes the name Hevea rubber Rubber is obtained from a latex that exudes from the bark of the Hevea tree when it is cut Latex is an aqueous dispersion of rubber, containing 25-40% rubber hydrocarbon, stabilized by a small amount of protein material and fatty acids. The latex is gathered, coagulated, washed, and dried Two different processes are used Crepe rubber results if a small amount of sodium bisulfite is added to bleach the rubber. The coagulum is rolled out into sheets about 1 mm thick and dried in air at about 50C If smoked sheets are to be made, the bleach is omitted and somewhat thicker sheets are rolled These are dried in smokehouses at about 50C in the smoke from burning wood or coconut shells. Mastication. It was discovered by Hancock in 1824 that rubber becomes a soft, gummy mass when subjected to severe mechanical working This process is known as mastzcatzon The addition of compounding ingredients is greatly facilitated by this treatment, which is usually carried out on roll mills or in internal mixers or plasticators (similar to extruders) Mastication is accompanied by a marked decrease in the molecular weight of the rubber Oxidative degradation is an important factor in mastication, since the decrease in viscosity and the other property changes do not take place if the rubber is masticated in the absence of oxygen After mastication is complete, compounding ingredients are added, and the rubber mix is prepared for vulcanization These processes are discussed in Chapter 19 Properties. The properties of vulcanized natural rubber form the model for the ideal elastomeric properties discussed in Chapter 1 l B , including rapid extensibility to great elongations, high stiffness and strength when stretched, and rapid and
RUBBERS DERIVED FROM BUTADIENE
373
complete retraction on release ofthe external stress. The properties of natural rubber are discussed further in comparison with those of' the major synthetic elastomers in Chapter 19
Stereoregular Synthetic Polyisoprene
In 1955, production began of essentially cis-.I,4polyisoprenes with structures closely duplicating that of natural rubber.. These polymers can be made by two processes that are almost identical except for the catalyst used: One is based on coordination polymerization (Chapter 4D), using a catalyst of titanium tetrachloride and an aluminum alkyl such as triisobutyl aluminum; the other is an anionic polymerization (Chapter 4C) with butyl lithium as the catalyst In a typical operation, isoprene (derived from petroleum) is mixed with a hydrocarbon solvent such as n-pentane The catalyst is added, and the reaction allowed to take place at about 50C and moderate pressures until a solids content of about 2.5% is reached.. At this point the reaction mixture is a highly viscous "cement." A catalyst deactivator and an antioxidant are added, and solvent is removed in an extruder or a drum dryer. The polymer made by coordination polymerization has a molecular weight distribution similar to that of masticated natural mbber and requkes no further processing before compounding, whereas that made with butyl lithium is higher in molecular weight and requires even longer mastication than natural rubber. The prope~ties the cis-l,4-polyisoprenes are, commensurate with their strucof' ture, very nearly identical with those of natural rubber, and the synthetic polymers are not only complete replacements fbr the natural product, but are often prefer~ed because of thelI greater cleanliness and uniformity. Production has, however, remained relatively small, dropping from around 175 million Iblyear in the early 1970's to about 120 million Ib in 1982..
GENERAL REFERENCES
Bean 1967; Kennedy 1968-,1969; Barnard 1970; Brydson 1978, Chapters 2 and 5; McGrath 1979; Tucker 1979; St. Cyr 1982; Blackley 1983.
E.
RUBBERS DERIVED FROM BUTADIENE
Although also made from alcohol during World War 11, butadiene is now derived exclusively from petroleum. Fractionation of the products of cracking petroleum, either for producing olefins or for obtaining high-octane gasoline, yields a cut containing largely hydrocarbons of'the butane and butene fkmily. 1-Butene is separated and catalytically dehydrogenated in the vapor phase to butadiene.
3 74
Styrene-Butadiene Rubber (SBR)

Production of SBR (then known as GR-S) was begun in the United States during World War 11.. The product was designed to be similar to the German Buna-S (Chapter 19) but lower in molecular weight for easier processing. The rubber was made by emulsion polymerization using the so-called Mutual recipe (Table 6-2), at 50C. After the war, product quality was improved by carrying out the polymerization at 5C (41F) with some being made at temperatures as low as - 10" 18C. These changes were brought about by the use of more active initiators, or -such as cumene hydroperoxide and p-menthane hydroperoxide, and the addition of antifreeze components to the mixture The product is known as cold rubber. Anionic solution copolymerization of butadiene and styrene with alkyllithium catalysts is used to produce so-called solution SBR. This product has a narrower molecular-weight distribution, higher molecular weight, and higher ci~-l,4-polybutadiene content than emulsion SBR. Tread wear and crack resistance are improved, as is economy because oil extension and carbon-black loading can be 8 increased Consumption of SBR in the United States has remained for many years at about 50% of all rubber use. It reached almost 3 billion Iblyear in 1979, but the reduction in the production and size of automobile tires, which account for about 75% of SBR use, reduced the production figure to around 2 . 2 billion Ib in 1982. List prices were in the range $0.60-$0.80 per Ib, but were heavily discounted.
Structure of SBR. By virtue of its free radical polymerization, SBR is a random copolymer. The butadiene units are found to be about 20% in the 1,2 configuration, 20% in the cis-1,4, and 60% in the trans-1,4 for polymer made at 50C, with the percentage of tmns-1,4 becoming higher fbr polymer made at lower temperatures In consequence of its irregular structure, SBR does not crystallize Branching reactions due to chain transfer to polymer and to polymerization of both double bonds of a diene unit become extensive if conversion is allowed to become too high or a chain transfer agent is not used in SBR polymerization However, SBR has been shown to have exactly one double bond per butadiene unit. Thus no extensive side reactions occur during its formation, at least up to about 75% conversion Processing of SBR. In general, the differences in mastication and vulcanization between SBR and natural rubber are minor . A reinforcing filler is essential to the achievement of good physical properties in SBR. However, some fillers other than carbon black reinforce it moderately well. SBR is compatible with the other major elastomers and can be used in blends The techniques of oil extension and masterbatching are widely employed. These techniques, as well as the processing of SBR in general, are discussed further in Chapter 19 Properties of SBR. Tire tread stocks made from regular SBR are inftrior in tensile strength to those from natural rubber (3000 versus 4500 psi), whereas those from
RUBBERS DERIVED F R O M BUTADIENE
375
"cold rubber" are almost equivalent to Hevea (3800 psi) At elevated temperatures, however, regular and "cold7' SBR lose almost two-thirds of' their tensile strength whereas natural rubber loses only 25%. The ozone resistance of' SBR is superior to that of' natural rubber, but when cracks or cuts start in SBR they grow much more rapidly Perhaps the most serious defect of both types of' SBR fbr tire uses, however, is its poorer resilience and greater heat buildup. Tread wear of the synthetic material is at least as good as that of natural rubber. The weatherability of SBR is better than that of' natural rubber "Cold" SBR is superior to the standard product because it contains less lowmolecular-weight (nonreinf'orceable) rubber, less chain branching and crosslinking, and a higher praportion (70%) of' the trans-1,4 configuration around the double bond. Originally, SBR was used in tires only of necessity, but "cold" SBR appears equal in most respects to natural rubber, especially for lighterduty tire use. It is inferior to the natural product for truck tires For many mechanical goods SBR is superior to natural rubber and is preferentially used because of its easier processing and good-quality end product Such items include belting, hose, molded goods, unvulcanized sheet, gum, and flooring Rubber shoe soles are made of SBR Extruded goods and coated fabrics are other fields in which SBR offers advantages in processability It is widely used for electrical insulation, although its properties are not as good as those of butyl rubber
Nitrile Rubbers Applications of SBR.
The nitrile rubbers are polymers of butadiene and acrylonitrile, having ratios of the two monomers similar to the ratio of butadiene to styrene in SBR They are noted for their oil resistance but are not suitable for tires The first commercial nitrile rubbers were made in Germany The oil resistance of the nitrile rubbers varies greatly with their composition Several grades are available commerciall.y, ranging in acrylonitrile content between 18% (only fair oil resistance) and 40% (extremely oil resistant) The nitrile rubbers are prepared in emulsion systems similar to those used for SBR Because the monomer reactivity ratios are quite different, the compositions of the feed and the polymer differ markedly This fact is usually taken into account by adjusting the monomer composition during the polymerization to achieve the desired polymer composition Nitrile rubber is used primarily for its oil resistance, by which is implied its low solubility, low swelling, and good tensile strength and abrasion resistance after immersion in gasoline or oils Swelling of nitrile rubbers is greater in polar solvents than in nonpolar solvents, but they can be used in contact with water and antifreeze solutions Resistance to ethylene glycol is good. The rubbers are inherently less resilient than natural rubber Their heat resistance is good; and, if properly protected by antioxidants, they show satisfactory resistance to oxidative degradation as well
376
Nitrile rubbers are extensively used for gasoline hoses, fuel tanks, creamery equipment, and the like.. In addition they find wide application in adhesives and, in the fbrm of' latex, for impregnating paper, textiles, and leather. In all, about 165 million Ib was used in the United States in 1982
Stereoregular Polybutadienes cis-l,4-Polybutadiene. This polymer is made by coordination or anionic polymerization in the same processes used fbr cis-l,4-polyisoprene (Section D) and has
similar properties. It is utilized almost entirely in tires, blended with SBR and natural rubber. A small amount of' natural rubber appears to be required to prevent the polybutadiene from crumbling during processing and to improve tack. Polybutadiene has high elasticity, low heat buildup, and good resistance to oxidation It imparts outstanding abrasion resistance to truck and passenger tires, but cannot be used at Ievels higher than 40-50%, above which its major deficiency of' poor skid resistance becomes too apparent. Use in 1982 was about 0..8billion lb in t$e United States
GENERAL REFERENCES
Saltman 1965; Kennedy 1968-1969; Rubin 1968; Brydson 1978, Chapters 6-9 and 14; Bauer 1979; Kuzma 1979; McGrath 1979; Robinson 1979; Hsieh 1981; Blackley 1983
F OTHER SYNTHETIC EL.ASTOMERS . Polyisobutylene and Butyl Rubber

Butyl rubbers are copolymers of' isobutylene with a small amount of isoprene added in order to make them vulcanizable. Since the amount of comonomer is small, the methods of' polymerization and the properties of the unvulcanized copolymer ase similar to those of poIyisobutylene itself.
Polymerization. The polymerization of'isobutylene and its mixtures with diolefins typifies the industrial application of low-temperature cationic polymerization (Chapter 4 B ) . Isobutylene polymerizes rapidly at - 80C with Friedel-Crafts catalysts. In a bulk system at - 80C, rapid polymerization is induced by bubbling BF, gas through isobutylene. The heat of reaction can be absorbed by adding solid carbon dioxide to the monomer or by adding a low-boiling diluent such as pentane or ethylene which is refluxed. In a typical process, butyl rubber is manufactured by mixing isobutylene with 1.5-4.5% isoprene and methyl chloride as diluent.. This mixture is fed to stirred reactors cooled to -95C by liquid ethylene. Catalyst solution, made by dissolving anhydrous aluminum chloride in methyl chloride, is
OTHER SYNTHETIC: ELASTOMERS
377
added Polymer forms at once as a finely divided product suspended in the reaction mixture. This slurry is pumped out of the reactor continuously as monomer and catalyst are added The product mixture is passed into a large volume of hot agitated water in a tank where the volatile components are flashed off and recovered. An antioxidant and some zinc stearate to prevent agglomeration of the polymer particles are added at this point The polymer is then filtered off, dried, and extruded
Structure. Polyisobutylene and butyl rubber are amorphous under normal conditions, but crystallize on stretching. Most, if not all, of the isoprene units are present in the 1,4 structure It is usually assumed that the polymers are linear, although the possibility of a small amount of branching has not been investigated in detail The molecular weights of the polymers made by low-temperature ionic polymerization can be quite high; chain transfer agents such as diisobutylene are often added to control molecular weight at the 200,000-300,000 level Unstabilized polyisobutylenes are degraded by heat or llght to sticky low-molecular-weight products. The usual rubber antioxidants or retarders of free radical reactions stabilize the polymers well Low-molecular-weight polyisobutylenes are liquids As molecular weight in creases, they change to balsamlike solids, and then to rubberlike polymers Unless low-molecular-weight material is removed, even polyisobutylene of 100,000 molecular weight is sticky Unlike natural rubber, polyisobutylene and butyl rubber do not crystallize on cooling and hence remain flexible to as low as -50C The response of isobutylene polymers and copolymers to stress is quite different from that of natural rubber Polyisobutylene is sluggish, showing large viscoelastic and viscous components in its response (Chapter 11C). The polymer has been widely used in the study of viscoelasticity. The strong tendency toward cold flow in polyisobutylene prevents its direct application as an elastomer It is used in adhesives, caulking compounds, pressuresensitive tapes, and coatings for paper
I
I
Properties. The propefiies of butyl vulcan~zatesare compatible with their structure The very low residual unsaturation of the rubbers leads to outstanding chemical inertness The closepacked linear paraffinic chains result in unusually low permeability to gases. Steric hindrance of the methyl groups on the chains causes high internal viscosity and viscoelastic response to stresses A property of great importance in elastomers is aging in the presence of oxygen It has been found that the presence of a double bond in the skeletal structure of a polymer is very important in enhancing the rate of absorption of oxygen, and that the presence of methyl side groups is also significant but less so Butyl rubber is therefore, as expected, less sensitive to oxidative aging than are most other elaschain scission tomers except the silicones Since methyl side groups appear to f a v o ~ whereas double bonds favor crosslinking, butyl rubber becomes soft rather than brittle on oxidative degradation Butyl has much better ozone resistance than natural rubber Its solvent resistance is typical of that of hydrocarbon elastomers Its acid resistance is quite good
378
HYDROCARBON PLASTIC:S A N D ELASTOMERS
The stress-strain properties of butyl rubber are similar to those of natural rubber. Both show the importance of' crystallization in obtaining high tensile strength. Crystallization does not take place in butyl, however, until higher elongations are reached The tear resistance of butyl is quite good and is retained well at high temperatures and fbr long times, in contrast to natural rubber. The electrical properties are quite good, as predicted from its nonpolar, saturated nature. The dynamic and elastic properties of butyl are masked by sluggishness over the temperature range - 30" to t 40C, characteristic of a polymer with high internal friction and high damping power. Rebound is slow and heat buildup is high
Applications. About 75% of the butyl rubber produced is used for inner tubes for tires. This usage was substantially reduced when tubeless passenger tires were introduced in 1953 (Herzegh 1981), but most nonpassenger tires still use tubes Other uses fbr butyl ase mainly in the area of mechanical goods. Consumption of butyl rubber in the United States was about 290 million Ib in 1982
3
Polychloroprene (Neoprene)
The generic term neoprene denotes rubberlike polymers and copolymers of chloroprene, 2-chloro-l,3-butadiene. Neoprenes were the first synthetic rubbers developed in the United States. Although they are primarily known for their oil resistance, they are good general-purpose rubbers that can replace natural rubber in most of its uses and are satisfactory in a wide variety of applications. About 270 million Ib was used in the United States in 1982 Chloroprene is prepared by the catalytic addition of' hydrogen chloride to vinylacetylene, which in turn is made by the catalytic dimerization of acetylene. This is an expensive process, however, and interest is turning to the production of chloroprene from butadiene One process utilizes chlorination to 3,4-dichlorobutene1, fbllowed by dehydrochlorination. The neoprenes are produced by emulsion polymerization. Some types are polymerized in the presence of sulfur, which introduces some crosslinking in the polymer.. In these cases, the latex is allowed to age in the presence of' an emulsion of' tetraethylthiuram disulfide, which restores the plasticity of the polymer. The latex is then coagulated by acidification followed by freezing. Polymerization appears to take place almost entirely in the t m s - 1 , 4 form As a result the neoprenes are crystallizable elastomers The vulcanization of' neoprene is different from that of other elastomers in that it can be vulcanized by heat alone..Zinc oxide and magnesium oxide are the preferred vulcanizing agents. The mechanism by which they cause crosslinking is not known Sulfur vulcanizes neoprene very slowly, and the usual rubber accelerators are in general not effective.-some, in h c t , are potent retarders of the cure. A few chemicals are known to accelerate the vulcanization, however, among them antimony sulfide..
Unlike many other elastomers, neoprene vulcanizates have high tensile strength (3500-4000 psi) in the absence of carbon black. No reinforcing effect is found with any filler Suitably protected neoprene vulcanizates are extremely resistant to oxidative degradation Weathering resistance and ozone resistance are quite good Neoprene is slightly inferior to nitrile rubber in oil resistance, but markedly better than natural rubber, butyl, or SBR The dynamic properties of neoprene are superior to those of most other synthetics and only slightly inferior to those of natural ~ubber They are less affected by elevated temperature than those of natural rubber Neoprene has been shown to make excellent tires but cannot compete with other elastomers in price Its major uses include wire and cable coatings, industrial hoses and belts, shoe heels, and solid tires Gloves and coated fabrics are made from neoprene latex
Thermoplastic Elastomers
The term thermoplastzc elastomers is currently used to describe a wide variety of materials that have elastomeric properties at ambient temperatures, but process like thermoplastics, obviating the need for the vulcanization step to develop typical rubberlike elasticity The best known and most widely used thermoplastic elastomers, to which the name originally applied, are block copolymers of styrene (S) with butadiene (B) or isoprene (I) with the block structure S-B-S or S-I-S The morphology and properties of these materials are described in Chapter 1 7 0 Other polymers commonly termed thermoplastic elastomers are polyurethanes and copolyesters (Chapter 15) and blends of ethylene-propylene copolymers (Section C) with polypropylene As a family, the thermoplastic elastomers are beginning to replace other specialty rubbers in a wide variety of uses, for example, replacing neoprene in adhesives and and w i ~ e cable insulation Including all types, about 400 million Ib was sold in 1982
Other Elastomers
Several other elastomers have the status of specialty rubbers, at relatively low production volume and high price. Among these are chlorosulfonated polyethylene (Section C), the acrylate and various fluorocarbon elastomers (Chapter 14), polyurethane and polysulfide elastomers (Chapter 15), and silicone and epichlorhyd~.in rubbers (Chapter 16)
GENERAL REFERENCES
Buckley 1965; Friedlander 1965; Hargreaves 1965; Kennedy 1968-1969; Brydson 1978, Chapters 10, 11, and 17; West 1978; Baldwin 1979; Finelli 1979; Johnson 1979b; McGrath 1979; Bonk 1982; Ellerstein 1982; Blackley 1983
380
HYDROCARBON PL.ASTICSA N D ELASTOMERS
1. Write the chemical reactions postulated fbr the formation of' long- and shortchain branching in polyethylene..
2. Describe the relation between long-chain branching and molecular-weight distribution typical of' low-density polyethylene
3. Discuss the properties of' LLDPE in relation to its branched-chain structure

4.
Compare the glass transitions in polyethylene and polypropylene and relate of them to the impact st~engths the two polymers
5. Account fbr the high transparency and optical clarity of' polymethylpentene
6 . Describe the crosslink bonding in ionomers in terms of' molecular structure, chemical sensitivity, and physical properties of' the polymers
7. Discuss the functions of the three chain constituents in (a) chlorosulfonated polyethylene and (b) EPDM rubber in producing elastomeric properties r
8.
Write chemical f'ormulas fbr typical chain structures of' the three major types of EPDM rubber..
9. Name and show by chemical formulas all the ways in which isoprene can add to a growing chain during polymerization. Identify those most frequently fbund in commercially valuable polyisoprenes
BIBLIOGRAPHY
Baker 1982 George Baker, "Ethylene-Methyl Acrylate," p 81 In MPE, f 1982 Baldwrn 1979 F B Baldwin and R I-I Schatz, "Butyl Rubber," pp 470-484 In ECT,$ Vol 8, 1979 Barnard 1970 D Baxnard, J I Cunneen, P B Lindley, A R Payne, M Porter, A Schallamarch, P W A Southo~n, McL Swlft, and A G Thomas, "Rubber, Natural," pp 178-256 in EPST,S: Vol 12, 1970 Bauer 1979 R G Bauer, "Styrene-Butadlene Rubber," pp 608-625 in ECT, Vol 8, 1979 Bean 1967 Arthur R Bean, Glenn R Hines, Geoffrey Holden, Robert R Houston, John A Langlon, and Roger H Mann, "Isoprene Polymers," pp 782-855 In EPST, Val 7, 1967 Blackley 1983 D C Blackley, Synthetic Rubbers Thelr Chemistry and Technology, Elsevier, New York, 1983 Bonk I982 H W Bonk, "Thermoplastics Elastomers," pp 120-122, 127 in MPE, 1982 Bonotto I982 Sergio Bonotto, "Ethylene-Ethyl Acrylate," pp 80-81 In MPE, 1982
t Joan Agranoff, ed ,Modern Plaslics Encyclopedta 1982-1983, Vol 59, No 10A, McGraw-Hill, New York, 1982
$Martin Grayson, ed ,Krrk-Othmer Encyclopedra ojChemica1 Technology, 3rd ed , Wiley-Interscience, New York Herman F Maxk, Norman G Gaylord, and Norbert M Bikales, eds , Encyclopedia ojPolymer Science and Technology, Wiley-Interscience, New York
BIBLIOGRAPHY
381
1
I
Borg 1979 E L Borg, "Ethylene-Propylene Rubbers," pp 492-500 in ECT, Vol 8, 1979 Polyethylene," pp Bovsen 1981 Robert L Boysen, "High Pressure (Low and Intermedrate Dens~ty) 402-420 in ECT, Vol 16, 1981 Appl~edSc~ence, London, 1978 Brydson 1978 J A Brydson, Rubber Chem~srry, Buchanan 1981 P R Buchanan, "Olefin Fibers," pp 357-385 in ECT, Vol 16, 1981 Buckley I965 D I Buckley, "Butylene Polymers," pp 754-795 In EPST, Vol 2, 1965 Buckley 1968 D J Buckley, B S Dyer, M R Day, and W R Bergenn, "Olefin Polymers," pp 440-458 in EPST, Vol 9, 1968 Carol 1983 Frederick J Carol, "The Polyethylene Revolut~on," Chemtech 13, 222- 228 (1983) Chatterjee 1982 A M Chatterjee, "Polybutylene," pp 58, 60 in MPE, 1982 C ~ e s p ~ G~ovanni 1981 Crespi and Luciano Lucian~,"Polypropylene," pp 453-469 in ECT, Vol 16, 1981 Drrncan 1982 R E Duncan, "Ethylene-Vinyl Acetate," pp 81-82 in MPE, 1982 Ellerstein 1982 S M Ellerstein and E R Bertozz~,"Polysulfides," pp 814-831 In ECT, Vol 18, 1982 Flnellr 1979 A F Finelli, R A Marshall, and D A Chung, "Thermoplastic Elastomers," pp 626640 in ECT, Vol 8, 1979 Frank 1968 H P Frank, Polypropylene, Gordon and Breach, New York, 1968 Frredlander 1965 Hemy Z Friedlander, "Chemically Resistant Polyme~s," pp 665-683 in EPST, Vol 3, 1965 Hargreaves 1965 C A Hargreaves, 11, and D C Thompson, "2-Chlo~obutadiene Polymers," pp 705-730 In EPST, Vol 3, 1965 Henkel I982 R N Henkel, "H~gh and Low Density Polyethylene," pp 73-74, 76 in MPE, 1982 Herzegh 1981 Frank Herzegh, "The Tubeless T~re," Chemtech 11, 224-228 (1981) Hogan 1981 J Paul Hogan, "Linear (High Density) Polyethylene," pp 421-433 in ECT, Vol 16, 1981 Holohan 1980 J F Holohan, JI , J Y Penn, and W A Vredenburgh, "Hydrocarbon Resins," pp 852-869 in ECT, Vol 12, 1980 Holtgrewe 1982 D A Holtgrewe, "Polypropylene," pp 94,96,98 in MPE, 1982 Hs~eh1981 Henry L Hsieh, Ralph C F a ~ ~ aand Kishore Ud~pi,"Anronic Polymerizat~on:Some r, Commercial Applicat~ons,"Chemtech 11, 626-633 (1981) Hug 1982 D P Hug, "UHMW Polyethylene," pp 79-80 in MPE, 1982 Jezl 1969 James L Jezl and Earl M Honeycutt, "Propylene Polymers," pp 597-619 In EPST, Vol 11, 1969 Johnson 1979a Paul R Johnson, "Chlorosulfonated Polyethylene," pp 484-491 m ECT, Vol 8, 1979 Johnson 19796 Paul R Johnson, "Neoprene," pp 515-534 in ECT, Yo1 8, 1979 Kennedy 1968-69 Joseph P Kennedy and Enk Tdrnquist, eds , Polymer Chemistry oj Synthetic Elastomers, W~ley-Interscience,New York; Part I, 1968; Part 11, 1969 Kinsey 1969 Roy H Klnsey, "Ionomers, Chemistry and New Developments," Appl Polym Symp 11, 77-94 (1969) Kochhar 1981 R K Kochhar, Yurl V Klssln, and David L Beach, "Polymers of H~gher Olefins," pp 470-479 in ECT, Vol 16, 1981 Kresser 1969 Theodore 0 G Kresser, Polyolejin Plasncs, Van Nostrand Reinhold, New York, 1969 Kuzma 1979 L J Kuzma and W J Kelly, "Polybutadiene," pp 546-567 In ECT, Vol 8, 1979 McGrath 1979 James E McGrath, "Elastomers, Synthetic-Survey," pp 446-459 in ECT, Vol 8, 1979
'I
I I
,
382
HYDROCARBON PLASTICS AND ELASTOMERS
McGrew 1958 Frank C McGrew, "The Polyolefin Plastics Field--Present Technical Status," Modern Plastics 35 (7), 125,126,128,132-133 (1958) Ojstead 1979 Ellert A Ofstead, "Polypentenamers," pp 592-607 in ECT, Vol 8, 1979 Oh1 1982 H Ohi, "Polymethylpentene," p 93 in MPE, 1982 Paschke 1981 Eberhard Paschke, "Ziegler P~ocessPolyethylene," pp 433-452 In ECT, Vol 16, 1981 Person 1982 W J Pe~son, "Llnea~ Low Density Polyethylene," pp 76,78 in MPE, 1982 Rafl1965-1966 R A V Raff and K W Doak, eds , C ~ s f a l l i nOlefn Polymers, WiIey-Interscience, e New York; Part I, 1965; Part 11, 1966 Ruff 1967 R A V Raff, P E Campbell, R V Jones, E D Caldwell, Herbert N Friedlander, and Peter J Canterino, "Ethylene Polymers," pp 275-454 in EPST, Vol 6, 1967 Rebenfeld 1967 Ludwig Rebenfeld, "Fibers," pp 505-573 in EPST, Vol 6, 1967 Robinson 1979 H W Robinson, "NltriIe Rubber," pp 534-546 in ECT, Vol 8, 1979 Rubzn 1968 I D Rubin, Poly(1 Butene). Rs Preparation and Propertzes, Gordon and Breach, New York, 1968 St Cyr 1982 Dav~d St Cyr, "Rubber, Natural," pp 468-491 in ECT, Vol 20, 1982 R Saltman 1965 W M Saltman, "Butadiene Polymers," pp 678-754 In EPST, Vol 2, 1965 Short 1980 James N Short, "Processes for Polypropylene Commerciallzat~on," Ind Res Dev 22, (9) 109-1 12 (1980) Short 1 9 8 1 ~James N Short, "Polypropylene: Processes, Catalysis, Economics," Chemtech 11, 238243 (1981) Short 1981b James N Short, "Low Pressure L~near(Low Denslty) Polyethylene," pp 385-401 In ECT, Vol 16, 1981 Snell 1982 T P Snell, "HMW-Hlgh Density PolyethyIene," p 79 In MPE, 1982 Staudinger 1936 H Staudmger, "The Formation of Hlgh Polymers of Unsaturated Substances," Trans Faraday Soc 32, 97-121 (1936) Tucker 1979 Harold Tucker and S E Home, Jr , "Poly~soprene," pp 582-592 in ECT, Vol 8, 1979 West 1978 James C West and Stua~t Cooper, "Thermoplast~cElastomers," Chapter 13 In Frederick L R Elrich, ed , Science and Technology oj Rubber, Academic Press, New Yo~k,1978
In this chapter we extend the discussion of hydrocarbon polymers (Chapter 13) to the consideration of other macromolecules with a carbon-carbon backbone chain Included here are polymers made from vinyl monomers, CH,=CHX, and vinylidene monomers, CH2=CY2, and the fluorinated polymers Discussed elsewhere are thermosetting resins made in par1 f ~ o m vinyl monomers, such as "polyester" resins (Chapter 16C) It should be noted that in most of this chapter the term vinyl is used in the sense of monosubstituted ethylenic, while in the plastics industry (and in Section D) the term vznyl resins commonly refers to poly(viny1 chloride) and related plastics
A.
POLYSTYRENE A N D RELATED POLYMERS
The family of styrene polymers includes polystyrene, copolymers of styrene with other vinyl monomers, polymers of derivatives of styrene, and mixtures of polystyrene and styrene-containing copolymers with elastomers Total production of these polymers exceeded 3 2 billion Ib as early as 1969, and reached about 4 6 billion Ib in 1982, down some 20% from a peak in the late 1970s Prices have about tripled since the early 1970's, however, to a level around $0 45 per Ib in 1982. In compa~ison the other ma'or high-volume resins, the outlook for polyto styrene seems relatively unpromising The major application field for polystyrene is packaging, accounting for a third of its end markets Most of the polymer is fabricated by injection molding or extrusion, with the foamed or expandable-bead product accounting f o ~ about 10% of the total and growing
384
OTHER CARBON-CHAIN POLYMERS
Polystyrene is a thermoplastic with many desirable properties It is clear, transparent, easily colored, and easily fabricated It has reasonably good mechanical and thermal properties, but is slightly brittle and softens below 100C.
Monomer.
Styrene (vinylbenzene),
is made from benzene and ethylene In one process, ethylene is passed into liquid benzene under pressure, in the presence of aluminum chloride catalyst:
The resulting ethylbenzene is dehydrogenated to styrene by passing it over an iron oxide or magnesium oxide catalyst at about 600C The styrene is then refined by distillation
Polymerization. Although solution or emulsion polymerization may occasionally be use, most polystyrene is made either by suspension polymerization or by polymerization in bulk. All but the latter process are typically carried out as described in Chapter 6 The bulk polymerization of' styrene is begun in a "prepolymerizer," a stir~ed vessel in which inhibitor-free styrene is polymerized (usually with a peroxide initiator) until the reaction mixture is as concentrated in polymer as is consistent with efficient mixing and heat transfer. Normally, a solution containing about 30% polymer is as viscous as can be handled. The syrupy mixture from the prepolymerizer then enters a cylindrical tower (about 40 ft long by 15 f't diameter) maintained essentially full of' fluid.. By cooling the upper part of the tower and heating the lower part, the polymerization is controlled to prevent runaways but to proceed to essentially pure molten polymer at the bottom. This melt is discharged through spinnerets or into an extruder producing small-diameter rod that is chopped, after cooling, into short lengths to provide the finished molding powder Structure and Properties. Polystyrene is a linear polymer, the commercial product being atactic and therefore amorphous Isotactic polystyrene can be produced, but offers little advantage in properties except between the glass transition (about 80C) and its crystalline melting point (about 240C), where it is much like other crystalline plastics. Isotactic polystyrene is not of commercial interest because of increased brittleness and more difficult processing than the atactic product
'
6
pO1.YSTYRENE A N D RELATED POI.YMERS
385
Like most polymers, polystyrene is relatively inert chemically. It is quite resistant to alkalis, halide acids, and oxidizing and reducing agents It can be nitrated by fuming nitric acid, and sulfbnated by concentrated sulfuric acid at 100C to a watersoluble resin (see ion-exchange resins, below). Chlorine and bromine are substituted on both the ring and the chain at elevated temperatures. Polystyrene degrades at elevated temperatures to a mixture of low-molecular-weight compounds about half' of' which is styrene. The characteristic odor of' the monomer serves as an identification fbr the polymer. As made, polystyrene is outstandingly easy to process,. Its stability and flow under injection-molding conditions make it an ideal polymer fbr this technique. Its optical properties-color, clarity, and the like--are excellent, and its high refractive index (1..60) makes it useful for plastic optical components. Polystyrene is a good electrical insulator and has a low dielectric loss fBctor at moderate frequencies Its tensile strength reaches about 8000 psi. On the other hand, polystyrene is readily attacked by a large variety of' solvents, including dry.-cleaning agents. Its stability to outdoor weathering is poor.; it turns yellow and crazes on exposure. Two of' its major defects in mechanical properties are its brittleness and its relatively low heat-deflection temperature of' 82-88"C, which means that polystyrene articles cannot be sterilized Many of these defects can be overcome by proper formulating, or by copolymerization and blending as described below. For example, the addition of ultravioletlight absorbers improves the light stability of' polystyrene enough to make it useful in lighting fixtures such as fluorescent,-light diffusers. Flame-retardant polystyrenes have been developed through the use of additives
Copolymers of Styrene Butadiene Copolymers. The most important of all copolymers of styrene in terms of volume are the styrene-butadiene synthetic rubbers discussed in Chapter 13E. Another group of styrene-butadiene copolymers is widely used in latex paints; these fall in the composition range 60 styrene: 40 butadiene by weight. Styrene-butadiene block copolyme~s also a major constituent of thermoplastic elastomers (Chapter are 17C). Heat- and Impact-Resistant Copolymers. A number of' copolymers of' styrene with a minor amount of' comonomer have enhanced heat and impact resistance without loss of' other desirable properties of' polystyrene. Typical comonomers are those that increase intermolecular forces of' attraction by introducing polar groups, or those that stiffen the chain and reduce rotational freedom through the steric hindrance of' bulky side groups. In the first class are acrylonitrile (CH,=CHCN), fumaronitrile (trans-NCCH=CHCN), and 2,5-dichlorostyrene Comonomers with bulky side groups include N-vinylcarbazole and N,N-diphenylacrylamide The products of' major commercial interest in this group contain, typically, 76% styrene and 24% acrylonitrile. Such copolymers have heat-deflection temperatures
386
of 90-92C and more resilience and impact resistance than polystyrene.. Their color, however, is slightly yellow
Ion-Exchange Resins. Cationic type ion-exchange resins are produced by making a suspension polymerization of' styrene with several per cent divinylbenzene. The product, in the form of unifbrm spheres, is sulf'onated to the extent of' about one -SO,H group per benzene ring. Anionic ion-exchange resins are made by copolymerizing styrene with divinylbenzene and vinylethylbenzene The polymer is treated with chloromethyl ether to put chloromethyl groups on the benzene rings Quaternary ammonium salts are formed by reaction with tertiary amines such as trimethylamine, methyldiethanolamine, and dimethylpropanolamine Copolymers of' styrene with methyl methacrylate are discussed in Section B Polymers of Styrene Derivatives
Many styrene derivatives have been synthesized, but only a few have found corn-. mercial importance. Poly(or-methyl styrene), with better heat resistance than polystyrene, has failed to survive because of' brittleness and difficulty in fabrication due to a low ceiling temperature of' 61C. Poly(p-tert-butylstyrene) is used as a viscosity improver in motor oils, polychlorostyrene is valuable because of its selfextinguishing characteristics, and poly(sodium styrenesulfbnate) is used as a watersoluble flocculating agent. It is predicted that poly(p,-methyl styrene) may become of commercial interest because of' a new monomer synthesis from toluene and ethylene that could result in lower prices; some key properties of' the polymer are improved over those of' polystyrene
Rubber-Modified Polystyrene
Rubber is incorporated into polystyrene primarily to impart toughness The resulting materials consist of'a polystyrene matrix with small inclusions of'the rubber (usually 5-10% polybutadiene or copolymer rubber). They are termed impact polystyrene, and account for over half' of' the polystyrene homopolymer produced Grafting of' the rubber to the polystyrene may occur if' the rubber is present during the styrene polymerization; these materials are the most effective in enhanced impact strength, particularly if the rubber is slightly crosslinked; but the mechanical blends are also used.
ABS Resins
Like the rubber,-modifiedpolystyrenes, ABS resins are two-phase systems consisting of' inclusions of' rubber in a continuous glassy matrix. In this case the matrix is a styrene-acrylonitrile copolymer, and the rubber a styrene-butadiene copolymer, the name ABS deriving from the initials of' the three monomers. Again, development of' the best properties requires grafting between the glassy and rubbery phases. The ABS resins have higher temperature resistance and better solvent resistance than
ACRYLIC POLYMERS
387
the high-impact polystyrenes and are true engineering plastics, particularly suitable for high-abuse applications. They can easily be decorated by painting, vacuum metalizing, and electroplating.. In addition to fabrication by all the usual plastics techniques (Chapter 17), the ABS resins can be cold fbrmed, a technique typical of' metal fabrication. About 700 million Ib of ABS resins was sold in 1982
Polystyrene Foams
Around 760 million Ib of styrene homopolymer was used in 1982 to make a wide Most of' these are based on foamed-in-place beads, variety of foamed p~oducts. made by suspension polymerization in the presence of a foaming agent such as pentane or hexane, liquid at polymerization temperature and pressure. Subsequent heating softens the resin and volatilizes the fbaming agent.
GENERAL REFERENCES
Basdekis 1964; Abrams 1967; Boyer 1970; Platzer 1977; Radosta 1977; Morneau 1978; Evans 1982; 1,antz 1982; Peppin 1982; Swett 1982; Platt 1983
B.
ACRYLIC POLYMERS
In this section axe discussed the acrylate and methacrylate plastics, of which the most important is poly(methy1 methacrylate), and the acrylic fibers, based largely on polyacrylonitrile
Poly(rnethy1 methacrylate)
Poly(methy1 methacrylate) is a clear, colorless transparent plastic with a higher softening point, better impact strength, and better weatherability than polystyrene. It is available in molding and extrusion compositions, syrups, and cast sheets, rods, and tubes..
Monomers. The principal commercial processes for the production of acrylate esters are based on ethylene cyanohydrin (much like the acetone cyanohydrin route to methyl methacrylate described below), the carbonylation of' acetylene, or the polymerization and subsequent depolymerization of P-propiolactone to give acrylic acid or (in the presence of an alcohol) an acrylate ester. Methyl methacrylate is made by heating acetone cyanohydrin (from the addition of'hydrocyanic acid to acetone) with sulfuric acid to form methacrylamide sulfate
388
The latter is reacted (without separation) with water and methanol to give methyl methacrylate:
Polymerization. Poly(methy1 methacrylate) fbr molding or extrusion is made by bulk or suspension polymerization, as described in Chapter 6 . The production of' cast sheets, rods, and tubes is carlied out by bulk polymerization, starting in most cases with a syrup of' partially polymerized methyl methacrylate with a convenient viscosity for handling. Shrinkage and heat evolution during polymerization are reduced by the use of a syrup Sheets of' poly(methy1 methacrylate) are commonly made by extrusion Alternatively, they may be cast in cells made up of' two sheets of' heat-resistant glass separated by a coated rubber gasket. The cell is filled with syrup and sealed, and polymerization is carried out at 60-70C in an air oven or water bath, with a finishing treatment at 100C. Peroxide or azo initiators may be used Properties. Poly(methy1 methacrylate) is a linear thermoplastic, about 70-75% syndiotactic. Because of its lack of' complete stereoregulasity and its bulky side groups, it is amorphous Both isotactic and syndiotactic poly(methy1 methacrylate) have been prepared but have not been offered commercially. Poly(methy1 methacrylate) is resistant to many aqueous inorganic reagents, including dilute alkalis and acids It is quite resistant to alkaline saponification, in contrast to the polyacrylates. It undergoes pyrolysis almost completely to monomer by a chain reaction Perhaps the outstanding property of poly(methy1 methacrylate) is its optical clarity and lack of color. Coupled with its unusually good outdoor weathering behavior, its optical properties make it highly useful in all applications where light transmission is important The mechanical and thermal properties of' the polymer are good also. Tensile strength ranges as high as 10,000 psi. Impact strength is about equal to that of the impact-resistant styrene copolymers. Heat-deflection temperatures are above 90C for the heat-resistant grades of' poly(methy1 methacrylate) molding powder.. Electrical properties are good but not outstanding. Fabricability is quite good; only slightly higher temperatures are needed fbr molding poly(methy1 methacrylate) than for polystyrene. Poly(methy1 methacrylate) is less susceptible to crazing than is polystyrene. A limitation to the optical uses of' the material is its poor abrasion resistance compared to glass. Despite considerable effort, attempts to improve the scratch resistance or surface hardness of' poly(methy1 methacrylate) have so far been accompanied by deterioration in other properties, such as impact strength.. Applications. Injection-molded acrylic articles include automotive lenses, reflective devices, instrument and appliance covers, optical equipment, and home
ACRYLIC POLYMERS
389
furnishings. Acrylic sheeting is used for signs, glazing (in particular, aircraft windows), furniture, partitions, and lighting-fixture diffusers About 0.5 billion lb of acrylic plastics was sold in 1982 Copolymers of methyl methacrylate, ethyl acrylate, and monomers containing reactive functional groups are widely used as thermosetting resins in baked enamel applications The functionality can be derived from amides (acrylamide, methacrylarnide), acids (acrylic acid, methacrylic acid, or others), hydroxyls (hydroxyalkyl acrylates or methacrylates), or oxiranes (glycidyl methacrylate).
Other Acrylic Plastics Higher Methacrylates. The monomers of higher alkyl methacrylates are most conveniently prepared from methyl methacrylate by alcoholysis The reaction is carried out at 150C with an excess of the alcohol whose methacrylate ester is desired and a small amount of sulfuric acid catalyst As described in Chapter 12B, the higher alkyl methacrylate polymers have glass transition temperatures lower than that of poly(methy1 methacrylate). Poly(1auryl methacrylate) is widely used as a pour-point depressant and improver of viscosity-temperature characteristics for lubricating oils. Polyacrylates. The rubbery, adhesive nature of the lower acrylates makes both suspension polymerization and casting less feasible than with the methacrylates Consequently, solution and emulsion polymerizations are used commercially Since the polyacrylates contain an easily removed tertiary hydrogen atom, they undergo some chain transfer to polymer when polymerized to high conversion This leads to highly branched, less soluble materials Since the lower acrylate polymers have glass transition points below room temperature, they are typically soft and rubbery As the size of the ester group decreases, the polymers become harder, tougher, and more rigid The polyacrylates have been used in finishes, in textile sizes, and in the manufacture of pressure-sensitive adhesives Copolymers of ethyl acrylate with methyl methacrylate (to provide hardness and strength) and small amounts of hydroxyl, carboxyl, amine, or amide comonomers (to provide adhesion and thermosetting capability) are used to produce high-quality latex paints for wood, wallboard, and masonry in homes Other copolymers are used for automotive and appliance coatings and a variety of specialty coatings, finishes, polishes, and adhesives Copolymers of ethyl acrylate with a few per cent of a chlorine-containing monomer such as Zchloroethyl vinyl ether have elastomeric properties Vulcanization reactions apparently involve the chlorine atom, ester group, and a-hydrogens on the chain Numerous rubber vulcanization agents and accelerators vulcanize these polymers They are of interest because of their heat resistance and their excellent resistance to oxidation, allowing their use to over 180C Poly(acrylic acid) and Poly(methacry1ic acid). Acrylic acid, CH,=CHCOOH, can be prepared directly from ethylene cyanohydrin Methacrylic acid can be prepared from acetone cyanohydrin Salts of the acids can be polymerized directly,
390
OTHER C:ARBON-CHAIN POLYMERS
then acidified to give poly(acry1ic acid) and poly(methacry1ic acid) Alternatively, poly(methy1 acrylate) or poly(methy1 methacrylate) may be saponified to form the acids Or polymethacrylonitrile may be treated with hydrogen chloride to give poly(methacry1ic acid). The polymers are insoluble in their monomers and in most organic liquids, but are soluble in water and very soluble in dilute bases. Poly(acry1ic acid) and poly(methacry1ic acid) are too water sensitive to serve as plastics They are brittle when dry, and on heating do not become thermoplastic but crosslink, char, and decompose In solution they show typical polyelectrolyte behavior, including abnormally high viscosities Because of this property they are useful as thickening agents for latices and for adhesives
:
t
Cyanoacrylate Adhesives. Methyl cyanoacrylate monomer is an extremely powerful adhesive Adhesion occurs when the liquid monomer is spread in a thin layer between the surfaces to be bonded Traces of bases (even as weak as alcohols or water) on the surfaces catalyze polymerization by an anionic mechanism Adhesion arises in part from mechanical interlocking between polymer and surface, and in part from strong secondary bond forces Polyacrylamide.
as a flocculant This water-soluble polymer is used as a thickening agent and
Acrylic Fibers
The acrylic fibers are polymers containing at least 85% acrylonitrile Other monomers are often used in small amount to make the polymer amenable to dyeing with conventional textile dyes Common comonomers are vinyl acetate, acrylic esters, and vinyl pyrrolidone The last may be used as the homopolymer in a blend or grafted onto a copolymer backbone A generic class of fibers closely related to the acrylics is the modacrylic fibers, containing 35-85% acrylonitrile and, usually, 20% or more vinyl chloride or vinylidene chloride
Polyacrylonitrile. Acrylonitrile can be made either by the direct catalytic addition of' HCN to acetylene, or by the addition of' HCN to ethylene oxide to give ethylene cyanohydrin, fbllowed by dehydration. The monomer is soluble in water to the extent of about 7.5% at room temperature, and polymerization is usually carried out in aqueous solution by means of redox initiation. The polymer precipitates fiom the system as a fine powder. During the time that other vinyl polymers were being developed commercially as plastics as well as fibers, polyacrylonitsile was considered a useless material because it could not be dissolved or plasticized.. It softens only slightly below its decomposition temperature, and, because the polymer is insoluble in the monomer, it could not be polymerized into useful shapes by casting. This intractability was fbr a time attributed to crosslinking, but better knowledge of the requirements fbx. solubility of' polymers led to an extensive search (Houtz 1950) fbr molecules that might interact with the highly polar - e N groups and cause solution of the polymer. Several such solvents were discovered, among them dimethyl fbrmamide
POLY(V1NYL ESTERS) A N D DERIVED POLYMERS
391
and tetramethylene sulfbne. Some concentrated aqueous solutions of salts, such as calcium thiocyanate, also dissolve polyacrylonitrile . With these solvents available, both wet- and dry-spinning techniques (Chapter 18) have been used fbr this polymer..
Properties of Acrylic Fibers. The acrylic fibers exhibit the p~opertiesof high strength, stiffness, toughness, ab~asion resistance, resilience, and flex life associated with the synthetic fibers as a class They a e relatively insensitive to moisture and have good resistance to stains, chemicals, insects, and fungi Their weatherability is outstandingly good In continuous filament form, they are considered to have superior "feel," while as crimped staple they are noted for bulkiness and a woollike "hand " In many respects the acrylic fibers are similar to the chlorine-containing fibers described in Section D, and to the condensation fibers discussed in Chapter 15 Further information on fibers and fiber technology is found in Chapters 15 and 18
GENERAL REFERENCES
Bamford 1964; Davis 1964; Fram 1964; Luskin 1964; Miller 1964; Thomas 1964; Rebenfeld 1967; Beevers 1968; Kennedy 1968; Moncrieff 1975, Chapters 28, 32--37; Elias 1977, Chapter 25 8; Brydson 1978, Chapter 14; Cooper 1978; Hobson 1978; Kine 1978, 1981; Peng 1978; Morxis 1978; Vial 1979; Krenz 1982
C.
POLY(VINY1. ESTERS) A N D DERIVED POLYMERS
The most widely used polymer of a vinyl ester is poly(viny1 acetate). It is utilized not only as a plastic, primarily in the form of emulsions, but also as the precursor for two polymers which cannot be prepared by direct polymerization, poly(viny1 alcohol) and the poly(viny1 acetals) None of these polymers is useful for molding or extrusion, but each is important in certain special applications
Poly(vinyl acetate) Monomer.
Vinyl acetate,
is prepared by the vapor.-,phaseaddition of' acetic acid to acetylene:
Almost 2 billion Ib/yr was produced in 1979-1982, virtually all being used as polymers..
392
Polymerization. Bulk polymerization of vinyl acetate is difficult to control at high conversions, and in addition the properties of' the polymer may deteriorate because of' chain branching. Bulk or solution polymerization is usually stopped at low to medium conversion (20-50%), after which the monomer is distilled off'. If a solvent (such as methanol, ethanol, ethyl acetate, or benzene) is present, the polymer may be precipitated out or further reacted in solution to one of the derived polymers. The polymerization may be done batchwise or continuously Poly(viny1 acetate) may also be made in suspension or emulsion systems. In suspension polymerizations, various additives must be used to coat the beads of polymer to prevent their sticking together during the dr;ying operation. Most of the poly(viny1 acetate) emulsions commercially available are made by emulsion polymerization mechanisms (Chapter 6 B ) . Properties and Uses. This polymer is atactic and hence amorphous; stereoregular polymers have not been offered commercially The glass transition temperature of poly(viny1 acetate) is only slightly above room temperature (29C) As a result the polymer, while tough and form stable at room temperature, becomes sticky and undergoes severe cold flow at only slightly elevated temperatures Lower-molecularweight polymers are brittle but become gumlike when masticated and in fact are used in chewing gums Poly(viny1 acetate) is water sensitive in certain physical properties, such as strength and adhesion, but does not hydrolyze in neutral systems Aside from the production of poly(viny1 alcohol), one major use of poly(viny1 acetate) is in the production of water-based emulsion paints The low cost, stability, quick drying, and quick recoatability of emulsion paints have led to their wide acceptance Poly(viny1 acetate) is often copolymerized with dibutyl fumarate, vinyl stearate, 2-ethylhexyl acrylate, or ethyl acrylate, or is plasticized, to obtain softer compositions for emulsion use Poly(viny1 acetate) is also widely used in adhesives, both of the emulsion type and of the hot-melt type Poly(viny1 alcohol)
Since vinyl alcohol is unstable with respect to isomerization to acetaldehyde, its polymer must be prepared by indirect methods
Preparation. Poly(viny1 alcohol) is prepared by the alcoholysis of poly(viny1 acetate) (the less accurate terms hydrolysis and saponlficatlon are also used):
pOL.Y(VINYL ESTERS) AND D E R I V E D POLYMERS
393
Ethanol or methanol can be used to effect the alcoholysis, with either acid or base as catalyst.. Alkaline hydrolysis is much more rapid. The acid hydrolysis is more likely to give some ether linkages in the chain by a mechanism involving the loss of' a molecule of' water from two adjacent hydroxyl groups.. This is an undesirable side reaction. The alcoholysis is usually carried out by dissolving the poly(viny1 acetate) in the alcohol, adding the catalyst, and heating. The poly(viny1 alcohol) precipitates from the solution.
Structure and Properties. Although poly(viny1 alcohol) is amorphous when unstretched, it can be drawn into a crystalline fiber, the hydroxyl groups being small enough to fit into a crystal lattice despite the atactic chain structure Poly(viny1 alcohol) does not melt to a thermoplastic, but decomposes by loss of water fxom two adjacent hydroxyl groups at temperatures above 150C Double bonds axe left in the chain, and, as more of these are formed in conjugate positions, severe discoloration takes place Poly(viny1 alcohol) is water soluble It dissolves slowly in cold water, but more rapidly at elevated temperatures, and can usually be dissolved completely above 90C The aqueous solutions are not particularly stable, especially if traces of acid or base are present The solutions can undergo a complex series of reversible and irreversible gelation reactions FOI example, crossllnking can occur at ether linkages, resulting in incxeased viscosity through the formation of insoluble products
Applications. The major use of poly(viny1 alcohol) is as a textile size, particularly for cotton-polyester blends Other used fall into two categories In one type of application, use is made of the water solubility of the polymer It serves as a thickening agent for various emulsion and suspension systems, and as a packaging film where water solubility is desired A further outlet is wet-strength adhesives In the second type of end use, the final form of the polymer is insoluble in water as a result of chemical treatment The use of poly(viny1 alcohol) as a textile fiber (vmal fiber) is the major example The polymer is wet spun from waIm water into a concentrated aqueous solution of sodium sulfate containing sulfuric acid and formaldehyde The polymer is insolubilized by the formation of formal groups:
-CH2-CH-CH2--CH--
OH
OH
+ HCIIO
394
About one-third of' the hydroxyl groups is reacted to insolubilize the fiber.. Some interchain acetalization is desirable to reduce shrinkage of'the fiber, but the amount must be carefhlly controlled. Poly(viny1 alcohol) fibers have higher water absorption (about 30%) than other fibers. They can thus replace cotton in uses where the fiber is in contact with the body. The hand of' the fabric can be varied from wool-like to linenlike. The fiber washes easily, dries quickly, and has good dimensional stability. Tenacity and abrasion resistance are good
Poly(viny1 acetals)
An important use of poly(viny1 alcohol) is as the starting material for poly(viny1 acetals). By far the most important of' these polymers is poly(viny1 butyral), used as the plastic interlayer for automotive and aircraft safety glazings Lower poly(viny1 acetals), especially poly(viny1 formal), are utilized in enamels fbr. coating electrical wire and in self-sealing gasoline tanks
Poly(viny1 butyral). This polymer is made by condensing poly(viny1 alcohol) with butyraldehyde in the presence of' an acid catalyst, usually sulfuric acid:
The reaction can be carrled out by starting with an aqueous solution of poly(viny1 alcohol) and adding the butyraldehyde and catalyst The poly(viny1 butyral) precipitates as it is formed Alternatively, one may start with a suspension of poly(viny1 alcohol) in a water-alcohol mixture, the poly(viny1 butyral) dissolving as it is formed Poly(viny1 butyral) produced for safety-glass use requires some hydroxyl groups in order to have adequate strength and adhesion to glass Consequently the condensation reaction is stopped when about onequarte~ the hydroxyl groups is of * left
Safety Glass. The earliest safety-glass interlayer was cellulose nitrate. Its poor weathering qualities led to its general replacement in 1935 by cellulose acetate, and in 1940 by poly(viny1 butyral). Some of' the advantages of the latter resin as an interlayer are its superior adhesion to glass, toughness, stability on exposure to sunlight, clarity, and insensitivity to moisture
CHLORINE-CONTAINING P L M R OY E S
395
Poly(viny1 butyral) must be plasticized for safety-glass use The usual plasticizers are high-molecular-weight esters such as dibutyl sebacate and triethyleneglycol di7,-ethylbutyrate. About 40-45 parts plasticizer per 100 parts resin by weight is used The plasticizer may be incorporated in the resin in several ways One common method is to mix the resin, plasticizer, and a solvent such as ethanol to form a plastic mass This is extruded in sheet form into a saltwater bath where the solvent IS leached out, leaving the plasticized sheet. Since the plasticized resin sticks to itself, it must be powdered with talc or sodium bicarbonate before being rolled up for storage. Safety-glass laminates are made by washing and drying the poly(viny1 butyral) sheet, and placing it between two pieces of glass These are subjected to mild heat and pressure to seal the assembly, and are then autoclaved at higher temperature and pressure to complete the lamination process The usual automotive safety glass contains one plastic layer about 0 030 in thick
GNR L RFRNE E E A EE E CS
Pritchard 1970; Lindemann 1971a,b; Elias 1977, Chapter 25..5
D. CHLORINE-CONTAINING POLYMERS
Polymers and copolymers of vinyl chloride and vinylidene chloride are widely used as plastics and, to a much smaller extent, fibers In the trade, these polymers are known as "the vinyls" or "vinyl resins " Once the largest group of thermoplastic materials, the vinyl resins have long since been surpassed in volume by the olefin polymers Production of the vinyls in 1982 was about 5 35 billion Ib, down from a 1979 peak of 6 1 billion Ib Prices were in the range $0 32-$0 50 per lb.
Poly(viny1 chloride) Monomer. Vinyl chloride is a gas boiling at - 14'C It is produced by the dehydrochlorination of ethylene dichloride, which is made by reacting ethylene with chlorine. Almost all the vinyl chloride produced is used in polymerization, with great care being taken throughout because of' the carcinogenic nature of' the monomer. Polymerization. Suspension polymerization (Chapter 6B) is used for the production of well over 80% of poly(viny1 chloride) Small amounts are made by solution, emulsion, and bulk polyme~ization,despite difficulties in the latter process ~esulting from the insolubility of the polymer in its monomer Struc,ture. Poly(viny1 chloride) is a partially syndiotactic material, with sufficient irregularity of'structure that crystallinity is quite low. Its structural characterization
is complicated by the possibility of chain branching and the tendency of the polymer to associate in solution
Stability. Poly(viny1 chloride) is relatively unstable to heat and light. Thermal
f I
I
I
initiation involves loss of a chlorine atom adjacent to some structural abnormality which reduces the stability of the C-Cl bond, such as terminal unsaturation (Grassie 1964). The chlorine radical so fbrmed abstracts a hydrogen to fbrm HCI; the resulting chain radical then reacts to form chain unsaturation with regeneration of' a chlorine radical. The reaction can also be initiated by ultraviolet light which is absorbed at unsaturated structures with liberation of an adjacent chlorine atom In the presence of oxygen, both chain reactions are accelerated, and ketonic structures are fbrmed in the chain. Stabilizers are almost invariably added to improve the heat and light stability of the polymer. Metallic salts of lead, barium, tin, or cadmium are used. Oxides, hydroxides, or fatty-acid salts are most effective. Epoxy plasticizers (Chapter 160) aid materially in stabilizing the resin. Free radical acceptance appears to be a prominent mechanism of stabilization HCI acceptors have been used, but the degradation reactions are not autocatalytic as had previously been supposed.
Vinyl Resins
! I
i
1
'
Rigid Compounds. The term rrgzd vzayls usually refers to unplasticized poly(viny1 chloride), or compositions with only a few percent of a plasticizer such as an epoxy resin (Chapter 160) Often other polymers are mixed physically with the poIy(viny1 chloride) to improve impact resistance (nitrile rubber, chlorinated polyethylene, ABS resins, or methyl methacrylate-butadiene-styrene terpolymer) or processability (styrene-acrylonitrile copolymer or poly(methy1 methacrylate)).
Copolymers. The advantages in poIymer properties resulting from the copolymerization of small amounts of vinyl acetate with vinyl chloride were discovered around 1928. The lower softening point and higher solubility of' the copolymers make fabrication very much easier. Stability is improved, if anything, over that of the homopolymer. Color and clarity are also better. Polymerization methods are similar to those for vinyl chloride homopolymer except that emulsion polymerization has had less success. Commercial compositions containing 5-40% vinyl acetate are available. Polymers containing around 10% vinylidene chloride have better tensile properties than pure poly(viny1 chloride). Copolymers containing 10-20% diethyl fumarate or diethyl maleate have improved workability and toughness and retain the high softening point of poly(viny1 chloride). Acrylic esters have also been used to impart improvements in solubility and workability. Plasticization. Many properties of poly(viny1 chloride) and vinyl chloride-vinyl acetate copolymers are improved by plasticization (Chapter 17) The large majority of the commercial production of vinyl resins is in the form of plasticized compo-
C:HL.ORINE-C.ONTAININGPOLYMERS
397
/
i
sitions The first important plasticizer for the vlnyls was tricresyl phosphate, which has since been replaced by other esters because of its tendency to cause lowtemperature brittleness in plasticized compounds Dibutyl phthalate, dibutyl sebacate, and tributyl phosphate have also been used and in turn replaced Dioctyl phthalate, trioctyl phosphate, dioctyl sebacate and adipate, and various low-molecular-weight polymers such as poly(propy1ene glycol) esters are now widely utilized as plasticizers for the vinyls The plasticizers are usually added to the polymers on hot rolls or in a hot mixer such as a Banbury The plasticizer content varies widely with the end use of the material, but typically may be around 30% by weight
Plastisols and Organisols. These liquid compositions are produced by spraydrying latices obtained from the emulsion polymerization of vinyl chloride The latex particles are dispersed into plasticizers to make plastisols, or into mixtures of plasticizers and volatile organic liquids to make organisols Other ingredients, such as stabilizers, fillers, colorants, surfactants, and possibly blowing agents or gelling agents, are also present The polymer particles do not dissolve in the liquids, but remain dispersed until the mixture is heated Fusion (plus loss of solvent from the organisols) then yields the final plastic object Applications. About 55% of poly(viny1 chloride) is used as rigid resins, the remainder plasticized The single largest use, in the rigid field, is pipe, accounting for 40% of production It is used for water supply and distribution, agricultural irrigation, chemical processing, drain, waste, and vent pipe, sewer systems, and conduits for electrical and telephone cables The building construction market accounts for about another 30% of poly(viny1 chloride) production, including siding, window frames, gutters, and interior molding and trim (12%); flooring (5%); wire and cable insulation (7%); and wall coverings, upholstery, shower curtains, gaskets, and so on (5%) Automotive uses, meat and food packaging, bottles, footwear, outerwear, phonograph records, sporting goods, and toys are other significant markets The major plastics application for poly(viny1idene chloride) is saran film Chlorine-Containing Fibers Compositions. The use of' poly(viny1 chloride) as a fiber was patented as early as 1913, and the material was commer.cialized in 1931. It was soon withdrawn in favor of modified polymers with better solubility, including chlorinated poly(viny1 chloride) and copolymers of' vinyl chloride with vinylidene chloride and with vinyl acetate Cur~ently, materials of most interest in the class of' chlorine-containing the fibers are copolymers of' about 60% vinyl chloride and 40% acrylonitrile, and of' about 8.5% vinyl chloride and 15% vinyl acetate. Other comonomers are occasionally used.. The generic classes of fibers in this group include vinyon, containing more than 85% vinyl chloride; saran, containing more than 80% vinylidene chloride; and modacrylic, defined in Section B .
398
Properties and Uses. The chlorine-containing fibers are usually moderately crystalline and are drawn (typically while hot) 5- to 1.5-fold to increase strength and reduce extensibility Like the acrylic (Section B) and condensation (Chapter 15) fibers, they are insensitive to moisture and have high wrinkle resistance and good resistance to chemicals, insects, fungi, and the like They have poorer dimensional stability at high temperatures, and somewhat less strength, elasticity, and abrasion resistance than the condensation fibers
GENERAL REFERENCES
Kaufinan 1969; Koleski 1969; Sarvetnick 1969; Brighton 1971; Wessling 1971; Moncrieff 1975, Chapters 27,29, and 30; Nass 1976-77; Burgess 1982; Brown 1982; .Jeziorski 1982
E.
FLUORINE-CON'TAINING POLYMERS
Fluorine-containing polymers represent in many respects the extremes in polymer properties Within this family are found materials of high thermal stability and concurrent usefulness at high temperatures (in some cases combined with high crystalline melting points and high melt viscosity), and extreme toughness and flexibility at very low temperatures Many of the polymers are almost totally insoluble and chemically inert, some have extremely low dielectric loss and high dielectric strength, and most have unique nonadhesive and low friction properties In this section we distinguish between fully fluorinated fluorocarbon plastics, including only polytetrafluoroethylene and its fully fluorinated copolymers, and fluoroplastics, containing hydrogen or chlorine in addition to fluorine on the carboncarbon backbone
Polytetrafluoroethylene Monomer. Tetrafluoroethylene is a nontoxic gas boiling at -.76"C. Its modern synthesis starts with the fluorination of chlosoform by HF in a two-step process yielding CHCIF, This is then dimerized by pyrolysis with loss of' HCI to provide tetrafluosoethylene Polymerization. The first polymer of tetrafluoroethylene was discovered by
1
i
R. I. Plunkett (Garrett 1962) when an apparently empty cylinder of the gas was
cut open to see why more material had not been obtained from it. The cylinder was partly filled with a waxy white powder shown to be the polymer. Tetrafluoroethylene is usually polymerized with free radical initiators at elevated pressure in the presence of water Redox initiation may be used; persulfates and hydrogen peroxide have been employed as initiators
1!
1
FLUORINE-CONTAINING POLYMERS
399
Structure (Sperati 1962). Polytetrafluoroethylene is a highly crystalline, orientable polymer.. These facts indicate a regular structure, which implies the absence of' any considerable amount of' crosslinking.. Branching is presumed to be absent, since branching mechanisms would involve breaking C-F bonds, which are estimated to be too strong to be ruptured easily. The chances are, therefore, that polytetrafluoroethylene consists of linear -CF,--CF,-chains The crystal structure and crystalline-.phase transitions in polytet~afluoroethylene are discussed in Chapter 10B. These transitions, occurring near Ioom temperature, involve a 1..3%volume change having an important effect on the mechanical p r o p erties of' the polymer for some applications. The degree of' crystallinity of the polymer as fbrmed from the monomer is generally quite high, 93-98% The crystalline melting point is 327C. The crystalline density of polytetrafluoroethylene is 2..30 gicm3. Upon melting the polymer and subsequently cooling it, lower densities are obtained. Subsequent thermal treatment (annealing) generally increases the crystallinity of the solid. A standardized molding and annealing technique under conditions carefully controlled to prevent the formation of' voids in the sample gives "standard" specific gravities for various samples ranging between 2..15 and 2..28. This range of densities has been correlated with the molecular weight and thus the melt viscosity ofthe samples.. Higher-molecular-weight polymers have more viscous melts and hence crystallize more slowly and reach lower crystallinities under standardized thermal treatment.. Despite the reputation of' polytetrafluoroethylene as a totally insoluble polymer, solvents for it have been found at temperatures not f8r below its crystalline melting point.. Solution viscosities have been measured in perfluorinated kerosene fractions at 300C, the measurements giving clues as to the molecular weight of the polymer. Other infbrmation as to molecular. weight is derived from the kinetics of polymerization of' tetrafluoroethylene using radioactive initiators, and from the flow properties of' the polymers (see the next paragraph). All the evidence points to unusually high number-average molecular weights of many millions for this polymer.. Polytetrafluoroethylene does not flow easily above its crystalline melting point.. The viscosity of the polymer is very high because of restricted rotation about the chain bonds and high molecul~.weight Upon the continued application of stress to the amorphous polymer above the crystalline melting point, the strength ofthe polymer is exceeded, causing fracture befbre the stresses necessary to induce rapid flow are reached This unusual behavior necessitates unconventional fabrication techniques, as discussed below Polytetrafluoroethylene decomposes at elevated temperatures. In vacuum, man,omer is the chief' p~oduct. vapor pressure of monomer in equilibrium with the The polymer at 500C is 1 mm Hg. At low temperatures (250-350C) degradation appears to start at chain ends; at higher temperatures random cleavage becomes more im.portant. In the presence of air the degradation is more complicated. Properties. Polytetrafluoroethylene is extremely resistant to attack by corrosive reagents or solvents. Of many hundreds of reagents tested up to their boiling points, only alkali metals, either molten or dissolved in liquid ammonia, attack the polymer,
400
presumably by removing fluorine atoms from the chain Fluorine itself degrades the polymer on prolonged contact under pressure. For all practical purposes the polymer is completely unaffected by water.. Its thermal stability is such that its electrical and mechanical properties do not change for long intervals (months) at temperatures as high as 250C Molded polytetrafluoroethylene articles have high impact strength but are easily strained beyond the point of' elastic recovery. In a tensile test molded articles begin to cold draw at 1500--2000 psi, elongating (with orientation) to 300-450% before breaking at a load of' 2500-4500 psi. Highly oriented fibers have tensile strengths as high as 50,000 psi. Polytetrafluoroethylene is subject to cold flow under stress but exhibits some delayed elastic recovery. The polymer is not hard, but is slippery and waxy to the touch, and has a very low coefficient of' friction on most substances Its density is unusually high (2.1-2 3) and its refiactive index unusually low (1 375) Polytetrafluoroethylene has extremely good electrical properties. Its dielectric constant is low (2 O), and its loss factor for all frequencies tested, including those used for television and radar, is one of' the lowest known for solids. The properties of' polytetrafluoroethylene can be varied widely as a function of molecular structure and hbrication, as described in Chapter 12.
Fabrication. Since polytetrafluoroethylene is almost completely insoluble and has impracticably low melt flow rates, most of the fabrication techniques ordinarily used with polymers (Chapter 17) are not suitable for this material. Several unusual techniques have been developed for putting polytetrafluoroethylene into usable shapes. Most of these techniques are variants of two important processing steps: (a) pressing the polymer cold into the desired shape, fbllowed by (b) sintering at a temperature above the crystalline melting point (say, 380C) to yield a dense, strong, homogeneous piece. Techniques for molding polytetrafluoroethylene resemble those of' powder metallurgy or ceramics processing more than those of' polymer hbrication. Granular polytetrafluoroethylene powder is compressed at room temperature and 2000-10,000 psi to give a prefbrm Sintering of the pref'orm follows Polytetrafluoroethylene may be extruded at slow rates in a ram or screw extruder In these devices the ram or screw serves merely to compact the molding powder and feed it to the die, within which the material is heated above the crystalline melting point, shaped, and allowed to solidify. The high degree of' cohesion between cold pressed particles of' polytetrafluoroethylene is utilized in a calendering process for making tape and coating wire. Molding powder is ftd to calender rolls, which compress the particles into a solid structure.. This passes into a sintering bath or oven Polytetrafluoroethylene is available in the fbrm of aqueous dispersions of' ultimate particles about 0.2 pm in diameter.. They can be used fbr casting films or for dip coating or impregnating porous structures. Mechanical properties of' polytetrafluoroethylene, including resistance to wear and to deformation under load, stiffness, and compressive strength, can be enhanced by the use of fillers. Most of the desirable properties of the polymer (heat resistance,
F~~~RINE-CONTAINING POLYMERS
401
low friction, weatherability, etc ) are retained Short glass fibers, graphite, carbon, and bronze are prefer~edfillers pojytetrafluoroethylene can also be provided in a microfibrous form, which can be spun from aqueous dispersions to provide fibers that can be sintered at around 385C to develop low but useful strength
~ ~ ~ l i c a t i o nThe uses of polytetrafluoroethylene are largely those requiring exs. cellent toughness, electrical properties, heat resistance, low frictional coefficient, a combination of' these. Among the electrical applications of the polymer are ,ire and cable insulation, insulation fbr motors, generators, transfbrmers, coils, and capacitors, and high-frequency electronic uses. Chemical equipment such as gaskets, pump and valve packings, and pump and valve parts is made from the polymer. Low-friction and antistick applications include nonlubricated bearings, linings for trays in bake~iesand other hod-processing equipment, mold-release devices, and covers fbr heat-sealer plates of packaging machines. Low-molecularweight polymers of' tetrafluoroethylene dispersed as aerosols are effective dry lubricants. Uses fbr the fiber include gasketing, belting, pump and valve packing, filter cloths, and other industrial functions where essentially complete chemical and solvent resistance, combined with heat resistance to above 250C, is required
Fluorocarbon Copolymers
A copolymer of hexafluoropropylene and tetrafluoroethylene has a crystalline melting point near 265C It retains most of the properties of polytetrafluoroethylene, but has melt viscosity in the range allowing fabrication by conventional techniques This copolymer is tough at liquid-air temperatures, yet retains adequate mechanical strength for continuous service at temperatures up to 200C Like polytetrafluoroethylene, it is chemically inert and has zero water absorption It maintains a low d~electric constant and loss factor over the unusually wide range of 60 cycles to 60 megacycles It has excellent weatherability, nonstick and low friction properties, and very low permeability to gases The material can be processed by conventional extrusion, injection molding, compression, transfer, or blow molding, fluidized bed coating, or vacuum forming of sheets or film It finds application as wire jacketing, as extruded film, rods, tubes, or complex shapes, and as an encapsulating ~esin Important end uses, many of which are similar to those of polytetrafluoroethylene, include film for inner glazing in solar collectors Copolymers of per fluor oalkox y monomers,
with tetrafluoroethylene are very close in properties to polytetrafluoroethylene, with T, = 310C, but are melt processible by conventional molding and extrusion tech-
402
OTHER CARBON-CHAIN
POLYMERS
niques, All the outstanding properties of the homopolymers are retained, including a recommended upper use temperature of 260C; for properties related to defbrmation, the copolymer is superior.
Polychlorotrifluoroethylene
Monomer. Chlorotrifluoroethylene is made by dechlorination of trichlorotrifluoroethane The monomer is less subject to spontaneous explosive polymerization than is teaafluoroethylene Unllke tetrafluoroethylene, however, chlorotrifluoroethylene is itself toxic Polymerization. As in the case of tetrafluoroethy!ene, the polymerization of chlorotrifluoroethylene is best carried out in an aqueous system using a redox initiator The details of commercial polymerizations have not been disclosed Structure and Properties. Polychlorotrifluoroethylene is surpassed only by poly-
Ii
1 / i
i
iI
tetrafluoroethylene and tetrafluoroethylene-hexafluoropropylene copolymers in chemical inertness and resistance to elevated temperatures Differences in the properties of the polymers are a consequence of the lower symmetry of the chlorine containing polymer For example, the cr ystalline melting point of polychlor otrifluoroethylene is 21SC, and the polymer can be quenched to quite clear sheets in which crystallinity is absent Slower cooling provides opportunity for growth of crystallites and spherul~tesand results in quite different optical and mechanical properties Polychlorotrifluoroethylene is soluble in a number of solvents above 100C, and is swelled by several solvents at room temperature The polymer is tough at - 100C and xetains useful properties as high as 150C Although high compared to that of many other polymers, its melt viscosity is low enough that the usual fabrication techniques, such as molding and extrusion, are practicable The electrical properties of polychlorotrifluoroethylene are inferior to those of polytetrafluoroethylene, especially for high-frequency applications, because of the more polar nature of the polymer This material also has outstanding barrier properties to gases and is cornpatible with liquid oxygen The combination of these unusual properties makes polychlorotrifluoroethylene ir~eplacablein several applications
Poly(viny1 fluoride)
i
i
i
1 i
i
J1 1
1
;
1 1
E
L
PoIy(viny1 fluoride) is a highly crystalline plastic which is commercially available in the form of a tough but flexible film The polymer has the outstanding chemical resistance of the fluoroplastics and excellent outdoor weatherability It is extremely resistant to thermal degradation and maintains usable strength above 150C while remaining tough at - 180C. It has low permeability to most gases and vapors and resists abrasion and staining
,
1
FLUORINE-CONTAINING POLYMERS
403
The film has wide use as a protective coating in the building industry. In 0.0010.002 in. thickness, bonded to wood, metal, or asphalt-,based materials, it lasts many times longer than paints, enamels, or other surface coatings
~oly(viny1idene fluoride)
Poly(viny1idene fluoride) is a crystalline plastic with a melting point of' about 160C It has good strength properties and resists distortion and creep at both high and low temperatures. It has very good weatherability and chemical and solvent resistance.. This polymer can be processed by extrusion and by compression and injection molding. It is used as a coating, gasketing, and wire- and cable-jacketing material, and in piping, molded and lined tanks, pumps, and valves in the chemical and nuclear power industries
Fluoroplastic Copolymers
Copolymers of ethylene with tetrafluoroethylene and with chlorotrifluoroethylene have essentially 1 : 1 alternating structures (Chapter 5A) They have high melting points (270 and 245"C, respectively), and retain much of the superior chemical resistance and good weatherability of the fully fluorinated materials In other respects, however, their properties are more like those of the engineering thermoplastics, with good strength, toughness, and wear resistance over wide temperature ranges Copolymers of chlorotrifluoroethylene and vinylidene fluoride range from tough, flexible thermoplastics to elastomers, depending on composition One of their outstanding properties is their resistance to the attack and penetration of powerful ox~dizing agents such as propellant-grade red-fumlng nitric acid and 90% hydrogen peroxide, typical of the fluorocarbons, coupled with the toughness and flexibility required for use as valves, ring seals, caulking compounds, and so on They also find application in automobile transmissions and brake systems as gaskets and seals Copolymers of hexafluoropropylene and vinylidene fluoride are elastomers combining high resistance to heat and to fluids and chemicals with good mechanical properties They are cured with amine-type systems These materials can be compounded to remain serviceable for short periods as high as 300C, and retain useful properties indefinitely at 200C Their resistance to the lubricants, fuels, and hydraulic fluids used in jet aircraft is unequalled by other elastomers Their mechanical properties are respectable for any elastomer, and excellent among the oil-resistant
GENERAL REFERENCES
Brown 1967; Towler 1969; McCane 1970; Cohen 1971; Dohany 1971; Wall 1972; Moncrieff 1975, Chapter 40; Robertson 1976; Brydson 1978, Chapter 13, West 1979, 1980; Brasure 1980; Dohany 1980; Gangal 1980; Johnson 1980; Kim 1980; Sperati 1982
404
1.
2.
Compare the behavior of polystyrene, poly(viny1 chloride), and polyacrylonitrile on thermal degradation, writing appropriate chemical equations Write chemical equations for (a) the polymerization of poly(viny1 acetate), (b) the formation of poly(viny1 alcohol), (c) the crosslinking of poly(viny1 alcohol) fibers, and (d) the formation of poly(viny1 butyral) Postulate, and write equations for, long-chain branching in poly(viny1 acetate) arising at three different sites Illustrate the fate of these branches as the polymer is hydrolyzed and reacetylated
ii
3.
4. Discuss the relative physical properties of polystyrene, impact polystyrene, and ABS resins, and account for differences in terms of molect~larstructure
5. Discuss (a) thermal stability and stabilization and (b) the effect of plasticization on thermal and physical properties, in vinyl resins
6. Explain in terms of molecular structure the following properties of polytetrafluoroethylene: (a) melting point, (b) solubility, (c) melt flow behavior, (d) chemical inertness, and (e) coefficient of friction
BIBLIOGRAPHY
Abrams 1967 Irvrng M Abrams and Leo Benzera, "Ion-Exchange Polymers," pp 692-742 In EPST, i Vol 7, 1967 Bamford 1964 C H Bamford, G C Eastmond, Gerlitd P Ziemba, and A Levovits, "Acrylonrtr~le Polymers," pp 374-444 in EPST, Vol 1, 1964 Bardekrs 1964 C H Basdekrs, ABSPlasr~cr, Rernhold, New York, 1964 Beeverr 1968 R B Beevers, "The Physical Propertres of Polyacrylonrtrrle and Its Copolymers," Macromol Rev 3, 113-254 (1968) Boyer 1970 Raymond F Boyer, w ~ t h E Coulter, J L Duda, Edward F Gumee, Kon Sup Hyun, K H Kehde, Henno Keskkula, Alan E Platt, and J S Vrentas, "Styrene Polymers" pp 128-447 In EPST, Vol 13, 1970 Brasure 1980 Donald E Brasure, "Poly(Vmyl Fluoride)," pp 57-64 In ECTI Vol 11, 1980 Brrghton 1971 C A Brighton, wrth J L Benton, J P Dux, and G C Marks, "Vrnyl Chlonde Polymers," pp 305-483 in EPST. Vol 14, 1971 Brown 1967 Henry C Brown and Robert P Bnnger, "Fluorine Contarn~ngPolymers," pp 179-219 In EPST, Vol 7, 1967 Brown 1982 W E Brown, "Vrnyl~deneChiorrde Polymers and Copolymers," pp 113-1 14 In MPE.9 1982
t Herman F Mark, Norman G Gaylord, and Norbert M B~kales, ,Encyclopedia ojPolymer Scienre eds and Technology, Wrley-Intersc~ence,New York f Martin Grayson, ed ,K~rk-Othmer Encyclopedia ojChemrca1 Technology, 3rd ed ,Wiley-Interscience, New York 5 Joan Agranoff, ed ,Modern Plartrcs Encjclopedia 1982-1983, Val 59, No IOA, McGraw-H111,New York, 1982
B~~~~~GRA*HY
405
Bndson 1978 J A Brydson, Rubber Chemistry, Appl~edSclence, London, 1978 Burgess 1982. R H Burgess, ed , Manufacture and Processing of PVC, Macmlllan, New Yolk, 1981 cohen 1971 Fredem S Cohen and Paul Kraft, "Vinyl Fluoride Polymers," pp 552-540 in ~ p ~ j - , Vol 14, 1971 Coob2er1978 H W Cower, Jr , and J M McIntyre, "2-Cyanoacrylic Ester Polymers," pp 408413 in ECT, Vo1 I , 1978 uul.ir 1964 C W I);iv*s and Paul Shapiro, "Acryl~cFibers," pp 342-373 In EPS7, Vol 1, 1964
I
uohany 1971 J E D o h n y , A A Dukert, and S S Preston 111, "Vlnyl~deneFluoride Polymers," pp 600-610 In EPST, Vol 14, 1971 Dohanv 1980 Jul~us Dohany, "PolylVlnyl~dene Fluoride)," pp 64-74 in ECT, Vol 11, 1980 E thus 1977 Hans-Gwrg Elias, Macromolecules 2 Synthecis and Materials (translated by John W Stafford), Plenum Press, New York, 1977 E ~ U 1982 T E Evans, "Styrene- Acrylon~tr~le," ~ S pp 116-1 17 In MPE, 1982 Fram 1964 Paul Fram, "Acrylic Elastomers," pp 226-246 in EPST, Vol 1, 1964 ~ ~ n g a l 1 9 8S V Gangal, "Polytetrafluoroethylene," pp 1-24, and "Fluonnated Ethylene-Propylene 0 Copolymers," pp 24-35 In ECT, Vol 11, 1980 Gurrett 1962 Alfred B Garrett, "The Flash of Genus, 2 Teflon. Roy J Plunkett," J Chem Educ 39, 288 (1962) Grassre 1964 N Grassie, "Thermal Degradat~on,"Chapter 8B In E M Fettes, ed , ChemrcalReactrons of Poljrners, Wlley-lntersc~ence, New York, 1964 Hobson I978 P H Hobson and A L McPeters, "Acryl~cand Modacryllc Rbers," pp 355-386 In ECT, Vol 1, 1978 Horrti 1950 R C Houtz, " 'Orlon' Acrylic Fiber: Chemistry and Proper.ties," Textile Res J 20, 786-801 (1950) Jeziorskr 1982 R J Jez~orskl R A Wenzler, "Polyv~nyland Vinyl Copolymers," pp 108, 110, and 112 In MPE, 1982 Johrtson 1980 R~chard Johnson, "'Ietrafluoroethylene Copolymers w ~ t h L Ethylene," pp 35-41 and Perfluorovlnyl Ethers," pp 42-49, in ECT, Vol 11, 1980 "I'etrafluoroethylene Copolymers w ~ t h Kaigman 1969 Morris Kaufman, 7he History of PVC, Maclaren, London, 1969 Kennedy 1968 R K Kennedy, "Modacryl~cFlbers," pp 812--839 In EPS7, Vol 8, 1968 , Klm 1980 Y K Klm, "Poly(Fluoros~l~cones),"pp 74-81 In EC7, Vol 11, 1980 Ktne 1978 Benjamin B Kine and Ronald W Novak, "Acrylic Ester Polymers," pp 386-408 In ECT, Vol 1, 1978 Kine 1981 Benjamin B Kl?e and R W Novak, "Methacrylic Polymers," pp 377-398 in ECT, Vol 15, 1981 Koleski 1969 J V Koleski and L H Wartmen, Poly(V1ny1Chloride), Gordon and Breach, New York, 1969 Kren: 1982 Heider Krenz, "Acrylic," pp 18, 21, 22 In MPE, 1982 Lantz 1982 J M Lantz, "ABS," pp 6-7 in MPE, 1982 Lindemann 1 9 7 1 ~Martin K Lindemann, "Vinyl Alcohol Polymers," pp 149-239 in EPST, Vol 14, 1971 Lfndemann 19716 Martin K Lindemann, "Vinyl Ester Polymers," pp 531-703 in EPST, Vol 15, 1971 Lidskin 1964 Leo S Luskin, Robert J Myers, J P Bruise, H W Coover, JI , and T H Wicker, JI , "Acrylic Ester Polymers," pp 246-342 in EPST, Vol 1, 1964 M~Cane1970 Donald I McCane, "Tetrafluoroethylene Polymers," pp 523-670 in EPST, Vol 13, 1970 Miller 1964 M L Miller, "Acrylic Acid Polymers," pp. 197-226 in EPST, Vol 1, 1964
406
OTHER CARBON-CHAIN P L M R OY E S
Motzcrreff 1975 R W Moncneff, Man-Made Fibers. Newnes-Butterworths, London, 1975 Morneau 1978 G A Momeau, W A Pavelich, and L G Roettger, "ABS Resins," pp 442-456 in ECT, Vol 1, 1978 Morris 1978 J D Morrls and R J Penzenstadler, "Acrylamlde Polymers," pp 312-330 In ECT, Vol 1, 1978 Nass 1976-1977 Leonard I Nass, ed , Encyclopedia ojPVC, Dekker, New York, Vol 1, 1976; Vols 2 and 3, 1977 Peng 1978 Fred M Peng, "Acrylonltnle Polymers," pp 427-442 in ECT, Vol 1, 1978 Pepprn 1982 A Peppln, "Styrene-Maleic Anhydr~de," pp 1 1 7-1 18 In MPE, 1982 Platt 1983 A E Platt and T C Wallace, "Styrene Plastics," pp 801-847 in ECT, Vol 21, 1983 Platzer 1977 Norbert Platzer, "Elastomers for Toughening Styrene Polymers," Chemtech 7, 634-641 (1977) Pritchard 1970 J G Prltchard, Poly(Vrnj1 Alcohol)-Barn Properties and Uses, Gordon and Breach, New York, 1970 Radosta 1977 Joseph A Radosta, "Improvlng the Phys~cals Impact Polystyrene," Plastrcs Eng 33 of (9), 28-30 (1977) Rebenfeld 1967 Ludwlg Rebenfeld, "F~bers," pp 505-573 In EPST, Vol 6, 1967 Robertson 1976 A B Robertson and E C Lupton, Jr , "Fluor~natedPlast~cs,"pp 260-287 In EPST, suppl Voi 1, 1976 Sarvetnrck 1969 Harold A Sarvetn~ck, Polpvrnyl Chloride, Van Nostrand Relnhold, New York, 1969 Speratr 1962 C A Sperat~ H W Starkweather, Jr , "Fluor~ne-Conta~ntng and Polymers I1 Polytetrafluoroethylene," Adv Polym Sci 2, 465-495 (1962) Speratr 1982 C A Sperat~,"Fluoroplast~cs," pp 35-39 In MPE, 1982 Swett 1982 R M Swett, "Polysty~ene," pp 98, 100, 102 In MPE, 1982 Thomas 1964 W M Thomas, "Acrylamlde Polymers," pp 177- 197 In EPST, Vol 1, 1964 Vral1979 7 M V ~ a l ,"Acrylic Elastomers," pp 458-469 In ECT, Vol 8, 1979 Wall I972 Leo A Wall, ed , Fluoropolymers, W~ley-Intersc~ence, Yolk, 1972 New Wesslrng 1971 R Wessling and F G Edwards, "Vlnylldene Chlonde Polymers," pp 540-579 m EPST, Vol 14. 1971 West 1979 Arthur C West and Allen G Holcomb, "Fluonnated Elastomers," pp 500-515 In ECT, Vol 8, 1979 West 1980 A C West, "Polychlorot~~fluoroethylene," 49-54 In ECT, Vol 11, 1980 pp
CHAPTER FIFTEEN
HETEROCHAIN THERMOPLASTICS
This chapter includes discussion of those thermoplastic materials with backbone chains in which the regula sequence of carbon atoms, seen in the materials discussed in Chapters 13 and 14, is interrupted by the presence of other atoms, notably oxygen and nitrogen Included are the polyamides and polypeptides characterized group (Section A), a variety of polymers containing oxygen in by the -CONH-the main chain (Section B), the plastics and fibers derived from cellulose (Section C), and a variety of other organic and inorganic polymers, mostly of interest for their superior high-temperature properties (Section D) Excluded are several types of polymers used exclusively, or nearly so, in crosslinked form. These materials, including the silicones and polyurethanes used as foams, are discussed in Chapter 16 The polymers discussed in the present chapte~have in common, for the most part, the high strength and toughness, stiffness and abrasion resistance, and retention of properties over wide temperature Ianges associated with fibers and with engineering plastics. It is therefore appropriate that their description be contained in a single chapter.
A.
POLYAMIDES A N D POLYPEPTIDES
Polyamides
The coined word nylon has been accepted as a generic term for synthetic polyamides The nylons are described by a numbering system that indicates the number of' carbon atoms in the monomer chains Amino acid polymers are designated by a single number, as 6-nylon for poly(w-aminocaproic acid) (polycaprolactam) . Nylons from diamines and dibasic acids are designated by two numbers, the first representing the diamine, as 66-nylon for the polymer of hexamethylenediamine and adipic acid,
408
and 610-nylon fbr that of' hexamethylenediamine and sebacic acid. Of' greatest commercial importance are 6,-and 66-nylon The production of nylon fibers in the United States was about 2..5 billion Ib, whereas production for plastics use was only about 0 2.5 billion Ib, in 1982. The drop in usage of both apparel and home furnishings in the early 1980s was reflected in a reduction in nylon production from a high of about 3 . 1 billion Ib in 1979
Development of the Nylons. The research leading to the production of synthetic fibers from polyamides began about 1928, when W H Ca~others(Mark 1940) undertook a series of researches dealing with the fundamentals of polymerization processes These studies were not aimed at the production of any specific product or process In his fundamental researches, Carothers studied the condensation reactions of glycols and dibasic acids He made a large number of polyesters ranging In molecular weight betweeh 2500 and 5000. He also studied polymers of o-amino acids, obtaining polyamides with a degree of polyme~izationof about 30 These were hard, insoluble, waxy materials Recognizing the equilibria involved in stepwise polymerization, Carothers and J W Hill introduced the use of a molecular still to Iemove the water liberated in the condensation process and to shift the equilibrium toward higher molecular weights In this way they synthesized materials with molecular weights up to 25,000 These substances had properties so different from those previously studied that Carothers coined the term superpolymers to describe those having molecular weights above 10,000. The superpolyesters obtained from glycols and dibasic acids were tough, opaque solids that melted at moderately elevated temperatures to clear, viscous liquids Filaments could be pulled from the liquids by touching the surface with a rod and withdrawing it When cool, these filaments could be drawn to several times their original length They became tough, transparent, lustrous materials of high strength and elasticity, possessing as high tenacity when wet as dry. Examination by x-ray diffraction showed them to be crystalline and highly oriented The properties of the aliphatic superpolyesters made them unsuitable for textile fibers because of their relatively low melting points and high solubility Mixed aliphatic polyester-polyamides had the same bad features as the polyesters. It was clear that success, if achieved at all, would involve the use of a polyamide fiber The next phase of the research was a concentrated effort to find polyamides which could be made into fibers A large variety of amino acids, diamines, and dibasic acids was studied Several of the products gave good fibers melting near 200C and equal to silk in strength and pliability In 1935 the polyamide of hexamethylenediamine and adipic acid was made. Like the others, it gave fibers by melt spinning OK by dry spinning from phenol solutions The fibers could be cold drawn to obtain high tenacity and elasticity. They melted at about 265C and were insoluble in all common solvents This polymer was selected for commercial development because of its good balance of properties and the possibilities foreseen for the manufacture of the raw materials. At that time, adipic acid was made only in Germany, and hexamethylenediamine was little more than a laboratory curiosity In Germany, the first
409
commercial nylon was polycaprolactam, 6-nylon For these historic reasons, the major fraction of nylon production in the United States has been 66-nylon, and that in Europe, 6-nylon. Despite some differences in upper use temperature resulting from the difierence in crystalline melting points (265C for 66-nylon, 225 for 6nylon), the production, properties, and applications of the two can be considered together.
Production. The key chemical in the production of both 6- and 66-nylon is cyclohexane Produced by catalytic hydrogenation of benzene, some 1 5 billion Ib was made in 1982, severely down from 2 4 billion Ib in 1979 Sixty percent was converted to adipic acid by oxidation, and 30% was converted to caprolactam In turn, a portion of the adipic acid can be converted to hexamethylenediamine via the ammonium salt, adipamide, and adiponitrile, although other syntheses can be used The polymerization of caprolactam is carlied out by adding water to open the rings and then removing the water again at elevated temperature, where linear polymer forms An autoclave or a continuous reactor can be used Polycaprolactam is in equilibrium with about 10% of the monomer, which must be removed by washing with water before the polymer can be spun At the spinning temperature, more monomer is formed to restore the equilibrium, and this must again be removed to achieve good properties in the yarn Caprolactam can also be polymerized by ionic chain mechanisms, as described in Chapter 4E The reaction can be carlied out below the melting point of the nylon and at atmospheric pressure, making the technique very attractive for the production of large cast articles In the polymerization of 66-nylon, the achievement of the stoichiometric balance needed to obtain high molecular-weight step-reaction polymers (Chapter 2) is simplified by the tendency of hexamethylenediamine and adipic acid to form a 1 : 1 salt that can be isolated because of its low solubility in methanol This salt is dissolved in water and added to an autoclave with 0 5-1 mol % acetic acid as viscosity stabilizer As the temperature is raised, the steam generated purges the air from the vessel Pressure is kept at 250 psi as the temperature is raised to 270280C Pressure is then reduced, and a vacuum may be applied After a total time of 3-4 hr, nitrogen pressure is used to extrude the nylon as a ribbon through a valve in the bottom of the autoclave The ribbon is subsequently cut into cubes, the finished product for the plastics-grade material For fiber use, the polymer (somewhat lower in molecular weight) is melt spun as described in Chapter 18B Properties. Both as plastics and as fibers, the nylons are characterized by a combination of high strength, elasticity, toughness, and abrasion resistance Good mechanical properties are maintained up to 150C, although a more conservative limit for plastics use is 125". Both toughness and flexibility are retained well at low temperatures The solvent resistance of nylon is good; only phenols, cresols, and formic acid dissolve the polymer at room temperature Strong acids degrade it somewhat The polymer discolors in air at temperatures of about 130C and is degraded by hydrolysis
/I
I
I
41 0
at elevated temperatures. Its outdoor weathering properties are only fair unless it is especially stabilized or pigmented with carbon black Nylon has a moderately low specific gravity, 1. 14. Its moisture resistance is f a i ~moisture acts as a plasticizer. ; to increase flexibility and toughness. The electrical uses of nylon are restricted to low frequencies because of the polar groups in the polymer
Fiber Applications. About 60% of nylon production goes into housing-related outlets, especially carpets. Another 20% is used for apparel, and only about 10% for tire cord, down only slightly in volume over the last 20 years and accounting for nearly one-third of the market Plastics Applications. The most important uses of the nylon plastics are as an engineering material-a substitute for metal in bearings, gears, cams, and so on. Nylon is well suited for such applications because of its high tensile and impact strength, fbrm &ability at high temperatures, good abrasion resistance, and selflubricating bearing properties. The ability to be injection molded to close dimensional tolerances, augmented by its light weight, frequently gives the plastic a cost advantage in appliances. Rollers, slides, and door latches are often made of nylon. Bearings and thread guides in textile machinery may be made of nylon, which requires no lubricant that could damage yarn or cloth. Electrical wire is jacketed with nylon, which provides a tough, abrasion-resistant outer cover to protect the primary electrical insulation A number of specialty grades of the nylons are finding increased plastics use: highes-molecular-weight polymers, copolymers with olef n-based ~ e s i n sfor increased impact r.esistance at low temperatures, and reinforced grades with increased strength, stiffness, and creep resistance. Fillers consist of either continuous mat or chopped glass, various minerals and glass, and minerals alone. Other Polyamides. Higher nylons (610,612, 11, and 12) have significantly lower stiffness and heat resistance than 6- and 66-nylons, but have improved chemical resistance and lower moisture absorption. They are used as specialty plastics Nylon copolymers whose monomers include 6, 66, and 610 starting materials, terephthalic acid, and alkyl substituted diamines, are amorphous and therefore transparent though slightly yellow.. These materials are also plastics. The name aramid is applied to aromatic polyamides, two of which are firmly established in the specialty fiber markets. To use their Du Pont trade names, Nomex,
i 5
i
; 1 1
41 1
is virtually nonflammable and achieved an early solid market because of' this p r o p erty; it is now progressing into other industrial applications where unusual heat resistance is required Kevlar,
more readily crystallized and oriented as can be infer~ed from its para structure, is noted for its outstanding strength-weight ratio It is stated (Layman 1982) that a Kevlar cable can match the strength of steel at the same diameter and 20% of the weight The major present use of the material is in premium-quality tire cords, but its impact on that market is still virtually negligible A nylon made from dodecanedioic acid and bis-para-aminocyclohexyl methane.
(Du Pont's Quiana), spun in trilobal form, has a variety of' good aesthetic qualities that lead to its use in luxury dress apparel applications
Polypeptides
Polypeptides, the basis of natural proteins, are polyamides of a-amino acids They are characterized by the repeat unit
where R varies widely among the 30 or so amino acids in proteins Although synthetic products can be made, the polypeptides of greatest interest are based on natu~ally occur~ing materials They include wool and silk and a variety of fibers and plastics made from proteins extracted from nonfibrous natural products The importance of wool and silk is discussed in relation to other fibers in Chapter 18 Other protein-based fibers and plastics are at present of little more than historic interest.
Wool. Wool is a complex polypeptide, a polyamide made up of' about 20 a-amino

acids. The acids glycine, leucine and isoleucine, proline, cystine, and arginine, glutamic acid, and aspartic acid make up about two-thirds of' its weight. About half' of wool's mass is in the main polymer chain and half in the side chains. Since some 30% of' the side chains have acid or amine groups along their length, wool
41 2
is a huge polymeric "zwitterion" whose properties and solubility depend markedly upon pH. Wool can exist in both the a- and P-keratin structures characteristic of' the polypeptides: helical and extended sheet structures, respectively. The high reversible extensibility of wool (about 30% compared to .5% fbr cotton) derives from the fact that it can exist in either of' these structures The microscopic appearance of' wool is easily recognized by the presence of' scales in the epidermal or surface layer of' the fiber.. These scales give wool the ability to felt or pack into dense stable mats Wool is noted f b r its high moisture absorption (up to 10-15%) and good insulating properties, its wrinkle resistance and elastic recovery, and its crimpiness. Its soil resistance, dyeability, and resistance to organic (dry-cleaning) solvents are good On the other hand, it is attacked by moths and mildew, mild alkalis, and bleaches It causes allergies. Its stsength and stiffness are low compared to those of' cotton It has poor temperature resistance and weatherability in sunlight, and is adversely aff'ected by hot water, with shrinkage and loss of' luster and strength Despite these drawbacks, the virtues of wool are so great that it probably will continue to find extensive use, as it has to date, in suitings and other wearing apparel, blankets, yarns, carpets, felts, and upholstery. So far, no artificial fiber has equaled it in resilience, hand, insulating properties, dyeability, and flame resistance The price of' wool is notoriously unsteady, however, and its future may well depend upon its price in a competitive market. Production in the peak year of' 1979 was only a little over 100 million Ib in the United States.. Wool is crosslinked though cystine links. This structure gives swelling but not solubility in the natural product On the addition of'thioglycolic acid, however, the crosslinks are broken with the formation of sulfhydryl groups. The tensile strength of' the fiber decreases, its stress-strain curve changes to resemble that of' a typical elastomer, and it shrinks readily and becomes soluble in polar solvents Wool can be recrosslinked by a number of' reagents, including alkyl dihalides Since the thioethers resulting from the treatment are much more stable than the original disulfides, the new fiber has much better resistance to laundering and chemicals. Since moth damage usually begins with rupture of' the disulfide links, the treated material has improved moth resistance. The scission and crosslinking can be made so mild that the appearance and hand of'the material are not changed
Silk. Silk is a polypeptide made up of' only four amino acids: glycine, H,NCH,COOH; alanine, H,NCH(CH),)COOH; serine; and tyrosine. Glycine and alanine, present in a 2: 1 ratio, account fbr over 75 mol.% of' the material. Silk crystallizes in an antipasallel-chain pleated sheet P-keratin structure, in which only glycine and alanine residues occur. The other amino acids are present only in the amorphous regions of' the fiber. There is no structure in silk equivalent to the a-keratin spiral in wool and other polypeptides.. Thus the silk fiber has an elongation (20-25%) lower than that of' wool but higher than that of cotton The silkworm spins a double fiber from which the two single fibers are easily separated as continuous filaments 400-700 yards long. Silk is thus more nearly
I
I
POLYESTERS, POLYETHERS, AND RELATED POLYMERS
413
analogous to continuous filament than to stable synthet~c fibers Silk has been noted for thousands of years for its strength, toughness, and smooth, soft feel It is resilient and shapes well. It is a poor conductor of heat and electricity Its moisture regain is 11%; it withstands heat and hot water better than wool Some disadvantages of silk are those associated with its cultivation and gathering, unevenness of the fibers, loss of strength on dyeing, and poor launderability of the dyed fabr-ics It lacks the strength of nylon for hosiery use but would otherwise be preferred for its warmer feel There is relatively little effort in progress to improve the properties of silk
GENERAL REFERENCES
Polyamides Rebenfeld 1967; Schule 1969; Snider 1969; Sweeney 1969; Kohan 1973; Moncrieff 1975, Chapters 19-22; Elias 1977, Chapter 28 3; Jacobs 1977; Preston 1978; Fisher 1982; Galanty 1982; Putscher 1982; Reitano 1982; Saunders 1982; Welgos 1982 Polypeptzdes Elias 1977, Chapter 30
B. POLYESTERS, POLYETHERS, A N D RELATED POLYMERS

Polyesters
The p~operties low melting point and high solubility caused Carothers (see Section of A ) to reject the linear aliphatic polyesters as fiber-formingcandidates, and to this day no commercial product is based solely on these polymers. The stiffening action of the p-phenylene group in a polymer chain, however, leads to high melting points and good fiber-forming properties, as discussed in Chapter 12A Commercially important polyesters are based on such polymers, of which poly(ethy1ene terephthalate) is the major product These polymers are used for fibers, films, and blown bottles About 4 3 billion lb was consumed as fibers in 1982, about 400 million Ib as blown bottles, and about 240 million Ib in extrusion and film uses.
Polymerization and Properties. The production of high-molecular-weight polyesters differs somewhat from that of similar polyamides In the case of the nylons, the chemical equilibrium favors the polyamide under readily achieved polymeriequivalence is easily achieved by the use of salts, zation conditions. Stoichiomet~ic and arnide interchange reactions are slow In polyester formation, however, the equilibrium is much less favorable, and equivalence is more difficult to achieve, since the salts do not form Furthermore, the dibasic aromatic acids are very difficult to purify because of low solubility and high melting point This situation has been met by taking advantage of the ~apidityof ester-interchange reactions. The acid, such as terephthalic acid, is converted to the dimethyl ester, which can easily be purified by distillation or crystallization This is then
41 4
HET'EROC,HAIN THERMOPLASTICS
allowed to react with the glycol by ester interchange. In practice a low-molecularweight glycol is used and the reaction takes place in two steps. First, a lowmolecular-weight polyester is made with an excess of glycol to ensure hydroxyl end groups. Then the temperature is raised and pressure lowered to effect condensation of these molecules by ester interchange with the loss of the glycol In the production of poly(ethy1ene terephthalate), terephthalic acid is made by the oxidation of p-xylene. The polymerization step is similar to that for the polyamides in so f i r as the equipment and conditions are concerned. The polymer coming from the autoclave is quenched from the molten state to below its glass transition point of about 80C and is amorphous. Crlystallinity develops on heating; the crystalline melting point is 265C The polymer is melt spun Because of its high crystalline melting point and glass transition temperature, poly(ethy1ene terephthalate) retains good mechanical properties at temperatures up to 150-175OC. Its chemical and solvent resistance is good, being similar to that of' nylon..
Fiber Applications. The properties of poly(ethy1ene terephthalate) fiber that influence its uses are its outstanding crease resistance and work recovery and its low moisture absorption.. These properties arise from the stiff' polymer chain and the resulting high modulus, and the h c t that the interchain bonds are not susceptible to moisture. As a result, garments made from the polyester fibers are very resistant to wrinkling and can be washed repeatedly without subsequent ironing The higher modulus of the polyester fibers and their wrinkle resistance are reminiscent of those of wool, but fully oriented polyester fibers are too stiff to have the crisp but soft hand of wool.. It is possible, however, to impart this property to the polyesters by control of' crystallinity and orientation so that their stress-strain curve is similar to that of wool. This is done in the production of staple polyester fiber. The polyester fibers are seldom used alone, but are blended with cotton or wool to make summer- and medium-weight suiting and other goods. They have set new standards fbr the perfbrmance of wearing apparel through the "wash and wear" concept . Polyester fibers are unique among the synthetics in their ability to form felts These products are better than wool felts in resistance to temperature, to abrasion, and to further felting. popular as a tire cord, with about 25% of the market Polyester fiber is cur~ently for the last decade.
The tensile strength of poly(ethy1ene terephthalate) film is about 25,000 psi, two to three times that of cellophane or cellulose acetate film (Section C) If the area of the specimen at the break point is considered, the tensile strength of this plastic is about twice that of aluminum and almost equal to that of mild steel The stiffness of poly(ethy1ene terephthalate) film is comparable to that of cellophane and other cellulosic films, but its resistance to failure on repeated flexing is unusually high In one flex test it lasts for over 20,000 cycles compared to a few
Plastics Applications.
i ! 1
i
i i
t
i
POLYESTERS, POLYETHERS, A N D RELATED POLYMERS

/
41 5
/:
!
hundred for cellulosic films Its tear strength is also better than that of the cellulosics The impact strength of this material is three to four times that of any other plastic film. This toughness is a major advantage in typical applications, such as magnetic recording tape. The major new plastics use fbr poly(ethy1ene terephthalate) in the last few years has been in the development of' blown bottles for soft-drink use, with early emphasis on the 2-liter bottle size.. Molding is usually done in two stages, with a prefoxm injection molded, then blown to final bottle shape. Both are biaxially oriented to provide the necessary strength for the application, to withstand dropping (though an additional cap on the bottom of' the bottle is requir.ed for extra protection) and the internal pressure of casbon dioxide.. Poly(buty1ene terephthalate) and various copolymers have minor plastics uses
Ether and Acetal Polymers

Polyethers, by definition, contain the ether linkage --R-0--R-0as part of their chain structure By convention, the polymers in which R is a methylene group, -CH,--- (i e , polyoxymethylenes), are described by the generic term acetal reszns, the name polyethers being applied to polymers in which R is more complex Epoxy ~esins,also polyethers, are discussed in Chapter 16C.
Polyoxymethylene. Polyoxymethylene is a polymer of formaldehyde or of trioxane Although polymeric products of formaldehyde have been known for over 100 years, and were studied in detail by Staudinger (1925), thermally stable polymers of formaldehyde were only recently prepared The improved stability of these acetal resins allows them to be fabricated into useful articles Copolymers, thought to contain small amounts of ethylene oxide as the comonomer, are also produced About 80 million Ib of acetal resins was produced in the United States in 1982 Exceptionally pure formaldehyde (better than 99 9% CH20) is polymerized by an anionic mechanism in the presence of an inert solvent (e g , hexane) at atmospheric pressure and a temperature, preferably between - 50 and -t 70C, where the solvent is liquid A wide variety of anionic catalysts is suitable, including amines, cyclic nitrogen-containing compounds, arsines, stibines, and phosphines The polymer is insoluble in the reaction mixture and is removed continuously as a slurry. Thermal stability is improved by acetylation of the hydroxyl end groups of the polymer using acetic anhydride:
Acetal resins have properties characteristic of partially crystalline high-molecular-weight polymers Typical specimens are about 75% crystalline, with a melting point of 180C The impact strength of the polymers is high, and both their stiffness
41 6
(450,000 psi) and yield stress (10,000 psi) are greater than those of' other crystalline polymers. Moisture absorption is negligibly small, and the polymers are insoluble in all common solvents at room temperature. They can be processed by conventional molding and extrusion techniques.. Major uses for the acetal resins are as direct replacements for metals. Their stifmess, light weight, dimensional stability, and resistance to corrosion, to wear, and to abrasion have led to their replacing brass, cast iron, and zinc in many instances. Typical applications include automobile parts, such as instrument panels, door hardware, and pump housings and mechanisms; pipe, especially fbr oil field systems; and a wide vziety of' machine and instrument parts
Polyglycols. Polymers made from ethylene oxide and its higher homologues
can be considered to be the condensation products of glycols, although they are in fact produced by ring-scission polymerization of the corresponding oxides (Chapter 4E) The product of greatest commercial interest is the polymer of ethylene oxide This polymer combines the thermoplastic behavior and mechanical properties of a highly crystalline, high-molecular-weight polymer with complete water solubility
Polyurethanes
Polyurethanes are polymers containing the group
formed typically through the reaction of a diisocyanate and a glycol:
The polymers formed in this way are useful in four majo~types of product: foams, fibers, elastomers, and coatings Polyurethane foams are invariably crosslinked; hence they are discussed with thermosetting resins in Chapter 16C
!
a
Elastomers. Polyurethane elastomers are made in several steps A basic intermediate is first prepared in the form of a low-molecular-weight (1000-2000) polymer with hydroxyl end groups This may be a polyester, such as that made from ethylene glycol and adipic acid, a polyether, or a mixed polyester-polyamide The basic intermediate, which is here designated B, is then reacted with an aromatic diisocyanate to give a prepolymer Typical isocyanates are 2,4-tolylene-
POLYESTERS, POLYETHERS, AND RELATED POlYMERS
41 7
diisocyanate, 4,4-benzidenediisocyanate, and 1,5-naphthalenediisocyanate: Using the former as an example, a typical prepolymer can be represented as
I
NCO
NHCOOBOCONH NHCOOBOCONH
The elastomer is vulcanized through the isocyanate groups by reaction with glycols, diamines, diacids, or amino alcohols If water is used, carbon dioxide is eliminated during crosslinking, as in the production of urethane foams (Chapter 16C) Polyurethane elastomers are noted for extremely good abrasion resistance and hardness, combined with good elasticity and resistance to greases, oils, and solvents They make tire treads with unusually long life and are widely used in applications requiring outstanding abrasion resistance, such as heel lifts and small industrial wheels
Fibers. Polyurethane fibers with unusually high elasticity are used for lightweight foundation garments and swimsuits They have replaced rubber latex thread in this Coatings. Coatings based on polyurethanes have very good resistance to abrasion and solvent attack plus good flexibility and impact resistance They can be applied by dip, spray, or brush and adhere well to a wide variety of materials They are suitable in applications for which unusual impact and abrasion resistance is required, such as gymnasium and dance floors and bowling pins, in magnet wire coatings, and in a variety of outdoor and marine uses because of their good weatherability
+
Molding Compounds. With the development of methods for molding thermosetting materials (reaction injection molding), there has been a resurgence in the use of polyurethane molding compounds These are discussed in Chapter 16C
Polycarbonates
A polycarbonate plastic, characterized by the -0COO-heterochain unit, can be made either from phosgene and bisphenol A (4,4'-dihydroxydiphenyl-2,2'-propane) or by ester exchange between bisphenol A and diphenyl carbonate It has this structure:
41 8
HETEROCHAlN THERMOPLASTICS
Like the nylon, acetal, and polyether resins, this polymer is a crystalline thermoplastic of very good mechanical properties. It has unusually high impact strength, even at low temperatures, ascribed in part to a combination of relatively high ordering in the amorphous regions and considerable disordering in the crystalline regions It has low moisture absorption, good heat resistance (useful to 140C), and good thermal and oxidative stability in the melt It is transparent and selfextinguishing It can be processed by conventional injection molding and extrusion Typical applications include telephone parts, business machine housings, machinery housings, automobile tail-light lenses, and unbreakable glazing applications About 210 million lb of polycarbonate resins was produced in the United States in 1982
i
i
'
Sulfur-Containing Polymers
Polysulfide elastomers are the reaction products of ethylene dihalides and alkali sulfides It was discovered in 1920 that a rubberlike material could be made from these reagents Commercial production began in 1930, and the products were quite successful despite some disadvantages (notably odor) because of their oil resistance The polysulfide rubbers are linear condensation polymers: xRC1,
+ xNa2S,
f
k
-(R--S,-),
+ 2xNaC1
; 4
The physical properties of the materials depend on the length of' the aliphatic group and the number of'sulfur atoms present.. With four sulfurs per monomer all products are rubbery, whereas with only two sulfurs at least fbur methylene groups are needed in the dihalide to obtain elastomeric properties. The halides ordinarily used ase ethylene dichloride, P,P1-dichloroethyl ether, and dichloroethyl formal Cocondensation can be carried out The polysulfide elastomers ase outstanding in oil and solvent resistance and in gas impermeability.. They have good resistance to aging and ozone.. On the other hand, they have disagreeable odors, poor heat resistance, poor abrasion ~.esistance in tire treads, and low tensile strength (1500 psi). Most of their applications as rubbers depend on solvent resistance and low gas permeability.. They include gasoline hoses and tanks, diaphragms and gaskets, and balloon fabrics. Liquid polysulfide elastomers can be cured at room temperature and without shrinkage to tough, solvent-resistant rubbers. These liquid compounds are widely used as gasoline tank sealants and liners for aircraft and in a variety of other sealing and impregnating applications. These liquid polymers, when combined with oxidizers, burn with great intensity and generate l u g e volumes of gas. They form the basis of solid-fuel rocket propellants. Inorganic oxidizers, in finely ground form, are mixed with the liquid elastomeric binder and cast directly into the socket motor, binding to the walls of the vessel fbr support and insulation of the motor case..
I
I
CELLULOSIC. POLYMERS
41 9
Other sulfur-containing polymers are engineering plastics with good property retention over wide temperature ranges They include polyphenylene sulfides, discussed with the cor~esponding oxygen compounds in Section D; polysulfones,
and polyethersulfones,
Polyesters Rebenfeld 1967; Farrow 1969; Goodman 1969; Hawthorne 1969; Moncrieff 1975, Chapter 24; Elias 1977, Chapter 26 4; Katz 1977; Davis 1982; Jaquiss 1982; Pengilly 1982; Scales 1982; Yarger 1982 Polyethers Akin 1962; Bevington 1964; Elias 1977, Chapter 26.2; Persak 1978, 1982; Vandenbe~g1979; Braun 1982; Dreyfuss 1982; Newton 1982; Royal 1982; White 1982 Polyztrethanes Saunders 1962, 1964; Ibrahim 1967; Blokland 1968; Bruins 1969; David 1969; Pigott 1969; Wright 1969; Moncrieff 1975, Chapter 26; Backus 1977; Brydson 1978, Chapter 16; Dominguez 1982 Polycarbonates Schnell 1964; Bottenbrvch 1969; Fox 1982; Page 1982 Sulfur-Containing Polymers. Berenbaum 1969; Eichhorn 1977. Elias 1977, Chapter 27.2; Ballintyne 1982; Boeke 1982; Hill 1982; Rigby 1982
C.
I
CEL.L.ULOSIC POLYMERS
1
,
1
The class of polymers based on cellulose includes the following: native cellulose, including wood and most plant matter, but of interest here primarily because of the widespread popularity of cotton as a fiber; ~egeneratedcellulose, used as a fiber (viscose rayon) and a film (cellophane); chemical derivatives of cellulose, of which the organic esters, particularly the acetate, are most important; and minor polymers with structures similar to that of cellulose
I
420
Cellulose
Cellulose is a polysaccharide made up of P-D(+)-glucose residues joined in linear chains having the structure
-0-HC
C-0 /H C-C
\H H / C ~ * -
With its three hydroxyl groups, cellulok has the opportunity of forming many hydrogen bonds The resulting highsiirtermolecular forces plus the regular structure of the polymer result in its having an unusually high degree of crystallinity The crystalline melting point of cellulose is far above its decomposition temperature The solubility of the polymer is very low; it is doubtful that solution ever takes place unless a chemical derivative is formed. Cellulose does swell, however, in hydrogen-bonding solvents, including water The swelling is, of course, restricted to the amorphpus regions of the structure When native cellulose is dissolved via chemical reaction and then reprecipitated as pure cellulose, the product is known as regenerated cellulose
Cotton
Cotton is about 95% cellulose, with small amounts of' protein, pectin, and wax The fiber can easily be recognized under the microscope because of its flattened, twisted shape. Cotton fabxics launder well and show excellent resistance to alkalis. Since the strength of cotton is 25% greater wet than dry, it withstands repeated washings well, in contrast to most animal and synthetic fibers which lose strength when wet or moist. Cotton is less stiff' than flax, hemp, or jute. Fabrics of' cotton can be folded repeatedly without loss of strength. With proper construction and treatment, shrinkage and stretching can be controlled. Cotton is a good conductor of' heat, better than silk or wool but not as good as linen. This property and its high moisture absorption give it comfbrt in wearing apparel. Cotton withstands heat better than the other natural fibers. It can be dyed readily, and the colors ase good in lightfastness and washfastness. It is not affected by solvents and usually not attacked by moths About 5.4 billion Ib of cotton was grown in the United States in 1978
Mercerization. The chemical treatment of cotton known as mercerizatlon to improve its luster has been practiced for many yeais The treatment consists in swelling the cotton by concentrated alkali solution and then washing out the alkali
CELLULOSIC: POLYMERS
421
The chemical changes during the process are similar to those in the formation of' alkali cellulose At the end of'the treatment there is no molecular difr'erence in the product, but it is somewhat more amorphous than native cotton and, unless the fibers were held under tension during the process, somewhat less oriented As a result of' these changes, mercerized cotton is slightly lower in density than the untreated material and has increased water absorption, better dyeability, lower tensile strength, and higher extensibility
Regenerated Cellulose
The term regenerated cellulose describes cellulose which has been dissolved by virtue of the production of a soluble chemical derivative, cellulose xanthate, and subsequently reprecipitated When prepared as a fiber, regenerated cellulose is known as viscose or (viscose) rayon Traditionally both viscose and cellulose acetate have been known as rayons, but the term is now restricted to viscose in order to avoid confusion As a film, regenerated cellulose is known by the generic t e ~ m cellophane.
Manufacture. The first step in the production of regenerated cellulose, starting
with purified cellulose usually obtained from wood pulp, is the formation of alkall or soda cellulose This is produced by the reaction of cellulose with aqueous NaOH at room temperature for 15 min to 2 hr The excess alkali is then pressed out and the mass is shredded into "crumbs" and aged for several days at 25-30C to promote oxidative degradation of the chains to the desired degree of polymerization The alkali cellulose is then converted to cellulose xanthate by the addition of carbon disulfide: S R-OH
t CS2
+ NaOH
R-0-C-.-S--Na
II
t H20
where R is a cellulose residue The average degree of xanthation is about one CS, for every two cellulose residues Since the reaction takes place heterogeneously by the addition of liquid or gaseous CS, to the alkali cellulose crumbs, the xanthation in some regions may correspond to the dixanthate or higher The mass retains most of its physical form but becomes more sticky or gelatinous and takes on a deep orange color and a characteristic and highly unpleasant odor. It is held at 25-30C for 1-3 hr with about 10% CS2 by weight and then evacuated to remove excess CS, It is now dissolved in dilute NaOH, becoming completely soluble in this reagent for the first time The solution is known as viscose. Its color and odor are due to products of side reactions such as Na2CS,, NaSH, H,S, sulfides, and polysulfides A fresh viscose solution cannot easily be coagulated and must be allowed to "ripen" for a few days The changes taking place during ripening are complex,
422
HETEROCHAIN T.HERMOPLAS.TICS
and the reaction must be casefully controlled by time and temperature to avoid premature coagulation Cellulose xanthate is essentially unstable and decomposes gradually during the ripening process by hydsolysis and saponification. At the end of the ripening period, enough cellulose residues have been regenerated so that coagulation is imminent The viscose is then spun into a bath containing sulfuric acid to effect the regeneration of the remaining cellulose residues and coagulate the polymer:
S
R-0-C-S-Na
II
+ H'
R--OH
+ CS2 + Na'
Viscose Rayon Normal Tenacity. Viscose rayon normally has a tenacity of 2-2 5 gldenier, somewhat lower than that of cotton (3-5 gldenier) Its wet tenacity is only about half the dry value but is still adequate for laundering The moisture absorption of viscose is about twice that of cotton The elasticity 1s not high (typically 15% elongation at break), and the fiber shows considerable viscoelastic character Viscose of normal tenacity is primarily an apparel fiber, but its popularity is declining Several one-time major manufacturers have ceased producing the material, and the others aIe operating well below capacity High Tenacity. In contrast to apparel rayon, a high-tenacity product (3-6 g/ denier) is produced by stretching the fibers just short of their breaking point in the spinning bath Considerable orientation (and crystallinity) is so ~ntroduced Hightenacity rayon has lower elongation and moisture absorption than the apparel material and was used almost exclusively for tlre cord This market has, however, virtually disappeared with the increased use of nylon, polyester, and steel for that purpose Cellophane
In the manufacture of' cellophane the viscose solution is extruded as film and then immersed in a bath of' ammonium and sodium sulfate and dilute sulfuric acid, which removes the xanthate groups and precipitates the cellulose. The film is passed through various washing, bleaching, and desulf'urizing baths, and finally through a bath of glycerol, glucose, or a polyhydric alcohol, which is imbibed and acts as a plasticizer. The film is then dried.. For applications in which low moisture permeability is desired, it is coated with a mixture of' nitsocellulose and various plasticizers and waxes. Cellophane is a thin (0.001-0.002 in .) film of fair physical properties. Its tensile stsength is good, but its tear strength, impact strength, and resistance to flexing are poor compared to those of newer film materials.. The permeability of' the uncoated film to water vapor and to water-soluble gases is extremely high. The coated material passes about 0.3 g of' water vaporlin2 per hour at 40C. Cellophane is widely used as a wrapping and packaging material
C~~.~ULOSIC POLYMERS
42'3
Cellulose sponges are made by stir~ing lumps of' a salt such as sodium sulfate into viscose, coagulating, and washing out the salt to leave a porous product Sp'wes may also be made by incorporating a blowing agent in the material.
cellulose Acetate ~anufacture. As a polyhydric alcohol, cellulose can undergo esterification reaction~ In the presence of a strong organic acid such as formic acid, equilibrium in the reaction
acid f alcohol
F=
ester
+ water
lie3 partly to the right, and some ester is formed With other organic acids, including acetic acid, equilibrium is well to the left and ester formation does not take place under normal circumstances The easiest way to shift the equilibrium to the right is by the removal of water as it is formed in the reaction To accomplish this, part of the acid is replaced by an acid chloride or anhydride Sulfuric acid is also utilized in the reaction mixture It acts as catalyst and also assists in removing water Purified cotton linters or wood pulp form the raw material for cellulose acetate This cellulose is partially dried from its natural moisture content of 5-10% to reduce the water content of the subsequent reaction mixture The end of the acetylation is indicated by complete solution of the cellulose At this point complete acetylation to the triacetate has taken place The mixture is now held at about 50C until the desired chain degradation has been achieved, as indicated by the viscosity of the mixture. To obtain a more easily spinnable composition, it is necessary to reverse the acetylation reaction to the point where the proper solubility relations are obtained This cor~espondsto the acetone-soluble diacetate, or an acetyl content of about : 30% This stage of the reaction can be car~ied out in homogenous solution, thus preserving the maximum amount of chemical homogeneity in the final product At the end of the reaction to produce the diacetate the mixture is cooled and added to a large amount of 25-35% acetic acid This precipitates the polymersin finely divided form After the polymer is washed and dried, it is dissolved in acetone for the spinning operation
Fiber Properties and Applications. The traditional acetate fiber (diacetate) has a tenacity of 1.2--1 5 gldenier and about the same moisture absorption as cotton It tends to soil less and wash more easily than viscose Since acetate is thermoplastic, pleats and creases can be permanently set On the other hand, acetate is less susceptible to creasing and wrinkling in use than viscose because its resilience is greater With the advent of spinning processes using rnethylene chloride, cellulose triacetate became popular as an apparel fiber Its properties are improved over those of the diacetate in moisture absorption, wrinkle resistance and creep retention, dyeability, and speed of dryin2
424
HETEROCHAIN IHERMOPLASTICS
About 1 . 3 billion Ib of cellulose acetate and viscose rayon, combined, was produced in the United States in 1979..
Plastics Properties and Uses. Cellulose acetate was introduced in sheet, rod, and tube form (as well as in lacquers and dopes) around the time of World War I Its popularity as a substitute for cellulose nitrate was not high, and it was not widely used until injection molding became common in the 1930's. It has been fbr many years a very popular injection-molding material.. The usual molding material has 50-55% acetyl content by weight. It is somewhere between the diacetate and the triacetate in composition. Higher acetyl content gives poorer flow but better moisture xesistance; lower acetyl content gives better impact strength. A plasticizer is ordinarily used Higher-softening grades are preferred fbr compression molding, and softer grades for extrusion. Cellulose acetate molded and extruded articles are used where extreme moisture resistance is not required Cellulose acetate is also utilized in sheet form. Thin films are cast from solution onto polished metal surfices. The films are used in photography, fbr wrapping, and fbr making small envelopes, bags, and boxes fbr packaging. The plastics uses of cellulose acetate plus the mixed organic esters described below amounted to about 100 million Ib in 1982 Mixed Organic Esters of Cellulose Cellulose AcetateButyrate. The mixed ester cellulose acetate-butyrate has several advantages in properties over cellulose acetate These include lower moisture absorption, greater solubility and compatibility with plasticizers, higher impact strength, and excellent dimensional stability..The material used in plastics has about 13% acetyl and 37% butyryl content.. It is an excellent injection-molding material Moldings of cellulose acetate-butyrate are used in automobile hasdware.. Motion picture safety film is cellulose acetate-butyrate or cellulose propionate-butyrate The outdoor weathering prope~ties cellulose acetate-butyrate are superior to those of of the other cellulose esters, but not nearly as good, for example, as those of the acrylic resins Cellulose Propionate and Acetate-Propionate. These materials are similar to the acetate-butyrate in properties and uses The acetate-propionate has some application in surface coatings Cellulose Nitrate
Although cellulose nitrate accounts today for only a minute fraction ofthe volume of cellulose plastics, it has considerable historical interest (Chapter 1B). The major use of plastics-grade cellulose nitrate is in the coating field. The properties and applications of the material are dependent on its degree of nitration; the plastics and lacquer grades contain 10.5-12% nitrogen, corresponding roughly to the dinitrate. Explosives grades contain 12.5-1 3.5% nitrogen
CEILUL.OSIC POLYMERS
425
Manufacture. The esterification reaction of the hydroxyl group on cellulose with
nitric acid is an equilibrium: ROH t HONO, RONO,
+ H20
which can be shifted to the ester side by the removal of water The usual reagent for this purpose is sulfuric acid; the nitration mixture ordinarily used consists of sulfuric and nitric acids and a limited amount of water
Fabrication. The temperature sensitivity of cellulose nitrate excludes it from all fabrication methods involving heat, such as molding and extrusion. The plastic is handled by so-called "block" methods Cellulose nitrate, alcohol, and camphor, the plasticizer always used, are mixed in a dough mixer until homogeneous. Here for the first time the fibrous shape of the original cellulose is lost The colloid is filtered under pressure and rolled on hot rolls (65-80C) where much of the alcohol is evaporated It is then sheeted into slabs that are pressed while hot into a homogeneous block about 6" x 5' x 2' in size Sheets of any desired thickness are subsequently sliced from the block They must be aged by storage at about 30C until the last traces of solvent have diffused out This may take from 2 hr to 6 months, depending on the thickness of the sheet The final operation consists of finishing the surface of the sheet by pressing in contact with polished metal sheets or by buffing Rods, tubes, and other shapes can be formed by solvent extrusion of the plas ticized cellulose nitrate, followed by removal of solvent as above Properties and Uses. Despite its disadvantages of flammability, instability, and poor weathering properties, cellulose nitrate is one of the cheapest and most highly impact--resistantplastics It still has some applications as a plastic, but is now used mostly for lacquers Cellulose Ethers
The cellulose ethers are products of the reaction of an organic halide with cellulose swollen by contact with an aqueous base: ROH
+ R'CI + NaOH ---+
ROR'
+ NaCl
-t H,O
where ROH represents a cellulose residue and one of' its hydroxyl groups, and R'CI is the organic halide. Chlorides are preferred over bromides or iodides, despite their lower reactivity, because oftheir higher diffusion rate in the heterogeneous reaction system used.
Manufacture. Cellulose ethers are made by reacting cellulose, alkali, and the organic chloride at about 100C. The ether content and the viscosity ofthe product
426
HETEROCHAIN THERMOPL.ASTICS
ase controlled by temperature, pressure, reaction time, and composition of' the reacting mixture..
Ethyl Cellulose. The most important of the cellulose ethers is ethyl cellulose The commercial material has 2 4-2.5 ethoxy groups per glucose residue It is a molding material which is heat stable and has low flammability and high impact strength It is flexible and tough at temperatures as low as -40C. Its electrical, mechanical, and weathering properties are good in comparison with those of other cellulosics, but not generally outstanding. Its softening point is low, and it has a high water absorption (but not as high as cellulose acetate) and some tendency to cold flow GENERAL REFERENCES
Ott 1954--1955; Bruxelles 1965; Haskell 1965; Hill 1965; Nissan 1965; Savage 1965; Rebenfeld 1967; Mitchell 1969; Bikales 1971; Elias 1977, Chapter 31 5: Andrews 1979; Bogan 1979; Greminger 1979; Serad 1979; Turbak 1979; Daul 1981; Brown 1982; Lundberg 1982; Rich 1982
D. HIGH-TEMPERATURE AND INORGANIC POLYMERS
The requirements of' modern technology, including those related to the "space age," place increasing demands on the high-temperature behavior of all mater.ials, including polymers. For the last few years there has been an increased effort to define and produce polymeric structures capable of demonstrating good mechanical properties for long periods of time at higher and higher temperatures Mark (1967) has pointed out several ways in which the melting or softening points of polymers can be raised, with concomitant improvement in high-temperature properties. These are: stifiening the chains by the addition of ring structures or other stiff' elements, crosslinking the chains, and inducing crystallization. With our present capabilities, it has not been possible to devise structures in which all three of these approaches are combined; for example, crystallinity and crosslinking are still largely mutually exclusive. But the application of' one or two of these principles has proved fruitful. The results of' adding aromatic rings to polymer chains have been described in Chapter 12A The addition of other ring structures, usually heterocyclic, to form fused-ring groups even larger and stiffer than the phenylene moiety, is a logical extension of'this idea.. Beyond this, two approaches are suggested: One, attributed (Fraser 1969) to Marvel, is to produce ladder polymers whose structure is such that breaking one or a few bonds can not sever the chain; the other is to abandon polymers based on the casbon-carbon bond and go to other structures known collectively as inorganic polymers C. S. Marvel has stated (private communication) that the upper temperature limit for the stability of carbon-chain polymers in a nitsogen atmosphere is probably around 550-600C, unless the structure is very similar to that of graphite itself' (as
HIGH-TEMPERATURE AND I N O R G A N I C POLYMERS
427
"black orlon" may be-see below) If the polymer is completely free from hydrogen, the stability in air and in nitrogen should be essentially the same. These limits are conventionally determined by dynamic TGA (Chapter 4E); isothermal weight loss experiments at high temperatures sometimes yield other results. The synthesis of' ladder polymers, anticipated with enthusiasm a decade ago, has led to little more than laboratory curiosities (Overberger 1970)..These polymers are defined as those consisting of an uninter~uptedseries of' rings connected by links around which rotation cannot occur except by bond breaking. If' their structure were perfect, the chain of' a ladder polymer could be broken only if at least two bonds on the same ring were broken In a degradation process in which all bonds are equally strong, this would be fir less probable than a single rupture breaking a single-chain polymer Thus ladder polymers should have far greater thermal stability. Such polymers may be fbrmed by reactions of existing polymer chains; there are several well-known examples: One is the cyclization of'poly(methy1 vinyl ketone):
A second is the well-studied production of "black orlon" by the simultaneous cyclization and oxidation at 160-300C of polyacrylonitrile, originally reported by Houtz (1950) The final material, whose structure may be
or may be an oxidized and aromatized reaction product of' an intermediate of' this structure, glows red but maintains its fbrm in a blowtorch flame (as do some of' the heterocyclic polymers described below), withstanding temperatures of' 700-. 800C without loss in properties (which are admittedly not too good). This material is, however, the precursor of the graphite fibers described in Chapter 1 7 0
High-Temperature Polymers Polyimides. Polyimides with structures similar to
428
HErEROCHAlN THERMOPLASTICS
are made by the polycondensation of' aromatic dianhydrides and aromatic amines.. The first step is a soluble polyamic acid, which is converted to the polyimide by further condensation These materials are available as films and as fabricated solid parts. The polyamic acid is also sold as a coating solution. The polyimides retain usable properties at 300C fbr months, and at 400C for a few hours, and withstand exposures of' a few minutes to temperatures well over 500C. Depending on degree of' condensation and structure, the polyimides can be used either as thermoplastic materials for compression molding, or. thermosetting fbrmulations. Related commercial polymers include polyamide-.imides,
and polyetherimides
Poly(pheny1ene oxide). The polymer
is made by oxidative coupling (Chapter 2B) It is most useful as blends with polystyrene, often filled, which have excellent dimensional stability at elevated temperatures, good electrical properties, and good resistance to aqueous environments Water absorption is exceptionally low As far as is disclosed, these are the most widely used of the high-temperature polymers described in this section, 2bout 100 million Ib being produced in 1982 A related material, poly(pheny1ene sulfide), does not have the methyl substituents on the rings Its use temperatures are higher than those of the modified phenylene oxide compositions.
1
f
Conductive Polymers. Although no commercial products are yet known, there is considerable interest in polymers with a wide range of electrical conductive properties, produced for the most part by doping with inorganics such as arsenic
HIGH-TEMPERATURE AND INORGANIC POLYMERS
pentafluoride or iodine, polymers such as poiyacetylene,
---CH=CH-CH=CHpoly-p-phenyiene,
i f
l 1
and the polyphenylene sulfide mentioned above (MorZ 1980, Seymour 1981)
Miscellaneous High-Temperature Polymers. Minor commercial polymers and others promoted for the purpose but not known to be commercially available are listed in Table 15-1
TABLE 15-1.
Structures of Miscellaneous High-Temperature Polymers
1i
i
3
Polybenzimidarole
mixed isomers Polyaryletherketone

-
Poiyarylethersulfone
I
i
430
Inorganic Polymers
Fraser (1968) has described three major classes of' inorganic polymers on which research has been concentrated. 1. Organic-inorganic polymers in which o~ganicsubstituents are placed on Inorganic chains The outstanding example, and the only family of inorganic polymers to date, is the silicones, discussed in Chapter 16D Research on other families of this type has centered on replacing the silicon in silicone-like stIuctures with other elements, such as aluminum, tin, titanium, or boron The objective is to improve the already good thermal stability of the silicones 2 Metal chelate polymers, sometimes called coordination polymers (not to be confused with the entirely different conventional organic polymers produced by coordination polymerization-Chapter 4 0 ) These materials can be prepared in several ways: by linking polydentate ligands by metal ions, as in metal acetonylacetonates; by polymer formation in the presence of metals, as in the production of polyphthalocyanines; by incorporation of metal ions in preformed polymers; by reacting chelates containing functional groups, as in the polymerization of basic beryllium ca~boxylates;and by the preparation of polymers containing fer~ocene groups, such as biscyclopentadienyl iron 3 Completely inorganic linear polymers based on silicon-nitrogen, phosphorous-nitrogen, or boron-nitr ogen chains Two examples of the last-named class deserve mention First, a wide variety of' polymers has been prepared based on modifications of polydichlorophosphazene, -NPCI,-, (itself' hydr.olytically unstable) by nucleophilic substitution reactions By way of example, the polymer OCH2CF3
--N=P--
OCHzCFj 1s a stable, film-forming thermoplastic with T,,, = 242OC and T, = -66C It is sald to be more water-repellent than either the fluorocarbon or silicone polymers Second, boron nitride (BN) fibers can be produced by treatment with ammonia of a B,O, precursor fiber The BN fibers are stated to be stable and chemically inert at temperatures of 1800-2000C.
GENERAL REFERENCES
High-Temperature Polymers. Fraser 1968, 1969; Hale 1969; Johnson 1969; 1969, 1976; Braunsteiner 1974; Marvel 1975; Kovacic 1969; Levine 1969; S~oog Elias 1977, Chapter 28 5; Hensel 1977; Schildknecht 1977; Cassidy 1980, 1982; Mort 1980; Ohm 1980; Preston 1980; Seymour 1981; Stening 1981; Boeke 1982;
Cekis 1982; Dickinson 1982; Feth 1982; Floryan 1982; Recchia 1982; Seanor 1982; Serfaty 1982; Sullo 1982. Inorganic Polymers. Stone 1962; Hunter 1964; Andrianov 196.5; Teach 1965; Block 1966; Allcock 1967, 1972, 1975, 1976, 1981, Chapter 7; Arledter 1967; Rebenftld 1967; Venezky 1967; Neuse 1968; Sander 1969; Elias 1977, Chapter 33; Car~aher1978; Economy 1980; Peters 1981
I
I
DlSCUSSlON QUESTIONS AND PROBLEMS 1.
Write chemical equations for the polymerization, as practiced industrially, of' (a) 66-nylon, (b) 6-nylon, (c) poly(ethy1ene terephthalate), (d) the polyurethane made from hexamethylene diisocyanate and 1,4-butanediol
2. Describe and explain the relation between moisture content and toughness in 66-nylon
3.
4.
Compare 66-, 610-, and 1 1-nylons with respect to water sensitivity. What post-.polymerization treatment is needed to develop the best properties in 6-nylon? Why? In what respects does silk ~esemble major synthetic fibers more than other the natural fibers?
5.
for 6. Write chemical fb~mulas (a) a repeat unit of'cellulose, (b) the fbrmation of' cellulose acetate, (c) the reaction of' cellulose with alkalis, (d) the formation of' cellulose xanthate, (e) the regeneration of'viscose, (f) the formation of ethyl cellulose.
7. Write the chemical formula for the presumed structure of the pyrolysis product of polyacrylonitrile
8.
Write chemical equations for the polymerization of (a) a polyimide, (b) poly(pheny1ene oxide), (c) a polycarbonate, (d) a polybenzimidazole, and (e) boron nitride fiber.
Akin 1962 R B Akm, Acetal Resins, Reinhold, New Yoxk, 1962 Allcock 1967 Harry R Allcock, Heteroatom Rtng Systems and Polymers, Academ~c Press, New York, 1967 Allcock 1972 Harry R Allcock, Phosphorus-Nitrogen Compounds Cyclic. Linear, andHtgh-Polymerrc Systems, Academic Press, New York, 1972 Allcock 1975 Hany R Allcock, "Poly(organophosphazenes)," Chemtech 5, 552-560 (1975) Allcock 1976 Harry R Allcock, "Polyphosphazenes New Polymers wlth Inorganic Backbone Atoms," Sc~ence 193, 1214- 1219 (1976)
432
HET'EROC,HAIN THERMOPLASTICS
Allcock 1981 Harry R Allcock and Frederick W Lampe, Contemporary Polymer Chemistry. PrenticeHall, Englewood Cliffs, New Jersey, 1981 Andrews 1979 B A Kottes Andrews and Ines V De Gruy, "Cotton," pp 176-195 in ECT, f Vol 7, 1979 Andrianov 1965 K A Andrianov, Metalorganic Polymers, Wiley-Interscience, New York, 1965 Arledter 1967 H F Arledter, F L Pundsack, and W 0 Jackson, "Fibers, Inorganic," pp 610-690 In EPST, f Vol 6 , 1967 Backus I977 John K Backus, "Polyurethanes," Chapter 17 In Calvm E Schlldknecht, ed , w~th Irving Ske~st, Polymerizatron Processes, Wiley-Interscience, New York, 1977 Balllnryne I982 Nicolaas J Balhntyne, "Polysulfone Resins," pp 832-848 In ECT, Vol 18, 1982 Berenbaurn 1969 M B Berenbaum, "Polysulfide Polymers," pp 425-447 in EPST, Vol 11, 1969 Bevlngton I964 J C Bevington and H May, "Aldehyde Polymers," pp 609-628 in EPST, Vol 1, 1964 Bikales 1971 N M Blkales and L Segal, Cellulose and Cellulose Dericatives, John Wiley & Sons, New York, 1971 Block 1966 B B Block, "Coordination Polymers," pp 150-1 65 In EPST, Vol 4, 1966 Blokland 1968 R Blokland, Ela~trcity and Structure of Polvurethane Networks, Gordon and Breach, New York, 1968 Boeke 1982 P J Boeke, "Polyphenylene Sulfide," pp 93-94 in MPE, 1982 Bogan I979 R T Bogan, C M Kuo, and R J Brewer, "Cellulose Denvatlves, Esters," pp 118142 In ECT, Vol 5, 1979 Bottenbruch I969 L Bottenbruch, "Polycarbonates," pp 7 10-764 in EPST, Vol 10, 1969 Braun 1982 David B Braun and D J De Long, "Ethylene Oxide Polymers," pp 616-632 In ECT, Vol 18, 1982 Braunsteiner I974 E E Braunstelner and H F Mxk, ''Aromatlc Polymers," J Polym Sci D9, 83126 (1 974) Brown 1982 R Malcolm Brown, Jr , ed , Cellulose and Other Natural Polvmer Systems Biogenesls, Structure, & Degradation, Plenum Press, New York, 1982 Bruins I969 Paul F Bruins, ed , Polyurethane Technology, Wiley-Intersclence, New York, 1969 Bruxelles 1965 G N Bruxelles and V R Grassie, "Cellulose Esters, Inorgan~c," pp 307-325 in EPST, Vol 3, 1965 Brydson 1978 J A Brydson, Rubber Chemrstry, Applled Sclence, London, 1978 Carraher 1978 Charles E Carraher, TI , John E Sheats, and Charles U P~ttman, , Organomefallic eds Polymers, Academic Press, New York, 1978 Cassidy 1980 Patrick E Cassidy, Thermally Stable Polymers Synthesis and Properties, Marcel Dekker, New York, 1980 Cassidy I982 Patrick E Cassidy, "Polyimides," pp 704-719 in ECT, Vol 18, 1982 Cekis 1982 G V Cekls, "Polyamide-Imide," pp 56, 58 in MPE, 1982 Daul1981 George C Daul, "Rayon Revisited," Chemrech 11, 83-87 (1981) Davis 1982 Gerald W Davis and Eric S Hill, "Polyester Flbers," pp 531-549 in ECT, Vol 18, 1982 ?Martin Grayson, ed ,Kirk-Othmer Encyclopedia ojChemica1 Technology, 3rd ed , W~ley-Interscience, New York $Herman F Mark, Norman G Gaylord, and Norbert M Bikales, eds ,Ent vclopedra ofPolymer Scrence and Technology, Wiley-Interscience, New York Joan Agranoff, ed ,Modern Plast~cs Encyclopedta 1982-1983, Vol 59, No 10A, McGraw-Hill, New York, 1982
~rcklnson 1982 B L Dicklnson, "Polyarylate," pp 61-62 In MPE, 1982 Dominguez 1982 R J G Dominguez, "Polyurethane," pp 102, 104, 106, 108 In MPE, 1982 Dreyfuss 1982 P Dreyfuss and M P Dreyfuss, "Tetrahydrofuran and Oxetane Polymers," pp 645670 in ECT, Vol 18, 1982 Economy 1980 James Economy, "Now That's an Interesting Way to Make a Flber!," Chemtech 10, 240 247 (1980) Elchhorn 1977 Robert M Eichhorn, "Polysuifone Comlng on Strong for Electncal Applications," Plastics Eng 33 (2), 53-56 (1977) Elias 1977 Hans-Georg Elias, Macromolecules 2 Synthesis and Materials (translated by John W Stafford), Plenum Press, New York, 1977 Farrow 1969 G Farrow, E S H111, and P L Weinle, "Polyester F~bers," pp 1- 41 in EPST, Vol 11, 1969 Feth 1982 George Feth, "Modified Phenylene Oxide," pp 44, 49 In MPE, 1982 Flsher 1982 William B F~sherand Lamberto Crescentini, "(Polyamides) Caprolactam," pp 425--436 In ECT, Vol 18, 1982 Floryan 1982 D E Floryan and I W Serfaty, "Polyetherim~de. More Information on a New High Performance Resm," Modern Plastics 59 (6), 146, 151 (1982) Fox 1982 D W Fox, "Polycarbonates," pp 479-494 In ECT, Vol 18, 1982 Frazer I968 A H Frazer , Hlgh Temperature Resistant Polymers, Wiley Intersc~ence,New York, 1968 F razer 1969 A H Frazer, "High-Temperature Plastlcs," Scf Am 221 (I), 96-100, 103-105 (1969) Galanty 1982 P G Galanty, E C Caughey, and P P Salatlello, "Nylon, Crystalline," pp 49-50, 52 In MPE, 1982 Goodman I969 I Goodman, "Polyesters," pp 62-128 In EPST, Vol 11, 1969 Greminger 1979 G K Gremlnger, Jr , "Cellulose Der~vatlves,Ethers," pp 143-163 in ECT, Vol 5, 1979 Hale 1969 Warren F Hale, "Phenoxy Resins," pp 111-122 In EPST, Vol 10, 1969 Haskell 1965 V C Haskell, "Cellophane," pp 60-79 in EPST, Vol 3, 1965 Hawthorne 1969 1 M Hawthorne and C J Heffelflnger, "Polyester Fllms," pp 42-61 in EPS7, Vol 11, 1969 Heme11977 Joseph D Hensel, "When, Where, and How to Use Polyimldes," Plastlcs Eng 33 (lo), 20-23 (1977) Hlll 1965 Roy 0 Hlli, Jr , B P Rouse, Jr , B Sheldon Sprague, Lawrence I Horner, and Davld I Stanonis, "Cellulose Esters-Organic," pp 325-454 in EPST, Vol 3, 1965 Hill 1982 H Wayne Hill and D G Brady, "Poly(pheny1ene Sulfide)," pp 793-814 in ECT, Vol 18, 1982 Houtz 1950 R C Houtz, " 'Orlon' Acryllc Fiber Chem~stryand Propertles," Textile Res J 20, 786-801 (1950) Hunter 1964 D N Hunter, Inorgati~cPolymers. John Wlley & Sons, New York, 1964 lbrahim 1967 S M Ibrahim and A J Ultee, "Fibers, Elastomenc," pp 573-592 in EPST, Vol 6, 1967 Jacobs 1977 Donald B Jacobs and Joseph Zimmerman, "Preparat~on of 6,6-Nylon and Related Polyamides," Chapter 12 In Calvin E Schlldknecht, ed , w~thIrvlng Skelst, Polymerization Processes, Wiley-Intersc~ence,New York, 1977 Jaqu~ss1982 Donald B G Jaqmss, W F H Borman, and R W Campbell, "Polyesters, Thermoplastic," pp 549-574 In ECT, Vol 18, 1982 Johnson 1969 R N Johnson, "Polysulfones," pp 447-463 in EPST, Vol 11, 1969 Karz 1977 Manfred Katz, "Prepa~at~on Linear Saturated Polyesters," Chapter 13 in Calvln E of
434
HETEROCHAIN THERMOPL.ASTICS
Schildknecht, ed , wlth Irving Skeist, Polymerrzatron Processes, Wiley-Intersclence, New York, 1977 Kohan 1973 Melvin I Kohan, ed , Nylon Plastrcs. Wlley-Interscience, New York, 1973 Kovacrc 1969 Peter Kovaclc and Fred W Koch, "Poly(phenylenes)," pp 380-389 In EPST, Vol 11, 1969 Layman 1982 Patricia L Layman, "Aramlds, Unlike Other Fibers, Continue Strong," Chem Eng News 60 (6), 23 (1982) Levrne 1969 H H Levine, "Polybenz~m~dazoles," 188-232 In EPST, Vol 11, 1969 pp Lundberg 1982 John Lundberg and Albin Turbak, "Rayon," pp 855-880 In ECT, Vol 19, 1982 Mark 1940 H Mark and G Stafford Whltby, eds , Collected Papers of Wallace Hume Carothers on High Polymeric Substances, Interscience, New York, 1940 Mark 1967 Herman F Mark, "The Nature of Polymeric Materials," Scr Am 217 (3), 148-154, 156 (1967) Marvel 1975 C S Marvel, "Trends In Hlgh-Temperature Stable Polymer Synthesis," 3 Macromol SCI Rev Macromol Chem C13, 219-233 (1975) Mitchell 1969 R L M~tchelland G C Daul, "Rayon," pp 810-847 m EPST, Vol 11, 1969 Moncrieff 1975 R W Moncneff, Man-Made Frber~, Newnes-Butterworths, London, 1975 Mort I980 1 Mort, "Conductive Polymers," Scrence 208, 8 19-825 (1980) Neuse 1968 Eberhard W Neuse, "Metallocene Polymers," pp 667-692 In EPST, Vol 8, 1968 Polymers and Hlgher 1,2-Epox~dePolymers," Newton 1982 Robert A Newton, "Propylene Ox~de pp 633-645 m ECT, Vol 18, 1982 N~ssan1965 Alfred H N~ssan,Gunther K Hunger, and S S Sternstern, "Cellulose." pp 131-226 in EPST, Vol 3, 1965 Ohm 1980 Robert F Ohm, "Polynorbomene, the Porous Polymer," Chemfech 10, 183-187 (1980) Ott 1954-55 Em11 Ott, Harold M Spurlln, and Mildred W Grafflln, eds , Cellulose and Cellulose Derrvatrves, Intersc~ence,New York, Parts I and 11, 1954, Part 111, 1955 Overberger 1970 C G Overberger and J A Moore, "Ladder Polymers," Adv Polym Scr 7, 113I50 (1970) Page 1982 S L Page, "Polycarbonate," p 60 In MPE, 1982 Pengilly 1982 B W Pengilly and J W H111, Thermoplastic Polyester: PET," pp 64, 69, 70 In MPE, 1982 Persak 1978 K J Persak and L M Blalr, "Acetal Reslns." pp 112-123 In ECT, Vol 1, 1978 Persak 1982 K J Persak and W G Lofthouse, "Acetal Homopolymer," pp 8, 17 In MPE, 1982 Peters 1981 Edward N Peters, "Inorganic H ~ g h Polymers," pp 398-413 In ECT, Vol 13, 1981 Prgotr 1969 K A Plgott, "Polyurethanes," pp 506-563 In EPST, Vol 11, 1969 Preston 1978 J Preston, "Aramld F~bers," pp 213-242 In EC7, Vol 3, 1978 Preston 1980 J Preston, Heat-Resistant Polymers," pp 203-225 In ECT, Vol 12, 1980 Putscher 1982 Rlchard E Putscher, "Polyamides, General," pp 328-371 in ECT, Vol 18, 1982 Rebenfeld 1967 Ludwig Rebenfeld, "Fibers," pp 505-573 in EPST, Vol 6, 1967 Recchla 1982 F P Recchia and W J Farnssey, "Thermoplastic Poly~mide," pp 82, 84 In MPE, 1982 Reitano 1982 P A Reitano, "Nylon, Amorphous (Transparent)," pp 52, 56 in MPE, 1982 Rich 1982 R P Rich, "Cellulos~c," pp 29, 30, 32 In MPE, 1982 Rigby 1982 R B Rigby and D F Dakin, "Sulfone Polymers," pp 118-1 19 in MPE, 1982 Royal 1982 C A Royal, "Acetal Copolymer," pp 17-18 In MPE, 1982 Sander 1969 Manfred Sander and H R Allcock, "Phosphorus-Contaln~ng Polymers," pp 123-144 In EPST, Vol 10, 1969
BIBLIOGRAPHY
435
sounders I962 J H Saunders and K C Fnsch, Polrurethnnes Chemrstry and Technologv Part I Chemrstry, Wiley-Intersc~ence,New York, 1962 Saunders 1964 J H Saunders and K C Fnsch, Polyurethanes Chemrstry and Technology Part 2 Technology, Wiley-Intersc~ence,New York, 1964 Saunders I982 J H Saunders, "Polyam~deFibers," pp 372-405 In ECT, Vol 18, 1982 Savage I965 A B Savage, E D Klug, Norbert M B~kales,and David J Stanonls, "Cellulose Ethers," pp 459-549 in EPST, Vol 3, 1965 Scales 1982 R E Scales, "'Ihermoplast~c Copolyester," p 70 In MPE, 1982 Schrldknecht 1977 Calvln E Schildknecht, "Synthesis of Some High-Temperature Polymers," Chapter 19 In Calvin E Schildknecht, ed , wlth Irving Skelst, Polymerization Processes, Wiley-Interscience, New York, 1977 Schnell 1964 Herman Schnell, Chemistry and Physrcs of Polytarbonates, Wlley-Interscience, New York, 1964 Schule 1969 E C Schule, "Polyamide Plastics," pp 460-482 rn EPST, Vol 10, 1969 Seanor 1982 Donald A Seanor, ed , Electrrcal Propertres of Polymers, Academlc Press, New York, 1982 Serad I979 George A Serad and J R Sanders, "Cellulose Acetate and Triacetate Fibers," pp 89117 in ECT, Vol 5, 1979 Serfaty I982 I W Serfaty and J R Bartolomucc~,"Polyethenmide," pp 72-73 In MPE, 1982 Seymour I981 Raymond B Seymour, Conductzve Polymers, Plenum Press, New York, 1981 Snrder I969 0 E Snider and R J Richardson, "Polyamide Rbers," pp 347-460 In EPST, Vol 10, 1969 Sroog 1969 C E Sroog, "Poly~m~des," 247--272 in EPST, Vol 11, 1969 pp Sroog 1976 C E Sroog, "Polylmldes," I Polym Scr D11, 161--208 (1976) Staudinger 1925 H Staudlnger, "The Constitution of Polyoxymethylenes and Other High-Molecular Compounds" (in German), Helv Chrm Acta 8, 67-70 (1925) Stenrng I981 T C Stenlng, C P Smith, and P J Kimber, "Polyaryletherketone: Hlgh Performance in a New Thermoplast~c," Mod Plastics 58 ( l l ) , 86-87, 89 (1981) Stone 1962 F G A Stone and W A G Graham, eds , Inorganic Polymers, Academ~c Press, New York, 1962 Sullo I982 N A Sullo, "Thermoset Polyimide," pp 84, 93 in MPE, 1982 Sweeny I969 W Sweeny and J Zimmerman, "Polyam~des," pp 483-597 in EPST, Vol 10, 1969 Teach I965 W C Teach and Joseph Green, "Boron Containing Polymers," pp 581-604 in EPST, Vol 2, 1965 Trtrbak 1979 Albln F Turbak, Donald F Durso, 0 A Batt~sta,Henry I Bolker, J Ross Colvin, Nathan Eastman, Theodore N Kleinert, Hans Krassig, and R St John Manley, "Cellulose," pp 70- 88 in ECT, Vol 5, 1979 Vandenberg 1979 E J Vandenberg, "Polyethers," pp 568-582 in ECT, Vol 8, 1979 Venezky I967 Davld L Venezky, "Inorgan~cPolymers," pp 664-691 in EPST, Vol 7, 1967 Welgos I982 R J Welgos, "Polyamide Plastics," pp 406-425 In ECT, Vol 18, 1982 Whzte I982 Dwain N Whlte and Glenn D Cooper, "Aromatic Polyethers," pp 594-615 in ECT, Vol 18, 1982 Wr~ght I969 P Wrlght and A P C Cummlng, Solid Polyurethane Elastomers, Gordon and Breach, New York, 1969 Yarger 1982 S B Yarger, "Thermoplastic Polyester: PBT," pp 62, 64 in MPE, 1982
CHAPTER SIXTEEN
THERMOSETTING RESINS
Thermo~etting resins are those that change irreversibly under the influence of heat from a fusible and soluble material into one which is infusible and insoluble through the formation of' a covalently crosslinked, thermally stable network. In contrast, thermoplastic polymers, discussed in Chapters 13-15, soften and flow when heat and pressure are applied, the changes being reversible In some of the systems considered in this chapter, network fbrmation occurs with little or no heat required, as in the production of urethane h a m s and the use of' unsaturated polyester resins. Furthermore, vulcanized rubbers consist of' covalently crosslinked network polymers, usually formed by the application of heat. However, this network is generated in a separate postpolymerization step. With the exception of' the silicone rubbers, thermosetting resins discussed in this chapter are those in which crosslinking occurs simultaneously with the final stages of' polymerization, regardless of the amount of' heat required in this step. The vulcanization of' rubbers is discussed in Chapter 19A. The most important thermosetting resins, both from a historic standpoint and in current commercial application, are formaldehyde condensation products with phenol (phenolic resins) or with urea or melamine (amino resins). Other thermosetting types include the epoxy resins, the unsaturated polyester resins, urethane foams, the alkyds widely used for surface coating, and minor types.
A.
PHENOLIC A N D A M I N O RESINS
Phenolic Resins
Phenolic resins have been in commercial use longer than any other synthetic polymer except cellulose nitrate In contrast to the latter, however, the sales of phenolic
436
PHENOLIC AND AMINO RESINS
437
resins continued to rise at about 15% per year, reaching a peak of about 1 3 billion Ib in 1979 Severely hit by the depression because of major uses in plywood and other construction uses, sales dropped 44% to about 0.97 billion Ibf in 1982, at an average price of $0 45 per pound
Reactions of Phenol and Formaldehyde. Phenols react with aldehydes to give condensation products if' there are free positions on the benzene ring ortho and para to the hydroxyl group. Formaldehyde is by far the most reactive aldehyde and is used almost exclusively in commercial production. The reaction is always catalyzed, either by acids or by bases.. The nature of' the product is greatly dependent on the type of' catalyst and the mole ratio o f t h e reactants The first step in the reaction is the formation of' addition compounds known as methylol derivatives, the reaction taking place at the ortho or para position:
These products, which may be considered the monomers for subsequent polymerization, are formed most satisfactorily under neutral or alkaline conditions In the plesence of acid catalysts, and with the mole ratio of formaldehyde to phenol less than 1, the methylol derivatives condense with phenol to form, first, dihydroxydiphenyl methane:
and, on further condensation and methylene bridge formation, fusible and soluble
?Beginning in 1981, phenolics production and sales volumes wexe reported as gross weights including solvents, xoughly 85% higher than the figures cited
438
linear low polymers called novolacs with the structure
~ - C H 2 f i c H 2 f i C H 2 f i - C H 2 ~
' ' 4
etc.
where ortho and para links occur at random Molecular weights may range as high to as 1000, cor~esponding about ten phenyl residues. These materials do not themselves react further to give crosslinked resins, but must be reacted with more formaldehyde to raise its mole ratio to phenol above unity In the presence of alkaline catalysts and with more formaldehyde, the methylol phenols can condense either through methylene linkages or thr.ough ether linkages In the latter case, subsequent loss of'f'ormaldehyde may occur with methylene bridge f'o~mation:
Products of this type, soluble and fusible but containing alcohol groups, are called resoles. If the reactions leading to their formation are carlied further, large numbers of phenolic nuclei can condense to give network formation In summary, the four major reactions in phenolic resin chemistry are (a) addition to give methylol phenols, (b) condensation of a methylol phenol and a phenol to give a methylene bridge, (c) condensation of two methylol groups to give an ether bridge, and (d) decomposition of ether bridges to methylene bridges and formaldehyde, the latter reacting again by the first mode.
PHENOLIC:AND AMINO RESINS
439
production of Phenolic Resins. The formation of resoles and novolacs, respectively, leads to the production of phenolic resins by one-stage and two-stage processes In the production of a one-stage phenolic resin, all the necessary reactants for the final polymer (phenol, formaldehyde, and catalyst) are charged into a resin kettle and reacted together. The ratio of formaldehyde to phenol is about 1 25 : 1, and an alkaline catalyst is used. Two-stage resins are made with an acid catalyst, and only part of the necessary formaldehyde is added to the kettle, producing a mole ratio of about 0 8 : 1 The rest is added later as hexamethylenetetramine, which decomposes in the final curing step, with heat and moisture present, to yield formaldehyde and ammonia which acts as the catalyst for curing The procedures for one- and two-stage resins are similar, and the same equipment is used for both The reaction is exothermic and cooling is required The formation of a resole or a novolac is evidenced by an increase in viscosity. Water is then driven off under vacuum, and a thermoplastic A-stage resin, soluble in organic solvents, remains. This material is dumped from the kettle, cooled, and ground to a fine powder At this point fillers, colorants, lubricants, and (if a two-stage resin) enough a hexamethylenetetramine to g ~ v e final forma1dehyde:phenol mole ratio of 1 5 : 1 are added The mixture is rolled on heated mixing rolls, where the reactions are carried further, to the point where the resin is in the B stage, nearly insoluble in organic solvents but still fusible under heat and pressure The resin is then cooled and cut into final form The C stage, the final, infusible, crosslinked polymer, is reached on subsequent fabrication-for example, by molding
Properties and Uses. Especially when combined with suitable fillers, phenolic resins have good chemical and thermal resistance, dielectric strength, and dimensional stability Products made with these resins are inherently low in flammability, are creep resistant, and have low moisture absorption By far the largest use of the phenolics (some 43% of production) is in heatsetting adhesives for plywood Related uses in insulation and fibrous and granulated wood products account for almost another 25% of production For impregnating papel, wood, and other fillers, about 8% of the phenolic resins is produced as alcoholic solutions of one-stage resins (essentially varnishes), called laminating resins These are used to produce decorative laminates for counter tops and wall coverings and industrial laminates for electrical parts, including printed circuits. The impregnated filler is dried in an air oven to remove volatiles and then is hot pressed between polished platens Shaped products, including cutlery handles and toilet seats, are similarly made A number of industrial applications are based on the excellent adhesive properties and bonding strength of the phenolics These include the production of brake linings, abrasive wheels and sandpaper, and foundry molds (sand filled) Phenolics are also widely used in the production of ion-exchange resins, with amine, sulfonic acid, hydroxyl, or phosphoric acid functional groups
440
T H E R M O S E T T I N G RESINS
Today less than 10% of'the phenolics is used as injection-molded products, once their most popular outlet. These products are used primarily in the electrical and electronic industries, with other significant uses in appliances and housewares
Amino Resins
The two important classes of amino resins are the condensation products of urea and of melamine with fbrmaldehyde They are considered together here because of the similarity in their production and applications In general, the melamine resins have somewhat better properties but are higher in price Production of amino resins in the United States in 1982 was about 1 billion Ib
Chemistry and Production. Both melamine (I), a trimer of cyanamide, and urea
react with formaldehyde, first by addition to fbrm methylol compounds and then by condensation in reactions much like those of phenol and fbrmaldehyde. The methylol reaction takes the fbrm
These may polymerize to crosslinked resins by loss of water. Other reactions that may enter into the condensation include the formation of' methyleneurea,
1
!
i
pHENOL.IC AND AMINO RESINS
and the following crosslinking reactions:
The production of the amino resins is similar to that of phenolic resins Since the A-stage resin is water soluble, it is only partially dehydrated, the water solution being used to impregnate the filler The molding resins are almost always filled with cellulose obtained from good-quality sulfite-bleached paper Impregnation is car~ied in a vacuum mixer, and the subsequent drying step carries the resin to out the B stage It is then ground to the desired particle size in ball mills
Properties and Applications. A distinct advantage of the amino resins over the phenolics is the fact that they are clear and colorless, so that objects of light or pastel color can be produced. The tensile strength and ha~dness of'the amino resins are better than those of the phenolics, but their impact strength and heat and moisture resistance are lower, although still characteristic of thermosetting resins. The melamine resins have better hardness, heat resistance, and moisture resistance than the ureas. Almost three-quarters of' the amino resins are used for adhesives, largely fbr plywood and furniture. The melamine resins give excellent, boil-resistant bonds, but for economy are usually blended with the ureas. The remaining quarter of production is divided roughly equally among uses as textile treatment and coatings resins, protective coatings, molding compounds, and paper treatment and coating resins Practically all urea molding compounds are cellulose filled, whereas the melamines, although predominantly cellulose filled, are also used with asbestos, glass or silica, and cotton fabric. Because of their poorer flow characteristics, the urea resins
442
are usually compression molded, but injection molding is now common with both resins. Both resins can be preheated by high-frequency curxent because of their high polarity Because of' their colorability, solvent and grease resistance, surface hardness, and mar resistance, the urea resins are widely used fbr cosmetic container closures, appliance housings, and stove hardware. The production of high-quality dinnerware from cellulose-filled compounds is the largest single molding use for the melamine resins
Other Applications. The amino resins modify textiles such as cotton and rayon by imparting crease resistance, stiffness, shrinkage control, fire retardance, and water repellency. They also improve the wet strength, rub resistance, and bursting strength of paper. Alkylated resins, in which butyl- or amyl-substituted monomethyl01 ureas or melamines are used, are combined with alkyd resins (Section E) to give baking enamels. The urea-based enamels are used for refrigerator and kitchen appliances, and the melamine fbrmulations in automotive finishes
GENERAL REFERENCES
Widmer 1965; Kentgen 1969; Elias 1977, Chapters 26 3 and 28 2; Updegraff 1977, 1978; Turxo 1978; Brode 1982; Keegan 1982; Lichtenberg 1982
1'
i
B.
UNSAT'URAT'ED POLYESTER RESINS
Polyesters of several diverse types are useful as polymers This section is concerned only with those in which the dibasic acid or the glycol, or both, contain doublebonded carbon atoms It is further restricted to cases in which radical chain polymerization involving these double bonds and a vinyl monomer, usually styrene, is made to take place in the presence of a fibrous filler, generally glass Other polyester systems are the fiber-forming saturated polyesters (Chapter 15B), the polyester intermediates in the production of urethane elastomers (Chapter 15B) or foams (Section C), and the alkyd paint and molding resins (Section E) The polyester systems discussed here have been known from time to time as low-pressure laminating resins, contact resins, polymerizable polyesters, and styrenated polyesters The term reinforced plastics, however, is more general and includes the use of thermosetting resins other than the unsaturated polyesters, among them the phenolic, epoxy, melamine, diallyl phthalate, and alkyd resins, all discussed elsewhere in this chapter. Badly depressed by the recession in marine products, construction, and transportation, the sales of unsaturated polyester formulations was about 860 million Ib in 1982, down 12% from 1981 and at the lowest level in 6 years; this is less than half of estimated production capacity Prices were about $0 60 per Ib
U N S A T U R A T E D POLYESTER RESINS
443
Composition of Reinforced Polyester Systems

Much of the versatility of the reinforced polyester systems lies in the wide variation in resin composition and fabrication methods possible, allowing the properties of' the product to be tailored to the requirements of' the application. This variability is manifested in the large number of resin components used. Most of' the material produced is, however, based on a few key components The unsaturation in the polyester is usually supplied by the inclusion of maleic anhydride as one component. In addition, a saturated acid or anhydride is often used, such as phthalic anhydride. A higher proportion of' unsaturated acid gives a more reactive resin, with improved stiffness at high temperatures, while more of' the saturated components give less exothermic cures and less stiff resins. Propylene glycol is the most widely used dihydroxy component of'the polyester, and styrene is the most popular by far of' the crosslinking components. A typical fbrmulation of' a general-purpose polyester may contain 43 lb of phthalic anhydride, 19 Ib of' maleic anhydride, and 38 Ib of' propylene glycol To these 100 Ib the fbrmulator will add about '70 Ib of' styrene fbr. csosslinking. The amount of' glass-fiber filler added to this may vary between 100 and 300 lb, depending upon the end use.
Formulation
The acid and glycol components of the polyester resin are mixed in a resin kettle and polymerized by step reaction to a molecular weight of 1000--5000, which is in the highly viscous liquid range After cooling, the mixture is thinned down to apourable liquid by the addition of the monomer An inhibitor such as hydroquinone is then added to prevent premature polymerization When kept cool, the mixture is stable for months to years Cure is begun by adding an initiator, usually an organic peroxide, such as benzoyl peroxide, or a hydroperoxide Typically, promoters or accelerators are used to pIomote the decomposition of the initiator at room temperature and, thus, rapid low-temperature curing Common accelerators are cobalt naphthenate or alkyl mercaptans Cure takes place in two stages: The initial formation of a soft gel is followed by rapid polymerization. The heat evolved can lead to quite high temperatures in large masses of resin
Fabrication
The fabrication of reinforced plastic articles is traditionally divided into laminating and molding processes, but this classification has been modified as newer techniques have been developed The following discussion applies to the processing of a variety of reinforced plastics and composites in addition to the polyesters
Laminating. Laminating processes are those in which separate plies of reinforcing material are coated or impregnated with resin and pressed together until c u e d to
444
T H E R M O S E T T I N G RESINS
a single reinforced structure The operation can be done batchwise, but most processing is now done continuously on an in-line conveyor on which the reinforcements are impregnated with resin, cured, and trimmed
Molding. The simplest molding process fox reinforced plastics is contact moldlng or open or hand lay-up molding, a process closely resembling laminating A single mold is used; the reinforcing material is placed on the mold, impregnated with resin, and allowed to cure in air. Variations ~nclude bag molding, where a bag or blanket is used to apply low pressure to the open surface of the material, and sprayup techniques, in which the resin and sometimes chopped glass are sprayed onto the mold surface. Matched-die molding includes processes involving two dies that, together, match or conform to all the dimensions of the finished piece The material can be placed in the die in the form of mats or preforms of reinforcing material with the resin separately added; "pre-pregs, " where the preform has been impregnated; or premixes, where the resin, reinforcement, and other extenders have been premixed Conventional molding techniques, described in Chapter 17A, can also be used with reinforced plastics if the reinforcement consists of short lengths of fiber, for example In injection and transfer molding, some damage to the reinforcement may occur, but this is minimized in compression molding Stampzng is a variation in which the heated impregnated reinforcement is placed in an open mold, which quickly closes 01 strikes the materials, holds for 5-15 sec for cooling, then opens again for part removal and reloading Very fast cycles are obtained Unsaturated polyester resins also form the basis of sheet-moldzng compounds, described in Chapter 17C
Filament Winding. In this process, continuous filaments (fibers) are wound onto or in a fbrm.. Usually either the fbrm or the supply of filament revolves in the process. The resin is used to bind or encapsulate the fibers, resulting in a solid structure.. Filament winding is used for pipes for gases and chemicals, tanks, drive shafts, and bearings, and also for nonsymmetrical objects such as automobile leaf' springs and helicopter blades..
Pultrusion. Originally devised fbr making glass fiber-reinfbrced fishing rods, this process resembles extrusion (Chapter 178) except that the materials are pulled rather than pushed through a die. Continuous filament bundles, saturated with resin, are pulled through a curing die that shapes the profile while the resin is polymerizing. Properties and Applications
The most important properties of the unsaturated polyester systems include ease of handling, rapid curing with no volatiles evolved, light color, dimensional stability, and generally good physical and electrical properties The major applications of' glass-reinfbrced polyester resins fall in the fbllowing categories: boat hulls, whose popularity has risen spectacularly; transportation,
including passenger car parts and bodies and truck cabs; consumer products, including such diverse items as luggage, chairs, and fishing rods; trays, pipe, and ducts; electrical applications; appliances; construction applications, largely sheet and paneling; and missile and radome uses
GENERAL REFERENCES
Boenig 1969; Parker 1977; Clavadetscher 1982; Ewald 1982; Jensen 1982; McLarty 1982; Menzer 1982; Raymes 1982.
C.
EPOXY RESINS AND POLYURETHANES
These thermosetting resins are considered together because of their similar chemistry, resulting from the reaction of a glycol with a reagent other than a diacid, as seen in the resins just discussed
Epoxy Resins
The epoxy resins are fundamentally polyethers, but retain their name on the basis of their starting material and the presence of epoxide groups in the polymer before crosslinking About 310 million Ib of epoxy resins was sold in the United States in 1982, at an average price of $1.25 per Ib
Chemistry of Preparation and Curing. The epoxy resin used most widely is made by condensing epichlorohydrin with bisphenol A, diphenylol propane An excess of epichlorohydrin is used, to leave epoxy groups on each end of the lowmolecular-weight (900--3000) polymer:
aq NaOH
446
T'HERMOSETTING RESINS
Depending on molecular weight, the polymer is a viscous liquid or a brittle highmelting solid. Other hydroxyl-containing compounds, including resorcinol, hydroquinone, glycols, and glycerol, can replace bisphenol A . No epoxides other than epichlorohydrin are available at attractive prices, however.. The epoxy resins are cured by many types of materials, including polyamines, polyamides, polysulfides, urea- and phenol-formaldehyde, and acids or acid anhydrides, through coupling or condensation reactions.. The reaction with amines involves opening the epoxide ring to give a P-hydroxyamino linkage:
Acids and acid anhydrides react through esterification of the secondary hydroxyl groups on the epoxy resin as well as with the epoxide groups The phenolic and amino resins may react in several ways, including condensation of methyl01 groups with the secondary hydroxyls ofthe epoxy, and reaction of the epoxide groups with phenolic hydroxyls and amino groups Epoxy resins can also be cured by cationic polymerization, using Lewis acid catalysts such as BF, and its complexes, which fbrm polyethers from the epoxide groups.
Properties and Applications. The major use of the epoxy resins is as surfacecoating materials, which combine toughness, flexibility, adhesion, and chemical resistance to a nearly unparalleled degree In addition to curing systems of the types described above, the epoxies can be esterified with drying or nondrying oil fatty acids and then cured by either air drying 01 baking Epoxy resins can be used in both molding and laminating techniques to make glass fiber-reinforced articles with better mechanical strength, chemical resistance, and electrical insulating properties than those obtained with unsaturated polyesters Only the higher p ~ i c e the epoxies prevents their wider use in this field. of Casting, potting, encapsulation, and embedment are widely practiced with the epoxy resins in the electrical and tooling industries Liquid resins are often utilized, while hot-melt solids have some application. Other important uses include industrial flooring, adhesives and solders, foams, highway sur.facing and patching materials, and stabilizers for vinyl resins Polyurethanes
As indicated in Chapter 15B, urethane polymers contain the group -NHCOOand are formed through the reaction of a diisocyanate and a glycol.. In the production of' urethane foams, excess isocyanate groups in the polymer react with water or carboxylic acids to produce carbon dioxide, blowing the foam, at the same time that crosslinking is effected..
EPOXY RESINS A N D POLYURETHANES
447
Urethane foams can be made in either flexible or rigid form, depending on the nature of the polymer and the type of crosslinking produced The flexible foam is more widely used, however, accounting for about 65% of the 1 5 billion Ib produced in the United States in 1982. Thermosetting polyurethane molding compounds fabricated by reaction injection molding (RIM) provide a new use for this polymer type, fabricated by what may become a major new processing technique
Urethane Chemistry Like polyurethane elastomers, urethane foams are made in several steps A basic intermediate of molecular weight around 1000 is a polyether made from poly(l,4-butylene glycol), sorbitol polyethers, or others. The basic inte~mediate bifunctional if flexible foams are desired and polyf'unctional if rigid is foams are to be made. As in elastomer production, the intermediate is reacted with an aromatic diisocyanate, usually tolylenediisocyanate, to give a prepolymer of structure simila~to that illustrated in Chapter 15B Catalysts based on tertiary amines or on stannous soaps axe added to achieve rapid production of foam. Crosslinking takes place through the formation of urea linkages:
RNCO R'NCO
+ H 2 0.--+ + RNH,
RNH2 + C 0 2
RNHCONHR'
The use of 1ow.boiling ine1.t liquids, in particular fluorocarbons, to augment or replace the chemical blowing action described above has led to certain property advantages in the final foams, such as low thermal conductivity characteristic of the entrapped fluorocarbon gas The ingredients of the foam may be partly expanded (frothed) with an inert gas in a preliminary step, and subsequently expanded to form the final object in a separate second step
Flexible Foams. The use of flexible urethane foams for cushions for furniture and automobiles has displaced rubber foam in these applications because of improved strength, lower density, and easier fabrication The "one-shot'' process, wherein polyether intermediate, tolylenediisocyanate, and catalysts are mixed just before foaming, is widely used for the production of flexible urethane foams. Most of the material is made in the form of slab stock in a continuous process, the foam being up to 8 ft wide and 3-4 ft high It is cut into 10-60 ft lengths and, after curing 10-24 hr, is further cut up for sale to fabricators. The foam can also be produced in molding processes, with or without external heating
Rigid Foams. Rigid urethane foams are resistant to compression and may be used to reinforce hollow structural units with a minimum of' weight In addition, they consist of closed cells and so have low rates of heat transmission. They develop excellent adhesion when they are formed in voids or between sheets of material
448
THERMOSETTING R SN EI S
Finally, they are resistant to oils and gasoline and do not absorb appreciable amounts of water. These properties make the rigid foams valuable for sandwich structures used in prefabrication in the building industry, for thermal insulation in refrigerators, portable insulated chests, and so on, and fbr imparting buoyancy to boats.
RIM Urethanes. Reaction injection molding (Chapter 17A) is a process that allows polymerization and crosslinking to take place simultaneous with fbxming of' a part into its final shape..Because ofthe rapid curing ofpolyurethanes, compatible with the fBst cycle times of RIM, these polymers seem exceptionally well suited to RIM processing. The major application to date has been in producing microcellular elastomers. Fillers can be used to produce quite stiff'products. Most of' the technology is currently used in the automotive industry, to produce bumper covers, with larger exterior parts such as fenders under development
GENERAL REFERENCES
Saunders 1962-1964; Lee 1967; Bruins 1968, 1969; David 1969; Pigott 1969; Backus 1977; Elias 1977, Chapter 26 2 3; Salva 1977; Bauer 1980; Sherman 1980; Breitigam 1982; Dominguez 1982
D. SILICONE POLYMERS
Like carbon, silicon has the capability of forming covalent compounds. Silicon hydrides (silanes) up to Si,H,, are known. The Si-Si chain becomes thermally unstable at about this length, however, so that polymeric silanes are unknown The siloxane link
is more stable, and is the one fbund in commercial silicone polymers. Unlike carbon, silicon does not form double or triple bonds. Thus silicone polymers are usually formed only by condensation-type reactions. Silicone polymers became available commercially during World War 11. They are particularly noted f b r their stability at temperatures as high as 150C..The variety of' products available ranges from liquids (lubricants, water repellents, release agents, defoamers) through greases and waxes to resins and rubbers
Chemistry of the Silicones
The study of the chemistry of silicon and its compounds began with the discove~y of the element in 1824. Soon afterwards SiCl, was prepared by reacting silicon
SILICONE POLYMERS
449
and chlorine, and ethyl silicate was first made in 1844 by the reaction of the tetrachloride with ethanol The compounds of silicon were studied intensively from that time on Silicone polymers are made from organosilicon intermediates prepared in various ways from elemental silicon, which is produced by reducing quartz in an electric furnace The intermediates ("monomers") are compounds of the type SiR,X,-,, where R is an alkyl or aryl group and X is a group which can be hydrolyzed to --SiOH, such as chlorine or alkoxy. The intermediates are made by a direct synthesis in which the R and X groups are attached simultaneously to the silicon by a hightemperature reaction of a halide with silicon in the presence of a metal catalyst The chief reaction is, for example,
but a number of side reactions occur.
Polymerization
Silicone polymers are produced by intermolecular condensation of' silanols, which ase formed from the halide or alkoxy intermediates by hydrolysis:
--Sic1 4- H z 0 --+ -SiOH
-t HCI
The desired siloxane structure is obtained by using silanols of different functionality, the alkyl (R) groups in the intermediate being unreactive.
Silicone Fluids
The silicone fluids are low polymers produced by the hydrolysis reaction mentioned above, in which a predetermined mixture of chlorosilanes is fed into water with agitation. In many cases, the cyclic tetrames predominates in the resulting mixture These compounds, not polymers in the sense of this book, are used as cooling and dielectric fluids, in polishes and waxes, as release and antifoam agents, and fbr paper and textile treatment.
Silicon Elastomers
Silicone elastomers are high-molecular-weight linear polymers, usually polydimethylsiloxanes They can be cured in several ways:
450
THERMOSElTlNG RESINS
By free-radical crosslinking with, for example, benzoyl peroxide, through the formation of' ethylenic bridges between chains. b. By crosslinking of vinyl or ally1 groups attached to silicon through reaction with silylhydride groups:
a.
R'
c.
R'
R'"
By crosslinking linear or slightly branched siloxane chains having reactive end groups such as silanols. In contrast to the above reactions, this yields Si-0-Si crosslinks..
The latter mechanism forms the basis of the curing of room-temperature vulcanizing (RTV) silicone elastomers Th'ese are available as two-part mixtures in which all three essential ingredients for the cure (silanol-terminated polymer, crosslinking agent such as ethyl silicate, and catalyst such as a tin soap) are combined at the time the two components are mixed, and as one-part materials using a hydrolyzable polyfunctional silane or siloxane as crosslinker, activated by atmospheric moisture Silicone elastomers must be reinforced by a finely divided material such as sillca if useful properties are to be obtained These materials are outstanding in lowtemperature flexibility (to - 80C), stability at high temperatures (to 250C), and resistance to weathering and to lubricating oils. They are used as gaskets and seals, wire and cable insulation, and hot gas and liquid conduits They are valuable in surgical and prosthetic devices The RTV elastomers are very convenient for caulking, sealing, and encapsulating
Silicone Resins
In contrast to the silicone fluids and elastomers, silicone resins contain Si atoms with no or only one organic substituent.. They are therefore crosslinkable to harder and stiffer compounds than the elastomers, but many must be handled in solution to prevent premature cure. They are, in fact, usually made by hydrolysis of' the desired chlorosilane blend in the presence of' a solvent such as mineral spirits, butyl acetate, toluene, or xylene. These materials are usually cured with metal soaps or amines.. The silicone resins are used primarily as insulating vanishes, impregnating and encapsulating agents, and in industrial paints. A part to be coated is typically dipped into the resin solution, and drained or scraped free of excess resin. The solvent is allowed to evaporate, and the resin is cured in an oven.
MISCELLANEOUS THERMOSETTING RESINS
451
GENERAL. REFERENCES
Lichtenwalner 1970; Brydson 1978, Chap 15 6; Owen 1981; Kookootsedes 1982; Hardman 1982
E.
MISCELLANEOUS THERMOSE'T'T'ING RESINS
Alkyd Resins
Alkyd resins (the name deriving from alcohol + acid) are polyesters used primarily m organic paints, with some molding applications About 480 million Ib was produced for paints, varnishes, and lacquers in 1982. Among many possible compositions for alkyd resins, perhaps the most common IS based on phthalic anhydride and glycerol ("glyptal") Other polyhydric alcohols commonly used are glycols, pentaerythritol, and sorbitol; other acids include maleic anhydride, isophthalic, and terephthalic Many alkyd resins are modified by the addition of fatty acids derived from mineral and vegetable oils If the acids are unsaturated, the resulting resins are the air-drying type which cure by oxidation of the (saponified) acids (drycng oils) ~aking type alkyds cuIe by heat alone, or by cocondensation with alkylated amino resins Surface coatings may be classified as lacquer types, in which drying involves only the evaporation of solvent, and varnish types, in which chemical reactions also take place on drying These reactions may involve free-radical crosslinking of drying oils, the triglycerides of unsaturated acids such as oleic, linoleic, and ~icinoleic Alkyd resins are applied in surface coating uses as follows
Plasticizing Resins. These alkyds are of the glycerol-phthalic acid or glycerolsebacic acid type and are used in lacquers along with natural resins such as shellac to impart flexibility-a plasticizing action. Drying Resins. Dr:ying alkjrds contain, in addition to the glycerol phthalate components, some d~yingoils or acids of drying oils. They are, of course, used in varnish-type coatings. Here phenol-.aldehyde or, better, urea- or melamine--aldehyde resins may be added to improve hardness Hard Resins. The hard resins are usually made from maleic anhydride combined with glycerol and rosin. Their function is to improve film hardness and gloss in both lacquer and varnish coatings.
Ally1 Resins Although allylic monomers are not used directly in chain-reaction or addition polymerization because of' the stability and consequent low reactivity of the ally1
452
radical, diallyl esters can be crosslinked by polymerization through their double bonds to give thermosetting resins. Two major types are of' commercial interest.
Diallyl Phthalate. Prepolymers (i . e . ,partially polymerized but still thermoplastic resins) of' diallyl phthalate and diallyl isophthalate are used as molding compounds and in the production of' glass fiber-reinforced laminates. They are cured by peroxide catalysts to heat- and chemical-resistant products with good dimensional stability and electrical insulating properties. Allyl Diglycol Carbonate. The ester diethylene glycol bisallyl carbonate, (CH2=CHCH20COOCH2CH,),0, is used directly fbr casting clear glass-like products similar to cast poly(methy1 methacrylate) but much harder The castings are cured with peroxide-type catalysts. They are used for special glazing applications and as spectacle lenses and for other optical purposes Furane Resins
Furane resins are based on furfuraldehyde,
HC-CH
II
II
H c ~ ~ C
--iH
0
which is derived from waste vegetable matter.. It is used in a thermosetting resin in combination with phenol, or converted to furfuryl alcohol, which itself can be thermoset by acids, or reacted with aldehydes or ketones to give polymerizable products The resins are all dark in color but are strong, have good chemical resistance, and penetrate porous surfaces well.
GENERAL REFERENCES
Patton 1962; Mraz 1964; Lanson 1978; Schildknecht 1978; Har~ington1982; Mckillip 1982; Sare 1982.
DISCUSSION QUESTIONS A N D PROBLEMS
1. Describe the molecula~species present in each of the following stages in phenol-formaldehyde polymerization, using structural formulas, giving approximate degrees of polymerization, and so forth: (a) resole or A-stage resin. (b) novoiac, (c) B-stage resin (alkaline catalyst), and (d) final product (Cstage)
)
I
I
BIBLIOGRAPHY
45 3
2. Suggest two methods of fabrication for (a) an amino resin, (b) an unsaturated polyester resin, (c) an epoxy resin, (d) a polyurethane foam, and (e) a silicone rubber
3. Write chemical formulas for each type of repeat unit found in a typical polymer of' each kind mentioned in Question 2
4.
State one advantage and one disadvantage of amino resins in comparison to phenolic resins.
BIBLIOGRAPHY
Backus I977 John K Backus, "Polyurethanes," Chapter 17 In Calvln E Sch~ldknecht,ed , w~th Irving Skeist, Polymerzzatzon Processes, W~ley-Interscience,New York, 1977 Bauer 1980 Ronald S Bauer, "The Versatile Epox~es," Chemtech 10, 692-700 (1980) Boenrg 1969 H V Boen~g,"Polyesters, Unsaturated," pp 129-168 In EPSI, 1 Vol 11, 1969 Brertzgam 1982 W V Breitigam, "Epoxy," pp 32, 34-35 In MPE,$ 1982 Brode 1982 Geo~ge Brode, "Phenoltc Reslns," pp 384-416 In ECT, Vol 17, 1982 L Brulns 1968 Paul F Brulns, ed , Epoxy Resrns Technology, W~ley-Intersc~ence, New York, 1968 Bruzns I969 Paul F Brums, ed , Polyurethane Technology, Wlley-Intersc~ence,New York, 1969 Brydson I978 1 A Brydson, Rubber Chemistry, Appl~edScience, London, 1978 Clavaderscher 1982 Dave Clavadetscher , "Closed Mold Processing," pp 357- 358, 360-361 In MPE, 1982 Davld 1969 D J Davld and H B Staley, Analytical Chemzrrry of the Polyurethanes, W~ley -Inter sclence, New York, 1969 Domlnguez 1982 R J D Domlnguez, "Polyu~ethane," pp 102, 104, 106, 108 in MPE, 1982 EIzar I977 Hans-Georg Ellas, Macrotnolecules 2 Synrherzs and Materralr (translated by John W Stafford), Plenum Press, New York, 1977 Ewald 1982 G W Ewald, "Pultius~on," pp 365-367 In MPE, 1982 Hardman 1982 Bruce B Hardman and Arnold Torkelson, "S~licones," pp 922 962 In ECT, Vol 20, 1982 Harrzngton 1982 H J Harrington, "Alkyd Polyester," p 61 In MPE, 1982 Jenren I982 J C Jensen and J F Keeney, "Unsaturated Polyesters," pp 71, 72 In MPE, 1982 Keegan 1982 J F Keegan, "Phenol~c," pp 42-44 ~nMPE, 1982 Kentgen 1969 W A Kentgen, "Phenol~cRes~ns,"pp 1-73 In EPST, Vol 10, 1969 Kookootredes 1982 G J Kookootsedes, "S~llcone," pp 114, 116 in MPE, 1982 Lanron I978 H J Lanson, "Alkyd Reslns," pp 18-50 In ECT, Vol 2, 1978 Lee I967 H Lee and K Nevllle, "Epoxy Resms," pp 209-271 In EPST, Vol 6, 1967 Lzchtenberg 1982 D W Llchtenberg, "Amlno," pp 23-24, 29 In MPE, 1982
f Herman F Mark, Norman G Gaylord, and Norbert M Bikales, eds , Encyclopedra ojPolymer Sclence and Technology, W~ley-Interscience, New Yolk :Joan Agranoff, ed , Modern Plastics Encytlopedza 1982-1983, Vol 59, No lOA, McGraw Hill, New York, 1982 $Martin Grayson, ed ,Krrk-Othmer ~ n c ylopedm of Chemr~al t Technology, 3rd ed ,Wiley-Interscience, New York
454
Llchtenwalner 1970 H Lichtenwalner and M V Sprung, "Silicones," pp 464-569 in EPST, Vol 12, 1970 Mckillip 1982 W J Mckillip, "Furan," p 39 in MPE, 1982 McLarfy 1982 J L McLarty, "Filament Winding," p 363 in MPE, 1982 Menzer I982 A B Menzer, "Continuous RP Laminating," p 362 in MPE, 1982 Mraz 1964 Richard G Mraz and Raymond P Silver, "Alkyd Resins," pp 663-734 in EPST, Vol 1, 1964 1 Owen 1981 M T Owen, "Why Silicones Behave Funny," Chemtech 1 , 288-292 (1981) Parker 1977 Earl E Parker and John R Peffer, "Unsaturated Polyester Resins," Chapter 3 in Calvin E Schildknecht, ed , with Irving Skeist, Polymerization Processes. Wiley-Interscience, New York, 1977 Panon I962 T C Patton, Alkyd Resln Technology, Wiley-Interscience, New York, 1962 Prgott I969 K A Pigott, "Polyurethanes," pp 506-563 in EPST, Vol 11, 1969 Raymer 1982 J F Raymer, "Open Mold Processing," pp 364-365 in MPE, 1982 Salva 1977 Manilal Salva and Irving Skeist, "Epoxy Resins," Chapter 16 in Calvln E Schlldknecht ed , with Irvlng Skeist, Polymerlzation Processes, Wiley-Intersc~ence,New York, 1977 Sure 1982 E J Sare, "Allyl," pp 22-23 in MPE, 1982 Saunders 1962-1964 J H Saunders and K C Frisch, Polyurethanes Chemistry and Technology Part I Chemistry (1962) and Part 2 Technology (1964), Wiley-Interscience, New York, 1962, 1964 Schlldknecht 1978 C E Sch~ldknecht,"Allyl Monomers and Polymers," pp 109- 129 in ECT, Vol 2, 1978 Sherman 1980 Stanley Sherman, John Gannon, Gordon Buchi, and W R Howell, "Epoxy Resins," pp 267-290 in ECT, Vol 9, 1980 Turro I978 N J Turro, Urea-Formaldehyde Reszns, Benjamin Cummings, Menlo Park, Cal~fornia, 1978 Updegraff 1977 Ivor H Updegraff and T J Sven, "Condensations w ~ t h Formaldehyde," Chapter 14 In Calvln E Schlldknecht, ed , with Irving Ske~st, Polymerlzation Processes, W~ley-Interscience, New York, 1977 Updegraff 1978 Ivor H Updegraff, Sewell T Moore, William F Herbes, and Philip B Roth, "Amino Resins and Plastics," pp 440-469 in ECT, Vol 2, 1978 W~dmer1965 Gustave Widmer, "Amino Resins," pp 1-94 in EPST, Vol 2, 1965
PART SIX
POLYMER PROCESSING
CHAPTER SEVENTEEN
PLASTICS TECHNOLOGY
As the earlier sections of this book have suggested, the unique physical and mechanical properties of polymers are responsible in large part for their important place in our modern life But these properties can be developed and utilized only by fabricating the polymer into useful articles or shapes The methods by which this fabrication is carried out are described in this chapter of Fabrication methods are largely determined by the rheological pr~perties the polymer in question A primary consideration is whether the material is thermoplastic, that is, retains the ability to flow at elevated temperatures for relatively long times, or thermosettzng, that is, subject to (controlled) crosslinking reactions at the temperatures necessary to induce flow, so that the ability to flow is rather quickly lost in favor of form stability Other considerations of importance in selecting fabrication methods are softening temperature, stability, and, of course, the size and shape of the end product Tadmo~ (1979) points out that methods of processing polymers, including plastics, fibers, and elastomers, have several elementary steps in common: handling of particulate solids, melting, pressurization and pumping, mixing, and often a final step of devolatilization and stripping off undesired components The only way in which the various processing methods differ is in the step of forming, in which the material is given its final shape In this chapter fabrication methods pertinent to plastics as d~stinguishedfrom fibers or rubbers are considered. Techniques specific to the latter classes of materials are discussed in Chapters 18 and 19. The first three sections of this chapter describe various forming methods Section D discussed compositions involving more than one polymer type, or one polymer and a reinforcing material that is an integral part of the smcture of the composite, while Section E is devoted to other additives, including inert fillers and plasticizers, and to some of the other steps mentioned by Tadmor Finally, Section F provides an abridged table of polymer properties
457
458
PLASTICS TECHNOLOGY
A.
MOLDING
Molding processes are those in which a finely divided plastic is forced by the application of heat and pressure to flow into, fill, and conform to the shape of' a cavity (mold). One of' the oldest methods of' polymer processing, molding can be car~ied in many different ways. We start with the simplest and oldest of these, out compression molding
Compression Molding
In compression molding, the polymer is put between stationary and movable members of a mold (Fig. 17-1). The mold is closed, and heat and pressure are applied so that the material becomes plastic, flows to fill the mold, and becomes a homogeneous mass. The necessary pressure and temperature vary considerably depending upon the thermal and rheological properties of' the polymer. For a typical compression-molding material they may be near 150C and 1000-3000 psi.. A slight excess of' material is usually placed in the mold to insure its being completely filled The rest of' the polymer is squeezed out between the mating surfaces of' the mold in a thin, easily removed film known as flash..
Injection Molding
I
I
Most thermoplastic materials are molded by the process of' injection molding. Here the polymer is preheated in a cylindrical chamber to a temperature at which it will flow and then is forced into a relatively cold, closed mold cavity by means of' quite high pressures applied hydraulically, traditionally through a plunger or ram, but today almost invariably by means of a reciprocating screw that serves the dual
Platen Mold plunger Guide pins Mold cavity Platen
Hydraulic "es'ure
I 1 ,-,-% l----k
I
plunger
I
FIG. 17-1. Diagram of a compression-molding press and mold
MOLDING
Ram
Rec~procating screw
fb)
FIG. 17-2. Sketch of (a) a ram-fed and (b) a screw-fed injection-molding machine (Reprinted with pe~missionfrom Holmes-Walker 1975 Copyright 1975 by John Wiley & Sons.)
purposes of providing the molten polymer mass (as it does in extrusion, Section
8 ) and forcing it into the mold With reference to Fig 17-2, the screw rotates to
pick up the particulate polymer, compact and melt it, mix the melt, and deliver it to the entrance to the mold The screw then moves forward (to the left in the figure), to force a fixed volume of the molten polymer into the closed mold. The melt temperature may be considerably higher than in compression molding, and pressures of hundreds to thousands of tons are common After the polymer melt has solidified in the cool mold, the screw rotates and moves backward to ready the charge of polyme~for the next cycle Meanwhile the mold is opened and the molded article is removed An outstanding feature of injection molding is the speed with which finished articles can be produced Cycle times of 10-30 sec are common, as are multicavity molds allowing the production of many parts per cycle Articles weighing up to many kilograms can be produced
Other Molding Techniques
Blow Molding. This operation can be carried out either on an extruder (Section B) or a reciprocating-screw injection machine. A section of molten polymer tubing Iparlson) is extruded into an open mold (Fig 17-3) By means of compressed air or steam the plastic is then blown into the configuration of the mold. This technique is widely used for the manufacture of bottles and similar articles In the case of large articles, such as 2-liter beverage bottles, the parison may previously have been injection molded and oriented to provide additional strength to the final blown piece
PLASTICS TECHNOLOGY
Extruded ParisonMold Open
Mold Closed and Bottle Blown
Finished Bottle Removed from Mold
FIG. 17-3. Sketch of blow-molding process (courtesy of the Society of the Plastics Industry, Inc )
Reaction--Injection Molding (RIM). This relatively new process (Fig. 17-4) starts with the unreacted components that lead to a polymer product. They are pumped in predetermined amounts into a mixing head, where they are thor.oughly mixed and injected into a warm (-65OC) mold under relatively low pressures (50-100 psi). Polymerization takes place in the mold. Meanwhile, the mixing head is cleaned, with unused monomers being recycled. To date, RIM has been used almost entirely fbr the molding of polyurethanes (Chapter 16C), but commercialization of' nylon and epoxy RIM is anticipated Rotational Molding. In this technique powdered polymer is loaded into a relatively inexpensive closed mold, which is intensively heated while being rotated biaxially The polymer coats the inner walls of the mold to a uniform thickness and is fused there The method has advantages fbr producing large hollow parts, and can be used to produce multiwall constructions by successive steps
FIG. 17-4. Basic components of a RIM processing system (Hall 1982 Reprinted with permission from Modern Plastics Enc3clopedia 1982-1983 Copy~ight1982 by McGraw-Hill, Inc )
EXTRUSION
461
Thermoset Molding. The injection molding of thermosetting resins differs only slightly from the injection molding process described above for thermoplastics First, the ingredients (B-stage resin, Chapter 16A) are heated just enough to make them fluid but not to initiate cure. In this stage their viscosity is considerably lower than that of molding-grade thermoplastics. Molds are usually heated, and this plus the frictional heat generated as the resin is injected into the mold raise its temperature enough to initiate cure The part may be removed from the mold when it is form stable but befbre the cure is completed; stored heat in the part allows the cure to be completed in a minute or so, before the part can cool Transfer Molding. In this variation of compression molding, which evolved into
the forerunner of the injection molding of' thermosetting resins just described, the resin is placed in a separate chamber, called the pot, preheated to just below cure temperature It is then injected or transferred into the heated mold where cure takes place
GENERAL REFERENCES
McKelvey 1962; Hull 1968; Rubin 19'73, 1977; Han 1976; Middleman 1977; Tulley 1977; Dym 1979; Tadmor 1979; Olmsted 1982; Richardson 1982; Rosen 1982; and specifically the f'ollowing:
Blow Moldzng Fredrickson 1982; Irwin 1982; Rainville 1982 RIM Becker 1979; Kubiak 1980; Schneider 1981; Hall 1982; Sneller 1982a Rotatzonal Moldzng Ramazzotti 1979; Fair 1982 Thermoset Moldzng Kiesau 1980; Hull 1982 Transfer Molding O'Brien 1982
B.
EXTRUSION
In the extrusion process, polymer is propelled continuously along a screw through regions of high temperature and pressure where it is melted and compacted, and finally forced through a die shaped to give the final object (Fig 17-5) A w ~ d e variety of shapes can be made by extrusion, including rods, channels and other structural shapes, tubing and hose, sheeting up to several feet wide and one-quarter in or more thick, and film of similar width down to a few thousandths of an inch in thickness The screw of an extruder is divided into several sections, each with a specific purpose (Chung 1977) The feed section picks up the finely divided polymer from a hopper and propels it into the main par.[ of the extruder In the compression section, the loosely packed feed is compacted, melted, and formed into a continuous stream of molten plastic Some external heat must be applied, but much is generated
PLASTICS TECHNOLOGY
L-
//////I///I////LLLUI//I,I,/N///,/I/I.I
Die
\ screw
FIG.17-5. Diagram of a plastics extruder
by friction. The metering section contributes to unifbrm flow rate, required to produce uniform dimensions in the finished product, and builds up sufficient pressure in the polymer melt to force the plastic through the rest of the extruder and out of the die. Since viscous polymer melts can be mixed only by the application of shearing forces (their viscosity is too high to allow turbulence or. diffusion to contribute appreciably to mixing), an additional working section may be needed befbre the die Modern trends in extruder usage include the twin-screw or multiple.-screw extruder, in which two screws turn side by side in opposite directions, providing more working of the melt, and the vented extruder, having an opening or vent at some point along the screw that can be opened or led to vacuum to extract volatiles from the polymer melt.
Coextrusion
Films or sheets consisting of' layers of two or more different polymers can be produced by mixing the molten streams from a like number. of extruders in a multimanifold die. This process can be used to combine materials to provide combinations of' properties that cannot be obtained in a single polymer For example, a film for packaging food may consist of' three layers imparting, respectively, high strength, low oxygen permeability, and heat scalability.
Film Extrusion
The extrusion of film can be carried out either by casting as in sheet extrusion (see below) or by the blown-film process. In the latter a tubular die is used, from which a hollow tube of product is extruded vertically upward toward a film tower, as shown in Fig. 17-6. The tube is blown into a thin cylindrical film by air introduced through the die and trapped in the film bubble. At the top of the tower, the bubble (by now cool) is collapsed, and subsequently slit into a flat film
EXTRUSION
FIG. 17-6. Schematic view of film blowing: (A) film die, (B) die inlet from extruder, (C) air inlet and valve, (D) plastic tube (bubble), (E) air ring for cooling, (F) guide rolls, (G) collapsing frame, (H) pull rolls, and (I) windup roll (Reprinted with permission from Richardson 1974 Copyright 1474 by John Wiley & Sons )
Pultrusion
In this variation of extrusion, material is pulled rather than pushed through a die
mold The process is used to produce continuous lengths of fiber-reinforced thermosetting resin (Fig 17-7) The reinforcing materials are continuously pulled through a bath of liquid resin that saturates each individual fiber. Excess resin is
01
Heated mold
G la
FIG. 17-7. Schematic diagram of the pultrusion process (Reprinted from Martin 1979, by cou~tesy
of Plastic r Engineering )
464
PLASTICS TECHNOLOGY
removed by a forming guide, and the mix is then pulled through a heated mold that shapes and cures it
Other Extrusion Processes

In sheet extrusion, the product from a sheet die is passed between rolls that control its thickness and apply a desired surface finish by polishing and embossing Wire and cable extrusion employs dies through which bare copper or aluminum wire is passed, and in which it is coated with plastic insulation Extrusion rates can be as high as 8000 ftlmin
GENERAL REFERENCES
McKelvey 1962; Westover 1968; Brydson 1973; Richardson 1974; Fisher 1976; Han 1976; Batiuk 1977; Klein 1977; Middleman 1977; Tadmor 1978, 1979; Levy 1981; Haisser 1982; Richardson 1982; Rosen 1982; and specifically the following:
Coextruszorz Brown 1981; Sneller 19826 Pultrusion Martin 1979; Anderson 1981
C;. OTHER PROCESSING METHODS
Calendering
Calendering is a process used for the continuous manufacture of sheet or film Granular resin, or thick sheet, is passed between pairs of highly polished heated rolls under high pressure (Fig 17-8) For the production of thin film, a series of pairs is used, with a gradual reduction in roll separation as the stock progresses through the unit Proper calendering requires precise control of roll temperature, pressure, and speed of rotation By maintaining a slight speed differential between a roll pair, it is often possible to impart an exceedingly high gloss to the film or sheet surface An embossed design can be produced on the surface by means of a calender roll, appropriately engraved By calendering a mixture of granular resln chips of varying color, it is possible to produce unusual decorative effects (e g , marblization) in the product; this technique is widely employed in the manufacture of flooring compositions A large fraction of the resin used in calendering is poly(viny1 chloride) (Chapter 1 4 0 ) The products range from thin films to thick, semirigid sheets It should be noted that, as with virtually all processing techniques, much more than just the calender itself is required to transform the polymer from raw material to finished resin (Fig 17-9)
I I
OTHER PROCESSING METHODS

Bank
FIG. 17-8. Schematic diagsam of four-roll inverted-L calendel (Reprinted from Perlberg 1977, p 1378, by courtesy of Marcel Dekker, Inc )
Casting
In casting processes, a liquid material is poured into a mold and solidified by physical (e g , cooling) or chemical (e g , polymerization) means, and the solid object is removed from the mold Casting utilizes low-cost equipment (molds can be made out of soft, inexpensive materials such as rubber and plaster) but is a relatively slow process
Casting of Thermosetting Resins. Thermosetting resins are cast by stopping the polymerization at the A stage, where the resin is still fusible and fluid After the mold is filled, the resin is cured in an oven Casting of Vinyl Polymers. Vinyl polymers, primarily the acrylic resins, are cast by preparing a syrup of monomer and polymer, and polymerizing the monomer in the mold Sheets, rods, and tubes are prepared as described in Chapter 143 Film Casting. Films, including photographic film and cellophane, are made by flowing a solution of the polymer onto an extremely smooth surface in the form of a large polished wheel or, occasionally, a metal belt or band After the solvent has
PL.ASTICS TECHNOLOGY
FIG. 17-9. Schematic diagram of typical poly(viny1 chloride) calendering system showing the princ~pal
elements that are required to convert powdered resin into a finished sheet (Reprinted from Watkins 1976, by courtesy of Plastltr Englneermng )
evaporated (or, in the case of cellophane, the polymer has coagulated) the film is stripped from the casting surface
Epoxy Resins. Cast epoxies are widely used for dies because of' their good dimensional stability and high impact strength Polyesters. Clear (unreinforced) polyester resins (Chapter 168) are cast in open molds of' ceramic, wood, plastic, or metal fbr both industrial and hobby craft work
I
I
OTHER PROCESSING METHODS
467
01
Nylon. Casting from the monomer is used for the anionic polymerization caprolactam to thick-walled parts of nylon-6
Urethane Elastomers. Liquid urethane prepolymers (Chapter 15B) can be used in several casting methods, including hand batch, centrifugal or rotational, solvent, or spray techniques Coating
The technology of coating fabrics and paper, a major outlet for some plastics, is a subject whose magnitude and complexity place it outside the scope of this book The plastic may be utilized as a melt, solution, latex, paste, or enamel or lacquer It may be applied to the substrate by spreading with a knife, brushing, using a roller, calendering, casting, or extrusion Coating processes include dipping, in which a form (such as for a rubber glove) is dipped into a suspension or latex of polymer and then into a bath of coagulating agent After several dips to build up the desired thickness, the film is st~ippedfrom the form and subjected to heat treatments to cure or c~osslinkthe resin In "slush" molding, a hollow mold is filled with a more viscous latex or "slush" of partly plasticized material such as a vinyl plastisol (Chapter 1 4 0 ) The excess is poured out, leaving a film which is heat treated and removed The processes are very flexible and involve inexpensive equipment
Foaming
The production of plastic foams is accomplished by generating a gas in a fluid polymer, usually at an elevated temperature The chemical production of a gas during polymerization to form urethane foams is described in Chapter 16C Thermoplastics are foamed by incorporating either a blowing agent, which decomposes to a gas at elevated temperature, or an inert gas The references cited at the end of this section provide comprehensive reviews
Forming Postforming Thermosetting Resins. Laminated sheets of' thermosetting resins are fbrmed into various shapes by a process resembling that used for sheet metal The sheet is heated before the final thermosetting reaction, shaped quickly in a mold or around a form, and held in place with light pressure until it sets up. This type of processing is widely practiced with sheet-molding compound (SMC), an inexpensive mix of' about one-third unsaturated polyester resin (Chapter 16B), one-. third filler (usually CaC03), and one-third chopped glass fiber. The liquid resin and filler, together with thickeners (MgO or CaO, that form weak ionic crosslinks with the acid groups on the polyester), and small amounts of curing agent (usually an organic peroxide), are mixed and spread onto layers of' polyethylene film as shown in Fig 17-10. Chopped glass fibers, about 1 in long, are added, and the
PLASTICS TECHNOLOGY
Glass rovings
'.
11,:
Resin paste
,, ..
Resin
, / ,.,
\ .
Compaction rolls
SMC
FIG. 17-10.. Sheet-molding compound line (Reprinted with permission f ~ o m Rosen 1982 Copylight 1982 by John Wiley & Sons )
sheet is compacted to wet out the glass. The polyethylene is subsequently removed from the hardened (but not polymerized) sheet, which is formed and heated to effect the cure SMC is widely used for automobile body parts such as front ends and fender extensions, and for other structural parts
Forming 'Thermoplastic Sheets. Vacuum forming is used widely for the manipulation of' cellulose acetate and acrylic resin sheeting. A sheet of' the plastic is warmed and laid across a hollow mold cavity, and a vacuum is drawn on the cavity Atmospheric pressure fbrces the sheet to conform with the mold; after cooling of the sheet, the vacuum is released, and the fbrmed object removed Many modifications of'this basic process are in use. In one, an inflatable rubber bag forces the sheet to confbrm with the mold contour. In another var.iation the sheeting is first dished by vacuum to a curvature greater than ultimately desired, a convex mold is placed into the mold cavity, and the vacuum is released while the sheet is still warm.. The sheet recovers sufficiently to conform to the shape of the convex mold and is frozen in that shape by cooling Laminating and Low-Pressure Molding
Both lamination, a high-pressure process, and low-pressure molding involve (a) the impregnation of sheets (wood, paper, hbric) with a liquid or dissolved thermosetting resin which acts as an adhesive, (b) assembly of' the individual sheets, and ( c ) compression and curing. In low-pressure molding or laminating, the sheets of impregnated material ar'e laid over a mold and held in place by a rubber mattress or bag which is inflated
MUL.TIPOLYMER S S E S AND COMPOSITES YTM
469
with steam to provide heat and pressure to hold the laminate in place and effect the cure. Many va~iationsof' the process are practiced Lamination differs from low-pressure molding in that standard shapes are produced in sufficient quantity to allow economical application of hot-press methods. Resins requiring higher temperatures for curing can also be used
GENERAL REFERENCES
McKelvey 1962; McDonagh 1977; Middleman 1977; Tadmor 1979; Crump 1980; Richardson 1982; Rosen 1982; and specifically the following:
Calendering. Meinecke 1965; Perlberg 1977; Eighmy 1982 Casti'ng Wallis 1965; Metz 1977; Schaer 1977; Luskin 1982. Coating. Higgins 196.5, 1967; Metz 1977; Werner 1977; Kiernan 1982. Foaming Skotchdopole 196.5;Benning 1969; Suh 1980; Cargile 1982; Gordon 1982; Herbert 1982; Schraf'ft 1982. Forming. Monroe 1965; Foster 1982; Rosen 1982, Chapter 20.4; Teal 1982. Laminating Power 1968; Pasquale 1982
D. MU1,TIPOLYMER SYSTEMS A N D COMPOSITES
It has been stated in this book and elsewhere that advances in producing highperformance polymer-based compositions will in the future more likely result from the utilization of new mixtures of polymers and mixtures of polymers with reinforcing components, rather than from new polymer compositions There is already much evidence to support this view, and here we provide a few final remarks on the two types of compositions just described, which for convenience we denote multipolymer systems and composites, respectively (Terminology in this area is not fully established, as will be seen )
Multipolymer Systems
I
I
It was pointed out in Chapter 7 that mixtures of two (or more) polymers do not in general form miscible systems All but a small minority form two-phase systems, the mechanical properties of which are very poor except in special circumstances Alfrey (1980) suggests that the circumstance lead~ng good, indeed often superior, to properties in immiscible systems is the development of strong interfacial adhesion between the two phases If this does not occur naturally, it may be possible to bring it about by the addition of a third polymer that adheres well to both phases, or by the use of appropriate block or graft copolymers Some examples of useful multipolymer systems, several of which are discussed elsewhere in the text, are given below
470
PLASTICS TECHNOLOGY
Multilayer Films. The production of' multilayer films by coextrusion was de.. scribed briefly in Section B . If' there is good adhesion between the layers, the mechanical properties of these laminates are generally additive, but there are some striking exceptions. For example, the presence of a high-elongation layer may prevent tensile failure due to transverse crack formation in a hard, brittle layer, but occasionally the opposite effect can be seen Polymer Blends. Both miscible and immiscible blends are known and utilized comme~cially. The best known commercial example of a miscible blend is that of polystyrene and poly(pheny1ene oxide) (PPO).. The latter is a tough, temperatureresistant engineering plastic, but difficult to process. The blend is both easier to process than PPO and higher in heat resistance and toughness than polystyrene Probably the best-known examples of' immiscible polymer blends (sometimes called alloys, but the exact meanings and differentiation of' the terms is not clear) a1.e high-impact polystyrene and the ABS resins (Chapter 14A). The former is produced by dissolving rubber in styrene monomer, which is then polymerized Phase separation takes place, and the structure is controlled to have a continuous polystyrene phase with dispersed rubber particles which themselves have inclusions of' polystyrene. In ABS resins, inclusions of butadiene-based rubber exist in a continuous phase of' styrene-acrylonitrile copolymer; the rubber molecules are graft copolymerized to styrene-acrylonitrile copolymer in order to enhance the interfacial adhesion. Simila~considerations dictate the block copolymer structure of the thermoplastic elastomers desc~ibedin Chapter 13F Interpenetrating Polymer Networks. A new class of multipolymer systems has recently been synthesized in the form of two independent interpenetrating polymer networks. One can, for example, take a lightly crosslinked polymer A, swell it with a second monomer B, plus a crosslinking agent, and polymerize B, or one can select the A and B systems to polymerize together by different mechanisms, fbr example, a styrene-divinyl benzene mixture polymerizing by radical chain mechanism together with a polyurethane network forming by a step-reaction mechanism.. Phase separation will in general take place, but the structure greatly restricts the sizes of the domains of each phase. Novel synergistic properties have been observed in some cases of interpentrating polymer networks. No commercial applications are yet known. Composites
The term composites is generally applied to fiber-reinforced engineering structural materials, in which the fibers are continuous or long enough that they can be oriented to produce enhanced strength properties in one direction.. Several examples of such systems have been mentioned in earlier chapters, for example the polyester resins reinforced by continuous glass fibers by filament winding (Chapter 168) and by pultrusion (Section B of this chapter) Other fibers providing outstanding strength in this application (albeit still at very high costs) are graphite and asamid (aromatic
AI)DITIVES AND COMPOUNDING

I
y , ~ y m i d e , Chapter 15A) The epoxy resins (Chapter 16C) are favored as the matrix for best properties 'I he graphite fibers used in high-performance composites are produced by further !{,atment of the ladder polymer made by pyrolyzing polyacrylonitrile (Chapter 15D) l a ~ l h ~ ~ g h precursors can be used, such as rayon) If the polyacrylonitrile is other ,,afrjed through the first oxidation stage (at 250-400C) under tension, and then Irr'ruJyzed at 1500-2500C to drive off all atoms but carbon, fibers with a graphite ?,tructure to a greater or lesser degree are obtained In an epoxy matrix, continuous ~aphite fibers yield composites with nearly the tensile strength and stiffness of +,lwJ,but of course with a much lower maximum use temperature The weight of rjlr: plastic composite is about a quarter that of the metal structure with comparable I,fop"rties particularly if higher production results in lower prices, an important application f ~these composites is anticipated in the automobile industry, where, for example, r they can be used to produce leaf springs for both cars and trucks, drive shafts (by fjjament winding) and many other parts, and in the aircraft industry
Multlpolymer Systems Alfrey 1980 Polymer Blends Manson 1976; Paul 1978; Fenelon 198 1; Jalbert 1982; Kamal 1982; Olabisi 1982 lnterpenetratlng Polymer Networks Sperling 1981 Composites Agarwal 1980; Beardmore 1980; Piggott 1980; Tsai 1980; Delmonte 1981; Hancox 1981; Hull 1981; McGarry 1981; Hearons 1982; Lubin 1982; Miner 1982; Peterson 1982; Sheldon 1982.
E. ADDITIVES AND COMPOUNDING
Fillers
Many plastics are virtually useless alone but are converted into highly serviceable products by combining them with particulate or fibrous solids Phenolic and amino resins are almost always filled with substances like wood flour, pure short-fiber cellulose, or powdered mica These materials greatly enhance dimensional stability, impact resistance, tensile and compressive strength, abrasion resistance, and thermal stability The use of glass fiber as a reinforcing filler for polyester resins is another important illustration Soft thermoplastics like coumarone-indene and hydrocarbon resins are usually blended with very large amounts (over 80% by weight) of mineral solids such as crushed quartz, limestone, or clay In compounds of this type, the resin functions as an interparticle adhesive; these products often have poor tensile properties but,excellent compressive strength, abrasion resistance, and dimensional
472
PLASTICS TECHNOLOGY
stability.. A final and extremely important example of' the beneficial effect of' fillers is the reinfbrcement of' rubber, discussed in Chapter 19B The principal fillers used in plastics can be divided into two types: particulate and fibrous. Among the particulate fillers are silica products, including sand, quartz, and diatomaceous earth; silicates, including clay, mica, talc, asbestos, and some synthetic silicates; glass, including granules, flakes, and solid and hollow spheres (the latter in syntactic foams); inorganic compounds, including chalk, limestone, alumina, magnesia, and zinc oxide, barytes, silicon carbide, and others; metal powders; and finely divided cellulosics (wood flour) and synthetic polymers (fluorocarbons and others) Fibrous fillers, some quite old (cellulosics) and some new "space.-age" products (metallic f'ibers, whiskers) include cellulosic fibers, such as alpha-cellulose and cotton flock; synthetic f'ibers, including nylon, polyester, acrylic, and poly(viny1 alcohol); carbon fibers made by pyrolizing materials such as rayon; boron filaments made by depositing boron from a BC1,-H, mixture onto tungsten wire, and having tensile strengths approaching one-half' million psi; and single-crystal f'ibers of' aluminum or beryllium oxide, silicon or boron carbide, or others In recent years the use of' mineral fillers has increased markedly, with emphasis on calcium carbonate, which dominates the market Custom compounding companies now offer "filler concentrates," much like color concentrates, to provide an easy way for small processors to incorporate fillers into their products
Plasticizers
Plasticizers are added to plastics to improve flow and, therefore, pracessability, and to reduce the brittleness of the product. This is achieved by lowering the glass transition temperature below room temperature, thus achieving a change in properties from those of' a hard, brittle, glasslike solid to those of a soft, flexible, tough material (Chapter 1 2 0 ) . An example is the plasticization of' poly(viny1 chloride) and vinyl chloride-acetate copolymer~s (Chapter 1 4 0 ) . similar changes in properties can, of course, be brought about by altering the molecular structure of the polymer (e..g., by copolymerization, sometimes called internal plasticization) The basic requirements that must be met by a plasticizer are compatibility and permanence. The plastizer must be miscible with the polymer. This implies a similarity in the intermolecular fbrces active in the two components, and explains why compatibility is difficult to achieve with a nonpolar polymer such as polyethylene. Permanence requirements demand low vapor pressure and a low difusion rate of the plasticizer within the polymer, both of which are obtained by the use of' high-molecular-weight plasticizers The efficiency of the plasticizer in bringing about the desired changes in prop-, erties is important in determining the proportion in which the plasticizer must be added to the resin. Plasticizer efficiency may be evaluated by a number of' different semiempirical tests.. Some of these measure the amount of nonsolvent needed to cause phase separation when added to polymer-plasticizer solutions (dilution ratio), the viscosity of dilute solutions of the polymer in the plasticizer, polymer-solvent
interacr~afi: constants measured on these solutions, the depression of the glass transjtion ~ ~ ~ ~ ~ r d t u r e viscosity of' the plasticized polymer, the electrical or the melt , mechac;i:d properties of' the plasticized polymer, or the molecular size and shape vjs~i;cyr'.;; oi'the plasticizer itself',.Not all these tests, needless to say, rate plasticizer ,-,ndjdeft5?~in the same order.. The selection of a particular plasticizer still depends a larz.5 ~ z t e n upon empirical results rather than theoretical predictions t 'The :v!k~"ing types of plasticizers are in common use: Phthalate esters, accounting for over half of the total volume of plasticizers Di(2-ethylhexyl) phthalate (dioctyl phthalate, DOP) is the dominant ;~lalticizer poly(viny1 chloride) for Yhchphate esters, chiefly tricresyl phosphate, valued primarily for their b. farneproofing characteristics c . ,'?dipates, azelates, oleates, and sebacates, used chiefly in vinyl resins fbr jmproving low-temperature flexibility
a.
d. f;p)xy plasticizers, produced by reacting hydrogen peroxide with unsaturated

;,egetable oils and fatty acids e. f-atty acid esters from natural sources, used primarily as extenders to reduce co\t (secondary plast~czzers) f. Glycol derivatives, employed mainly as lubricants and mold-release agents, and as plasticizers for poly(viny1 alcohol) g. Sulfonamides, used to plasticize cellulose esters, phenolic and amino resins, and amide and protein plastics h. Hydrocarbons and hydrocarbon derivatives, serving as secondary plasticizers
Other Additives
: Antioxidants.
The role of antioxidant in preventing or inhibiting the oxidation of polymer5 is usually filled by a substance which itself is readily ox~dized, although in some cases the antioxidant may act by combining with the oxidizing polymer to form a stable product Common antioxidants fa11 in the classes of phenols, aromatic amines, and salts and condensation products of amines and aminophenols with aldehydes, ketones, and thio compounds
Colorants. Colorants for plastics include a wide variety of inorganic and organic mater.ials. A few are molecularly dispersed (oil-soluble dyes) or have small particle size and a refractive index near that of the plastic (organ~c pzgments such as the
phthalocyanines) and lead to transparent colored products when incorporated into transparent plastics Others, including inorganic pigments, impart opacity to the plastic. Common colorants for plastics include, among many others, titanium dioxide and barium sulfate (white), phthalocyanine blues and greens, ultramarine blues, chrome greens, quinacridone reds and magentas, molybdate oranges, cadmium reds and yell^^^, iron oxide and chrome yellows, carbon black, flake aluminum for a silver metallic effect, and lead carbonate or mica for pearlescence. The coloring of
474
PLASTICS T E C H N O L O G Y
plastics is normally carried out by adding the colorants to the powdered plastic, Other techniques are tumbling, and compounding on hot rolls or in an ext~uder. occasionally used: For example, colored castings of poly(methy1 methacrylate) are produced by dissolving or dispersing the colorants in the syrup before polymerizing Compounding companies provide color concentrates, consisting of up to 70% pigment in resin, that can be "let down" with up to 200 times as much uncolored resin by the plastics process0r.. Available in a wide variety of pigments and of' resins, they provide an easy (but somewhat more expensive) means fbr the small plastic processor to obtain a well-dispersed pigment mixture for satisfactory coloring.. Ultraviolet-light absorbers are not colorants in the above sense, but have similar light-modifjring effects
Flame Retardants. The most useful material imparting flame retardance to plastics is antimony trioxide. It must be used with a source of available chlorine to be effective; it is presume that antimony oxychloride is the active flame-retarding agent. Aside from this compound and other antimony derivatives, the phosphate ester plasticizers are widely used for seducing flammability, especially in vinyl resins. The major factors in reducing the flammability of materials appear to be (a) elimination of' volatile fuel, as by cooling; (b) production of a thermal barrier, as by char~.ing, thus eliminating fuel by reducing heat transfer; and (c) quenching the chain reactions in the flame, as by adding suitable radical scavengers Stabilizers. In addition to antioxidants and flame retardants, certain polymers require stabilizers to achieve and maintain utility. An important example is the vinyl resins, whose stabilization is discussed in Chapter 140 Other applications of' stabilizers include the use of carbon black to prevent photochemical degradation (by excluding light), as in polyethylene, and of ultraviolet light absorbers, such as hydroxybenzophenones, to improve the light stability of both plastics and their colorants Compounding
The term compounding is applied both to the selection of additives to modify the properties of a polymer, and to their incorporation with the polymer to give a homogeneous mixture, in a form suitable for efficient use in the subsequent processing or fabrication step. It is in the latter sense that compounding is here considered The traditional compounding device in the plastics industry is the two-roll mill, which looks and operates much like the top half' of a calender (Fig 17-8). By proper selection of temperatures and speeds of' rotation, the plastic is made to adhere to the front roll, except as it is cut off' by the operator. The compounding ingredients are added to the plastic mass as it passes between the rolls. The roll mill has been supplanted in many operations by the compound or mixer extruder, an extruder in which the function of the mixing section of' the screw is emphasized. The use of the extruder for compounding has many advantages: con-
DlSCUSSlON QUESTIONS AND PROPERTIES
475
tamination is reduced, inert atmospheres or vacuum may be utilized, continuous processes are more readily achieved, etc Other compounding devices in common use are internal mixers such as kneaders, masticators, and paddle blenders, tumblers, and blenders
GENERAL REFERENCES
Fillers Frissel 1967; Katz 1978; Blumberg 1980; Stayner 1982; Tapper 1982; Solomon 1983 Pla~tlczzers Buttery 1960; Darby 1969; Krauskopf 1976; Sears 1982a,b; Beeler 1982 Additives Mascia 1974 Antzoxidant~ Paolino 1982 Colorants Hopmeir 1969; Webber 1979; Owen 1982; Faulks 1982 Flame Retardants Blake 1982 Stabzlners Brilliant 1982; Chasar 1982 Compoundzng Bajaj 1982; Eise 1982; Mathews 1982; Reising 1982
F.
TABLES O F PLASTlCS PROPERTIES
Tables 17-1 and 17-2 list comparative properties for typical examples of the major commercial thermoplastic and thermosetting resins, respectively It should be noted that the properties of a given plastic can vary widely, depending on compounding, fabrication, the~malhistory, and many other variables The numbe~s the tables in may represent only a fraction of the total range of values attainable and should be used only for compa~ative purposes
DlSC:USSION QUESTIONS AND PROBLEMS

1.
Compare compression and injection molding fbr speed, cost per part, investment cost, and flexibility in types of' material that can be handled
2. Describe the relative advantages and disadvantages of transfer molding, compression molding, and injection molding of thermosetting materials
3. Under. what circurnst.ances can reaction injection molding be used in place of one of' the methods listed in Question 2, and what are its advantages when it can be used?
4.
Describe the blow molding and the blown film extrusion processes, indicating where orientation takes place
TABLE 17-1.
Typical Properties of Commercial Thermoplastic Polymersa

Acrylic Fluoroplast~cs
Fluorinated
Property
ABS 1.02-1.04
Acetai Res~n 1.42 1.48 10,000 25-75 5.2 1.4-2.3 255
Poly(methy1 methacrylate)
Impact Acrylic 1.08-1.18
Cellulose Acetate
Polytetrafluoroethylene 2.14-2.20 1.35 2000-5000 2W-400 0.58 3.0 250" c2.1 <0.0002
Ethylene-Propylene Copolymer
Poly(v~nylidene fluoride)
Specific gravrty (glcrn') Refractive Index (11;~) Tensile strengih (psi) I<long:\l~ot~ (5%) Tensilc I I I O ~ U ~ U(I@ psi) S Impact strength (ft-lblin. of notch) Heat-deflect~ontemperature CF. 264 psi) Dielectnc constant (1000 cycles) Dielectnc loss (10M) cycles) Water absorpt:on, (onee~ghthIn. bar, 24 hr, %) Burnmg rate Effect of sunlight Effect of strong acids or bases Effect of organlc solvents Clar~ty
3500 -6200
s-ta
2-3.5 3-8 200-218 2.4-4.5 0.004-0.007 0.2-0.45 Slow Yellows Attacked, acids Soluble Opaaue
5000-9000 > 15-50 2-4 0.5-4.5 165-215 2.5-3.5 0.02-0.035 0.3-0.4 Slow None Attacked Soluble Transparent 0.2-0.4 Slow Slight Attacked, acids Soluble Opaque
I .7-6.5 Slow to selfext~ngulshing Slight Decomposes
,
No break
3.7 0.0098
0.25 Slow Chalks Attacked Res~stant Opaque
0.00 None Note Very reslstant Very reslstant Opaque
0.00 None None Very reslstant Very reslstant Transparent
0.04-0.06 None None Very resistant Very reslstant Transparent
Soluble Transparent
anbedo
~uelsrsaa
s p p e 'paqsvnv
sazelj slung 9Z0'0-10'0>
8 I M)'O-t(M)O'O 91 'E-P9'Z S9Z S SS'E 09 0096
t0'0-ZO'O S'P-6's 0zz-0s I O'Z-0' f Z'f-8'1 00s-09 000'2 1-0006 ES'! SI'I-1'1
~ua~edsuei~ alqnlo~ luelsrsan lq8!1s Zu!qs~n%u~lxa-~la~ ZZ'O 100'0 El' SPE rE'1 9'E 00 1-0s JOOZ'OI EE9' 1 PZ'I auoj1nsblod
anhedo 3,08 ~ o l a lue~stsax s ~uelsisa~ uo11~a101d sailnha~ MOlS 10'0>

8 100'0-S000'0>
anbedo )uelsrsai KI~A S!leyle

'P~X~~IIV
~ua~edsue~~
anbedo 3,OR Molaq ~u~~s~sax ~ueis~sav ~uels~sax

I~%!IS
anhedo luelsrsav ~uelslsax uoll~alo~d sa~lnbax MOIS X l a ~
anbedo 3008 ~ueaisax ~uea~san u011~a)old sai~nbax MOIS X l a ~ 510'0> S000'0> SE'Z-SZ'Z 501-06 9Ii XF'O-PI '0 008-06 00~~-009 15'1 SZ6.0-016'0
~
AluB[~ sjuahlos ~ 1 u e 8 iJO lsaJJ3 o saseq lo s d ! ~ auolls ~o I

J ~ J J ~
auoN auoN ZE'O ZOO'O P'E OS9 1.1 S't. 0 ' L-0's OOS'OI
stuns Z'O-r ' 0 EO'O-200'0 S'P-9P'E 901-001 C9.0-SZ'O 0.9-O't OOE-DS 005'01-00S8 99' I 6E'I-PC'! (a]eleq)qda.ia~ aualLq~a)blod
~ue~s~sav u01)3aio~d sarrnha;y MOTS ha/\ 10'0> SOOo'O> SE'Z-OE'Z OEI-011 O'Z-S'0 S'!-9'0 000 I - O i 00SS-001 E PS' I S96'0-1P6'0
I ~ % ! ( u JO 1 ~ a j j 3 ~s
ale1 B u ~ l ~ l n a
(8 'J'! PZ "eq
S000'0> SE'Z-SZ'Z
9'Z-Z'Z OPI-SZI O'Z-S'O E'Z-9' 1 OOL-OOZ 00SS-00EP 6P' 1 906'0-206'0 aualhdo~dLlod
llll[8la-~~0) lll>lld.lOSqC JalL?M (sapAs 0001) ssol >ul~ala!a ( S ~ I J K J 0001) luelsuos slilsala!a
,111
0'6-0' 1
'a,)
SL'O-SCO 056-001 000P-0061 15'1 5E6'0-816'0
(!sd WZ
EP' 1 aprmrL!od
almeiadual u o l l s a ~ a p - i c a ~ (qSlOU JO 'u!/q1-1~)q ~ % u a ilsed~ul ~s (!sa sn!npo<u al!su;\~ ( 1 uot1a4uolz~ y (!sd) q l % u a ~al!suaL ~s (Siai~) xapul a n ~ t ~ e i j a x ( c u ~ j 8L11nei8s y ~ a d ~ ) Luadold
( J ~ ~ L I I ~ ) 9rsuaa M O Xusuaa q 2 ! ~ , , l e a u r ~,, aualbqlablod
(~~L~UBIQ) Ll~suaa o ~ m
(Panu!luoJ) L-L L\31SVl
Polystyrenes Styrene-Butadiene Property Specific gravlty (g/cm3) Refractive lndex (nDZS) Tensile strength (psi) Elongation (%) Tensile modulus (10' psi) Impact strength (ft-lblin. o l notch) Heat-Oellectlon temperature ( O F , 264 psi) Dielectric constant (1000 cycles) Dielectric loss (1000 cycles) Wa~erabsorption (one-e~ghtil111. bnr, 24 hr. %) Bum~ng rate Effect of sunlight Effect of strong acids or bases Effect of orgalilc solvents Clal.lry
"Moiirm Plusrrcs
P o l y ( v ~ n y lchloride) Poly(v~nylidene Rigld 1.35-1.45 1.52-1.55 5000-WOO 2.0-40.0 3.5-6 0.4-20 130-:75 3.0-3.3 0.009-0.017 0.07-0.4 Selfext~ngu~shing Slight Resistant Soluble Transparent Plastlc~zed 1.16-1.35 chloride) (Film) 1.65-1.70
Polystyrene i.04-1.09 1.59-1.60 5000-1 2.000 1.0-2.5 4-6 0.25-0.4" 220 2.4-2.65 0.0001-0.0003 0.03-0.10 Slow Yellows Attacked, acids Soluble Transparent
Impact Polystyrene 1.04-1.10
Thermoplastic
Elastomer 0.93-1.10 1.52-1.55 600-3000 300- 1000 0.008-0.5 No break - 150 2.5-3.4 0.001-0.003 0.1L)-0.3L) Slow Slight Attacked Soluble Transparen1
1500-7000 2-80 1 .5-5 0.5-1 1

210
1500-3500 200-450
2800-3500 250-400 0.5--0.8 0.3-i .0 130-150 3.4 0.015 0. i None None Reslstanr Soiuble Tr;lnspnrent
Var~es 4.0-8.0 0.07-0.16 0.15-0.75 Slow to selfextlngulshing Slight Resistant Soluble 'Trnnsp:~rent
2.4-4.5 0.0004-0.002 0.05-0.6 Slow Loses strength Attacked, acids Soluble Opaque
(1979-1982).
"66 psi.
'A[
yicid.
I".
"One-auaner
bar.
TABLE 17-2. Typical Properties of Commercial Thermosetting Resinsa

Res~n and Filler Phenol-Formaldehyde Epoxy, Cast None 1.11-1.40 1.55-1.61 4000-13.000 3.0-6.0 3.5 0.2-1 .O 1 15-550 3.5-4.5 0.002-0.02 0.08-0.15 Slow None Attacked Res~stant Transparent Wood or Cotton 1.34-1.45 Cast None 1.10-1.46 1.52-1.57 6000-13.000 4.0 3.0-6.4 0.2-0.4 140-400 2.8-5.2 0.005-0.025 0.15-0.60 Burns Yellows Attacked Attacked Transparent Polyester Glass Cloth 1.50-2.10 Silicone Cast / RTV None 0.99-1.50 1.43 350-1000 100-300 0.09 SheetMolding Compound 1.65-2.6 Polyurethane 1.05
Property Specific gravny (g1cm1) Refractwe lndex ( 1 1 ~ ~ ~ ) Tensile strength (psi) Elongatlon (%) Tensile ~nodulus (los psi) Impact strength (ft-lblin. of notch) Heat-deflect~on temperature ( O F . 264 psi) Dieiectr~cconstant (1000 cycles) Dielectr~closs (1000 cycles) Water absorpt~on(one-e~ghth ln. bar. 24 hr, '70) Burnlng rate Effect of sunlight Effect of strong acids or bases Effect of organlc solvents Clanty
"dfodenl Plosrtcs
None 1.25-1.30
7000-8(K)O 1.0-i.5 7.5-10 0.20-0.36 240-260 4.5-6.0 0.03-0.08 0.1-0.2 Very slow Darkens Attacked
Resistant Transparent
5000-9000 0.4-0.8 8-17 0.24-0.60 260-340 4.4-9.0 0.04-0.20 0.3-1.2 Very slow Darkens Attacked Res~stant Opaque
30,000-50.000 0.5-2 15-45 5-30
8000-25,000 3 14-25 7-22 375-500 4.15-5.17 0.008-0.017
10,000-1 1.000 3-6
2.7 0.01-0.002
0.4 190-200 3.43-4.22 0.005-0.0006 1.5 Burns Slight Reslstant Res~stant Transparent
4.2-6.0 0.01-0.06 0.05-0.50 Bums Slight Attacked Attacked Translucent
0.02 0.1-0.25 Self-ext~ngu~sh~ng None None Slight Attacked Attacked. acids Atr~cked
Transparent
Resistant Opaque
11979-1982).
5. Describe the various sections of an extruder screw, indicating the purpose of each one
6 . Why is it feasible to produce multilayered films or' sheets by coextrusion; why are blends not fbrmed instead'?
7. Discuss the conditions under which immiscible polymer blends can have useful properties
8.
Compare the structures of impact polystyrene and ABS resins
9. Define and describe processes for making interpenetrating polymer networks
10. Discuss the use of (a) fillers and (b) plasticizers to improve and tailor polymer properties
BIBLIOGRAPHY
Agarwall980 D B Agarwal and Lawrence J B~outman, Analvsrs & PerformanceofFrber Composites, Wlley-inter science, New York, 1980 A@ey 1980 Turner Alfrey, Jr , and Walter J Schrenk, "Mulipolymer Systems," Science 208, 813818 (1980) Anderron 1981 Roger A Anderson, "New Structural Pultrus~onsare Stronger, More Corrosion-Resistant," Plastic, Eng 37 (lo), 21-22 (1981) Bajaj 1982 1 K L Bajaj, "Llquid and Paste M~xers," pp 337, 340, 343, 344 In MPE, f 1982 Batrick 1977 Martin Batluk, "Polyvinyl Chlorlde Extrus~on," Chapter 23 in Leonard I Nass, ed , Encvclopedra of PVC, Vol 3, Marcel Dekker, New York, 1977 Beardmore 1980 P Beardmore, J J Harwood, K R Kisman, and R E Robertson, "Fiber-Reinforced Composites: Eng~neering Structural Materials," Screnre 208, 833-840 (1980) Becker 1979 Walter E Becker, ed , Reactron Injectron Molding, Van Nostrand Re~nhold, New York, 1979 Beeler 1982 A D Beeler and D C Finney, "Plasticizers," pp 193-194, 198 in MPE, 1982 Bennrng 1969 Calvin James Benning, Plastrc Foams, John Wiley & Sons, New York, 1969 Blake 1982 W P Blake, "Flame Retardants,': pp 171, 173 in MPE, 1982 Blrrmberg 1980 John G Blumberg, James S Falcone, Jr , Leonard H Sm~ley, Dav~d Nettlng, and I "Fillers," pp 198-215 In EC7.f Vol 10, 1980 Brrllrant 1982 S D Bnlliant, "Heat Stabihze~s,"pp 180-182 In MPE, 1982 Brown 1981 Randy J Brown and James W Summers, "Deslgnlng Profile Dies for Coextrus~on,' Plastrcs Eng 37 (9), 25-29 (1981) Brydson 1973 J A Brydson and D G Peacock, Prrncrples of Plastics Extrusron, Applied Sc~ence, London, 1973 2nd Buttery 1960 D N Buttery, Plast~c~zers, ed , Franklin, Pal~sades,New Jersey, 1960 Cargile 1982 H M Carglle and J M Tower, "Melt-Processible Structural Foam Molding," pp 275276, 278 in MPE, 1982 $Joan Agranoff, ed ,Modern Plastccs Encvtlopedra 1982-1983, Vol 59, No IOA, McGraw-H111, New York, 1982 *Martin Grayson, ed , Kirk-Othmer Encvclopedra ofChemrca1 Technologv, 3rd ed , Wlley-Intersclence, New York
t9
482
PLASTICS TECHNOLOGY
Chasar 1982 D W Chasar, J T Lal, and P N Son, "Ultrav~olet Stablltzers," pp 209-210 In MPE, 1982 Chung 1977 Chan I Chung, "A Gulde to Better Extruder Screw Deslgn," Plastics Eng 33 (2). 3437 (1977) Crump 1980 E Lea Crump, "Film and Sheetlng Materials," pp 216-246 In ECT, Vol 10, 1980 Darby 1969 J R Darby and J K Sears, "Plast~c~zers," 228-306 In EPST, i Val 10, 1969 pp Delmonte 1981 John Delmonte, Technology of Carbon & Graphite Fiber Compositer, Van Nostrand Re~nhold, New York, 1981 Dym 1979 Joseph B Dym, Injection Molds & Molding A Practical Manual, Van Nostrand Relnhold, New York, 1979 Eighmy 1982 G W E~ghmy,JI , "Calendenng," pp 220, 222 In MPE, 1982 Else 1982 Kurt Else, "Fluxed Melt Mlxers." pp 335, 337-338 In MPE, 1982 Fair 1982 R L F a ~ r , "Rotat~onalMoldlng," pp 367- 368 In MPE, 1982 Faulks 1982 B F Faulks, "Color Concentrates," pp 143-144 In MPE, 1982 Fenelon 1981 Paul J Fenelon, "Mult~component Polymer Systems Could be Industry Bellwether," Plartlcs Eng 37 (9), 37-41 (1981) Fisher 1976 E G Flsher, Extruszon of Plastics, 3rd ed , John W~ley Sons, New York, 1976 & Forter 1982 Joe Foster, "Sol~dPhase Pressure Forrnlng," p 376 In MPE, 1982 Fredrickron 1982 R B Fredrickson, "Stretch-Blow Moldlng," pp 219, 221 In MPE, 1982 Frirsell 1967 W J Frrssell, "Fillers," pp 740-763 In EPST, Vol 6, 1967 Gordon 1982 J B Gordon, "Expandable Polystyrene Moldlng," pp 272-273 In MPE, 1982 Hairrer 1982 A S Halsser, "Extrus~on," pp 246, 250, 254, 258, 260, 262, 264, 267, 268 In MPE. 1982 Hall 1982 C M Hall, "React~onInject~onMold~ng,"pp 355-357 In EdPE, 1982 Han 1976 Chang Dae Han, Rheology in Polymer Procetsmg, Acadern~c Press, New York, 1976 New York, 1981 Hancox 1981 N L Hancox, ed , Fibre Composite Hybrid Matenalr, Macm~llsn, Hearonr 1982 J S Hearons, (Fibrous Reinforcements) Glass," pp 153-154 In MPE, 1982 Herbert 1982 Victor Herbert, "Mutt~component L~quldFoam Processing," pp 278, 289 In MPE, 1982 Hlgglns 1965 D G H~gglnsand Arthur H Landrock, "Coatlng Methods," pp 765--830 In EPST, Vol 3, 1965 Higgms 1967 D G H~gglns,"Fabrics, Coated," pp 467-489 In EPST, Val 6, 1967 Holmes-Walker 1975 W A Holmes-Walker, Polymer Conversion, Halsted Press, John Wlley & Sons, New York, 1975 Hopmelr 1969 A P Hopmelr, L M Greenstem, and Anthony J Petro, "P~gments," pp 157-219 In EPST, Vol 10, 1969 Hull 1968 John L Hull, Lee J Znkor, G E P~ckering,Rlchard E Duncan, Davld R Ell~s,Robert A McCord, and A B H~tchcock,"Moldlng," pp 1-157 In EPST, Vol 9, 1968 Hull 1981 Derek Hull, Introduction t o Composite Materials, Cambr~dge Unlverslty Press, Cambridge, 1981 Hull 1982 J L Hull, "Inject~onMoldiqg The~mosets," pp 314, 316 In MPE, 1982 Irwin 1982 Chrrstopher Irwin, "Extrus~on-Blow Moldlng," pp 212-214, 217 In MPE, 1982 Jalbert 1982 R L Jalbert, "Alloys," pp 127-128 rn MPE, 1982
?Herman F Mark, Norman G Gaylord, and Norbert M Bikales, eds , Encyclopedia of Polymer Scrence and Technologv, Wiley-lntersc~ence,New York
~~IBLIOGRAPHY
483
h'[~mol 1982 Musa R Kamal, "Upgrading Plastics Performance-Part 2," Plartlc Eng 38 (12), 3 135 (1982). Kolz 1978 Harry S Katz and John V M~lewski,Handbook oj Frllers and Retnforcements for Plastlcs, Van Nostrand Relnhold, New York, 1978 h'rcrmn 1982 E F K~ernan,"Solvent Castlng of PVC Film," p 226 in MPE, 1982 h'ierau 1980 M~lfriedKiesau, "Thermoset Molding. Injection, Transfer or Compreas~on-Which IS Best for Your Job3," Plastrcs Eng 36 (8), 29-32 (1980) h'lern 1977 Imnch Kleln dnd Jules W Llndau, "Extrus~on," Chapter 9 In Herman S Kaufman and Joreph I Falcetta, eds , Introduction to Polymer Scrence and Technology-An SPE Textbook, John Wlley & Sons, New York, 1977 Kruurkopj 1976 Leona~d Krauskopf, "Plast~cizers," Chapter I I In Leonard I NISS, ed , EncyiloG pedta of PVC, Vol 1, Marcel Dekker, New York, I976 K~rhlak1980 Richard S Kub~ak,"Taking RIM Beyond the Urethanes," Plastlcr Eng 36 (3), 55-61 ( 1 980) Levy 1981 Sidney Levy, Plastzcs Extrusion Technology Handbook, Industrial Press, New Yolk, 1981 Lrthln 1982 George Lubln, Handbook of Composrtes, Van Nostrand Retnhold, New York, 1982 Lurkzn 1982 L S Luskln, "Casting of Acrylic," pp 222- 224 in MPE, 1982 McDonagh 1977 John M McDonagh, "Polymei Fabncat~onProcesses," Chapter 11 In Herman S Kaufman and Joseph J Falcetta, eds , Introducrton to Polymer Sczence and Technology--An SPE & Textbook, John W~ley Sons, New York, 1977 McGarry 1981 Frede~~ck McGarry, "Laminated and Retnforced Plastlcs," pp 968-978 In ECT, J Vol 13, 1981 MtKelvey 1962 James M McKelvey, Plastrcs Processing, John Wiley & Sons, New York, 1962 Manson 1976 J A Manson and L H Sperllng, Polymer Blends and Comporltes, Plenum Press, New York, 1976 Martln 1979 Jeffrey D Martin, "Pultrusion. The Other Process," Plast~cr Eng 35 (3), 53-57 (1979) Marcla 1974 L Masc~a,The Role ojAddlttves rn Plastlcs, John Wlley & Sons, New York, 1974 Mathews 1982 George Mathews, Polymer Mrxlng Technology, Elsevier, New York, 1982 Melnecke 1965 Eberhard Meinecke, "Calendermg," pp 802-819 In EPST, Vol 2, 1965 Merz 1977 Hyman Metz and Carl Pett~grew,"M~scellaneous Moldlng, or Casting, and Coatlng Operatlons using Vinyl Powders and Liquids," Chapter 27 In Leonard I Nass, ed , Encyclopedra oj PVC, Vol 3, Marcel Dekker, New York, 1977 Mrddleman 1977 Stanley Middleman, Fundamentals ojPolymer Procesrlng, McGraw-Hill, New York, 1977 Miner 1982 L H Mmer, "(Fibrous Remforcementr) Ararnid," p 152, "AiamidlCabon," p 154, "Aiamid/Glass," p 155, "Aramid/Carbon/Glass," p 155 ln MPE, 1982 Monroe 1965 Sam Monroe, "Bag Molding," pp 300-316 In EPST, Vol 2, 1965 O'Br~en 1982 J C O'Bnen, "Tiansfe~ Molding," pp 388-389 In MPE, 1982 Olab~rr1982 Olagoke Olab~si,"Polyblends," pp 443-478 In ECT, Vol 18, 1982 Olmsted 1982 B A Olmsted, "Inject~on Molding 'Ihermoplast~cs," pp 296, 298, 302, 306, 31 1, 314 in MPE, 1982 Owen 1982 J E Owen, R A Charvat, and F A Waksmunsk~,"Colorants," pp 135, 136, 138, 140, 142, 143 ~nMPE, 1982 Paofzno I982 R Rankln Paolino, "Antioxidants," p 130, 132 in MPE, 1982 Pasquale 1982 J A Pasquale 111, "Laminating of Film," pp 318-320 in MPE, 1982 Paul 1978 D R Paul and Seymour Newman, eds , Polymer Blends, Academic Press, New York, 1978 Perlberg 1977 S Everett Perlberg and Pete1 R A Bumett, "Calender~ngand Calender Laminating,"
/
I I
484
PLASTICS TECHNOLOGY
Chapter 25 in Leonard I Nass, ed , Encyclopedia of PVC, Vol 3, Marcel Dekker, New York, 1977 Peter~on1982 H L Peterson, "(Fibrous Reinforcements) Carbon," pp 152-153 in MPE, 1982 Piggott 1980 Michael R Piggott, Load-Bearlng Composite Materials, Pergamon Press, New York, 1980 Power 1968 G E Power, "Laminates," pp 121-163 in EPST, Vol 8, 1968 Rainville 1982 Dewey Rainville, "Inject~on-Blow Molding," pp 217, 219 in MPE, 1982 Ramazzottr 1979 D J Ramazzotti, "Rotat~onalMolding: A Process Whose Time Has Come," Plastztr Eng 35 (lo), 47-49 (1979) Relsing 1982 Tom Reising, "Dry Solids Mixers," pp 333-335 In MPE, 1982 Richardson 1974 Paul N Richardson, Introduction to Extricslon, John Wiley & Sons, New York, 1974 Richardson 1982 Paul N Richardson, "Plastics Processing," pp 184-206 in ECT, Vol 18, 1982 Roren 1982 Stephen L Rosen, Fundamental Princtples of Polymeric Materralr. Wiley-Interscience, New York, 1982 Rubin 1973 Irvin I Rubin, Injection Molding Theory & Practice, Wiley-Interscience, New York, 1973 Rubin 1977 Irvin I Rubin, "Injection Molding," Chapter 10 in Herman S Kaufman and Joseph J Falcetta, eds , Introductton to Polymer Science and Technology-An SPE Tertbook, John W~ley & Sons, New York, 1977 Schaer I977 Leonard S Schaer, "Processing Plastics--Spin Casting Thermoset Parts," Plastics Eng 33 (2), 48-49 (1977) Schneider 1981 Frrtz W Schneider, "Step up to RIM," Plastlcs Eng 37 (3), 89-93 (1981) Schrafft 1982 Fred Schrafft, "Extruding Thermoplastic Foams," pp 274-275 In MPE, 1982 Sears 19820 J K Sears and N W Touchette, "Plast~cizers," pp 11 1-183 in ECT, Vol 18, 1982 Sears 1982b J Kern Sears and Joseph R Darby, Te~hnology Plast~cizers, of Wiley-Interscience, New York, 1982 Sheldon 1982 R P Sheldon, Composite Polymeric Materials, Elsevier, New York, 1982 Skotchdopole 1965 R E Skotchdopole, "Cellular Materials," pp 80-130 in EPST, Vol 3, 1965 Sneller 1982a Joseph Sneller, "Nylon RIM Surges So does RIM Epoxy The Surprise? They're Boosting Urethanes," Mod Plastics 59 (5), 46-49 (1982) Sneller 1982b Joseph Sneller, " A Sure Bright Spot in '83: Sheet and Fllm Banier Coextrusions," Mod Plastrcs 59 (12), 58-61 (1982) Solomon 1983 D H Solomon and D G Hawthorne, Chemistry of Pigments and Ftllers, John W~ley & Sons, New York, 1983 Sperling 1981 L H Spelling, Interpenetrating Polymer Networks and Related Materials, Plenum Press, New York, 1981 Stayner 1982 Vance Stayner, "Nonfibrous Property Enhancers," pp 156-158, 160, 163 in MPE, 1982 Suh 1980 K W Suh and R E Skotchdopole, "Foamed Plastics," pp 83-126 in ECT, Vol 11, 1980 Tadmor 1978 Zehev Tadmor and Imnch Kle~n,Engineerrng Prrncrples of Plasticating Extrusion. Krieger , Melbourne, 1978 Tadmor 1979 Zehev Tadmor and Costa G Gogos, Prlncrplec of Polvmer Processing, W~ley-Intersc~ ence, New York, 1979 rapper 1982 Michael Tapper, "Fillers/Extenders," pp 163, 166, 168, 171 in MPE, 1982 Teal 1982 Kevin Teal, "Thermoforming," pp 371, 374, 376 in MPE, 1982 Tsai 1980 Stephen W Tsai and H Thomas Hahn, introduction to Composite Materials. Technomic. Westport, Connecticut, 1980
Tulley 1977 F T Tulley and B C Hams, "Inject~on Moldlng of Polyvinyl Chloride," Chapter 24 tn Leonard I Nass, ed , Encyclopedm o/ PVC. Vol 3, %farce1 Dekker, New York, 1977 Wall~s1965 Benedlct L Wall~s,"Casting," pp 1-20 In EPST, Vol 3 , 1965 Wutkms 1976 W D Watktns, "Calendenng," Pluitlcs Eng 32 (61, 23-25 (1976) Webber 1979 Thomas G Webber, ed , Coloring of Plaszzcr, W~ley-Intersc~ence, Yolk, 1979 New Werner 1977 Arnold C Werner, "Spread Coating," Chapter 26 m Leonard I Nass, ed , Encyclopedza of PVC, Vol 3 , Marcel Dekker, New Yozk, 1977 Wertover 1968 R F Westovei, "Melt Extms~on," pp 533-587 In EPST, Vol 8, 1968
CHAPTER EIGHTEEN
FIBER TECHNOLOGY
The wide range of materials classified as fibers includes natural and synthetic, organic and inorganic products. Some polymers used as fibers, such as nylon and cellulose acetate, serve equally well as plastics In the final analysis, the classification of a substance as a fiber depends more on its shape than on any other property One common definition of a fiber requires that its length be at least 100 times its diameter Artificial fibers can usually be made in any desired ratio of length to diameter. Among the natural fibers, cotton, wool, and flax are often fbund with lengths 1000-3000 times their diameter; coarser fibers such as jute, ramie, and hemp have lengths 100-1000 times their diameter To be useful as a textile material, a synthetic polymer must have suitable chasacteristics with respect to several physical properties These include a high softening point to allow ironing, adequate tensile strength over a fairly wide temperature range, solubility or meltability for spinning, a high modulus or stiffness, and good textile qualities such as those defined in Section A . In addition to these primary requirements, many other properties of the material are important if' it is to be suitable for textile applications..Some of these are listed in Table 18-1 The desirable values of these properties and the molecular structure needed to obtain them are discussed in the fbllowing sections of this chapter The textile industry is quite complex, and many ofthe technologies that comprise it are intricate and obscure. A new fiber must conform to a set of desirable properties that have been largely decided upon in advance. Manipulation of molecular structure and synthesis variables does not allow ficedom to achieve all the desired properties, so compromise is inevitable. Even a successful compromise in physical properties does not ensure success for a new fiber. Other factors that must be favorable include all the processes leading to some knitted or woven cloth ready fbr wearing trials, a sound economic situation and competitive price structul,e, a good supply of raw materials, and, eventually, a good record of customer acceptance
T'EXTI1.E AND FABRIC PROPERTIES
487
TABLE 18-1.
Fiber Properties Important in Textile Uses
----
Chemical
Physical
Mechanical Tenacity Elongation Stiffness Flex life Abrasion resistance Work recovery Tensile recovery Thermal Melting point Softening point Glass transition temperature Decomposition temperature Elect~ical Surface resistivity
Biological
Toxicological Dermatological Resistance to Bacteria Molds Insects
Fabric Properties
. --
Stability toward Acids Bases Bleaches Solvents Heat Sunlight Aging Flammability Dyeability
Appearance Drape Hand Luster Comfort Warmth Water sorption Moisture retention Wicking Stability Shape Shrinkage Felting Pilling Crease resistance Crease retention
/ a1
I
United States consumption of fibers in 1982 was very close to 10 billion Ib, down some 25% from the peak year of 1979 About 75% of the 1982 total was synthetic fibers, with virtually all of the remaining 25% being one natural fiber, cotton Figures fxom earlier chapters show that polyester is by far the most widely used synthetic (3.1 billion lb in the United States in 1982), with the nylons (1 9 billion lb) and the acrylics (0 65 billion Ib) also major, followed by the rayons and polypropylene All others, including wool, play only a minor role
A.
TEXTILE AND FABRIC PROPERTIES
Definitions of Textile Terms Types of Fibers. Most artificial f i b a s may be obtained either' as a very long cont~nuousfilament, as staple, made by cutting continuous filament into relatively or short lengths Natural fibers, with the exception of silk, are obtained only in the
1
j
form Of Denier. The denier of a fiber, a measure of its size, is defined as the weight in grams of 9000 m of the fibel it is thus proportional to the density of the fiber and to its cross-sectional area At least ten other measuies of the size of fibers are in use, but the denier is the most widely accepted
Tenacity. The tensile strength of a fiber is usually expressed in terms of tenaczty Tenacity is defined as the strength per unit size number, such as the denier, where the size number is expressed as a weight per unit length Tenacity is thus a function
488
FIBER TECHNOLOGY
of' the density of the f'iber as well as its tensile strength. Two fibers having the same denier and the same breaking strength have the same tenacity, usually expressed in grams per denier; if' they have the same density, they have the same tensile strength as well. Tensile strength (pounds per square inch) = tenacity (grams per denier) x density X 12,791..
Moisture Content and Moisture Regain. Most textile fibers absorb some moisIf' ture from their sur~oundings. the amount of' moisture present at equilibrium under standard conditions (65% relative humidity and 70F) is expressed as a percentage of'the total weight of' the (moist) fiber, it is known as the moisture content; if it is expressed as a percentage of the oven-dry weight (1 10C) of'the f'iber, it is known as the moisture regain Crimp. Crimp is the waviness of' a fiber, a measure of the difference between the length of' the unstraightened and that of the straightened fiber Some naturally occurring fibers, notably certain wools, have a natural crimp. Crimp can be artificially produced in fibers by suitable heat treatment or by rolling them between heated, fluted rolls Fabric Property Terms. The esthetic qualities of' fabrics are defined in terms such as appearance, hand (or handle), and drape Although the qualitative meanings of these terms are obvious, their quantitative definitions and interpretations in terms of fiber properties are quite difficult. Hand and drape are largely determined by the tensile and elastic behavior of'the fiber Properties of Textile Fibers Electrical Properties. Since fibers are not normally used in electrical applications, their only electrical property of' great interest is their resistivity Too high resistivity leads to the development of static electrical charges, which cause the fabric to cling unpleasantly and to be difficult to clean. Some of' the synthetic fibers, including the nylons, polyesters, and acrylics, are poor in this respect
The mechanical properties of fibers are quite complex and have been the subject of' much experimental work. A stressed textile f'iber is a complicated viscoeIastic system, in which a number of irreversible processes can take place. Some typical fiber stressstrain curves are shown in Figs. 18-1 and 182 . They may be divided roughly into two groups: the silklike curves, featuring almost constant high modulus to the break point, and the wool-like curves, featuring a sharp drop in modulus at low stress fbllowed by a long period of' elongation at almost constant stress. Intermediate characteristics are shown by some of' the synthetics such as nylon: The fir.st region of' reversible elasticity is followed by an irreversible region in which drawing takes place.. When orientation is complete, another reversible elastic region is found. This may end at the point of' failure, or plastic flow may take place
Mechanical Properties.
FIG 18-1. Stress-strain curves of silklike fibers fHeckert 1953)
. .
Elongation, %
FIG. 18-2. Stress-strain curves of wool-like fibers (Heckert 1953)
490
FIBER TECHNOLOGY
The tenacity at break of typical fibers ranges from about 1 gldenier for the highly oriented polyester and over 20 for the strongest aramid rayons up to 10 f o ~ and carbon fibers The range of' 1-5 g/denier is considered suitable fbr textile applications.
Moisture Regain. The moisture regain of' the synthetic polyesters, nylons, and acrylics is so much lower than that of any of' the natural fibers that the synthetics may be classed as hydrophobic materials This is an advantage for rapid drying Dyeability. Hydrophobic fibers are in general difficult to dye The acrylics are particularly bad in this respect, and modification of the polymer structure plus the development of new classes of dyes was resorted to in order to provide a wide range of pleasing colors in this material Nylon and polyester are intermediate I n dyeability, while cellulose and the cellulosic fibers are eminently dyeable Chemical Stability. All textile fibers must be stable to water, dry-cleaning solvents, and dilute acids, alkalis, and bleaches, and all are reasonably adequate in this respect Silk and nylon are the least satisfactory fibe~s from the standpoint of weatherability, whereas the acrylics are outstanding in this respect Fabric Properties Esthetic Factors. (See preceding definitions of textile terms.) Silk is the out.standing example of' a fiber with good esthetic properties Filament acetate, which possibly ranks next, is replacing filament viscose rayon in many applications for which luster, hand, and drape are more important than good mechanical properties Among the staple fibers, the polyesters appear to be most nearly wool-like and most pleasing. Comfort. The comfort of a fabric is a highly important but little understood property. It is related to the structure of' the fabric in determining ventilation and heat-insulating characteristics and wicking ability. Comfort is not, however, dependent upon high moisture absorption in the fiber. Crease Resistance and Crease Retention. These two properties are related in that a fabric that can be creased only with difficulty tends to retain its crease. The thermoplastic fibers such as nylon, polyesters, and acrylics are very good in both respects. Creasing in wear becomes more important at high moisture contents Cellulosics are naturally poor in crease resistance, although they may be modified by resin finishes to give good crease resistance Fabric Stability. The stability of shape and dimensions of the synthetic fibers is outstandingly good Wool retains its shape well in garments, but felts and shrinks under wet treatment
SPINNING
491
The phenomenon of "pilling" has become important in some of the synthetics A small nodule or pill can be formed on some fabrics by gentle rubbing, as the surface fibers are raised and tangled. In wool these pills are not important because they wear off rapidly, but in nylon and the polyesters the pills do not break away and may become quite unsightly..
Wear Resistance. There is no laboratory test fbr abrasion in a fabric that is correlated with actual use. There is evidence that two or more independent parameters are required to define wear in a fabric. In general, acetate rayon is most easily abraded, with viscose, cotton, and wool following as a group (except that viscose when wet is not much better than acetate), while the polyesters, acrylics, and nylon are progressively better
GENERAL REFERENC,ES
Mark 1967-1968; McIntyre 1968; Goswami 1977; McGovern 1980; Peters 1980; Roberts 1980
B.
SPINNING
The conversion of bulk polymer to fiber form is accomplished by splnning In most cases, spinning processes require solution or melting of the polymer (an exception 1 is the spinning of fibers from an aqueous dispersion, as in the case of polytetrafluoroethylene) If a polymer can be melted under reasonable conditions, the production of a fiber by melt spznnzng is preferred over solution processes When melt spinning cannot be canied out, a distinction as to type of process is made depending upon whether the solvent is removed by evaporation (dry sp~nnzng) by leaching out or into another liquid which is miscible with the spinning solvent but is not itself a solvent for the polymer (wet spznnlng) Dry spinning has some advantages over wet spinning. The three processes have many features in common The conversion of the spun polymer melt or solution to a solid fiber involves cooling, solvent evaporation, or coagulation, depending on the type of spinning used The rates of these processes decrease in the order listed Cooling of a fine 4 filament is normally very rapid and can be controlled within relatively narrow limits Solvent evaporation involves simultaneous outward mass transfer and inward heat transfer, the rate-controlling step invariably being outward diffusion of solvent Coagulation involves two-way mass transfer, the coagulating agent (e g , acid) { diffusing inward, and the products of coagulation (e g , salts, H,S) diffusing out $ As a consequence of these facts, it is very easy to obtain a melt-spun fiber that possesses uniform properties throughout its cross section, but almost impossible to 1 do so with solvent-spun or coagulated fibers In addition, the rapid cooling of meltspun fibers tends to ploduce an almost circular cross section, whereas with solvent-
492
FIBER TECHNOLOGY
FIG. 18-3.
Photomicrographs (X250) showing cross-sectional shapes of f~bers: cotton, (b) melt(a) spun polyester, (c) dry-spun acetate, (d) wet-spun conjugate viscose ftber [(a)-(c) Riley 1956, (d) Hicks 19671
spun fibers the cross sections are usually elliptical, and with coagulated fibers (because of the absence of strong surface-tension effects) the cross sections are ordinarily highly convoluted, as shown in Fig 18-3 It is found, however, that convoluted fibers have desirable esthetic properties, particularly hand and luster, so that an effort is made to produce the effect at will This can be done in melt spinning by the use of noncircular orifices in the spinneret (see below), or in several types of spinning by extruding two different polymers through the same orifice to produce a single conjugate fibers (Hicks 1967), as also depicted in Fig 18-3. The production of such bicomponent and biconstituent fibers is becoming increasingly important, especially since they provide a means for producing the textured products described in Section C Melt Spinning The process of melt spinning is inherently simple Molten polymer is pumped at a constant rate under high pressure through a plate called a splrzneret containing a large number of small holes The liquid polymer streams emerge downward from the face of the spinneret, usually into air They solidify and are brought together
i i 1
i f
: ,
494
FIBER TECHNOLOGY
to form a thread and wound up on bobbins. A subsequent drawing step is necessary to orient the fibers The polymer is melted by contacting a hot grid in the fbrm of steel tubing, which is heated by electric current or by hot vapors (Fig. 18-4a) It is usually necessary to protect the polymer melt fiom oxygen by blanketing it with steam or an inert gas such as carbon dioxide or nitrogen.. If the viscosity of' the molten polymer is low, it may pass directly to the metering (constant-rate) pump.. For melts of'highe~ viscosity a booster pump may be used. Other methods of melting have been proposed, including the use of an extrusion-type screw, one section of which can serve as its own metering pump.. Methods in which the metering pump is replaced, f o ~ example, by a source of' gas pressure or other device, do not appear to offer sufficiently precise control to hold the denier of fine yams constant. The spinneret may consist of'a 2- to 3-in.. diameter steel disk about one-quarter in. thick with 50-60 countersunk holes 0 010 in. or less in diameter. The denier of' the filament is determined not by the diameter of the holes but by the rate at which polymer is pumped through the spinneret and the rate at which the filaments are wound up. The filaments emerge from the spinneret face into air and begin to cool An air blast may be used to speed up the cooling process After the filaments have traveled far enough to become solid (about 2 ft) they are brought together and wound up Speeds of' about 2500 f'tlmin are usually employed The filaments as spun are almost completely unoriented Most of'the stretching that occurs between the spinneret and windup does so while the filament is still molten, and there is sufficient time fbr molecular orientation to relax befbre the fiber cools and crystallizes. Consequently a separate drawing step is necessary to produce the orientation ofthe crystallites necessary for optimum physical properties. In the drawing step somewhat lower speeds were traditionally required than can be achieved in spinning, so the two steps were usually carried out separately; now they are often done in a single continuous operation. The drawing step utilizes two sets of rolls (Fig. 18-5), one to feed the undrawn yarn fiom a supply package at velocity u , and the other, moving about four times as fast, to collect the drawn yarn at velocity vz The filaments may pass over a metal pin between the two sets
Stretching zone
C:z::-z:--:-~:~I)
iCHeater (optional)
=--- -------------3 - - - - - - - - - -Control rolls Snubbing pin

u2 (u2
>ull
Drawn yarn to bobbin
Skewed idler roll
Undrawn pretwisted yarn

FIG. 18-5.. Schematic diagram of a typical drawing process (Riley 1956)
SPINNING
of rolls; drawing is localized in the neighborhood of the pin The yarn is then collected on a strong metal bobbin Freshly drawn yarn has a tendency to contract somewhat in length Heat is evolved during the drawing process, since work is done on the polymer Melt spinning is used for polyester and nylon fibers, among several others, and thus accounts for the vast majority of synthetic fiber production
Dry Spinning
In the dry-spinning process, the filament is formed by the evaporation of solvent from the polymer solution into air or an Inert gas atmosphere The solvent must, of course, be volatile Dry spinning has been used for many years for spinning cellulose acetate from acetone solution, and is now also employed for spinning polyacrylonitrile and the axamid fibers from solutions in dlmethyl formamide or dimethyl acetamide So far no process is used in which the polymer is made in the same solvent from which it is dry spun Therefore the first step In dry spinning is always the preparation of the polymer solution The choice of solvent for spinning is based on considerations of solvent power, boiling point, heat of evaporation, stability, toxicity, ease of recovery, and so on Nonpolar solvents are preferred because of convenient boiling points and nonhygroscop~c nature, but may cause hazardous buildup of static charge Low-boiling solvents with high heats of evaporation may cause polymer precipitation or coagulation on the surface of the fiber, in part due to condensation of moisture as the fiber cools when solvent evaporates This leads to loss of luster are and strength High solution concentrat~ons desired, and elevated temperatures are often used to keep viscosities from becoming too high Cellulose acetate IS spun from 20-45% solutions with 400-1000 poise viscosity at 40C About 22% of the from acrylic fibers is produced by spinnlng polyacrylon~tr~le solution in dimethyl formamide Because of the higher boiling point of this solvent, up to 50% residual solvent can remain In the fiber after it is spun; this must be washed out in a subsequent step As in other types of spinning, poiymer solutions are usually filtered just before dry spinning Spinning is done in a vertical tubular cell (Fig 18-4b), jacketed for temperature control, in which air, steam, or inert gas may be passed through either concur~ently or countercur~entlyas required Downward spinning is prefened for small-denier high deniers, to control draw better by eliminating fibers and upward spinning f o ~ the influence of gravity Dry spinnlng is carried out at rates as high as 2500-3000 ftlmin, about the same as melt spinning
Wet Spinning
In the wet-spinning process, a solution of a polymer or polymer derivative is spun
11
I
through a spinneret into a liquid that can coagulate the polymer or derivative (Fig 18-4c) The chemical reaction to convert a derivative to the final polymer can take place simultaneously or later Wet spinning is commonly used for viscose, cellulose, and some synthetic fibers such as polyacrylonitrile spun from salt solutions
496
FIBER TECHNOLOGY
The essential feature in wet spinning is the transfer of the mass of the solvent from the polymer to the coagulating bath. This transfer is not accompanied by heat of solidification of the polymer, as in melt spinning, or heat of evaporation of the solvent, as in dry spinning. A heat of chemical reaction may be present but is not an essential part of the process As in other types of spinning, the surface or interfacial tension forces around the filament are quite strong so that it is difficult to produce at the spinneret a filament with a cross section other than circular. These fbrces also tend to make the filament break up into drops This tendency is opposed by the viscosity of the polyme~;hence more viscous solutions are easier to spin. It is also advantageous to work with as high concentration of' polymer as possible, for obvious reasons. This again leads to high viscosity, and the practical limit may be reached above which the polymer cannot be filtered, pumped, or extruded For this reason some polymers ase wet spun at elevated temperatures where viscosity is lower. Higher temperatures usually promote crystallization and lead to denser, stronger fibers. Cellulose fibers may be spun at 50C, acrylonitrile fibers as high as 160-185"C, from either 20-28% solutions in dimethyl acetamide or 10-15% solutions in 4 5 4 5 % aqueous sodium thiocyanate. In each case the fiber is spun into a more dilute aqueous solution of the solvent During coagulation several processes take place simultaneously, including diffusion, osmosis, and salting out. Because of the interplay of these processes, coagulation occurs in rather different ways for different fiber-solvent systems Usually coagulation is rapid and the fibers cannot be stretched greatly during this step. The skin of the fiber sets to a gel, and later the volume of material in the core is reduced as the rest of the solvent is removed.. The skin then has to fold to accommodate the reduced volume. This gives a characteristic wrinkled cross section to some fibers, such as viscose rayon. It is difficult to produce by wet spinning fibers whose final cross section is circular, despite their initial production at the spinneret in this shape. As a result of the shr.inkage, the skin of'the fibers is oriented much more than the core In other cases, coagulation is much slower and the filaments may be stretched to as much as 30 times their original length before coaguIation (as in the spinning of' cellulose from cuprammonium solutions into water) The need for sufficient time for coagulation and other treatments depending on diffusion, and the considerable viscous drag of the coagulating baths, limit the rates of wet spinning to 150-300 ft/min in the usual cases.. These speeds are low enough that drawing can be done immediately after spinning in a continuous operation Wet-spun yarn may be collected on a bobbin or as a loose cake, which is subjected to chemical treatments, washing, and drying. Alternatively, these steps may be car~ied continuously as indicated in Fig. 18-4c out
Other Spinning Methods
Insoluble and effectively nonfusible polymers such as pol,ytetrafluoroethylenecan be spun by the fbllowing technique or others achieving the same purpose: To a
FIBER AFTER-TREATMENTS
497
solution of ripened cellulose xanthate, an aqueous dispersion of' polytetrafluoroethylene is added until the polymeric constituents are about 95% flu010 polymer and 5% cellulosic. This mixture is wet spun. The cellulose is then decomposed completely, and the polytetrafluoroethylene sintered into a continuous fiber, by contact with a metal roll heated to around 390C. Elastomeric fibers based on the polyurethanes can be reaction spun by the fbllowing method: A hydroxy-terminated polyeste~or polyether of' molecular weight 1000-2000 is reacted with a diisocyanate in a 1 :2 mole ratio to form an isocyanateterminated prepolymer.. This is spun into an aqueous solution of' a diamine, with polymer formation and fiber formation taking place simultaneously. More often, however, these polymers are synthesized, then either wet or dry spun from solutions in dimethyl formamide or dimethyl acetamide
1
t
GENERAL REFERENCES
Corbikre 1967; Mark 1967; Siclari 1967; Smith 1967; Ziabicki 1967; McIntyre 1968; Middleman 1977; IIobson 1978; Preston 1978; Peters 1980
/
11
1
C.
FIBER AFTER-TREATMENJS
'
j 1
1I
1
I
I j
Although it is beyond the scope of this book to treat in detail the production of fabrics from fibers, there are certain intermediate steps between spinning and weaving which are ultimately dependent on the physical, chemical, and molecular structure of the fiber All natu~aland some synthetic fibers must be washed or scoured to free them of natural oils, dirt, chemicals, and other foreign impurities They must be lubricated or sized or both for propel processing into cloth They will very likely be dyed for pleasing appearance, and they may have any of a variety of treatments applied to impart or control crease resistance, softening, water repellency, slipping, dimensional stability, shrinkage, or many other properties After some or all of these treatments, the fiber is ready to be transformed into a fabric The most important method of doing this is weaving, in which a set of yarns running lengthwise (warp) is interlaced with a second set at light angles (filling). Other methods of producing fabrics include knitting, in which a series of yarns is looped together, and the manufacture of nonwoven fabrics, papers, and felts, whereby the fibers are bonded together into flat sheets by heat, pressure, and, possibly, bonding agents Variations of the weaving process lead to pile fabrics, laces, braids, and so on The properties of the fabric depend on the fiber properties, the yam construction (lengths and diarnetets of the fibers, size of yams, amount of twist), the fabric construction (number of yarns per unit area and patterns in which they are combined), and the finishes applied to the fibers and the fabric
498
FIBER TECHNOLOGY
Scouring
The removal of' impurities from textile materials, called scouring, is carried out by the use of' surhce-active agents such as soaps and synthetic detergents. These materials have the property of reducing the surface tension of' water, and their main function in scouring is to reduce the interfacial tension of their solution toward fats and oils. Most textile f'ibers contain such fats, oils, or waxes, either naturally or inadvertently or purposely added during various operations. The principal task of' scouring is to remove these substances, since most of the other ingredients of' the soil are embedded in the f'atty material. On repeated laundering, however, some fine soil may be redeposited on the fat-free surface of the fiber, find its way into cavities on the surface, and become extremely resistant to removal. In contrast to initial scouring, one important function of detergents in household and 1aundr.y use is to prevent this redeposition of' soil. Soil release finishes are also used, particularly with permanent-press garments (see below)
Lubrication
Lubrication of' fibers is necessary to reduce their friction against themselves and against elements of the processing machinery In the lubrication of' yarns, it is necessary to preserve a high static coefficient of friction to keep the yarn in place on its spool or cone, while reducing the dynamic coefficient so as to obtain high speed of movement of'the yarn without the generation of heat. Lubricants may be vegetable or mineral oils or suitably refined petroleum products. Vegetable oils have been prefer~.ed because of' easy and complete removal by saponification. The best lubricants are water soluble, such as the poly(alky1ene glycols) Another function of the lubricant, which is especially important in the case of the moisture-resistant synthetic f'ibers, is to reduce the static electric charge on the fibers by lowering their surface resistivity .
Sizing
A size is a surface coating used to protect the yarn during weaving It makes the yarn smooth by binding protruding fibers onto the core of the yarn Since the fabric is usually dyed after weaving, the size must be easily removed Starch is used almost exclusively to size cotton and, to some extent, rayon and wool It is unsuitable for continuous-filament yarns of rayon, acetate, or the synthetics because o poor adhesion. Gelatin, poly(viny1 alcohol), or other polymeric f materials are used to size these yarns
Dyeing
Dyeing consists in placing fiber in an (aqueous) solution of' a dye and leaving it there until an equilibrium is established in which most of the dye is adsorbed in
FIBER AFTER-TREATMENTS
499
the fiber and only a small part remains in the dye bath. When equilibrium is reached, the dye concentration is uniform throughout the cross section of'the fiber To attain these conditions, the dye must be preferentially adsorbed in the fiber by means of some type of' intermolecular bond such as a van der Waals or hydrogen bond. It is clear also that because of' the size ofthe dye molecules they can penetrate only the regions of' the polymer Two requirements fbr successful dyeing are sites to which dye molecules can bond and amorphous regions in the fiber Natural cellulosic and protein fibers were known fbr centuries to dye well with acid dyes which bonded to the hydroxyl or amino groups in the fibers When cellulose acetate was introduced, a new problem arose which was solved by the use of' so-called dispersed dyes containing terminal amine or hydroxyl groups. They are water insoluble and are used as dispersions, hence the name. It seems probable that these dyes undergo hydrogen bonding to the carbonyl oxygen of' the acetate groups Synthetic fibers have often been very difficult to dye. Polyamide, polyester, and ac~ylicfibers contain sites for bonding dye molecules, but have such compact structures that dye absorption is extremely slow.. Rate of dyeing has been increased by going to higher temperatures, or by utilizing swelling agents for the polymers The dye sites can be modified, as, fbr example, in the treatment of' acrylic fibers with copper salts so as to form sites having affinity for anionic dyes. New classes of dyes have been synthesized for several of' the synthetics, greatly alleviating this problem. In addition, many fibers are colored by the incorporation of' pigments, often organic pigments, prior to spinning, leading to so-called spun-dyed products Aspland (1983) has reviewed methods for dyeing textile fibers Titanium dioxide is often added to the polymer befbre spinning as a delustering agent for the resulting fiber.
Finishing
5 .
'f
k
$
P
Finishes That Affect Hand. Under the heading of finishes which affect the feel or hand of fabrics come those which increase or decrease the natural friction between the fibers Thus lubricants make the fabrics feel softer or have increased pliability On the other hand, antislip finishes such as rosin, carboxymethyl cellulose, and hydroxyethyl cellulose impart a harsh feel to the fabric Conditioners. Ethylene and propylene glycols and their low polymers, with the
ability to absorb and retain moisture, act as softening, plasticizing, and antistatic agents for the hydrophilic textile fibers, and can counteract to some extent the harshness or stiffness which accompanies the use of antislip agents
% i
i 3
i
9 Z
Water-Repellent Finishes. Treatment with waxes such as paraffin emulsions imparts water repellency to cellulosic fabrics, but the effect is not permanent Permanence can be enhanced by subsequent treatment of the fabric with aluminum or zirconium salts in solution, but the best wate1 repellents at present are those which use stearoylamide pyridinium chloride. This molecule contains the stearoyl radical
500
FIBER TECHNOLOGY
to give the desired water repellency, the amidomethyl radical which can form a semiacetal with a hydroxyl group of cellulose, and the pyridinium group to solubilize the molecule in water so as to bring it into intimate contact with the fiber. Baking the fiber causes the molecule to split o f pyridine hydrochloride, leaving the stearoyl amidomethyl group attached to the cellulose. Silicones also impart water repellency to fabrics
Permanent-Press Finishes. One of' the major advances in textile technology in recent years has been the introduction of' resin finishes which impart the wrinkleand crease-resistant properties associated with "wash and wear" fkbrics Resin finishes we cumently applied to the majority of shirts, blouses, dresses, trousers, and retail piece goods produced. The resin finishes consists of aqueous solutions of urea--firmaldehyde or mel-. amine-formaldehyde precondensates, or of' cyclic ureas such as dimethylol ethylene urea It is thought that wrinkle resistance is imparted through the crosslinking of adjacent molecular chains in the fiber, rather than from resin formation in or between the fibers The deposition of the crosslinked polymer between fibers is specifically avoided, since it leads to stiffness and increased body of the fabric The fabric is treated with an aqueous solution of the monomers containing an acid or acid-generating catalyst Excess liquid is removed, and the fabric is dried and cured by heat treatment. Crosslinking is demonstrated by the reduced swelling the elongation of the of the fibers The treatments function by reduc~ng ir~eversible fiber: A greater force is needed to stretch the fibei, but a larger percentage of the elongation is recovered These thermosetting permanent-press systems work very well with polyester fabrics, but not as well with cotton, the properties of which are otherwise much in demand as an outstanding textile fiber Polyester-cotton blends were developed to provide a material combining the best features of both polymers, and have become by f a r the most popular fiber blend for permanent-press goods 'Textured Yarns
The ability to stabilize the physical form of'the thermoplastic fibers by heat setting has been utilized in the production of' a number of' "textured" yarns which have unusually high elasticity. These yarns are made in a number of' ways: The original method of' knitting a fabric, heat setting it, and then de-knitting it has long since been superseded by other methods in which the fibers are crimped by passing them over a hot knife edge or between hot gear teeth, stuf'f'ed into a tightly enclosed space or twisted tightly and then heat treated, or looped randomly by a turbulent air blast
Nonwoven Fabrics
A nonwoven hbric is a web or continuous sheet of' staple-length fibers, laid down mechanically The fibers may be deposited in a random manner or pref'erentially
501
oriented in one direction. The sheet is then bonded together Polyamides, polyesters, polypropylene, and polyethylene can be used, with 66-nylon most widely utilized The spun fibers, which may be drawn, are laid down directly onto a porous belt, often with the aid of an electrostatic charge They are then bonded in an oven or a calender, using either melt-bonding or a chemical adhesive Other methods reminiscent of paper making can be used to form the sheet The major uses of these spun-bonded nonwoven fabrics are for carpet backing, in bedding and home furnishings, as disposable apparel including diapers and sanitary goods, durable paper, coated fabrics, and in "hidden" fabric uses such as furniture components and garment interlinings
GENERAL REFERENCES
Bell 1965; Valko 1965, Ellis 1966; Lewis 1966; Steele 1966; Bikales 1968; Hearle 1969; Wray 1970; Goswami 1977; Drehlich 1981; Porter 1981
D. TABLE O F FIBER PROPERTIES Table 18-2 lists comparative properties for typical examples of the major textile fibers It should be noted that the properties of fibers of the same chemical type can vary widely depending on precise composition, heat treatment, yarn structure, and many other variables The values in Table 18-2 should be used only for comparative purposes
D I S C U S S I O N QUESTIONS AND PROBLEMS
1. Define (a) fiber, (b) staple, (c) denier, and (d) tenacity
2. What are typical values ofthe crystalline melting point and the glass-transitian temperature for textile fibers, and why?
3.
4.
What is the approximate denier of a fiber 0 02 mm in diameter if the specific gravity of the polymer is 1 2? Sketch typical stress-strain cuIves fbr silklike and wool-.like fibers
5. How do the stress-strain properties of typical fibers change with draw ratio?
6 . Name the technique(s) for spinning each of' the fbllowing fibers, giving typical temperatures and naming any other substances present: (a) polyester, (b) acrylic, (c) nylon, (d) aramid, (e) viscose rayon, (f) acetate rayon, (g) polytetr afluoroethylene .
7. Will the fbllowing changes from a linear structure be likely to improve fiber properties? (a) To a slightly crosslinked structure and (b), to a branched-chain structure. Give examples to support your conclusions
TABLE 18.2.
Properties of Fibersa
Polyester
Continuous
Property Tenac~ty.dry (gldenter) Tet~ilc~ty. (gldcl~ter) !vet Tensile strength (thousands psi) Elongat~on(%, at break) Elastic recovery (?+, fro111c l o t ~ ~ : ~ t ~ o n ) Stiffness (gldenter) Toughness (g-cm) Specific gravtty (g/crn3) Motsture regaln ( 7 c . 70F. 65% RH) Melt temperature ("C)h Chem~cal reststance Bleach res~stance Solvent res~stance Sunlight reststance Abrns~onresistance Regular Tenaclty 2.8-5.6 2.8-5.6 50-90 24-42 7613 10-30 0.4-1.1 1.38 0.4 250 Good Reststant Excel. Good Excei.
Cellulos~c Amlde Staple 2.4-7.0 2.4-7.0 39-106 12-55 8113 12117 0.2-1.1 1.38 0.4 250 Good Reststant Excel. Good Excel. Carbon 23 23 515 1.5 100 1500
-
High
Tenaclty 6.8-9.5 6.8-9.5 106-168 12-25 8813 30 0.5-0.7 1.39 0.4 250 Good Reststant Excel. Good Excel. 6-Nylon 4.0-7.2 3.7-6.2 73-100 17-45 98110 18-23 0.7-0.9 1.14 2.8-5 220 Fair Bleaches Good Fatr Excel.
66-Nylon 2.3-6.0 2.0-5.5 40- 106 25-65 8813 5-24 0.8-1.3 1.14 4.0-4.5 250 Fair Bleaches Good Fair Excel.
Viscose Rayon 0.7-3.2 0.7-1.8 28-47 15-30 8212 6-17 0.2 1.5 11-13 l80d Falr Res~stant Excel. Good Good
Acetate Rayon 1.2-1.4 0.8-1.0 20-24 25-45 6514 3.5-5.5 0.2-0.3 1.32 6.3-6.5 260 Fa~r Bleaches Falr Good Fa~r
Triacetate 1.1-1.3 0.8-1.0 18-22 26-35 6515 5.2 0.15 1.3 3.2 300 Falr Reststant Poor Poor Fair
1.77
320 Good Poor Excel. Excel. Poor
Aramtd Property Tenactty, dry (gldenter) Tenactty, wet (gldemer) Tensile strength (thousands psi) Elongatton (%, at break) Eias~~c recovery (90, from elonpatton) StifCness (gldenier) Toughness (g-cm) Specific gravlty (g/cm3) Moisture regal11 (%, 70F. 65% RK) Melt temperature Chernrcal reststance Bleach resistance Solvent reststance Sunlight resistance Abrns~on reststance
Acrylic
2.2-2.6 1.8-2. I 32-39 20-28 7313 10 0.40 1.16 i .5 235 Good Bleaches Excel. Good Good
Polyethylene 1.O-3.0 1.0-3.0 11-35 20-80 9515 2-i2 1-3 0.92 -0 110-120 Excel. Res~stant Excel. Fatr Good
Polypropylene 3.0-4.0 3.0-4.0 35-47 80-100 9615 20-30 ;-3 0.91 0.01 160-180 Excel. Excel. Excel. Rlr Good
Polyurethane 0.7-0.9
Kevlar 22 22 400 2.5 10012 975
Nomex 4.0-5.3 3.0-4.: 90 27-32
Polytetrafluoroethyiene 0.9-2.0 0.9-2.0 25-3 1 19-140
Cotton 2.1-6.3 2.5-7.6 42-125 3-10 4515 42-82
Silk
2.8-5.1 2.4-1.4 45-83 13-31 33120 76-1 17
Wool 1.0-1.7 0.9-1.5 17-18 20-50 63120 24-34
11-14 400-625 97/50 0.1-0.2 2.0 1.21
0.85 i .38 6.5 370d Good Ues~stant Exccl. Fair Good
:.2-8.8 0.15 2.1

0 290 Excel. Reststant Excel. Excel. Good
1.44 1.5 500d Excel. Poor Excel. Far Good
1 .50
1.75
1.30 17
1.3 230 Good Poor Good Fa~r Good
8.5
-
II
Good Bleaches Good Fa~r Good
Fatr Bleaches Good Good Fair
Fair Bleaches E~cLsI. Good Good
"Texrile World 138 ( 3 ) (1982). "d = decornDoses.
5 04
FIBER T E C H N O L O G Y
8. Describe briefly the functions of'the following after-treatments: (a) scouring, (b) iubrication, (c) sizing 9. Name two treatments or reagents used to p~.oduce of'the fbllowing finishes each on fibers: (a) soft hand, (b) antislip, (c) antistatic, (d) water repellent, (e) permanent press 10. Discuss briefly the production, composition, structure, and uses of nonwoven fabrics.
BIBLIOGRAPHY
Arpland 1983 J R Aspland, "Textlle Color Appllcat~onProcesses," Color Rer Appl 8, 205-214 (1983) Bell 1965 T E Bell and H de V Partndge, "Bleachtng," pp 438-484 tn EPST, t Vol 2, 1965 Brkaler 1968 Norbert M Btkales, "Nonwoven Fabrics," pp 345-355 In EPST, Vol 9, 1968 Corb3re 1'967 1 Corbrt.re, "Fundamental Aspects of Solution Dry-Sp~nn~ng," 133-167 In H F pp Mark, S M Atlas, and E Cernla, eds , Man-Made Flbers, Science and Technology, Vol 1, W~ley-Intersc~ence, York, 1967 New Drehlich 1981 Arthur Drehltch, "Nonwoven 'Iextlle Fabrlcs (Staple F~bek),"pp 109-124 tn ECT,S Vol 16, 1981 Ellis 1966 J R Ellts and G M Gantz, eds , and Emery I Valko, "Dyetng," pp 235-375 In EPST, Vol 5, 1966 Goswaml 1977 B C go swam^, J G Martindale, and F L Scardtno, Textile Yarns Technology, Structure and Applrcation, John Wtley & Sons, New York, 1977 Hearle 1969 John W S Hearle, Percy Grosberg, and Stanley Backer, Structural Mechanics of Frbers, Yarns and Fabrrcs, Wlley-Intersclence, New York, 1969 Heckert 1953 W W Heckert, "Synthetic Flbers from Condensatton Polymers," Chapter 5 In Samuel B McFarlane, ed , Technology of Synthetic Frbers, Fairchtld, New York, 1953 Hicks 1967 E M Hlcks, E A Tippetts, J V Hewett, and R H Brand, "Conjugate Fibers," pp 375-408 tn H F Mark, S M Atlas, and E Cernla, eds , Man-Made Fibers, Science and Technology, Vol 1, W~ley-Intersctence,New York, 1967 Hobson 1978 P H Hobson and A L McPeters, "Acryltc and Modacryllc Ftbers," pp 355-386 In ECT, Vol 1, 1978 Lewls 1966 Charles E Lewis, "Dyes," pp 376-405 tn EPST, Vol 5, 1966 McGovern 1980 John N McGovern, "F~bers, Vegetable," pp 182-197 tn ECT, Vol 10, 1980 Mclntyre 1968 J E McIntyre, "Man-Made Ftbers, Manufacture," pp 374-404 In EPST, Vol 8, 1968 Mark 1967 H F Mark and S M Atlas, "Principles of Splnnlng In Emulslon and Suspenston," pp 237-240 In H F Mark, S M Atlas, and E Cernla, eds , Man-Made Frbers, Science and Technology, Vol 1, Wiley-Inte~sctence,New York, 1967 Mark 1967-1968 H F Mark, S M Atlas, and E Cernla, eds , Man Made Frbers. Science and Technology, W~ley-Intersclence,New York, Vol 1, 1967, Vols 2 and 3, 1968 iHerman F Mark, Norman G Gaylord, and Norbert M B~kales, , Encyclopedra offolymer Science eds and Technology, Wiley-Interscience, New York $Martin Grayson, ed , K~rk-Othmer Encyclopedra of Chemrcal Technology, 3rd ed , Wiley-Intersctence. New York
,+/~l,~ldlemun 11/77 Stanley Middleman, fundamental^ of Polvmer P t o c e s ~ ~ nMcGraw-H111, New York, g,
1977
peters I ~ V J Timothy V Peters, "Fibers, Elastomer~c," pp 166-182 in E C T , Vol 10, 1980 porter 1921 K Porter, "Nonwoven Text~leFabrics (Spunbonded)," pp 72-104 in E C T , Vol 16,
1981
pr~jston/!08 J Preston, "Aramid Fibers," pp 213-242 in E C T , Vol 3, 1978 ~ilty I954 I L Riley, "Spinning and Drawing Fibers," Chapter I8 in Calvin E Schildknecht, ed , Polymer Processes, Interscience, New York, 1956 Robert? ILW William 1 Rober.ts, "Fibers, Chemical," pp 148-166 in E C T , Vol 10, 1980 ~ltlur~ 1957 F Tlclan, "Fundamental Aspects of Wet-Splnnlng Solutions," pp, 95-132 In H F Mark, S aM Atlds, and E Cernla, eds , Man-Made Flbers, S~lenceand Technology, Vol 1, Wlley
Interx~enLe,New York, 1967

~mrth I967 Arthur L Smith, "Fibers, Identification," pp 593-609 in EPST, Vol 6, 1967 Steele I966 Richard Steele and D D Gagliardl, "Crease Reslstance," pp 307-330 In EPST, Vol
4, I f Yh
Vulko I565 Emery I Valko and Guiliana C Teroso, "Antistatic Agents," pp 204-229 in W S T , Vol
2, 1965
Wray 1970 G R Wray, "Textile Processing," pp 692-727 in EPST, Vol 13, 1970 Ziablcki I567 Andrzej Zlablckl, "Pnnctples of Melt-Sprnnlng," pp 169-236 In H F Mark, S M Atla\, and E Cerma, eds , Man-Made Fibers, Science and Technology, Vol 1, Wlley-Intersclence,
New York, 1967
CHAPTER NINETEEN
ELASTOMER TECHNOLOGY
The final class of' high polymers to be considered in Part 6 is elastomers Like fibem, elastomers are considered apart from other polymeric materials because of their special properties. Unlike fibers, elastomers do not in general lend themselves to plastics uses; elastomers must be amorphous when unstretched and must be above their glass transition temperature to be elastic, whereas plastics must be crystalline or must be used below this temperature to preserve dimensional stability About 5 . 8 billion Ib of' rubber was used in the United States in 1982. Synthetics accounted for 72% of the total consumed, with SBR in largest volume by a wide margin, as indicated in Table 19-1. Automobile tires remain the largest single end use of' rubber
History of Synthetic Robber. The first attempts to produce synthetic rubbers centered around the homopolymerization of' dienes, particularly isoprene because it was known to be the monomer for natural rubber. It was found in the late nineteenth century that rubberlike products could be made from isoprene by treating it with hydrogen chloride or allowing it to polymerize spontaneously on storage. These materials could be vulcanized with sulfur, becoming more elastic, tougher, and more heat resistant. Around 1900 it was discovered that other dienes such as butadiene and 2,3dimethylbutadiene could be polymerized to rubberlike materials spontaneously by alkali metals or by free radicals Application was made of these facts during World War I in Germany, where 2,3-dimethylbutadiene was polymerized spontaneously After World War I this research on rubberlike products continued, with the emphasis shifting to butadiene because of its more ready availability, at that time from acetaldehyde via the aldol synthesis. Alkali metals were used as initiators, and the products were called buna rubbers from the first letters of butadiene and the symbol Na fbr sodium
506
VULCANIZATION
TABLE 19-1. Approximate United States Consumption of Rubber in 1982"
TY pe Natural rubber. Synthetics

SBR
Consumption (million Ib)

1600 2200 800 300 290 270 1'70 120 50 4200 5800
Polybutadiene EPDM Butyl Neop~ene Nitrile Polyisoprene Other Total synthetics Total
"Estimates f ~ o m previous chapte~sand Rubber World 187 ( 6 ) . 8 (1983)
j
:
,
1
Improvements in processing and properties were made in two ways: Copolymers of butadiene with vinyl monomers, notably styrene, were introduced, and emulsion polymerization was adopted The significance of initiators, reduction activators, and oxygen became known The importance of maintaining low conversion or utilizing modifiers such as CCI, and long-chain mercaptans was discovered By the beginning of World War 11, acceptable polymers (buna-S) were being produced in Germany containing 68-70% butadiene and 30-32% styrene In review of earlier chapters, it may be recalled that the unique properties of elastomers include their abil~ty stretch and retract rapidly, exhibit high strength to and modulus while stretched, and recover fully on release of the stress. To obtain these properties, certain requirements are placed upon the molecular structure of the compounds: They must be high polymers, be above the11 glass transition temperatures, be amorphous in the unstretched state (but preferably develop crystallinity on stretching), and contain a network of crosslinks to restrain gross mobility of the chains
A.
VULCANIZATION
The process by which a network of crosslinks is introduced into an elastomer is called vulcanization.The chemistry of vulcanization is complex and has not been well understood throughout the century of practice of'the process since its discovery by Goodyear in 1839 The profound effects of' vulcanization, however, are clear: it transfbrms an elastomer from a weak thermoplastic mass without useful mechanical properties into a strong, elastic, tough rubber (Table 19-2). The tensile strength, stiffness, and hyster.esis (representing loss of energy as heat) of natural rubber before and after vulcanization are shown in Fig. 19-1, and the effects of
508
ELASTOMER TEC:HNOlOGY
TABLE 19-2.
Properties Typical of Raw, Vulcanized, and Reinforced Natural Rubber

-
Property
Tensile strength, psi Elongation at break (%) Modulus (psi)b Permanent set Rapidity of ret~action (snap) Water absorption Solvent resistance (hydrocarbons)
Raw Rubber
300 1200
- 3000 800 400
Vulcanized Rubbera
Reinforced Rubber
Large Good Large Soluble
Small Very good Small Swells only
"Not reinforced bTensile stress at 400% elongation, a measure of stiffness
extent of vulcanization on these and other. praperties of elastomers are shown schematically in Fig. 19.-2. For many decades fbllowing Goodyear's first experiments in heating rubber with small amounts of sulfur, this process provided the best and most practical method for.bringing about the drastic property changes described by the term vulcanization, not only in natural rubber but also in the diene synthetic elastomers such as SBR, butyl, and nitrile ~ubbers. has been fbund since, however, that neither heat nor It
Unvulcanized rubber
Elongation, %
FIG. 19-1.
Stress-st~ain curves to 600% elongation and back, typical of unvulcanized and vulcanized natural rubber
f1 i
Increasing cure time
-+
FIG 19-2. Schematic representation of the effect of extent of vulcanization (cure time) on various physical properties of elastomers (after Gamey 1959) AT is heat buildup in a compressionflex test
sulfur is essential to the vulcanization process. Rubber can be vulcanized or cured without heat by the action of sulfur chloride, fbr example A large number of' compounds that do not contain sulfur can vulcanize rubber; these fill generally into three groups: oxidizing agents (selenium, tellurium, organic peroxides, nitro compounds), generators of free radicals (organic peroxides, azo compounds, many accelerators, etc.), and phenolic resins. Thus it is clew that there is not a single method or chemical reaction of vulcanization Since the chemical reactions associated with vulcanization are varied and involve only a ftw atoms in each polymer molecule, a definition of vulcanization in terms of' the physical properties of the rubber is necessary. In this sense, vulcanization may be defined as any treatment that decreases the flow of' an elastomer, increases its tensile strength and modulus, but preserves its extensibility. There is little doubt that these changes are due primarily to chemical crosslinking reactions between polymer molecules. As might be expected, such properties as tensile strength are relatively insensitive to the onset of these crosslinking reactions; tensile strength does indeed change tenfold during curing, but this is evidence of the profbund alteration of' polymer properties by the process Tests based on melt flow we more sensitive to initial crosslinking reactions and are widely used in the rubber industry Although vulcanization takes place by heat in the presence of sulfur alone, the process is relatively slow. It can be speeded many fold by the addition of small amounts of organic or inorganic compounds known as accelerators Many accel-
510
ELASTOMER 1EC:HNOLOCY
eratoxs require the presence of still other chemicals known as activators orpromoters before their full effects are realized. These activators are usually metallic oxides, such as zinc oxide. They function best in the presence of a rubber-soluble metallic soap, which may be formed during the curing reaction from the activator and a fatty acid. The most efficient combination of chemicals for sulfur vulcanization includes sulfur, an organic accelerator, a metallic oxide, and a soap
Chemistry of Vulcanization Sulfur Vulcanization. The curing of rubber with sulfur alone is quite slow and no longer commercially practiced. Even today the exact mechanisms of the curing reactions ase not fully understood. Both free-radical and ionic intermediates have been postulated Accelerated Sulfur Vulcanization.
Typical accelerators for sulfur vulcanization include thiazoles, for example, 2-mercaptobenzothiazole,
dithiocarbamates, for example, tetramethyl thiuram disulfide,
and amines, for example, diphenyl guanidine,
In addition, the retarder N-(cyclohexylthio)phthalimide,
has proved very effective in inhibiting premature vulcanization. These chemicals
VULCANIZATION
511
are usually used with zinc oxide and a fatty acid, a typical recipe being 0.5-2 parts accelerator, 0 1--1 part retarder, 2-10 parts zinc oxide, 1-4 parts fatty acid, and 0.5-4 parts sulfur per 100 parts rubber. Again, mechanisms are not known in detail, but it is thought that the accelerator reacts with sulfur to give sulfides of the type Ac---S,--Ac, where Ac is a fiee radical derived fiom the accelerator. These polysulfides then interact with the rubber to give intermediates of the type rubber-S,-Ac. These react further, with the final crosslinks being of' the general type
Many other possibilities exist The role of the zinc oxide, which produces zinc ions in the presence of the fatty acid, appears to be to form chelates with the accelerator and sulfides, retarding and thus smoothing the course of the vulcanization The double bonds in the elastomer are not saturated in this mechanism, but contribute by enhancing the activity of the allylic hydrogens, at which the crosslinking takes place
Nonsulfur Vulcanization. The same activation effect of the elastomer double bonds is postulated to play a role in nonsulfur vulcanization, according to the following scheme: a. A free radical R is formed by the decomposition or oxidation of the curing
agent, or as a step in the oxidative degradation of the rubber b. This free radical initiates vulcanization by abstracting a hydrogen atom fiom one of the a-methylene groups (in natu~al~ubberthe methyl side group directs the attack to the methylene group nearest it):
c. The rubber fiee radical then attacks a double bond in an adjacent polymer
chain. This results in the formation of' a crosslink and the regeneration of
51 2
a free radical in a reaction analogous to propagation in an addition polymerization (Chapter 3):
Vulcanization may continue by several such propagation steps Chain transfer may also occur Termination probably occurs by reaction of the rubber free radical with a free radical fragment of the curing agent In contrast to add~tionpolymer-ization in a fluid system, the termination reaction between two rubber free radicals is considered unlikely because of the low probability of two such radicals coming into position to react, owing to the high viscosity of the medium. Since free-radical intermediates also exist in accelerated sulfur vulcanization, similar reactions no doubt take place in that case also Some double bonds are saturated in this way, some by addition of hyd~ogensulfide (a known product of sulfur in vulcanization), and some by still other mechanisms
Physical Aspects of Vulcanization
Vulcanization with sulfur present takes place when 0 5-5 parts (by weight) of'sulfur is combined with 100 parts of' rubber. If the reactions are allowed to continue until considerably more sulfur has combined (say 30-50 parts per 100 parts rubber), a rigid, nonelastomeric plastic known as hard rubber or ebonite is formed.. Tensile strength reaches a maximum, with elongation remaining high, in the early soft cure region; then both fall to much lower values with increasing amount of combined sulfur.. The region of maximum tensile properties in the neighborhood of 5-.10 parts combined sulfur is known as the region of' optimum cure for this mix. Beyond this point the stock ( i . e . , mixture) is said to be overcured and is likely to be leathey, that is, stiffer and harder than at the optimum cure but weaker and less extensible At high proportions of combined sulfur the strength of the material increases again, and the elongation becomes very low as the hard rubber region is reached Many properties of soft vulcanizates, including tensile and tear strength, stiffness, and hardness, go through a shallow maximum as the cure time is increased. Others.
I
1
REINFORCEMENT
51 3
1
i
1 ;
1
I
i
i
I
3I
such as elongation and permanent set, drop continuously The effect of increasing temperature is, as expected, to move the curves to shorter times Experiments in which the composition of the mix is varied show that the initial rise in tensile strength is accompanied by the rapid incorporation of sulfur in the rubber, and that, as the rate of sulfur addition decreases, the rate of increase of tensile strength begins to decline. It is also known that the total amount of combined sulfur does not measure the extent of change in physical properties during cure, and that the ratio of sulfur combined to double bonds used up va~ies widely from one mix to another These facts suggest that processes of crosslinking and degradation occur simultaneously during vulcanization Additional evidence for this view arises from the fact that badly overcured stocks may undergo revernon, that is, become soft and tacky with notable loss in strength Such stocks resemble oxidatively degraded rubber Reversion differs from overcuring in that it usually takes place on prolonged heating when not enough sulfur is present to get into the region of leathery cures
GENERAL REFERENCES
Wolfe 1971; Coran 1978, 1983; Morton 1981
6. REINFORCEMENT
For many uses, even vulcanized rubbers do not exhibit satisfactory tensile strength, stiffness, abrasion resistance, and tear resistance Fortunately, these properties can be enhanced by the addition of certain fillers to the rubber before vulcanization Fillers f o lubber can be divided into two classes: ~nertfillers, ~ such as clay, whiting, and barytes, which make the rubber mixture easier to handle before vulcanization but have little effect on its physical properties; and reznforcwzg fillers, which do improve the above-named unsatisfactory properties of the vulcanized rubber Carbon black is the outstanding reinforcing filler fpr both natural and synthetic rubbers The added effects of reinforcement with carbon black on the properties of rubbers over the effects of vulcanization alone are illustrated in Table 19-2 Although the nature of reinforcement is not completely understood, it appears to add a network of many relatively weak "fix points" to the more diffuse network of strong primary bond crosslinks introduced by vulcanization Vulcanization restrains the long-range movements of the polymer molecules but leaves their local segmental mobility high; reinforcement stiffens the mass and improves its toughness by restricting this local freedom of movement Nomeinforced vulcanized rubbers, sometimes called pure-gum vulcanizates, are relatively soft, pliable, and extensible, and are most useful for such items as rubber bands, tubing, and gloves. Rubber would be of little value in modern industry were it not for the reinfo~cing effects of carbon black
'Types of Fillers
Only a few fillers are reinforcing for even one or two of the important mechanical properties of natural rubber, while some weaken the vulcanizates in one or more important respects The latter are known as ztlert fillers In general it is found that the reinforcing action of a filler depends upon its nature, the type of elastomer with which it is used, and the amount of filler present Carbon black is the only important reinforcing filler for most elastomers, although silica and silicate fillers can be used in some cases, as with the silicone elastomers (Chapter 160). Important nonreinforcing fillers include talc, zinc oxide, and magnesium carbonate
Nature of the Filler. The chemical composition of the filler is a primary determinant of its reinforcing action. Carbon blacks are always more efficient reinfbrcing agents than, for example, talc, even though parzicle size, surfice condition, and both Within a single chemical class, other properties can be varied widely f ' o ~ particle size and surfice condition are important variables Nature of the Elastomer. Although a good reinforcing filler such as carbon black increases the tear and abrasion resistance of all the curTent major elastomers, it has no effect in enhancing the tensile strength of neoprene or butyl rubber, in contrast to the marked tensile reinforcement it causes in natural rubber, SBR, nitrile rubber, and even the polysulfide elastomers These differences are ascribed to a conflict between two effects of the filler (a) A reinforcing filler stiffens and strengthens the structure by introducing a network of many relatively weak fix points, and (b) simultaneously, it may inter-fere with the ability of the polymer to crystallize at high elongations, simply by its bulkiness in the system Since stiffness and tensile strength are greatly enhanced by crystallization, the specific effect of the filler depends upon which of these actions predominates Amount of Filler. Increasing amounts of reinforcing filler cause continuing improvement in properties until a maximum is reached, representing the optimum loading for this composition Beyond this point, additional filler merely acts as diluent and the properties of the vulcanizate again deteriorate. Carbon Black
Various colloidal fbrms of' carbon constitute the most important reinforcing filler fbs rubber . Carbon blacks are the only inexpensive materials that reinfbrce all three of the important properties-tensile strength, tear resistance, and abrasion resistance. Many different carbon blacks are used, differing mainly in particle size, surface condition, and degree of' agglomeration. Roughly, the degree of' reinfbscement increases with decreasing particle size of the black down to the practical lower limit at which blacks can be made, about 100 A in diameter
IJ
3
REINFORCEMENT
Almost all carbon blacks used for reinforcing elastomers are made by burning gas or oil in a furnace in limited air and removing the carbon from the of-gases by a centrifugal and electrostatic precipitation. They range from 300 to 800 A in particle diameter., and are often designated by their effects on the vulcanizates, as high-abrasion, high-modulus, semireinfbrcing, or conductive furnace blacks. Electron microscopy shows that carbon blacks consist of irregular aggregates of' approximately spherical subunits, sometimes called nodules, fused together in branched, chainlike structures. The nodules are thought to be paracrystalline domains with a crystal structure similar to that of graphite except that the sheets of hexagonally close-packed carbon atoms are arranged randomly above one another.. This suggests that unsatisfied bonds are present which may be important in reinfbrcement. The sur.faces of the particles contain much adsorbed material, including hydrogenated and oxygenated structures As a result the pH of a water slurry of the black may range from 2 . 5 to 1 1
Effects of Carbon Black Structure on Reinforcement. The three parameters-particle size, pH, and structure zndex, that is, extent of agglomeration or chain structure--appear to be predominant in determining the reinforcing behavior of blacks. Tensile strength, abrasion and tear resistance, hardness, and toughness increase with decreasing particle slze, whereas rebound and ease of processing become poorer Blacks with a high structure index are difficult to dlsperse and give high stiffness and hardness but low tenslle strength, toughness, and electrical resistivity The pH influences the rate of vulcanization, depending upon the acidlc or basic nature of the accelerator The stiffness of a reinforced vulcanizate rises steeply with the loading of carbon black and is nearly independent of the particle size of the black These two features are unique for this property Other properties, such as tensile strength, go through a maximum as loading is increased They also depend upon particle size, as mentioned above The tensile strength of natural rubber can be increased about 40% by reinforcement In SBR, however, the properties of the pure-gum vulcanizates are very poor, since the elastomer does not crystallize The tensile strength of SBR can be raised about tenfold by reinforcement, making it equivalent to natural rubber for fully reinforced stocks The tear resistance of SBR is also poor for unloaded vulcanizates, but is equivalent to that of natural rubber after reinforcement. Abrasion resistance is highly important for heavy-duty ~ubbersuch as tire tread stock The abrasion resistance of both natural rubber and SBR can be improved at least fivefold by proper reinforcement Unfortunately, the resilience of the rubber decreases with increasing loading of filler As a consequence hysteresis loss and heat buildup increase Thus the use of a reinforcing filler represents a compromise between adequate abrasion and tear resistance and abnormal heat buildup Origin of Reinforcement. Only two conditions must be met for significant re-. inforcement of' a rubber by carbon black. First, the particle size ofthe black must be small, usually between 200 and 500 A . This ensures a large surface area and
516
ELASTOMER T'ECHNOLOCY
thus a large filler-rubber interface. Second, the rubber must "wet" the carbon black. This not only assures dispersion of'the black in the rubber, but implies that the rubber-black adhesion approaches in strength the cohesion of the rubber to itself This is necessary if the rubber-black bonds are to survive the large strains associated with high elongations. This adhesion is produced in part by secondary bond forces, which alone are enough to account for reinforcement, and in part by chemical grafting of rubber on to the surface ofthe black particle, almost certainly as the result of radical processes While primary chemical bonding between the rubber and a filler is not essential fbr reinforcement, it does lead to the unique combination of p~operties that result when the filler is carbon black
GENERAL REFERENCES
Burgess 1965; Garvey 1970: Kraus 1970, 1978; Morton 1981.
C.
ELASTOMER PROPERTIES A N D C,OMPOUNDlNG
Mechanical Properties of Elastomers
The stress-strain behavior of pure gum (i .e.. no~einfbrced)vulcanizates of the , various elastomers depends markedly on the molecular structure, polarity, and crystallizability of the polymers.. Nitrile rubber and SBR, being copolymers of' complex and irregular structure, do not crystallize at all, but natural rubber and butyl rubber crystallize on stretching at room temperature. In these elastomers the crystallites have a stiffening effect like that of a reinfbrcing filler with the result that their stress-strain curves turn up markedly at higher elongations..The similarity of' these stress-strain curves to those predicted from the kinetic theory of' rubber elasticity (Chapter 11B) should be noted. Unvulcanized rubber undergoes a very large amount of mechanical conditioning when first exposed to tensile stress. Figure 19-1 shows the first cycle of loading and unloading of a tensile specimen of unvulcanized natural rubber. As successive cycles take place, the changes in resistance to stretching, tensile strength, energy absorption, and permanent set become smaller..This mechanical conditioning is not accompanied by molecular weight degradation. Such degradation does occur, however, in the process of milling raw rubber to reduce its melt viscosity to the desired range before the incorporation of fillers, and so o n . Here the process is complicated by the presence of' oxygen. Vulcanized rubber undergoes a similar but much less extensive change in stress-strain properties on successive cycles of' test. The mechanical properties of' elastomers depend greatly upon the rate of testing Since service conditions for automobile tires (by far the most important end use of elastomers) involve rapid cyclic stresses, dynamic test methods are essential if the results are to correlate with the performance of the finished article
ELASPJ>MER PROPERTIES A N D COMPOUNDING
517
ox&&ve
R~~;TB:~
Aging of Elastomers
that retain double bonds in their vulcanized structure, such as natural rubbfi, SBR, and nitrile rubber, are sensitive to heat, light, and particularly oxygen. unlt:?,; potected with antioxidants (of which the phenyl a- and P-naphthylamines are G-Emost common) the rubbers age by an autocatalytic process accompanied by an jr/i.rease in oxygen content The results of the aging may be either softening or embq~ilement, suggesting competing processes of degradation and crosslinking of the y&mer chains The f i ~ sstep in the process is presumed to be the same as that t in the oxidation of any olefin, that is, the formation by free radical attack of a p e r r > ~ ~ 4 a carbon atom next to a double bond: at e
Subsequent steps may include crosslinking or chain cleavage or involvement of the double bond, probably initially through an epoxide group Many of the steps in the reaction resemble closely those of certain types of vulcanization Many rubbers are sensitive also to attack by ozone, requiring protection by antiozonants These are often derivatives of p-phenylene diamine, and are thought either to react with the ozone before it can undergo reaction with the rubber surface, " 1 to aid in reuniting chains severed by ozone I Natural rubbe~is far more sensitive than most of the other elastomers to both d ; oxygen and ozone attack
I
Compounding
Y
The term compoundzng describes the selection of additives and their incorporation into a polymer so as to give a homogeneous mixture ready for subsequent processing steps, Except for differences in the nature of the compounding ingredients, especially as required for vulcanization and reinforcement and described under those headmgs, the compounding of rubbers is similar to that of most plastics Typical equipment described in Chapter 17 is used for rubber as well as plastics compounding In the case of natural rubber, SBR, and the other synthetic rubbers produced by emulsion polymerization, the polymer is first available in the form of a latex The normal procedure in compounding is to coagulate and dry the latex, and then to masticate and compound the rubber on mills or in other equipment With SBR, however, two important compounding steps, ozl extending and masterbarching, are con~n~only carried out before coagulation of the Jatex
ELASTOMER TEC HNOLOCY
Oil Extending. The use of' hydrocasbon oils to dilute, or extend, rubbers has been known fbr many years. Since 1950, this practice has been applied to SBR polymerized to high molecular. weight. The oil serves as a plasticizer and softener, reducing the melt viscosity of the rubber to that normally required in compounding The process, which leads to a less expensive final product, is widely used: About 90% of' the tire tread stock produced in this country is oil extended to some extent The oil is added as an emulsion to the latex before coagulation
Masterbatching. A convenient method of mixing rubber and carbon black is the coprecipitation of a mixture of the rubber latex and an aqueous slur~y the black of This method gives adequate mixing when the particle size of the black is similar to that of the latex particle and the number of each type of particle per cubic centimeter before the coagulation can be made nearly the same This situation holds for SBR and the most useful blacks
GENERAL REFERENCES
Cox 1965; Maassen 1965; Cooper 1966; Stagg 1968; Barnard 1970; Garvey 1970; Studebaker 1978; Morton 1981
D.
TABLE OF ELASTOMER PROPERTIES
Table 19-3 lists values of' some physical and chemical properties of' major corn mercial elastomers. In comparison with the corresponding tables in Chapters 17 and 18 it should be noted that the properties of' elastomers depend in very large extent on the details of' vulcanization, reinforcement, and compounding It is, consequently, almost futile to assign "typical" values of' properties according to polymer type. Therefore most of' the data in Table 19-3 are of no more than qualitative significance, and comparisons should be made with extreme caution
DISCUSSION QUESiiONS A N D PROBLEMS
1. Write chemical equations fbr the following processes in natural rubber: (a) the initial attack in oxidative degradation, (b) propagation in a vulcanizalion occurring by a free-radical mechanism, and (c) a typical crosslinking step in accelerated sulfur vulcanization. 2. Identify and write chemical structures fbr (a) two organic acceIerators of different types, (b) a common activator system, (c) a common rubber antioxidant, (d) the bonds thought to be responsible f'or reinforcement, (e) an elastomer that does not require reinforcement, (f) an elastomer for which reinforcement is essential for the development of good properties
TABLE 19-3.
Typical Properties of Commercial Elastomers"
Property Tensile strength (psi) Elongat~on(%) Modulus (PSI, 300-400% elongat~on) Dy natn~c properties Permanent set Tear resistance Abras~onresistance Adhes~on Electr~calproperties Gas permeability Upper use temperature ("C) Lower use temperature ("C) Weather resistance Ozone resistance Oil resistance Gasoline res~stance Water swelling Adhesion to metal
Natural Rubber
4000 700 2500
SBR
3500 700 2500
Acrylate
Butyl
3000 700 1000
Chlorosulfonated Polyethylene
2800 500
EPDM
3000 300
Epichlorohydr~n
2500 400
Fluonnated Rubbers
2400 400 250
Excel. Low Excel. Fa~r Excel. Excel. High 80 - 50 Fan Poor Poor Poor Excel. Excel.
Good Low Good Good Excel. Fa~r High 110 - 50 Fa~r Fa~r Poor Poor Good Excel.
Good Low Fan Good Good Poor Low 150 - 20 Excel. Excel. Excel. Fair Poor Good
Poor Moder. Excel. Good Good Exccl. Low 100 - 50 Excel. Excel. Poor Poor Excel. Good
Good Excel. Fa~r Good Excel. Excel. Good
Poor High Fa~r
Fa~r Fa~r
Good Excel.
!20
- 50
Excel. Excel. Good Fa~r Excel. Exce!.
!50 - 40
Excel. Excel. Poor Poor Excel. Excel.
120 - 45 Good Good Good Good Good Excel.
230
- 40
Excel. Exccl. Excel. Excel. Excel. Excel.
TABLE 19-3 (Continued)

- - - - - --
Property Tensile strength (psi) Elongatron (%) Modulus (PSI,300-400% elongation) Dynainrc properties Pcnnimcnt sct Tear resistance Abraslon resistance Adhesion Electrical propcrtics Gas pcrmcability Upper use temperature ( O C ) Lower use temperature ('C) Weather resrstance Ozone reslstance Oil reslstance Gasoline resistance Water swelling Adhes~on metal to
Neoprene
Nitrile
Polybutadiene (as-1,4)
Polyisoprene
(cw- 1,4)
4000 750 2500
Exccl. Low Excel. Fair Excel. Good High 80 - 50 Fair Poor Poor Poor Excel. Excel.
Polysulfide
Silicone
Urethane
4000 700 1000

Fa~r Modcr. Good Good Excel. Good Moder. 100 - 50 Excel. Good Fax Good Good Excel.
4000 600 1500

Poor Modcr. Good
3000 700
1200 400 1400

Poor High Good Poor Poor Exccl. Low 80 - 50 Excel. Good Excel. Excel. Excel. Excel.
1500 800
-
Exccl.
-
Excel. Fa~r Modcr . 120 - 50 Poor Falr Excel. Good Excel. Excel.
Exccl. Good
-
Poor High Poor Poor Poor Exccl.

-
Good Moder. Excel. Exce!. Excel. Far

-
100 - 60 Fa~r Poor Poor Poor Excel. Excel.
230 - 80 Excel. Exccl. Fair Poor Excel. Excel.
100
- 50
Excel. Excel. Excel. Excel. Fair Excel.
"Studebaker (1978) and Morton (1981).
1
I
BIBLIOGRAPHY
521
I
I
3. Sketch graphs showing the follow~ng(for natural rubber except as noted): (a) stress-strain curves for unvulcanized, vulcanized, and re~nforcedstocks; (b) stress-strain curves for vulcanized and reinforced stocks of SBR and natural rubber; (c) tensile strength, tear strength, and permanent set versus extent of vulcanization; (d) tensile strength versus carbon-black loading; (e) modulus versus surface area of carbon black for two values of the structure index
4. Define the following terms: gum, loaded stock, mastication, masterbatching, oil extension, overcure, reinforcement, reversion, scorch, vulcanization. 5.
List (a) four physical properties whose change or lack of change in vulcanization is significant and (b) four physical properties that implove on reinforcement
6. List the ing~edientsin a typical vulcanization mix, identifying each one as to purpose, chemical type or structure, and approximate amount used
BIBLIOGRAPHY
Barnard 1970 D Barnard, J I Cunneen, P B Lindley, A R Payne, M Porter, A Schallamarch, W A Southorn, P McL Swift, and A G Thomas, "Rubber, Natural," pp 178-256 in EPS7, i Vol 12, 1970 Burgers 1965 K A Burgess, F Lyon, and W S Stoy, "Carbon," pp 820-836 In EPST, Vol 2, 1965 Cooper 1966 W Cooper, "Elastomers, Synthet~c," pp 406-482 In EPST, Vol 5, 1966 Coran 1978 A Y Coran, "Vulcantzat~on," Chapter 7 In Fredenck R E ~ r ~ c ed , Sclence and h, Technology ofRubber, Academ~c Press, New York, 1978 Coran 1983 A Y Coran, "The Art of Sulfur Vulcanizat~on," Chemtech 13, 106-1 16 (1983) Coy I965 W~lllamL Cox, "Ant~ozonants," pp 197-203 In EPST, Vol 2, 1965 Garvey 1959 B S Garvey, Jr , "H~storyand Summary of Rubber Technology," Chapter 1 In Maur~ce Morton, ed , Introduction to Rubber Technology, Re~nhold,New York, 1959 Garvey 1970 B S Garvey, Jr , "Rubber Compounding and Processing," pp 280-304 In EPST, Vol 12, 1970 Kraur 1970 G Kraus, "Re~nforcement," pp 42-57 In EPST, Vol 12, 1970 Kraus 1978 Gerard Kraus, "Re~nforcementof Elastomers by Pa~t~culate F~llers,"Chapter 8 In Fredenck R Eir~ch, , Sclence and Technology of Rubber, Academlc Press, New York, 1978 ed Maasren 1965 G C Maassen, R J Fawcett, and W R Connell, "Ant~oxldants," pp 171-197 In EPST, Vol 2, 1965 Morton 1981 Maunce Morton, ed , Rubber Technologv, 2nd ed , Van Nostrand Reinhold, New York, 1973, teprtnted by Kr~eger,New Y o ~ k ,1981 Stagg 1968 Rlchard Stagg, "Latexes," pp 164-195 in EPST, Vol 8 , 1968 StudebaXer 1978 M L Studebake~and J R Beatty, "The Rubber Compound and Its Compos~tion," Chapter 9 In Freder~ckR E ~ n c h ed , Science and Technology oj Rubber, Academic Press, New , York, 1978 Woye 1971 James R Wolfe, Jr , "Vulcantzation," pp 740-757 In EPST, Vol 14, 1971
t Herman F Mark, Norman G Gaylord, and Norbert M Btkalcs, eds , Encyclopedia ojPolymer Sclence and Technology, W~ley-Intersc~ence, York New
APPENDIX O N E
LIST OF SYMBOLS
Symbols in dimensional formulas have the fbllowing significance:
M L
= =
mass length
T 0
= =
time Q = electric charge temperature Chapter Where First Defined
Symbol
Dimension
Common Unit
Definition Angstrom unit Helmholz free energy, work content Collision frequency factor (fir st-order reaction) Collision frequency factor (second-order reaction) Frequency hctor for viscous flow First virial coefficient Second virial coefficient Third virial coefficient Arbitrary constant Exponent in modified Staudinger equation
poise
I
i
L.IST
O SYMBOLS F
i
i
Symbol Dimension
--
Common Unit
Definition
Chapter Where First Defined
Length of crystal unit cell axis .. , "Shift factor" for time-temperature superposition . ,, . Arbitsary constant Length of' crystal unit cell axis cm or A Size parameter of Gaussian distribution , .. Arbitrary constant , , ,, Transfer constant in chain polymerization Thermodynamic constant mole' 2/g"2 of' the Flory--Krigbaum dilute solution theory g/cm3 Concentration gldl Concentration Length of crystal unit cell axis cm'lsec Diffusion constant . ,. . Signifies the total derivative cm Diameter of' spherical particle U Energy content ( J - ~ m ~ ) " ~ l Molar attraction constant mole kJ Activation energy .. , Base of' natural
,,
..
.
, ,.
. ,. .
logarithms Polarity factor in the Alf'rey-Price equation Mole fractions in copolymer Denotes a functional relationship
LIST O F SYMBOLS
Symbol
Dimension
Common Unit
Definition Functionality Initiator efficiency Fraction of polymer in given phase Frictional coefficient Force Volume fraction Mole fractions in monomer feed Gibbs free energy Modulus of elasticity Ratio of sizes of branched and unbranched molecules Coefficient of r2c2in ex.pansion of' osmotic pressure Symmetry tensor in EPR analysis Heat content or enthalpy Light-scattering calibration constant Magnetic field strength Planck's constant Transmitted light flux Incident light flux Summation index Elastic compliance Summation index Equilibrium constant Rate constant for initiation (second order) Light-scattering calibration constant Constant in Hory viscosity equation
Chapter WhereFirst Defined
,
i
2C 3B
kJ psi or dyne/ cm2

,,
.,.
gauss erg-sec W W
...
LIST OF SYMBOLS
525
Symbol K', k ?
Dimension
Common Unit
Definition Empirical constants in modified Staudinger viscosity equations Arbitrary constants Rate constant (second order) Boltzmann's constant Rate constant for decomposition of' initiator (fist order) Coefficients in Huggins viscosity equation Length Abbreviation for natural logarithm Abbreviation for logarithm to the base 10 Molecular weight Number-average molecular weight Viscosity-average molecular weight Weight-average molecular weight z...Ave~age molecular weight Mass of' a molecule Denotes meso symmetxy Number of elastic chains per unit area Number-average molecu-, lar mass Number of' molecules Avogadxo's numbe~ Mole fraction Refractive index
ergldegmolecule sec - '
k'.
glmole glmole glmole
glmole
Symbol
Dimension
Common Unit
Definition Exponent in Avrami equation
psi or dyne/ cm2
Pxessure Reactivity of' radical in Alfrey-Price equation Probability of' an event Angular light-scattering function Extent of reaction Reduced pressme General reactivity factor in Alfrey-Price equation Fraction of crosslinkable monomers on a chain Gas constant
...
cm3-atm/mole deg, exglmole deg, or. .J/ mole deg
Ratio of' volumes of' dilute and precipitated phases Rayleigh ratio Ratio of' number of' A and B groups present in stepwise polymerization Denotes racemic SYmmetrY Distance from origin Root-mean-squase endto-end distance Monomer reactivity ratios Entropy
LIST' OF SYMBOLS
Symbol
DimjrJn
Common Unlt cm or
Definition
Radius of gyration or root-mean-square distance of chain segments f ~ o m center of gravity Sedimentation constant Stress or applied force per unit area Temperature Glass-transition temperature Crystalline melting point Reduced temperature Time, efflux time Size parameter for light scattering of spheres Volume, specific volume, molar volume Reduced volume
sec
psi or dyne/ cmz K or "C K or "C
K or "C
sec
cm3
..
mole/l~ter-sec Rate of reactlon .. Volume fraction Size parameter for light scatteting of ~ a n d o m coils Velocity Partial molar or specific volume Probability function Work Weight, weight fraction
Distance, space coordinate Chain length, number of segments in chain, degree of polymerization Size parametel for light
cmlsec cm3/mole
kJ cm or
..
A
... ...
scattering of rods (Fig 8-7 only)
L.IST OF SYMBOLS
Symbol
Dimension
Common Unit
Definition Number-average degree of polymerization Weight-average degree of polymerization Space coordinate Parameter of distribution functions Size parameter for light scattering of' discs (Fig. 8-7 only) Number of' atoms in polymer chain Space coordinate z-Average (see M,) Parameter of distribution functions Denotes first in a series Branching coefficient Expansion factor of dis-. solved polymer chain Volume expansion coefficient Denotes second in a series Magnetic moment of electron spin Angle of' crystal unit cell Gamma function Coefficient of' c in expansion of' osmotic pressure Strain or. deformation in response to applied stress Shear rate Signifies a change in or difference between the values of a thermodynamic function for two states
Chapter Where Fi~st Defined
Bok magneton degree
. ,. ,,
[.,ST' OF SYMBOLS
Symbol
Dimension
Common Unit
,..
Definition Ratio of' termination to propagation rate constants in copolymerization Solubility parameter Chemical shift (NMR) Phase difference due to energy absorption Cohesive energy density Function of' reactivity
..
.,
Chapter Where Fist Defined
MUZL- 112T-1
(J/cm3)l12
,.
...
(angle) ML-T2
,, ,.
..
deg .J/cm3
. .
...
,.
ML-IT-'
Ilmh
ratios Size parameter of crystallization theory dyne sec/cm2, Viscosity
...
M-'L3
,..
poise dl/g
...
dl/g
..,
Inherent viscosity Relative visocity Reduced viscosity Specific viscosity Intrinsic viscosity Flory temperature, where polymes-solvent inter actions are zero Fraction of' surface sites occupied Angle Thermodynamic constant of Flory -Krigbaum dilute solution theory Wavelength of' light in air Wavelength of' light in solution = Xln Parameter of' distribution functions Magnetic moment of nucleus Huggins polymer-solvent interaction parameter
-I SP '
,. ,,
[q]
M-'L3 0
dl/g
- 0
0
K
...
(angle)
,.
..
...
deg
"..
L
L
cm or cm or
A,
P P
Ph
A
...
...
ML3T-'Q -'I
Bohr magneton
,..
...
5 30
LIST O F SYMBOLS
Symbol Pm
Dimension L MLQ-2
Common Unit micrometer henrylm
Definition Abbreviation for micrometer Magnetic permeability of' vacuum Kinetic chain length Number of' items Frequency Osmotic pressurc Ratio of number of' functional groups on branch units to total number of such gsoups Parameter of distribution functions Rate of generation of radicals Density Summation sign Parameter of distribution functions Parameter of fractionation theory Reduced temperature in scaling theory Turbidity Relaxation time, retardation time Mean lifetime of a free radical Universal constant of Flory viscosity theory Cross-termination probability constant in copolymerization Polymer-solvent mteraction constant
9B 9B 3B 9B 7C
20
v v v
IT
... ...
T-'
ML-'T-2
,. ,
... ...
secatm
'
...
,,
..
sec -
...
3C
'
6B
7A
I
I
3C 70 70
8C 11C
cm-' sec sec
3B
1
i i
8E
5A
Ii
i
i
1
7C
E
LIST OF SYMBOLS
531
Symbol Dimension - - - - -
Common Unit
..
Definition
.-
,,
, .,
Thermodynamic constant of' Flory-Krigbaum dilute solution theory Denotes last in a series Angular velocity Denotes proportionality Signifies par.tial derivative Denote concentration of substance within brackets (except in symbol Degree sign Multiplication sign
7C
..
T-'
...
sec-'
, .
w
K
a
[1
... ,, . .
,
.-. ..
,
80
..
..
,,
.,
,.,
APPENDIX TWO
TABLE OF PHYSICAL CONSTANTS
Symbol
Name Angstrom unit Base of' natural logarithms Planck's constant Boltzmann's constant Avogadro's number Gas constant
Value
1A
m (not a preferred unit)
Absolute temperature Micrometer Nanometer
2 7183 6 6255 x erg-sec 1 3805 x 10-l6 ergldeg 6.0226 x mole-' 8 3 14 x lo7 erglmole-deg 8 314 Jlmole-deg 82 1 cm3 atmlmole-deg 0C = 273.16 K (Kelvin) 1 pm = mm = m 1 nm = pm = mm
APPENDIX THREE
TRADE NAMES AND GENERIC NAMES
This list is not intended to be exhaustive, but to include only a few of the more in common names in widespread or recent use Products are ar~anged the order of the chapters in which they are discussed Full names and addresses of manufacturers should be obtained from curIent directories, for example, the Modern Plastzcs Encyclopedia Name Probable Composition
Manufiicturer
Alathon Bakelite Chemplex Dylan Hi ..Fax Hostalen Marlex Noschem Poly-Eth Rexene Sclair Tenite Tyvek
Chapter 13A Polyethylene Polyethylene Polyethylene Polyethylene Polyethylene Polyethylene Polyethylene Polyethylene Polyethylene Polyethylene Polyethylene Polyethylene Polyethylene Spun-bonded polyolefin
Du Pont Union Car bide Chemplex Arco Hercules American Hoechst Phillips Chemicals Northern Petrochemicals Gulf Oil El Paso Polyolefins Du Pont of Canada Eastman Du Pont
534
'TRADE N A M E S A N D GENERIC. N A M E S
Name
Probable Composition
Manuhcturer
Malex Moplen Nor chem Profax Tenite
Chapter 1.3B Polypropylene Polyprop ylene Polypropylene Polypropylene Polypropylene Polypropylene
Phillips Chemicals Novamont Northern Petrochemicals Hercules Eastman
Alathon Elvax Eva1 Maslex Nucrel Oxytuf'
Chapter 1.3C. Other Ole$n Polymers Ethylene copo1,ymers Du Pont Ethylene-vinyl acetate copolymers Du Pont Ethylene-vinyl alcohol copolymers' Hercules Ethylene copolymers Phillips Chemicals Ethylene--methacrylic acid Du Pont cop01ymer EPDM-vinyl graft copolymer Occidental Chemical Chapter 13D-F Olefin-Based Elastomers Polyisoprene (czs-1,4) Goodrich Butadiene-acrylonitrile copolymer Goodyear Butadiene copolymers Goodrich Chlorosulfonated polyethylene Du Pont Nitrile and nitrilelbutadiene Polysar elastomers Polyisoprene (cis-1,4) Goodyea Polychloroprene Du Pont EPDM elastomer Du Pont Polysar Butyl, halobutyl, SBR, and EPDM elastomers Polybutadiene elastomers Polysa Thermoplastic elastomers Uniroyal Enjay Polyisobutylene Chapter 14A. Polystyrene and Related Polystyrene Styrene-maleic acid copolymer, impact ABS Polystyrene Polymers Union Carbide Monsanto
Marbon Foster-GI ant
Ameripol Chemigum Hycar flypalon Krynak Natsyn Neoprene Nordel Polysas Taktene TPR Vistanex
Bakelite Cadon Cycolac Fostasene
TRADE NAMES A N D GENERIC NAMES
535
Name Kralastic Lustran Lustrex Oxyloy Rove1 Styson
Probable Composition ABS ABS Polystyrene Alloy of' EPDM--vinyl graft copolymer with ABS Styrene-,based terpolymer Polystyrene
Manufacturer U..S. Rubber Monsanto Monsanto Occidental Chemical Uniroyal Dow Chemical
Acrilan Barex Creslan
Elvacite Hycar Implex Lucite Orlon Plexiglas
Chapter 14B Acrylic Polymers Ac~ylonitrile-vinyl acetate Chemstrand copol ymer Modified polyacrylonitrile Hercules Acrylonitrile-vinyl ester American Cyanamid copolymers Vinylidene cyanide-vinyl acetate Celanese copol ymer Du Pont Methyl, ethyl, butyl methacrylate polymers and copolymers Acrylic elastomer Goodr ich Rohm and Haas Poly(methy1 methacrylate), impact modified Poly(methy1 methacrylate) Du Pont Du Pont Polyacrylonitrile Poly(methy1 methacrylate) Rohm and Haas Chapter 14C. Poly(vzny1 esters) and Derived Polymers Du Pont Poly(viny1 butyral) Du Pont Poly(viny1 acetate) Du Pont Poly(viny1 alcohol) Monsanto Poly(viny1 butyral) Poly(viny1 acetate), poly(viny1 Union Carbide alcohol) Chapter 140. Chlorine-Containing Polymers Poly(viny1 chloride) and Union Carbide copolymers Acrylonitsile-vinyl chloride Union Carbide copolymer Poly(viny1 chloride) and Goodrich copolymers
Butacite Elvacet Elvanol Saflex Vinylite
Bakelite Dyne1 Geon
536
T R A D E N A M E S A N D GENERIC N A M E S
Name Saran Tygon Vinylite
Probable Composition Poly(viny1idene chloride) and copolymers Poly(viny1 chloride) and copo1ymer.s Poly(viny1 chloride) and copolymers Dow U . S . Stoneware Union Carbide
Aclar Fluon Halar Halon Kel-F Kynar Tedlar Teflon Teflon FEP Tefiel Viton
Chapter 14E Fluorine.-Containing Polymers Polychlorotrifluoroethylene Allied Polytetrafluoroethylene ICI Americas Ethylene-chlorotrifluoroethylene Allied copolymer Polytetrafluoroethylene Allied Polychlorot~ifluoroethylene 3M Poly(viny1idene fluoride) Pennwalt Poly(viny1 fluoride) Du Pont Polytetrafluoroethylene Du Pont TetrafluoroethyleneDu Pont hexafluoropropylene copolymer Tetrafluosoethylene-ethylene Du Pont copolymer Vinylidene fluoride-hexafluoroprop Du Pont ylene copolymer. Chapter 1SA . Polyamides 6-Nylon (polycaprolactam) 6-Nylon (mineral filled) 66-Nylon (polyhexamethylene adipamide) film 6 12-Nylon copolymer Poly(p-phenylene terephthalate) 66-Nylon (mineral reinforced) Poly(m-phenylene isophthalate)
Caprolan Capron Dartek Grilon Kevlar Minlon Nomex Quiana Rilsan Tynex Vydane
Allied Allied Du Pont of Canada Emser Industries Du Pont Du Pont Du Pont Du Pont Rilsan Du Pont Monsanto
Poly(bis-p-aminocyclohexylmethane dodecamide) Nylons 11 and 12 Nylons, undisclosed (monofilament) Nylon (glass and mineral reinforced)
5 37
Name
Probable Composition Nylons 6, 66, 610, and copolymers
Manuf'acture~
Zytel
Du Pont
Chapter 15B. Polyesters, Polyethers, and Related Polymers Polyether-based polyurethane Du Pont (elastomer) Ardel Polyarylate Union Carbide Bayblend Polycarbonate-ABS blend Mobay Celcon Acetal copolymer Celanese Chemigum Goodyeax Polyester-based polyurethane (elastomer) Dacron Poly(ethy1ene terephthalate) (fiber) Du Pont Delrin Acetal polymer (polyoxymethylene) Du Pont Goodrich Estane Polyester-based polyurethane (elastomer) Poly(buty1ene terephthalate) GAF Gafite Isoplast Thermoplastic polyurethane (impact Upjohn modified) Kodapak Poly(ethy1ene terephthalate) Eastman Kodar Poly(ethy1ene terephthalate) (glycol Eastman modified) Lexan Polycarbonate General Electric Lycra Polyurethane (fiber) Du Pont ~Merlon Polycarbonate Mobay Montac Monsanto Pol yester--amide block copolymer Mylar Poly(ethy1ene terephthalate) (film) Du Pont Polathane Polyurethane (elastomer) Polaroid Poly(ethy1ene oxide) Union Carbide Polyox Polyaryl sulfone Union Carbide Radel Poly(ethy1ene terephthalate) (glass Du Pont Rynite reinforced) Poly(pheny1ene sulfide) Phillips Ryton Tenite Poly(ethy1ene terephthalate) Eastman Thiokol Polysulfide Thiokol Udel Polysulfone Union Carbide Poly(buty1ene terephthalate) General Electric Valox Vibrathene Polyurethane prepolymers Uniroyal Victrex Polyether sulfbne ICI Americas
Adiprene
5 38
Name Xenoy
Probable Composition Poly(buty1ene terephtha1ate)-polycarbonate blend
ManufBcturer General Electric
'
Acele Ampol Arne1 Avisco Cordur a Forticel Fortisan Kodapak Tenite
Chapter 1.5C. Cellulose-Based Polymers Cellulose (di)acetate Du Pont Cellulose esters and ethers American Polymers Cellulose triacetate Celanese Viscose and acetate rayons American Viscose Viscose rayon, high tenacity Du Pont Cellulose propionate Celanese Saponified cellulose acetate Celanese Eastman Cellulose acetate, propionate, butyrate Eastman Various cellulosics Chapter 15D Hzgh-Temperature and Inorganic Polymers Carbon (fiber) Great Lakes Carbon Polyimide (film) Du Pont Poly(pheny1ene oxidetpol ystyrene General Electric blend Poly(pheny1ene ether) Borg Warne~ Pol yamide-imide Amoco Chemicals Polyether-imide General Electric Chapter 16A. Phenolic and Amino Urea-formaldehyde Phenol-formaldehyde Urea-formaldehyde Phenol-formaldehyde Phenol, melamine, and urea-. formaldehyde Phenol-formaldehyde Resins American Viscose Union Carbide American Cyanamide Hooker Allied Plastics Engineering
Fortafil Kapton Noryl Prevex Torlon Ultem
Avisco Bakelite Beetle Durez Plaskon Plenco
Gem-Glaze Selectron
Chapter 16B. Unsaturated Polyester Resins Unsaturated polyester General Tire Unsaturated polyester PPG Industries Chapter I6C Epoxies and Polyurethanes Ciba
Araldite
EPOXY
TRADE NAMES A N D GENERIC. NAMES
5 39
Name Bayflex Epon
Probable Composition RIM polyurethane EPOXY
Manufacturer .--Mobay Shell
Rimplast
Silastic
Chapter 160 Silicone Polymers Silicone-based interpenetrating polPetrarch Systems ymer networks, polyamide and polyurethane components Silicone Dow Chapter l6E. Miscellaneous Thermosetting Resins PPG Industries Ally1 carbonate Hooker Diallyl phthalate Alkyd, diallyl phthalate Allied
CR-39 Durez Plaskon
AUTHOR INDEX
f Page numbers in rtalzcs refer to references at end of chapters.

Abrams, I M ,387,404 Adam, G , 323, 327 Aggarwal, S L., 122, 123, 471, 481 Akin, R B , 419, 431 Alexande~, E , 240, 253 L Alfrey, T , J r , 110, 114, 117, 123, 124, 144, 145, 316,317,320, 326, 327, 337,340,343, 355, 469,471, 481 Allcock, H R , 32, 47, 56,68,71,75, 77, 144, 145, 252,253, 431, 431, 432, 434 Allen, G , 204, 223 Allen, P E M , 32, 46, 47, 56, 68, 75, 77, 204, 223 Allport, D C , 122, 123 Altgelt, K H ,2 17, 22 3 Anderson, C F , 21 9, 223 Anderson, R A ,464, 481 Andrews, B A., 426, 432 Andrianov, K A , 431, 432 Archibald, W J , 206, 223 Arledter, H F , 431, 432 Armeniades, C D , 323, 327 Armstrong, R W ,219, 223 Asmussen, F , 15 3, 181, 185 Aspland, J R , 499, 504 ASTM, 252,253 ASTM D2857,213, 223 ASTM D3132, 154, I81 ASTM D3536,217, 223 ASTM D3592, 197, 223 ASTM D3593,217, 223 ASTM D3750, 197,223 ASTM D4001,205,223 Atherton, J N , 109, I23 Atlas, S M., 20, 504 Avrami, M ,282,297 Backer, S , 504 Backus, J K , 419, 432, 448, 453 Bacskai, R , 124 Badgley, W J , 196,223 Baekeland, L H , 10, 19 Bae~, , 252, 253, 327 E Bagchi, S N ,255, 298 Bajaj, J K. L ,475, 481 Baker, C A , 177, 182 Baker, G , 372, 380 Baker, R , 88, 98 Bakkei, M , 354, 355 Baldwin, F B., 379, 380 Ballintyne, N. J , 41 9, 432 Barnford, C H , 56, 67, 68, 71, 78, 110, 120, 123, 134,145, 391, 404 Bannerman, D G , 333, 355 Baclay, R K , 297 Bare;, J , 48, 78, 98, 124, 146, 182, 225, 254 Barnard, D , 373, 380, 518, 521 BarxaU, E M , 227 Bartolomucci, J R , 435 Barton, J M , 146 Basdekis, C H , 387, 404 Bassett, D C , 268, 279, 281, 289, 297 Batiuk, M , 464, 481 Battaerd, H A J , 122, I23
AUTHOR INDEX
Battista, 0 A , 435 Bauer, H , 257 Bauer, R G , 376, 380 Bauer, R S ,448, 453 Bauer, W. H , 302, 327 Bawn, C E H , 164, 182, 335, 355 Beach, D L , 381 Beaman, R G , 88, 98, 337, 355 Bean, A R , 373, 380 Beardmore, P , 471, 481 Beatty, R , Jr , 521 Becker, W E ,461, 481 Beeler, A D , 475, 481 Beevers, R B , 339, 355, 391, 404 Bekkedahl, N , 320, 32 7 Bell, T E , 501, 504 Bellemans, A , 184 Benning, C 5 , 4 6 9 , 481 Benoit, H , 182, 204, 215, 223, 226, 327 Benton, J L , 404 Benzera, L , 404 Berenbaum, M B ,419, 432 Berne, B J , 204, 223 Berry, G C , 225, 356 Bertozzi, E R , 381 Bevington, J C , 51, 52, 56, 78, 419, 432 Biesenberger, J A , 136, 146, 1 4 7 Bikales, N M , 255, 426, 432, 435, 501, 504 Billingham, N C , 18, 19, 187, 194, 197,205, 208,213,217, 223 Billrneyer, F W , Jr , 7, 8, 18, 19, 48, 78, 80, 98, 124, 146, 156, 159, 178, 182, 186, 197, 202,203,204,205,214,223, 224, 225,226, 227,250,253,254,287,297, 352,354, 355, 356 Birbeck, M S C ,242, 253 Bird, R B , 134, 146 Birshtein, T M , 159, 182 Blackadder, D A , 274, 297 Blackley, D. C , 134, 146, 373, 376, 379, 380 Blair, L M , 434 Blake, W. P., 475, 481 Block, B B , 43 1, 432 Blokland, R , 41 9, 432 Blumberg, J G , 475, 481 Bly, D D , 228 Boeke, P J ,419, 430, 432 Boenig, H V , 343, 349, 355, 445, 453 Boerio, F J , 238, 253 Bogan, R T ,426, 432 Bohme, R E ,67, 78 Bohrer, J J , 123, 145 Bolker, H I , 435 Bonk, H W , 379, 380
Bonnar, R U , 197,224 Bonotto, S , 372, 380 Boor, J , JI , 94, 96, 98 Borg, E L , 372, 381 Borman, W F. H , 433 Bottenbruch, L , 419, 432 Bovey, F A , 8, 19, 144, 146, 157, 159, 182, 236, 238,253, 254, 263, 297 Bowden, M J , 32, 35,40,46, 48,68, 78, 88, 91, 96, 97, 98 Bowen, T J., 208,224 Boyd, R H , 300, 357 Boyer, R F , 238, 254, 337, 338, 355, 356, 387, 404 Boysen, R L , 368, 381 Brady, D G , 433 Brady, P , 227 Brand, R H , 504 Branson, H R , 125, 299 Brasure, D E , 403, 404 Brauer, G M ,230, 254 Braun, D , 55, 78, 229,230, 254, 419, 432 Braunsteiner, E E , 430, 432 Breil, H , 100 Breitigam, W V , 448, 453 Brenner, W , 146 Bresler, S E , 238, 254 Bressan, G , 297 Brewer, R J , 432 Brice, B A , 199, 202, 225 Brickman, R D , 8 0 Brighton, C A , 398, 404 Brilliant, S D , 475, 481 Bringer, R P , 404 Brode, G L ,442, 453 Bxookman, E F., 8 0 Broutman, L J , 481 Brower, F M , 356 Brown, D S , 240, 254 Brown, H C ,403, 404 Brown, R 5 , 4 6 4 , 481 Brown, R M ,426, 432 Brown, W E ,246,252,254, 398, 404 Bruins, P F ,419, 432, 448, 453 Bruise, J P , 405 Bruxelles, G. N , 426, 432 Bryan, W. P , 273, 297 Bryant, W M D , 72, 78, 277, 297 Brydson, J A , 311, 327, 372, 373,376, 379, 381, 391,403, 405,419, 432,451,453,464, 481 Buchanan, P R , 368,369, 381 Buchi, G , 454 Buckley, D J , 368, 372, 379, 381
AUTHOR INDEX
Bucknall, C B , 343, .3.56 Bueche, F..,323, ,327 Bullock, A. T , 238, 254 Bunn, C. W., 276, 291, 297, 332, ,356 Burge, D . E., 197, 225 Burgess, R. H , 399, 405, 516, 521 Burlant, W. J., 122, 124 Burnett, G. M., 67, 78, 107, 113, 124 Bu~netfP.R. A,, 48.3 Burrell, H , 153, 154, 182 Butler, G. B., 140, 146 Buttery, D . N., 475, 481 Bywater, S , 91, 98 Cabannes, J , 199, 225 Caldwell, E D , 382 Cameron, G G , 2 5 4 Campbell, P E , 382 Campbell, R W , 433 Campbell T W , 80, 335, 356 Canale, A J , 76, 78 Canterino, P J , 382 Cantow, M J R , 177, 180, 182, 184 Cargile, H M ,469, 481 Carley, J F., 307, 327 Carol, F J , 368, 381 Carothers, W H , 9, 19, 25, 48 Carpenter, D K , 166, 168, 182, 213, 225 Carlaher, C E , 32, 48, 431, 432 Casassa, E F , 176, 182, 205, 225, 228 Cassel, J M , 230, 254 Cassidy, H. G , 138, 146, 430, 432 Catsiff, E , 329 Caughey, E. C , 433 Cazes, J , 214, 225 Cekis, G V , 431, 432 Ceresa, R J , 122, 124 Ce~nia, , 504 E Chan, R K S , 124 Chapiro, A , 144, 146 Charlesby, A , 144, 146 Charvaf R A , 483 Chasar, D W , 475, 482 Chatterjee, A M , 372, 381 Chatterjee, S R , 8 0 Cherdon, H , 78 Chiang Y , 2 56 Chiu, J , 245, 254 Chu, B , 16,20, 204,225 Chung, C I , 4 6 1, 482 Chung, D A , 381 Clavadetsche~,D ,445, 453 Cohen, F. S ,403, 405 Cohen, M H , 323, 327
Coil, H , 197, 225 Collier, J R , 342, 356 Collins, E A , 32, 40, 48, 55, 56, 64, 68, 78, 88, 91, 98, 110, 124, 134, 146, 153, 182, 187,197,205,213,217,225,229,238,240, 242, 245, 252, 254, 327 Colvin, J. R , 435 Colwell, R E , 329 Comell, W R , 521 Cooke, H G , Jr , 79 Coopex, G D , 435 Cooper, H W , 391, 405 Cooper, S L , 3 8 2 Cooper, W , 134, 146, 391, 581, 521 Coran, A Y , 513, 521 Corbiere, J ,497, 504 Corey, R B , 299 Corradini, P , 20, 99, 273, 297, 299 Cossee, P , 92, 98 Cotton, J P , 182, 223, 303, 327 Cottrell, T L., 16, 20 Coulter, K E., 404 Cowie, J M G , 212,213,225, 339, 356 Cox, W L , 518, 521 Cozzens, R F , 146 Cragg, L H ,208, 225 Crescentini, L , 433 Crespi, G , 369, 381 Crick, F H C , 300 Crissman, T M , 357 Cross, L H ,233, 254 Crothers, D M , 228 Crouzet, P ,217, 225 Crump, E L ,469, 482 Cuming, A P C , 435 Cumins, H 2 , 2 0 4 , 225 Cunliffe, A V , 238, 254 Cunneen, J I , 380, 521 Dakin, D F., 434 Danielle, D..,22.3 Danusso, F., 20, 99, 299 Daoud, M., 167, 176,182 Darby, J. R., 475, 482, 484 Dart, E C., 91, 9 8 Daul, G. C., 426, 4.32, 4.34 David, D . J., 419, 448, 45.3 Davidson, H..R., 300 Davis, C. W., 391, 405 Davis, G T., 298 Davis, G. W., 419, 4,32 Dawkins, J. V , 217,225 Dealy, J . M., 304, 307, ,327 Debye, P., 198, 225
544
AUTHOR INDEX
Dec, J , 79 Decker, D., 327 D'Esposito, L., 231, 254 deGennes,P G , 158,159,167,168,176,182, 306, 327 De Grey, I V , 432 Delmas, G , 184 Delmonte, J , 47 1, 482 De Long D J , 432 des Cloizeaux, J , 167, 182, 223, 32 7 Desper, C R , 295, 297 Desreux, V , 298 de Than, C B , 223 Dewar, M J S , 78 Dickinson, B L , 4 3 1, 433 Di Marzio, E A , 323, 327, 328 Dimbat, M , 224 Doak, K W , 382 Dobry, A , 196, 225 Dohany, J E ,403, 405 Dole, M , 51, 78, 256, 300, 333, 336, 356, 357 Dominguez, R J G , 419, 433, 448, 453 Doolittle, A K , 323, 327 Dostal, H , 101, 124 Drehlich, A , 501, 504 Dreyfus, P , 419, 433 Duch, M W , 237, 254 Duck, E W , 134, 146 Duda, J L , 404 Duke, C B ,252,254 Dukerf A A , 405 Duncan, R E , 372, 381, 482 Duplessix, R , 182 Durso, D F , 435 Dux, J P , 404 Dym, J B , 461, 482 Eastham, A M , 88, 9 8 Eastman, N , 435 Eastmond, G C , 5 1, 71, 78, 145, 404 Ebdon, J. R , 59, 78 Economy, J , 43 1, 433 Edgar, 0 B , 334, 337, 356 Edwards, F G , 406 Ehrmantr aut, H C , 227 Eichhorn, R M , 419, 433 Eichinger, B E , 166, 182, 183 Eighmy, G W , JI ,469, 482 Eirich, F R , 307, 320, 327 Eise, K , 475, 482 Elias, H G , 8, 16, 20, 26, 32, 46, 48, 68, 71, 78, 85, 88, 91, 96, 98, 110, 120, 124, 144, 146, 159,176,183,205,207,208,213,225,
337, 340,343,349, 356, 391,395,405,413, 419,426,430, 431, 433, 442, 448, 453 Eliseeva, V I , 134, 146 Ellerstein, S M , 379, 381 Elliott, J H , 177, 183, 238, 254 Ellis, D R , 482 Ellis, J R , 501, 504 Eskin, V E ,205, 225 Evans, T E , 387, 405 Ewald, G W ,445, 453 Ewart, R H , 147, 210, 225 Eyring, H , 79, 328 Ezrin, M , 197, 226, 255 Far, R L , 461, 482 Falcone, J S , Jr , 481 Farber, E , 134, 146 Farina, M , 266, 297 Farnoux, B , 182, 223, 327 Far~ar, C , 98, 381 R Farrissey, W J , 434 Farrow, G ,419, 433 Faulks, B F ,475, 482 Fava, R A , 273, 281, 297 Fawcett, R J , 521 Feinberg, S C , 99 Felnland, R , 230, 254 Fenelon, P J , 4 7 1, 482 Feny, J D ,315,319,320,323,326, 328, 329 Feth, G , 431, 433 Fetters, L J , 90, 98, 99, 121, 124 Feughelman, M , 265, 297 Fiel, R J , 227 Fine, F , 225 Finelli, A F , 379, 381 Fimey, D C , 481 Fischer, E W , 242, 255, 273, 298 Fisher, E G ,464, 482 Fisher, W B , 413, 433 Fixman, M , 228 Fujita, H , 208, 226 Flory,P J,4,10,11,20,26,34,36,37,40,43, 44, 45, 46, 48, 53, 68, 70, 71, 75, 78, 154, 157, 158,159,162,164,166, 167,168,174, 175,176,182, 183, 184,211,212,226,227, 276,298, 303,305, 328,336,337,341, 356 Floryan, D E , 431, 433 Ford, R , 355 Foster, 1 , 469, 482 Fox, D W , 419, 433 Fox, R B , 146 Fox, T G , JI , 305, 328, 339, 356 Fram, P , 391, 405
AUTHOR INDEX
545
Frank, H P , 369, 381 Franta, W A , 357 Fraser, A H ,426,430, 433 F~ederickson,R B ,46 1, 482 F~eeman, I , 169, 183 P Freeman, R F J , 182 Friedlander, H N , 382 F~iedlander, Z , 51, 79, 379, 381 H Frisch, K C , 435, 454 Frissel, W J , 475, 482 Fuchs, 0, 179, 183 Funt, B L , 51, 79 Furukawa, J , 97, 98 Gagliardi, D D , 505 Galanty, P G , 41 3, 433 Gangal, S V , 403, 405 Gannon, J , 454 Gantz, G M , 504 Gardon, J L , 134, 146, 154, 183 Ga~mon, G , 187,226 R Gam, P D , 256 Garrett, A B , 398, 405 Garvey, B S , 509, 516, 518, 521 Gaylord, N G , 96, 98, 255, 271, 298 Gee, G , 163, 164, 183, 196, 226 Geil, P H ,242,255, 268,273,274,279,281, 298, 299, 300 Gent, A N , 311, 328 Ghosh, P , 80 Giannini, U , 99 Gibbs, J H , 323, 327, 328 Gibbs, W E , 140, 146 Gibson, H W ,254 Giesekus, H , 179, 183 Gilbert, R D , 79 Glasstone, S , 74, 79, 303, 328 Glover, C A , 197, 226 Goddar, H ,255 Gogos, C G , 484 Goldfinger, G , 54, 64, 79, 123, 124 Goode, W. E , 78 Goodman, I , 419, 433 Gordon, J B ,469, 482 Gordon, M., 70, 79 Go~dy, , 256 W Go~nick,F , 227 Goswami, B C ,491,501, 504 Graessley, W W , 213, 226, 303, 328 G r a i n , M W , 434 Graham, R K , 78 Graham, W A G , 435 Grant, D M , 254
Grassie, N , 144, 146, 396, 405 Grassie, V R , 432 Gratch, S , 328 Grebowicz, J , 300 Green, J , 435 Greenstein, L M , 482 Greg& R A , 62, 79 Greminger, G K , JI , 426, 433 Grosberg, P , 504 Gross, S T , 300 Grubisic, Z ,216, 223, 226 Gurnee, E F , 327, 355, 404 Guth, E , 309, 310, 328 Gutoff, R , 353, 354, 356 Guzman, G M , 177, 180, 183 Habn, H T , 484 Haisser, A 5 , 464, 482 Hale, W F ,430, 433 Hall, C M , 460,461, 482 Hall, R W , 177, 178, 180, 183 Halwer, M , 225 Ham, G E , 56, 79, 109, 110, 120, 124 Hamilton, I, D ,297 Hammett, L P , 118,124 Han, C D H ,461,464, 482 Hancox, N L , 47 1, 482 Harborth, G , 80 Hardman, B B ,45 1,45 3 Hargreaves, C A , 1 , 379, 381 1 Harrington, H J , 452, 453 Harris, B C , 485 Hartshorne, N H , 242, 255 Harwood, J J , 481 Haskell, V C ,426, 433 Haslam, J , 229, 255 Haven, A C , J I , 356 Hawkins, W L , 144, 146 Hawthorne, D G , 4 8 4 Hawthorne, J. M , 41 9, 433 Haywa~d, , 20 R Hearle, J W . S , 501, 504 Hearons, J S ,471, 482 Heckert, W. W , 489, 504 Heffelfinger, C. J , 433 Hein, G E , 2 9 7 Hellman, M., 187,226 Hemsley, D., 242, 255 Henderson, J F , 90, 98 Hendra, P J ,232,255, 274, 298 Henkel, R N , 368, 381 He~ci-Olive,G , 92, 94, 98 Hensel, J D ,430, 433
AUTHOR INDEX
Herbert, V , 469, 482 Herbes, W F , 454 Herglotz, H K , 240, 255 Herzegh, F , 378, 381 Heuer, W , 21, 228 Hewett, J V , 504 Hiatt, G D , 257 Hicks, E M., 492, 504 Higgins, D G ,469, 482 Higgins, J S , 223, 32 7 Higginson, W C E , 90, 98, 100 Hildebrand, J H , 152, 154, 159, 183 Hill, E S , 432, 433 Hill, H W , 419, 433 Hill, J W , 434 Hill, R , 332, 356 Hill, R 0 , Jr., 426, 433 Hines, G R , 380 Hitchcock, A B , 482 Hobbs, L N ,356 Hobbs, S Y , 281,289, 298, 336, 356 Hobson, P H , 391, 405, 497, 504 Hoffman, A S , 122, 124, 184 Hoffman, J. D , 283,289, 298 Hogan, J P , 368, 381 Holcomb, A G , 406 Holden, G , 380 Holland-Moritz, K , 256 Holleran, P M , 195, 226 Holmes-Walker, W A , 459, 482 Holohan, J F , JI , 372, 381 Holtgrewe, D A , 369, 381 Holzkamp, E , I00 Honeycutt, E M ,381 Hooper, C W , 297 Hopfinger, A J , 159, I83 Hopmeu, A P ,475, 482 Horne, S E , Jr , 382 Horner, L I , 433 Ho~owitz, , 252, 255 E Hosemann, R , 240, 255, 278, 298 Houston, R R , 380 Houtz, R C , 390, 405, 427, 433 Howard, G J , 38, 40, 46, 48 Howell, W R , 454 Hoy, K L , 153, 154, 183 Hsieh, H L 88, 90, 98, 376, 381 Huang, S Y , 99 Hug, D P , 368, 381 Huggins, M L , 162, 168, 183, 184, 209, 226, 261,298 Huglin, M B ,205, 226 Hull, D , 47 1, 482
Hull, H . H , 307, ,328 Hull, .J. L., 461, 482 Humrnell, D. 0 , 2 3 8 , 2 5 5 Hunger, G. K., 4.34 Hunter, D. N . , 431, 43,3 Hyde, A, J . , 205, 226 Hyun, K. S., 404 Ibrahim, S M , 419, 433 immerguf E H , 228 Ingram, P , 277,281,295,298 International Union, 208, 226 Irwin, C ,46 1 , 482 Ivanchev, S S , I46 Ivin, K J , 75, 79 Iwama, M ,22 7 Jackson, W 0 , 432 Jacobs. D B , 413, 433 Jaffe, M , 285,298 Jalbert, R L , 47 1, 482 James, H M , 328 Janes, W H , 123 Jannink, G , 182, 223, 32 7 Jaquiss, D B G , 419, 433 Jellinek, H H G , 144, 146, 147 Jenkins, A D , 63, 79, 134, 147 Jenkins, L T ,20, 256 Jensen, J. C , 445, 453 Jeziorski, R J , 398, 405 Jezl, J L , 369, 381 Johnson, J F , 180, 184, 227 Johnson, P R , 372, 379, 381 Johnson, R L ,403, 405 Johnson, R N ,430, 433 Jones, R V , 382 Joshi, R M , 72, 7 5 , 79 Kakudo, M ,240, 255 Kamada, K ,227 Kamal, M R , 471,483 Kamaliddin, A R , 182 Kargin, V A , 343, 349, 356 Kasai, N , 255 Katz, H S ,475, 483 Katz, M , 419,433 Kaufman, M , 398, 405 Kazbekow, E N ,254 Ke, B ,244,245,255 Keegan, J F ,442, 453 Keeney, J F , 453 Kehde, H , 404 Keinath, S E ,254
i 1
AUTHOR INDEX
Kelen, T , 99, 144, 147 Keller, A , 273, 274, 276,280, 291, 298 Kelley, R N , 215, 226 Kelly, W J , 381 Kelso, R L , 78 Kennedy, J P , 82,88,99, 373, 376, 379, 381 Kennedy, R K , 39 1 , 405 Kentgen, W A , 442, 453 Kerker, M., 204, 205, 226, 227 Kern, W , 78 Keskkula, H , 404 Ketelaar, J A A , 16, 20 Ketley, A D , 96, 99 Khoury, F., 281, 298 Kiernan, E F., 469, 483 Kiesau, M , 461, 483 Kim, K Y, 329,403, 405 Kimber, P J , 435 Kin, L , 356 Kine, B B , 391, 405 Kinsey, R H , 3'70, 381 Kinsinger, J B , 78, 204, 226 Kirkland, J J ,228 Kissin, Y V , 381 Kisman, K R , 481 Klein, 1,464, 483, 484 Kleine~t, I , 435 'N Kline, G M , 229,230, 254, 255 Klug, E D , 435 Koch, F. W , 434 Kochhar, R K , 372, 381 Koening, J L , 232, 253, 254, 255 Kohan, M L ,413, 434 Kohle, H , 80 r Kokle, V ,224 Kolcynski, J R , 79 Koleski, J V , 398, 405 . Komoto, H , 338, 356 Koningsveld, R , 170, 184 Kookootsedes, G J ,45 1 , 453 Kornberg, A , 265,298 Korus, R , 68, 79 Kossler, I , 238, 255 Kossoff, R M , 354, 3 56 Kotera, A , 177, 184 Kovacic, P ,430, 434 Kovacs, A J , 32 3, 328 Kraemer, E 0 , 20, 79, 210, 226 Kraft, P ,405 Kramer, 0 , 320, 328 Krassig, H , 435 Kratohvil J P , 203, 227
Kraus, G , 516, 521 Kraus, M , 138, 147 l a u s e , A , 176, 230, 255 Krause, S , 176, 184 Krauskopf, L G ,475, 483 Krenz, H , 391, 405 Kresser, 1 0 G , 368, 369, 372, 381 Krigbaum, W R , 183, 190, 197,211, 227 Kubiak, R S ,461, 483 Kuchanov, A I , 146 Kun, K A , 146 Kuo, C M , 432 Kupfer, W , 146 Kurata, M , 212, 213, 227 Kuzma, L J , 376, 381 Kwei, T K , 168, 176, 182, 184 Lagally, P , 80 Lai, J T , 482 Laidlei, K J , 79, 328 Lambelt, J M , 300 Lampe, F W , 47, 77, 145, 253, 432 Landel, R F , 329 Landlex, Y , 120, 124 Landrock, A H , 482 Lange, A , 255 Langlon, J A , 380 Langridge, R , 297 Lansing, W D , 10, 20, 70, 79 Lanson, H J , 452, 453 Lantz, J M , 387, 405 Lauritzen, J I , 283, 298 Layman, P L , 41 1 , 434 Lebedev, A V , 146, 237,255 Ledwith, A , 88, 96, 99 Lee, H ,448, 453 Leeper, H M ,300 Lenz, R W , 26,27, 32, 40, 46, 48, 56, 68, 79, 83,88,91,96,97,99, 120, 124, 128, 134, 144, 147 Leve~, E ,229,252, 255 A. Levi, D. W ,256 Levine, H H , 430, 434 Levine, H I , 203, 224, 227 I~vovits, , 404 A Levy, S ,464, 483 Lewis, C E , 501, 504 Lewis, F M , 124 Lewis, G N , 159, 184 Lichtenberg, D W ,442, 453 Lichtenwalner, H K , 45 1, 454 Lightfoot, E N , 146 Lindau, J W , 483
AUTHOR INDEX
Lindemann, M. K.., 395, 405 Lindley, P. B., ,380, 521 Livesey, P. J , 204, 224, 227 Loan, L,.D., 144, I47 Lofihouse, W. G.., 4.34 Long, V C., 340, ,356 Lontz, J F , ,357 Loshaek, S . , ,328, ,356 Lowry, G. G., 109, 124, 159, 184 Lubin, G., 471, 483 Luciani, L , 381 Lundberg, J . , 426, 4.34 Lupton, E. C., Jr., 406 Luskin, L. S . , 391, 405, 469, 483 Lyon, F , 521 Lyons, J. W..,213, 227, ,329 Maassen, G C , 518, 521 McCaffery, E M , 55, 79 McCall, J S ,208,227 McCane, D. I , 403, 405 McClellan, A L., 20 McCord, R A , 482 McCormick, H W ,341,356 McCrum, N G , 320, 326, 328 McCullough, R L , 326, 328 McDonaugh, J M ,469, 483 McGarry, F J , 471, 483 McGovern, J. N , 491, 504 McGrath, J. E , 125, 373, 376, 379, 381 McGrew, F. C., 364, 382 McIntyre, D , 197,205,227 McIntyre, J. E , 491, 497, 504 McIntyre, J. M., 405 MacKenzie, R C , 245, 255 McKelvey, J M ,461,464, 469, 483 Mckillip, W J , 452, 454 McKinley, C , 100 McLarty, J L ,445, 454 McPeters, A L , 405, 504 McPherson, J L , 357 McQuiston, H , 354, 356 Madorsky, S L , 144, 147 Magaf E E , 355 Mageli, 0. , 50, 79 L Magnin, P , 225 Mandelkern, L , 281, 287, 289, 298, 299 Manley, R , 435 Mann, R H , 380 Manson, J A , 47 1 , 483 Mantica, E , 20, 99, 299 Marechal, E , 99 Mark, H F , 4, 8, 11, 20, 26, 32, 43, 48, 98,
123, 145,223,246,252,255,298,328,408, 426, 434, 491,497, 504 Marks, G C , 404 Marshall, R A , 381 Martin, H , 100 Martin, J. D ,463,464, 483 Martindale, J G , 504 Marvel, C S , 1 1 , 20, 32, 48, 56, 59, 79, 430, 434 Mascia, L , 475, 483 Matheson, M S , 79 Mathews, G , 475, 483 Mathof V , 184 Matsurnoto, M , 128, 147 May, H , 432 May, J A , Jr , 217,227 Mayo,FR,58,63,76,79,101,107,112,113, I24 Mazzanti, G , 20, 99, 299 Meie~, J , 176, I84 D Meinecke, E , 469, 483 Melillo, L , 299, 300 Melville, H W , 78 Menzer , A B ,445, 454 Menett, F M , 74, 79, 347, 356 Merz, E , 109, 124 Metz, H , 469, 483 Meyer, V E , 107, 1 1 1 , I24 Meyerson, I , 228 Michaels, A S , I47 MicMey, E S , 133, 147 I Middleman, S., 461, 464, 469, 483, 497, 505 Milewski, J V , 483 Miller, G W , 245, 256 Miller, M L , 391, 405 Miller, R L , 238, 256, 273, 298 Millich, F , 32, 48 Miner, L H ,471, 483 Mitchell, R L., 426, 434 Moncrieff, R W., 391, 398, 403, 406, 41 3, 41 9, 434 Monroe, S , 469, 483 Moore, A L , 147 Moore, J A , 434 Moore, J C , 214, 227, 228 Moore, S T , 454 Moore, W R , 213,227 Moranglio, G , 20, 99, 299 Morawetz, H , 154, 159, 184. 197, 205, 208, 219, 223, 227, 266, 298 Morgan, P W , 30, 32, 48 Morneau, G A , 387, 406 Morrls, J D , 391, 406
1
4
I
$
AUTHOR INDEX
549
Mort, J ,252,256,429,430,434 Morton, M , 91, 96, 99, 122, 124, 513, 516, 518,520, 521 Moynihan, R E , 257 Mraz, R G ,452, 454 Mukherjee, A R , 80 Mulvaney, J E , 99 Munzer, M , 134, 147 Myers, R J , 405 Nass, L I , 398, 406 National Research Council, 354, 357 Natta, G , 10, 20, 91, 96, 99, 261, 262, 266, 298, 299 Nauman, E B , 136, 137, 147 Naylor, M A , 60, 80 Netting D I , 481 Neuse, E W , 43 1 , 434 Neville, K , 453 Newing, M J , 164, I84 Newman, S , 184, 483 Newton, R A , 419, 434 Nicholson, A E , 74, 80 Nielsen, L E , 307, 320, 323, 326, 328, 340, 343, 349, 357 Nissan, A H , 426, 434 Norrish, R G W , 60, 79, 80 North, A M , 51, 52, 60, 80, 109, 123, 124 Noshay, A , 122, 125 Novak, R W , 405 Noyes, R M , 5 1, 80 Ober, R , 22 3, 327 O'Brien, J C ,461, 483 Odian, G ,40, 46, 48. 64, 68, 71, 75, 80, 85, 88,91, 96,97, 99, 110, 114, 120, 125, 134, 144, 147 O'Driscoll, K F , 5 1 , 79, 80 Ofstead, E A , 372, 382 Ohi, H , 372, 382 Ohm, R F ,430, 434 Okaya, T , 147 Olabisi, 0 , 154, 176, 184, 471, 483 Olah, G A , 82, 99 Olby, R , ix Olive S., 98 Olmstead, B A , 461, 483 Oppenheimer, L E , 227 011, W J C , 183 Osgan, M ,299 Oster, G , 50, 80 Ostwald, W , 3, 20 Ott, E ,426, 434
Ouano, A. C , 217, 227 Overberger, C G , 91, 99, 427, 434 Owen, M T , 451, 454, 475,483 Pae, K D , 328 Page, S L , 419, 434 Palit, S R , 63, 80 Panchak, J R , 78 Paolino, R R 475, 483 Parker, E E ,445, 454 Parry, A , 9 1, 99 Partridge, H de V , 504 Paschke, E , 368, 382 Pasquale, J A 111, 469, 483 Passagtia, E , 298 Pasternak, R A , 191, 227 Patchornik, A , 147 Patrick, C R , 47, 77 Patterson, D , 168,184 PatLon, I C ,452, 454 ' Paul, D R , 176, 184,471, 483 Pauling, L , 16, 20, 272, 299 Pavelich, W A , 406 Payne, A R , 380, 521 Peacock, D G , 481 Pearce, E M , 337, 357 Pecora, R , 2 2 3 Pedersen. Y. 0 . 228 Peebles, L H , 18, 20, 71, 8 U Pen& F M , 391, 406 Pengilly, B W , 419, 434 Penn, J Y , 381 Pennings, A J , 179, 184, 290, 299 Penzenstadler, R J , 406 Pepper, D C , 120,125 Peppin, A , 387, 406 Peffer, J R , 454 Pe~Ibexg,S E ,465,469, 483 Persak, K J ,419, 434 Person, W J ,372,382 Perutz, M F , 297 Peterlin, A , 278, 281, 283,292, 293, i 298, 299 Peters, E N , 43 1, 434 Peters, T V , 491, 497, 505 Peterson, H L ,47 1 , 484 Petro, A J , 482 Pettigrew, C , 483 Phillips, J. C , 16, 20 P i c k e ~ g G ?? , 482 , Picot, C , 182, 223, 327 Pierce, P E , 307, 328 Piggott, M R , 47 1, 484
AUTHOR INDEX
Pigott, K A , 419, 434, 448, 454 Piirma, I , 134, 147 Pike, E R , 225 Pimental, G. C , 16, 20 Pimer, S H , 144, 147 Pino, P ,20, 99, 299 Pittman, C U , 432 Platt, A E , 387, 404, 406 Platzer, N , 354, 357, 387, 406 Plesch, P H , 88, 99, 100 Poland, D ,273,299 Porter, K , 501, 505 Portel, M , 380, 521 Po~ter, S , 177, 184 R Potter, B J , 227 Power, G E ,469, 484 Preston, J , 48, 413, 430, 434, 497, 505 Preston, S S nI, 405 Price, C C , 118, 123, 125, 266, 299 Price, F P ,289, 299 Prigogine, I , 166, 184 Prime, R B ,279, 299 Pritchrud, J G , 395, 406 Ptitsyn, 0 B , 182 Pundsack, F L , 432 Putscher, R E ,413, 434 Quinn, F A , Jr , 287, 298, 299 Rabeck, J F , 256 Radosta, J A , 387, 406 Raff, R A V., 368, 369, 372, 382 Rainville, D , 46 1, 484 Ramazotti, D J , 46 1, 484 mnby, B., 238, 256 Randall, M , 184 Rawe, A , 32, 48 Rayleigh, Lord, 155, 184 Raymer, J F ,445, 454 Read, B , 328 Rebenfeld, L , 369, 382, 391, 406, 413, 419, 426,431, 434 Recchia, F P., 431, 434 Reding, F. P , 338, 357 Reich, L , 96, 100, 245, 256 Reising, T , 475, 484 Reitano, P A , 413, 434 Remp, P ,223, 226 Reneke~, H , 275,276, 299 D Repka, B C , 100 Rexroad, H N ,237, 256 Rhys, J A , 255 Rich, D P , 426, 434 Richards, R B , 254
Richards, R W , 204,227 Richardson, M J , 245, 256, 287,299 Richardson, P N ,461,463,464,469, 484 Richardson, R J , 435 Rigby, R B ,419, 434 Riley, J L ,492,493,494, 505 Riley, M W , 354, 357 Ringsdorf, H , 128, 147 Ritchie, P D , 252, 256 Roberts, D E , 287, 298, 299 Roberts, W J , 491, 505 Robertson, A B ,403, 406 Robertson, R E , 481 Robeson, L M , 184 Robinson, H W , 376, 382 Rocad, Y , 2 2 5 Rodgers, C E , 80 Roe, R J , 70, 79, 80, 227 Roettger, L G , 406 Roig, A ,22 7 Rosen, S L ,307,320,328,461,464,468,469, 484 Roth, P B , 454 Rouse, B P , JI , 433 Rowlinson, J S , 183 Royal, C A , 419, 434 Rubin, I D , 372, 376, 382 Rubin, I I , 461, 484 Rudd, J F , 341, 357 Russell, K E , 88, 100 Saegusa, T , 98 St Cyr, D R , 373, 382 Salatiello, P P , 433 Saltman, W M , 376, 382 Saltzman, M , 253 Salva, M , 448, 454 Sanchez, I , 289, 300 Sander, M , 431, 434 Sanders, J R , 435 Sare, E J ,452, 454 Sarma, G , 182 Sarvetnick, H A , 398, 406 Saue~, A , 320, 328, 357 J Saunders, J H ,413,419, 435, 448, 454 Saunders, K 5 , 2 2 9 , 2 5 6 Savage, A B ,426, 435 Sawada, H , 75, 8 0 Scales, R E , 419, 435 Scardino, F L , 504 Schaefer, J , 238, 256 Schaer, L S ,469, 484 Schallamarch, A , 380, 521 Schatz, R H , 380
1$
d
!
a9
1 !
AUTHOR INDEX
Scheraga, H A ,299 Schildknecht, C E , 91, 100, 122, 125, 128, 134,147, 263, 300, 430, 435, 452, 454 Schille~, M , 99 A. Schindler, A , I00 Schlesinger, W , 273, 300 Schmieder, W , 183 Schrnitz, J V , 246,252, 256 Schnabel, W , 144, 147 Schneider, F W , 461, 484 Schneider, N S , 179, 180, 184 Schnell, H ,419, 435 Schoff, C K , 328 Scholte, T G , 208, 228 Schrafft, F , 469, 484 Schenk, W J , 481 Schule, E C , 413, 435 Schultz, J , 281, 289, 300, 326, 328 Schulz, D N , 120, 125 Schulz, G V , 39, 40, 48, 60, 80 Schulz, R C , 265,266, 300 Schwenker, R F , JI ,245, 256 Scott, R L , I83 sea no^, D A., 252,256, 431, 435 Sears, J K ,475, 482, 484 Sebastian, D H , 146 Sedgwick, R D ,230,256 Seeds, W E , 297 Segal, L ,223, 432 Ser ad, C A ,426, 435 i Serfaty, I W , 431, 433, 435 Seymour, R B , 11,20, 429,430, 435 Shapiro, P , 405 Sharpless, A , 289, 300 Shaw, M T , 184 I Sheats, J E , 432 Sheldon, R P , 471, 484 Shen,M, 311, 328 Sherman, S., 448, 454 Sherrington, D. C , 99 Shimoda, K ,204, 228 Short, J N , 368, 369, 382 Shultz, A R , 148, 171, 172, I84 Siclari, F , 497, 505 Siebert, L R, 182 Siesler, H W., 238, 256 Silver, R P., 454 Simha, R , 101, 125, 144, 147 Siow, K S , 169, 184 Sjostrand, F. S , 242, 256 Skeist, I., 11 1, 125, 454 Skotchdopole, R E , 469, 484 Slade, P E , Jr , 18, 20, 245, 256 Slonimsky, G L , 356 Small, P A , 16, 21 Smiley, L. H , 481 Smith, A L ,497,505 Smith, C P , 435 Smith, C W , 243,256 Smith, T L , 311, 328 Smith, W B ,227 Smith, W V , 130, I47 Smothers, W J ,245, 256 Snell, T P , 368,382 Sneller, J ,461,464, 484 Snidex, 0 E ,413, 435 Snyder, L., 154, 184 Solomon, D H , 32, 33, 35, 40, 48, 475, 484 Son, P N , 482 Sorenson, W R , 32, 48, 55, 80 Southorn, W A , 380, 521 Speiser, R , 225 Sperati, C A , 342, 344, 346, 357, 399, 403, 406 Sperling, I, H , 315, 329, 471, 483, 484 Sprague, B S , 433 Sprung, M V , 454 Spurlin, H M , 434 Sroog, C E , 430, 435 Stabin, J V., 194, 228 S t a g , R , 518, 521 Staley, H B , 453 Stanonis, D J , 43 3, 4 35 Starkweather, H W , 233, 257, 287, 300, 333, 336, 357, 406 Statton, W 0 , 284, 292, 300 Staudinger, H , 9, 10, 11,2I, 49,80, 101,123, 125,208,228,261,265, 300, 362,382,415, 435 Stayner, V ,475, 484 Steele, R , 501, 505 Stejskal, E 0 , 256 Stening, 1 C , 430, 435 Sternstein, S S , 320, 329, 343, 357, 434 Stuart, A , 255 Stevens, M P , 230, 257 Stewart, W E , 146 Stockrnayer, W H , 110, 125, 173, 182, 185, 190,212, 213, 227, 228 Stokes, A R , 297 Stone, F G A , 431, 435 Stoy, W S , 521 Strauss, U P , 223 Stross, F H , 224, 225 Studebaker, M L , 518,520, 521 Sudol, R S , 235,257 Suh, K. W , 469, 484 Sullo, N A , 431, 435
AUTHOR INDEX
Summers, J W , 481 Svedberg T., 208, 228 Sven, T J . , 4.54 Sweeney, W., 337, .3.57, 413, 4.35 Swett, R. M , 387, 406 Swift, P Mc L , ,380, 521 Szabolcs, I . , 300 Szabolcs, O., 295, 300 Szwarc, M., 90, 91, 98, 100, 121, 125 Tadmor, 2,136,146,147,457,461,464,469, 484 Tadokoro, H ,266, 273, 300 Tager, A , 168, 185, 320, 323, 326, 329 Taif P J. T , 95, 100 Takakura, K , 147 Tapper, M , 475, 484 Tate, D P , 125 Taylor, L , 143, 148 Taylor, R P , 78 Taylor, W J , 157, 185 Teach, W C , 431, 435 Teal, K ,469, 484 Tedder, J M , 56, 80 Temin, S C , 144, 148 Teroso, G C , 505 Thomas, A G , 380, 521 Thomas, P E , 346, 357 Thomas, W M , 391, 406 Thornrnes, G ,354, 357 Thompson, D C , 381 Thornton, P R , 242, 257 Tiers, G V D , 253 Till, P H., Jr , 273, 300 Tippers, C F H , 78, 123 Tippetts, E A , 504 Tobolsky, A V , 48, 67, 78, 80, 31 3, 314, 320, 325, 326, 329 Todd, Lord, v Toporowski, P M , 356 Torkelson, A , 453 Tomquisf E , 381 Touchette, N W , 484 Tower, J M , 481 Towler, D W , 403, 406 Traskos, R T , 184 Tregear, G W , 123 rreloar, L R G , 155, 185, 311, 329 Trommsdorff, E , 60, 80, 147 Tsai, S W , 471, 484 Tsunashima, Y , 227 Tsvetkov, I D , 255 Tucker, H , 373, 382 Iudos, F , 99
Tulley, F T , 46 1 , 485 Tung, L H ,70,80, 180, 185, 217, 228 Turbak, A , 426, 434, 435 Turi, E A , 245, 257 Turnbull, D , 327 Turro, N J , 442, 454 Udipi, K , 98, 381 Ueberreiter, K , 153, 181, 185 Ulbricht, J , 65, 81 Ulrich, R D , 197, 228 Ultee, A J , 433 Updegraff, I H ,442, 454 Valko, E I , 50 1 , 505 Van de Hulst, H C , ?04, 228 Vandenberg E J , 9 6 , 100, 419, 435 Vanderhoff, J W , 134, 148 Van Wazer, J R , 304, 307, 320, 329 Venezky, D L ,43 1 , 435 Vial, T M , 391, 406 Voevodskii, V V , 255 Volkenstein, M V , 159, 185 Vollmerf B , 26, 32, 48, 66, 68, 81, 159, 185 Vredenburgh, W A , 381 Vrentas, J S , 404 Wake, W C , 229, 257 Waksmunski, F A , 483 Walker, E E , 16, 21, 356 Wall, F T , 101, 125, 226 Wall, L A , 230, 257, 403, 406 Wallace, T C , 406 Walling, C , 68,71,75, 81, 108, 116, 124, 125 Wallis, B L , 469, 485 Walter, E R , 357 Walters, K , 307, 329 Ward, I M ,233,257,320,323,326,329,343, 349, 357 Wartmen, L H , 405 Watkins, W D , 466, 485 Watson, J D , 264, 300 Weale, K E , 75, 81 Webber, A C , 250, 257 Webber, I G , 475, 485 ' Weinle, P L , 433 Welch, F 5 , 357 Welgos, R J , 41 3, 435 Wenzler, R A , 405 Werner, k C , 469, 485 Wesslau, H , 70, 81 Wessling R , 398, 406 West, A C ,403, 406 West, J C , 379, 382
AUTHOR I?+UfX
Westerman, L . 225 Westover, R F , 464, 485 Wetton, R E.. 254 Whitby, G Z 20, 48, 434 White, D N .d l 9, 435 White, E F T . 355 White, J L 3V7, 329 Whorlow, R a , 304, 307, 329 Wicker, T R J r , 405 Widrner, G . 4 2 , 454 Wilchinsky, Z W , 295, 300 Wilkins, 297 Williams, G 328 Williams, 3 -A', 208, 228 Williams, M L , 3 12, 322, 329 Williams, R J P , 182 Wlllis, H A 254, 255 Wilson, G I.. 100 Wilson, H k-, 297 Wilson, J E . 51, 81 \Vinding, C. 6 . 246, 247, 257 Winslow, F K , 19, 147 Witt, D R , 96, 100 Wolfe, J R. J r , 513, 521 Wood, L A -282,283, 300, 340, 357 IVooding, N S ,90, 98, I00
Woodward, A E , 338, 357 Wray, G R , 501, 505 Wright, P ,419, 435 Wunderlich, B , 240, 245,257, 273,279,281, 287,289,298,299, 300, 332,337,356, 357 Yarger, S B , 41 9, 435 Yaq W W ,217,228 Yeadon, G , 225 Yee, W , 79 Young, L J , 53,64, 81, 103, 118, 123, I25 Young, R J , 320, 323, 326, 329 Z a n d R , 50, 81 Zbinden, R , 238,257 Ziabicki, A , 497, 505 Ziegler, K , 91, 100 Ziemba, G P , 404 Zilliox, J G ,223 Zimm, B H , 69,81, 159,185, 193,194,201, 210,213,228 Zimrnerman, J , 338, 357, 433, 435 Zoss, A 0 , 100, 300 Zuko~, J , 482 L Zwolinski, B J , 79
SUBJECT INDEX
Abrasion resistance, elastomers, 348, 371, 375-376, 417-418,513-515, 5 19-520 fibers, 394, 398,409, 414, 491, 502-503 measurement, 248-249 plastics, 353, 36'7, 370, 388, 402, 409-410, 416417,471 ABS resins, 11, 122, 342-343,386-387, 396, 404,470,476,481 Accelerators, 378, 389, 443, 509-512, 518 See also Vulcanization Acetal resins, 415-416, 476 See also Polyoxymethylene Acetate rayon, 423-424, 489, 491-492, 498, 501-502 ;Acetylation, cellulose, 423 polyoxymethylene, 415 Acrylate elastomers, 389, 5 19 Acrylic esters, 83, 389-390, 396 fibers, 390-391, 472, 487-491, 493-496, 499,501,503 polymers, 387-391, 468, 476 Acrylonitrile, 83, 103, 114, 117-1 18, 390 Activation energy: copolymerization, 115-1 16 degree of polymerization, 73, 76, 88 polymerization rate, 72- 74, 76 viscous flow, 303 Activators, 510, 5 18 See also Vulcanization Addition polymerization, 5, 49-75 See also Anionic polymerization; Cationic polymerization; Coordination catalysts; Ionic bond; Radical polymerization
Additives, 471 -475 Adhesion, 394, 469-470, 498, 519--520 Adhesives, acrylate, 389 amino resin, 441 cyanoacrylate, 390 epoxy, 446 phenolic, 439, 468, 471 polyisobutylene, 377 poly(viny1 acetate), 392, 501 poly(viny1 alcohol), 393 Adipic acid, 119-1 20, 408 Adsorption, 9 4 9 5 , 230 After-treatments, fiber, 497-501 Aging, rubbers, 418, 5 17 Alfin polymerization, 96 Alfrey-Price equation, 117--118 Alkali cellulose, 421, 431 Alkyd resins, 30, 137, 442, 451 Ally1 acrylate, 139 diglycol carbonate, 452 resins, 451-452 Alloys, 470 Alternation, 104, 110-1 14, 116-1 19, 122 See also Copolymerization Amino acids, 41 1-412 resins, 440-442, 453, 471, 473 Analysis, chemical, of polymers, 229-230 Angular dissymmetry, 200-204 Anionic polymertzation, 82-85, 88-90, 373, 376, 390,415,467 copolymerization, 119-120 Annealing, 283-285,297, 399 Antioxidants, 375-377, 473-474, 517-518 Antiozonants, 5 17
B
SUBJECT INDEX
Antistatic agents, 499 Approach-tc+equilibriumultracentrifugation, 206 Aramid fibers, 410-411,470, 495, 501, 503 Arc xesistance, 25 1 Aromatic electrophilic-substitution, 32 Axrhenius equation, 72 Artificial weathering, 252 Association, 9 A-stage thermosetting resins, 439, 441, 452, 465 Atactic configuxation, see Tacticity Athe~mal solutions, 161 Autoacceleration, 59-60, 69, 133 Avrami equation, 282 Azeotropic copolymerization, 105-106 Azo-compound initiators, 50, 73, 76, 388, 509 Bag molding 444 Bakelite, 10 Batch polyrne~ization,reactors, 127, 135-136, 144-145 Bead model, 159 polymerization, see Suspensionpolymerization Bicomponenf biconstituent fibers, 492 Binodal curve, 17 1 Birefringence, 279-280, 284, 294-295 Bisphenol A, 31, 417, 445 "Black orlon," 427 Blends, polymer; 175-176,386,396,414,428, 470,477,481,500 Block copolymers, 90, 101,104,120-122, 138, 340, 347,369,379,385,469 methods, 425 Blowing, 397,415,423,446, 462-463, 467 Blow molding, ,366,401,413, 415, 459-460, 475 Boiling-point elevation, see Ebulliometry Boltzmam superposition principle, 3 12, 324 Bond energy, length, 13 Bonds, chemical, 11-14 hydrogen, 14, 153, 269, 271, 273, 335, 337-338, 347,420,499 Boron nitride fiber, 430-431 Brabender Plastograph, 304 Branching 3, 5, 10, 152, 389, 392, 399, 501 in chain polymerization, 70 detection, by infrared spectroscopy, 231 by solution viscosity, 213, 220, 222 in diene and divinyl polymers, 138-140 and fiactionation, 179 and the glass transition, ,340 and GPC, 215-216 and melt viscosity, 340-341 molecular size and, 159, 213
and molecular-weight distribution, 4046,70 multichain polymers, 39-40 in polyethylene, 363-364, 380 in poly(viny1 acetate) and poly(viny1 alcohol), 392 in SBR, 374-375 side-chain, and ciystalline melting point, 335 in stepwise polymerization, 4 & 4 6 by transfer to polymer; 53, 61, 122 Brittleness, 321, 323, 325, 344-346, 348, 384-386, 392,472 See also Impact polystyxene, resistance, strength Brittle point, temperature, 248, 5 19-520 strength, 345-346 B-stage thermosetting resins, 439,441,452,46 1 Bulk fractionation, 177-1 78 polymerization, 127-128, 144, 384,388, ,392 Buna rubber, 374, 506 Butadiene, 75, 88, 103, 117-118, 123, 129, 138, 373 Butylrubber, 11, 141,266,376-378,508,514, 516.519 Cage effect, 5 1 Calender, calendexing, 400,464-467,474,501 Calorimeter, calorimeby, 242-243 Capillary viscometer, 209, 304-305 Caprolactam, 47, 97,409, 467 Carbanion polymerization, see Anionic polymerization Carbanions, 82-85, 88-90, 119 Carbenium-ion polymerization, see Cationic polymerization Carbeniurn ions, 82-88, 119 Caxbon black, 374, 410,474, 5 13-5 16, 5 18, 521 fibers, 472, 503 see also Reinforcement Carbon-13 NMR, 234-236,263 Carbonyl addition-elimination, 27 addition-substitution, 3 1 Casting 231, 400,409,446,465467,474 acrylic resins, 387-388, 465 cellulosic plastics, 424, 465-466 Catalysts: for anionic polymerization, 82, 452 for cationic polymerization, 82, 86, 376,446 for condensation, 35,437,447,450,452,500 coordinatlon, 82, 91-96, 128 Frtedel-Crafts, 32, 85, 376 supported metal oxide, 96 Z~egler-Natta, 1-94 9 Cationic polymerization, 82-88, 376, 446 copolymerizat~on,1 19-1 20
SUBJECT INDEX Ceiling temperature, 72, 386 Cellophane, 10, 194,421 -423,465-466 Celluloid, 10 Cellulose, 28, 347, 419-426, 431, 441, 471 -472,490,495,499 acetate, 10, 196, 431 fiber, 423-424,486,489-490,495-496, 499,502 plastic, 424, 468, 476 acetate-butyrate, acetate-propionate, 424 alkali, 421, 431 bacterial, 194 crystal structure, 267, 269 degradation, 141 esters, 423-426, 473 ethers, 425-426 history, 10 nitrate, 10, 194, 196, 394,424-425,436 osmotic membranes, 194- 195 polymers derived from, 423 -426 propionate, 424 reactions, 137-1 38 regenerated, 194,420-423. See also Viscose single crystals, 273 solubility of derivatives, 347 triacetate, 423, 502 xanthate, 421-422,431,497 Centrifugation, see Ultracentrifugation, ultracentrifuge Ceramics, 353-354 Chain branching: entanglement, 303 expansion factor, 158, 165, 180 folding, 274-278,283,291-293,296 flexibility, 154-157, 303, 333-336 polymerization, 5,25,49-74 See also Anionic polymerization; Cationic polymerization; Coordination catalysts; Ionic' bond; Radical polymerization stiffness, 333-336, 339 transfer, activation energy, 73 agent, 61-62,64 control of molecular weight by, 63 in coordination polymerization, 93, 95 degradative, 117, 142 in diene-divinyl polymerization, 139, 507, 512 to initiator, 61-63 in ionic polymerization, 83, 86-90 to monomer, 61-63 to polymer, 122, 363, 389 in radical polymerization, 53,6144, 75 to solvent, 61-63 see also Branching Characte~istic ratio, 157- 158 Charge stor age, 25 1 Charge-transfer complex, 110, 116 Chemical analysis, 229 -230 properties, measurement, 25 1-252 reactions of polymers, 136-144, 229-230,386, 392-394 block and graft copolymerization, 120-122 crosslinking, seecrosslinking Vulcanization degradation, 141-143, 209 with ionizing radiation, 143-144 of specific polymers, 385, 388, 392-394 resistance of polymers, 364-366, 369-371, 385, 398, 401-404, 412, 414, 439, 446,476-480,490,502-503 shifts, NMR, 234-236 Chlor.inated polyethylene, 267, 396 poly(viny1 chloride), 397 Chlorine-containing polymers, 371, 378- 379, 389, 395-398, 402-403 Chloroprene, 138, 378 Chlo~osulfonated polyethylene, 371, 380, 5 19 Chlorotrifluoroethylene, 402 Chromatography, gas, 230 gel permeation, 214-217 Clarification of polymer solutions, 203 Clarity of plastics, 385,388,394,441,476-480 Cloud point, 344 Coacervation, 170 Coating, 365, 367, 401, 443, 446, 467 Coatings, 365, 367, 385, 389, 392,394,403, 416.-417,422, 424,428, 439, 441-442 Cocatalyst, 82, 85-88 Coextrusion, 462,470, 481 Cohesive energy (density), 14- 16, 152-153, 161, 337,347, 349-350 Cold drawing, forming, 290-293, 297, 325, 350, 387 rubber, 374-375 Colligative properties, 160, 187-1 97, 218 Collision frequency factox, 33, 72-73 Colloid, colloidal state, 8-9 Color, 249-250,385--386,388,441-442,444, 464 Colorants, 397,439,473-474 See also Dyeing Column elution, fractionation, see Elution, fractionation Combination, see Termination Comfort, fabric, 420, 490 Complexes, catalyst-cocatalyst, 86-88 cha~ge-transfer,110, 116 Compliance, elastic, 3 18-3 19
'
558
SUBJECT INDEX
Composite structures, 351, 353,443,470-471 Compounding, plastics, 471-475 rubbers, 5 16-518 Compression molding, 231,367,401,403,424, 428,441,458,461,475 testing, 246 Concentrates, 472, 474 Condensation polymerization, 5 , 2 5 4 8 polymers, 2 W 5 , 407-431, 436442, 448-452 polymerization conditions, 127 reactions of, 136-1 37 see also Stepreaction polymerization Condis crystals, 273 Conditioners, fiber, 497 Conductive polymers, 428-429 Conductivity, electrical, 25 1 Configurations, 6, 10, 154,261-266, 295-296 with asymmetric carbon atoms, 261-266 in chain polymerization, 54-55 CIS vs trans, 264 determination by NMR, 233-237 in diene polymers, 264 and optical activity, 264-266 see also Tacticity Conformations, 6 , 2 6 1 of dissolved polymers, 154-1 59, 180 branched, 159 fully-extended, 155-156, 262-263 random coil, flight, 154-159 helical, 270-273 of solid polymers, 204 trans vs gauche, 157 and x-ray diffraction, 238-240 Conjugate fibers, 492 Contact electrification, 25 1 molding, 444 resins, see Polyester, resins Continuous filament, 391, 412, 487 polymerization, 127, 362 Contour length, 155, 180 Coordinate bond, 12 Coordination catalysts, 91-96 polymerization, 82, 91-96, 168, 376 Alfin, 96 copolymerization, 119 of ethylene, 91, 94, 128, 366 history, 91 of isoprene, 9 1, 373 kinetics, 94-96 mechanism, 92-94 supported metal oxide catalysts, 96, 366 Ziegler-Natta catalysts, 91-94, 97 polymers, 430
Copolyamides, 366367, 410 Copolymer equation, 102, 104, 106-108, 110-111,122, 139 Copolymerization, 101-122, 472, 507 Alfrey-PI ice equation, 1 17-1 18 alternating, 104, 110-1 14, 116-1 17 azeotropic, 105-106, 112 block, 104, 120-121 complex participation, 110, 116-1 17 copolymer equation, 102, 104, 106-108, 110-111, 122, 139 cross termination, 108-109 depropagation, 109 of diene, divinyl compounds, 138- 140 evaluation of reactivity ratios, 106-107 graft, 120-122 history, 101 ideal, 103-105, 111-112, 119 ionic, 119-1 20 kinetics, 101-1 10 mechanism, 114-120 monomer reactivity, 114-1 15 multicomponent, 109 penultimate effect, 109 polarity, effect of, 117-1 18 and polymer properties, 347-348 Q-e scheme, 117-1 18 radical reactivity in, 116 rate of, 107-109 reactivity ratios, 103-107, 117-119, 123, 252 remote-unit effect, 109 resonance stabilization in, 116 stepreaction, 119-1 20 steric effects in, 116 termination in, 107-109 Copolymers, 101-1 22 alternating, 104, 110-1 14, 116-1 19, 122 attaining homogeneity in, 112, 114 azeotropic, 105, 106, 112 block, 101, 120-122 composition of, 104-107, 110-1 14 glass transition temperatures, 339-340 graft, 101, 120-122 ideal 103-105, 110-112, 119 instantaneous composition of, 104-107 light scattering of, 204 mechanical properties of, 347-348 melting points of, 336-337 and glass transition temperatures, 338 sequence distribution, 110- 11 1,237 solubility, 348 see also speczJcpolymers In the form poly(monomer I co monomer 2)
1
1
SUBlECT INDEX
559
Cor~elation holes, 303 Cor~espondingstate theories, 165-1 66, 176, 181 Cotton, 420421,442,472,480,486487,489, 491492,498,500,503 Coumarone resins, 471 Counterions, 83-90, 119, 218 Coupling, 32, 138, 428 See also Termination Covalent bond, 12 Crazing, 252, 343, 385, 388 Crease resistance, retention, 350, 410, 414, 423, 442,490,497,500 Creep, 314, 316, 327, 370, 423, 439 Crepe rubber, 372 Crimp, 412,488 Critical miscibility temperature, see Lower critical solution temperature; Upper critical solution temperature Crosslinked polymers, 5-6, 151,367-368,386, 440-441 Crosslinking, 138-141, 144-145, 151, 324, 340-341, 348, 350-351, 353, 368, 371, 375, 378, 380, 399, 404, 412, 426,436453,470,500-501, 507-513, 517--518 Seealso Vulcanization Cross termination, 108-109 C r y ~ ~ ~ o160,187-188,192--193,195,197, py, 220 Crystal defects, 240, 274, 287, 339 structures, 238--240,263,266 -273,295-296 Crystalline melting, 281-288 Crystalline melting point: depression by solvent, 152, 286 determination of, 240, 252, 284-285, 287-288 effect of structure on, 33 1-337,354,404,426 and glass transition temperature, 32@-322, 337-338 melting range near, 284-285 property changes at, 284-285 solubility and, 152 see also spec$c polymers Crystalline polymers: fractionation, 179 mechanical properties, 323-326, 343-344 solubility, 152, 344 see also specfie polymers Crystallinity: degree of, 286-289 determination of, 232-233,240,252, 286-289 and polymer properties, 31 1, 343-347, 350, 353
structural requirements for, 266-267 see also Larnellae, lamellar structure; Paracrystals; Single crystals; Sphemlites Crystallizability, 266--267, 3 37- 338 Crystallization, 8, 281-290, 322, 325-326, 426, 494,5 14 Crystals, single, 241 -242 C-stage thermosetting resins, 439, 452 Cure, curing, 378, 403, 418, 443-444, 446, 448,450,452, 507-513 See also Vulcanization Cyanoacrylate adhesives, 390 Cyclopolymerization, 139-140 Dashpot, 3 16-3 18 Dead-end polymerization, 67 Debye equation, 199 Defects, crystal, 133, 240, 271-274,276,278, 287,292, 339 Degradation, 141-143, 145, 229-230, 385, 388, 393, 396, 399,409410,420, 423,513,517-518 Degree of crystallinity, 232-233,240, 252, 286-289, 342-344, 399,420 of o~ientation,293-294 of polymerization, 35, 38-40,44,46,47,61, 73-76, 87-88, 95, 110, 130-131, 139-141, 33 1 See also Molecularweight Denier, 487,494,501 Density, 281-282, 284-285, 296, 342 gradient, sedimentation in, 206 of polymers, 362, 364, 366, 399, 400, 410, 415,421,447,476-480,502-503 Deoxyribonucleic acid, 264-265 Depolymerization, 74, 141-142, 145 Depropagation, 109, 141- 142 Diallyl phthalate resins, 452 Dichroism, 232, 295 Dielectric: constant, 83, 86, 88, 250, 344, 347, 400-401,476,480 loss, 250, 321, 347,400- 401,476480 properties of polymers, 344-345, 347, 351, 364, 385,400-402,410,426,439, 444,446,473,488,515,519-520 strength, 251 Diene monomers, 83, 371 polymers, 138-140, 252 Dienes, polymerization of, 96 ~ieth~lene glycol bisallyl carbonate, 452 Diethyl fumarate, 118
SUBJECT INDEX
Differential scanning calorimetry, 242-243, 253, 286-287 thermal analysis, 244, 253, 285 Diffusion, autoacceleration, 5 9 4 0 constant, 204,207, 252 control, in radical polymerization, 109 in emulsion polymerization, 145 through osmotic membranes, 195, 197 permeability a n 4 252, 344, 347 sedimentation and, 207 in solution process, 151, 153 Diffusivity, 344 Dilatancy, 302-303 Dilatometer; 55 Dilution, entropy, free energy, and heat of; 160-168 Dip coating, dipping, 400, 450, 467 Dipole bonds, forces, 12-1 3, 336, ,347 moment, 295 Diradicals, 77 Disorder, crystalline, 240, 274, 276 Dispersion forces, 13, 3 36 polymerization, 1.32-133 spinning, 491, 496497 Disproportionation, see Termination Dissipation factor; see Dielectric, loss Dissymmetry, 201 Divinyl benzene, 138 polymerization, polymers, 138-140, 145 D A 264-265 N , Double-bond addition, 27, 3 1 carbon-carbon, 49 Double helix, 264-265 resonance, NMR, 234,236 Drape, 488,490 Drawing 290-293,297, 325, 350,398,400, 408,488,493494,496,501 Drying resins, 351 Dry spinning, 391, 408,491-492, 495 DSC, 242-243,253,286-287 DTA, 244,253,285 Dyeing 412413,420421,423, 473474, 490,498499 Dynamic light scatte~ing, 204 mechanical properties, 345, 378-379, 516, 5 19-520 measurement, 247-248, 3 14-315 viscoelastic, 3 11-3 14 Ebonite, 5 12 Ebulliometry, 160,187-188, 195, 197 Economics of' polymer uses, 352-354 Efficiency, initiator, 5 1, 58 plasticizer, 472
Efflux time, 208 Elastic compliance, 3 18-3 19 polyamides, 364 Elasticity, 301, 309, 398, 400, 408, 412, 417, 488, 502-503 rubberlike, 301,307-311,376,378,417,506 Elastomer, ideal, 308 Elastomers, 506-521 molecular-structure requirements for, 302, 326, 349-350,507 property requirements f o ~ 307, 326, , 349-350,507 see also Rubbers, synthetic, acrylic Electrical properties, see Dielectric, properties of polymers ' testing, 250-251 Electrochemical initiation, 51 Electron diffraction, 241, 274 microprobe, 241 microscopy, 241-242,274, 276,279-281, 291-292,515 paramagnetic resonance, 132--13 3, 143, 237-238 Electronic properties, 25 1 Electrophilic substitution, aromatic, 32 Elongation, 246,291, 307-3 11, 342, 345, 371-372, 398,400,412,422,470, 476-480,488,500,502-503, 512-513,519-520 Elution, fractionation, 177-178 in gel permeation chromatography, 2 14-216 Emulsion polymerization, 67, 128-1 32, 144-145,374-375, 378,392,507, 517 End-group analysis, 9, 186-1 87, 197 End-to-end distance, 155, 180 Energy, see Free energy Engineering plastics, 352-354, 403, 407-4 30, 470 Enthalpy, see Heat Entropy: of elasticity, 308 of fusion, 286-287,296, 331-333, 336 of long-range inter actions, 165 of mixing 152, 160-168 of polymerization, 71-72, 115 Environmentally degradable polymers, 143 Environmental stress cracking, 252, 364, 367, 369, 370 EPDM rubbers, 371, 380, 519 Epichlorohydrin, 3 1, 445-446,5 19 Epitaxial crystallization, 289 Epoxy resins, 11, 31,445-446,453,460,466, 471,480
i
i
1
i
SUBJECT' INDEX
EPR, 132-133, 143,237-238

Equation of state, 166 Equilibrium: esterification, 423-424 phase, 168- 176 polymerization-depolymerization, 142 72, stepr eaction polymerization, 33 see also Ultracentrifugation, ultracentrifuge Ester inter change, 27, 4 14 Esthetic factors, fabric, 496 Ethyl acrylate rubbers, 389 cellulose, 426, 431 Ethylene, 83, 362 glycol dimethacrylate, 137-1 38 a-olefm copolymers, 370 Excluded volume, 158, 164- 165 Expansion coefficient, 320, 322 factor, chain, 158, 165 Extended-chain crystals, 279, 285, 297 Extent of reaction, 34-35,46, 68-69, 139 Extruders, extrusion, 383, 387-388, 395, 400-403,409,416,418,422, 424-425,461-464,467,474-475, 481,494 Fabrication: elastomers, 374, 378, 417, 507-518 fibers, 350, 390-391, 412, 421423, 491 -497 plastics, 365, 367-368, 395-397, 40C-401, 409,415,422,425,439,443-444, 447448,453, 4 5 7 4 8 1 Fabric properties, 490-491 Fabrics: nonwoven, 497,500,504 properties, 490-491 textured, 492 Fatigue tests, 247-248 Felting, felts, 414, 490 Fiber, 486-.504 aftertreatments, 497-501, 504 definition, 486,501 finishing, 499-500 molecular -structure requirements for, 350-35 1 photomicrographs, 492 prices and production volumes, 408, 41 3, 420,424,487 processing, 49 1-501 properties, 350-351,487, 501-503 spinning, 408409,412,486,491497 structure, 292-293, 297, 3% 35 1 textile terms, 487-491 Fibers:
acrylic, 390-391, 472, 487-49 1, 495-496, 499, 501,503 ararnid, 410-411,470,495, 501,503 boron nitride, 430, 472 carbon, 472, 503 cellulose, 420.422, 490, 495496, 499 acetate, 423-424, 442, 4 8 U 8 7 , 489492,495, 498-499,501-502 chlorine-containing, 397--398 cotton, 420-421,442,487,489,491-492, 498,500,503 fluorocarbon, 401,496-497,501,503 graphite, 427 modacr ylic, 396-397 nylon, 407-411, 472, 486-491, 495, 499, 501-502 poly(ethy1ene terephthalate), 413-414, 472, 487492,495, 499-502 polypropylene, 369, 487, 501, 503 polyurethane, 417 poly(viny1 alcohol), 393-394, 472 rayon, see Fibers, cellulose, acetate; Fibers, viscose regenerated cellulose, see Fibers, viscose saran, 397 silk, 412413,489490, 503 triacetate, 423, 502 vinal, 393-394 vinyon, 397 viscose, 421 -422,43 1,442, 487, 489-490, 495--496,498,501- 502 wool, 411412,487,489-491,498,503 Fibrillar crystallization, 289-290 Fibrils, 291-293 Filament: polypropylene, 369 winding, 444,470-47 1 Fillers: elastomers, 371, 379, 513-516 plastics, 397, 400-401, 410, 439,441443, 448,463,47&472,481 see also Reinforcement Film, 401,430,461-466 blown, 365, 370, 462-463 cask 465-466 cellophane, 421-423 cellulose acetate, acetate-butyrate, 424 extruded, 462,475 multilayer, 462 polyethylene, 365, 367 poly(ethy1ene terephthalate), 413 4 15 polyimide, 428 polypropylene, 369 poly(viny1 fluoride), 402403
SUBJECT INDEX
Film (Contznued): saran, 397 Finishes, 499-500 See also Coatings First-order kinetics, 59 Flame retardants, 385, 474 Flammability, 385-386, 41 1-412, 418, 425426,439,442,473474, 476-480 testing, 249 Flash, 458 Flex life, 364, 371, 391, 414, 422 measuremenf 247 structure and, 345-346 see also Dynamic light scattering, mechanical pr operties Flor y-Huggins theory, 162-176, 181, 190 Flor y-Krighaum theory, 1 6 4 165, 18 1, 190, 212 Flory temperature, see Theta solvent, temper ature Flow, 301-307,424,426,472,488,509 instabilities, 306 Fluid-bed polymerization, 91, 96, 128 Fluor ine-containing polymers, 398403, 472, 476,519 Fluorine-19 NMR, 264 Fluor ocar bon plastics, 398-402 Fluoroplastics, 398, 402-403 Foaming, 467 of solutions, 192 Foams, 383,387,416,446-448,453,467 Folded-chain structure, 274-278, 285, 296 Fold testing, 247 Formaldehyde, 415, 437 Forming, 401,457469 Fourier-transform infrared spectroscopy, 231 N R 234-235 M, Fractionation, 173-175, 177--181, 195 Free energy, 15 of long-range inter actions, 165 of mixing, 152, 160-168 of polymerization, 71-72, 75 of solution, 152, 160-168 of stretching elastomers, 308 of surfaces of crystals, 285 Freely-jointed chain, 155-157 Free radicals, volume, 165-166, 169,322,339 theories of polymer solutions, 165-166 see also Radicals Freezing-point depression, see Cryoscopy Frequency factor, 72-74 Friction, frictional coefficient, 207, 364, 367, 400-401, 404,46 1-462 Friedel-Crafts catalyst, 32, 85, 376
Fringed micelle, 276-,278, 292, 296 Fully-extended chain, 155-156,218,262-263, 267-269, 322 Functional-group analysis, 186-1 87, 229 Functionality, 26, 39,4146, 140 Furane resins, 452 Fusion, see Crystalline melting poinc Entropy, of' fusion; Heat, of fusion Gamma distribution, 70 Gas chromatography, 230 permeability, 252 phase polymerization, 128 Gases, light scattering, 198 Gaussian distribution, 110, 156-157, 164, 200, 206, 308-309, 3 11 Gel, gel point, gelatin, 40-47, 139-141, 144145,443,496 cellophane, 194-1 95 effect, 59-60 filtration, permeation chxomatography, 214217,223,253 Glass, 214,216, 35 3, 394-395,410,441-443. 467,470-471 transition, transition temperature, glassy state, 7,243-244,252,307-308,312, 320-323, 337-340,349-350, 384, 389,414,472473,480, 501 and free volume, 322 measurement of, 321 molecular interpretation, 321-323 and polymer properties, 337-340 and rubber elasticity, 308 Gloss, 250, 368 GPC, 214-217,223,253 GI adient elution, 177-1 78 Graft copolymers, 101, 120-122, 138, 144, 340, 347, 386, 390,469,516 Granular polyme~ization,see Suspension polymerization Graphite fibers, 427, 470-471 GR-S, see SBR Gum rubber, 513,521 Gutta percha, 273,287,296 Hammett equation, 118 Hand, 412,414,488, 490,492,499, 504 Hardness, 321,345, 347-348, 364, 366, 368, 388,408,417,441-442,450,512, 515 testing 248-249 Hard resins, 45 1 rubber, 5 12 Haze, 250
I1
[
1
i
SUBJECT INDEX
Head-to head, tail-to-tail configuration, 55, 253,264 Head-tetail configuration, 10, 55, 86, 253 Heat: buildup, 248, 375-376, 378, 515 capacity, content, 145,284, 321 deflection (distortion) temperature, tesf 249, 385, 388,476-480 of dilution, 152, 160-168 and elongation, elastomers, 308 of fusion, 242,252, 286287,296, 331-333, 336 of long-range interactions, 165 of mixing, 152, 1 6 6 1 6 8 of polymerization, 7 1-72, 115-1 16, 144-145 resistance, stability, elastomers, 371, 375, 378, 389,403, 418,450, 506, 519-520 fibers, 350, 4 1 W 1 4 , 420, 490 plastics, 244-245, 367, 386, 396, 400-403,415,418, 426427,430, 439,441,471 of vaporization, 19 1 Helical conformations, sbuctures, 239-240, 264-265,270-273,412 Helix-coil transition, 27 3 Hetercyclic polymers, 427-429 Heterochain polymers, 5, 407-430 Heterogeneous polymerization, 127-1 33 See also Coordination catalysts; Emulsion polymerization; Suspension polymerization Hevea bras~lzenszs, Hevea rubber, see Rubber, natural : Hexamethylene diamine, 409 High density polyethylene, see Polyethylene, linear , High-molecular-weight polyethylene, 36 7 High-pressure polyethylene, see Polyethylene, branched High-temperature polymers, 426-429 Hindered rotation, 33 3-335 See also Conformations HMW polyethylene, 367 Homochain polymers, 5, 36 1-403 Homogeneous polymerization, 127-128 Homologous series, melting points, 331-333 Hooke's law, Hookeanelasticity, 309,3 15-31 6, 318 Huggins' equation, 209, 220 Hydrocarbon resins, 471 Hydrodynamic volume, 21 2 , 2 16 Hydrogen bonds, 14, 153,269,271,273,335,
337-338, 347, 420, 499 Hydropesoxides, 50, 362, 3'74, 443 Ideal copolymerization, 103-105, 111-1 12, 119 elastomer, 308 solution, 160 Impact polystyrene, 386, 404, 470, 479, 481 resistance, strength, 246, 321, 325,345--346, 364, 366-369, 371,380, 386388, 396, 400, 410, 415, 417, 422,424, 426,441,466,471,476-480 Incompatibility, 175-1 76 Indene resins, 47 1 Induction force, 13 period, 54 Infrared spectroscopy, 187,236233,252,295, 363-364, 366 Inhe~ent viscosity, 208 53-54,64-65,75,77,443 Inhibition, inhibito~s, Initiation, initiator, 5, 443 anionic polymerization, 83-84, 89-90, 506 cage effecf 5 1 cationic polymerization, 83-84 coordination polymerization, 94 copolymerization, 107 degradation, 141 efficiency, 5 1, 58, 107 fragments, end-group analysis for, 187 radical polymerization, 50-51, 56, 73,76, 507 rate, 56 thermal, 58-59 vulcanization, 5 11 Injection molding, 364, 366-367, 369, 383, 385, 388, 401,403, 410, 415, 418, 424,442,444,458-459,461,475 Inorganic fibers, 353 polymers, 430 Insertion polymerization, see Coordination catalysts, polymerization Inter action paameter, polymer-solvent, 162--172, 190, 472-473 Interchange, 27, 39,46, 41 3-414 Interfacial condensation, 30 Interference microscopy, 240 Intermolecular forces, 12-16, 339, 347, 350-351,420, 472 Internal friction, 314-315, 321, 378 plasticization, 472 Interpenetrating polymer networks, 470, 481 Intrinsic viscosity, 179, 208 Inverse emulsion polymerization, 131 Ion-exchange resins, 386
SUBJECTINDEX
Ionic bond, 11 polymerization, 82-90, 408 copolymerization, 119-1 20 kinetics, 84-85 vs. radical polymerization, 83 Ring-opening polymerization simila~itiesand contrasts in, 82-85 see also Anionic polymerization; Cationic polymerization; Coordination catalysts Ionization, 2 18 Ionizing radiation, 122 Ionomers, 4, 11, 370, 380, 477 Ion pair, 83-90, 97 Irradiation, 122 I~regular solutions, 16 1 Isobutylene, 75, 83, 1 17, 376 Isome~ism,isomers, 233, 237 Isoprene, 373 pIsopropeny1 styrene, 138 Isotactic configuration, polymers, 91, 93, 261--263. See also Tacticity Kelvin (Voight) element, 3 16-3 18 270-273, 412 Keratin structu~e, Kevlar, 41 1, 503 Kinetic chain length, 60-63 Kinetics: of' anionic polymerization, 84-85, 90 of cationic polymerization, 84-85, 87-88 of coordination polymerization, 94-96 of'copolymerization, 101-1 10 of crystallization, 281-283 of' degradation, 141-142 of' emulsion polymerization, 130-1 3 1, 144-145 of radical polyme~ization,56-68 of' rubber elasticity, 301, 307-3 11, 326, 5 16 of stepreaction polymerization, 33-35 Knitting, 497 Kraemer equation, 21 0 Lactams, lactones, 84, 97 Ladde~ polymers, 35.3, 426--427, 47 1 Lamellae, lamellar structure, 274-283, 291.-293, 296 Laminates, laminating resins, lamination, 365, 395,443444,446,468470 Langrnuir isotherm, 95 Latex, 372, 376, 378, 467, 517, 519 paints, 385, 392 Lattice defects, 240, 276, 278, 283-284 model, 161-162, 303 Layer lines, 239
Leathery cure, 5 12 Lewis acids, bases, 85, 11.7, 446 Lifetime, radical, 66-67 Light microscopy, 241 scattering, 198--204, 218-22 1, 345 Limiting viscosity number, 208 Linear-low-density polyethylene, 370,380,478 Linear polyethylene, polyme~;3 . See also Polyethylene, linear Liquids, light scattering in, 198 "Living" polymers, 70, 85, 90, 97, 121 LLD polyethylene, 370, 380, 478 Logarithmic normal distribution, 70 viscosity number; 208 Long-chain branching, 340-341, 363-364 Loss, dielectric, see Dielectric, loss; Dielectric, proper.ties of polymers Low-density polyethylene, see Polyethylene, branched Lower critical solution temperature, 169 Low-pressure laminating resins, see Polyester, resins molding, 468-469 polyethylene, see Polyethylene, linear Lubricants, lubrication, 410, 439, 448, 473, 498499,504 Luster, 408, 412, 420,490, 492,495, 499 Macr omolecular concept, 9, 3 30 "Magic angle" NMR, 236 Maleic anhydride, copolymers of, 218 Mass specbometry, 130 Masterbatching, 374,517-518, 521 Mastication, 121, 372, 374, 517,521 Matched-die molding, 444 Maxwell element, 3 16-3 18 Mechanical properties: of' amorphous polymers, 3 11-323 323-327, 41 8, 426 of' crystalline polyme~.~, of' fibers, 4 1 4 , 4 8 8 4 9 0 measurement, 246-249 of' polymers, 351, 384, 388, 426, 446, 469470,473 of' rubbe~s,307-310,507-520 and structure, 341-344, 347-348 Mechanism of' polymerization: anionic, 89-90 cationic, 86--87 coordination, 92-94 copolymerization, 114-1 20 emulsion, 128-1.30 radical, 49-55 stepreaction, 25-32 vulcanization, 5 10-5 12
SUBJECT INDEX
Melamine, 440 size, 154-159, 212-213, 218 -aldehyde polymels, 3 1, 440-442 weight, 16-1 8, 186-223 -formaldehyde resins, 440-442, 500 in chain polymerization, 60-63 Melt flow, 301-307, 400, 404 comparison of methods, 197 ~ndex, 304-305 control of, 38-39, 46-47, 63 spinning, 408-409, 414, 49 1-495 cor~elation with solution viscosity, tensile properties of, 341, 399 210-213 viscoelastic properties of, 341, 364 with melt viscosity, 305-306, 326 viscosity, 301-307,331-337, 340-342, 399, definition, 4 401402,473,494,518 in degradation, 141-143 Melting, 15,281-288,457469,486,502-503 determination, by colligative prope~ties, See also Crystalline melting point 187-197 Membrane osmometry, 188--190, 193-197, by cryoscopy, 191-1 92 218-221 by ebulliometry, 191 Mercerization, 420 by end-group analysis, 186- 187 Mesomerism, see Resonance by GPC, 214-217 Metal chelate polymers, 430 by light scattering, 198-204 oxide catalysts, 96, 510 by melt viscosity, 305-306, 326, 399 Metallic bond, 12 by membrane osmometry, 193-197 Metals, 353-354, 471 by solution viscosity, 210-212, 21 9, Methacrylate polymers, 323, 387-389 222, 399 83, Methacrylic este~s, 389 by ultracentrifugation, 205-207 Methyl acrylate, 65, 68, 117 by vapor-phase osmometry, 191 methacrylate, 55, 65, 68, 103, 114, in diene-divinyl polymerization, 139-140 117--118, 123, 133, 387-388 early estimates, 8, 10 Methyl01 reaction, 437438, 440 and the glass transition, 339 Micelle, 129--130, 144 in ionic polymerization, 85-86, 88 Microscopy, 241-242, 253, 284-285, 412, and polymer properties, 345-346 420,492,5 15 polymerization, 6 0 6 3 , 68-70 in ~adical Milling, 231,441,474,516-517 and solubility, 151-152, 158-175, Miscibility, 15, 168-176, 181, 469, 472 178-181 Mixing, entropy, free energy, heat of, 160-168 in Step reaction polymerization, 3340, in polymerization reactors, 134-1 37 42-47 Modacrylic fibers, 390, 397 temperature dependence in polymerization, Modulus of elasticity, 246, 310 See also 73,88 Stiffness and volatility, 14 f Moisture content, 488 Molecular-weight averages, 17, 19 fibers, 391, 394, 398, 410-414, 420-423, number-, 17, 19, 187--197 490 sig?ificance of weight.. and number-, 190,200 plastics, 365, 369, 392, 394, 400-401,416, viscosity-, 21 1 418,424,426,428,430-431, 439, weight-, 17-19,199-200 441442,448,476-480,499 Z-, 206 regain, 413,488,490, 502-503 Molecular-weight distribution, 5, 10, 214-217, resistance, sensitivity, elastomers, 375, 219 519-520 with autoacceleration, 60 Molar attraction constants, 153-154, 180 in branched polymers, 42-46 mass, 186 cumulative, 178, 180-1 81 Mold, molding, 387, 402-415, 426, 439, 444, determination, by fi.actionation, 171-175, 446447,458-461,465. See also 177-181 Injection molding by GPC, 216-217 Molding resins, 366--367 by melt viscosity, 306 Molecular motion, 233, 236, 252, 283, 296, fiom two aver ages, 204 319, 321-326, 349-350 206207 from ultracent~ifugation, shape, 204 differential, 180-1 8 1
566
SUBJECIT INDEX
Nitrocellulose, 10, 194, 196 NMR, 233-237,253,263-264, 321 Nomex, 410-41 1,503 Nonsulfur vulcanization, 5 11-5 12 Nonwoven fabrics, 497, 500, 504 Novolac, 438, 452 Nuclear magnetic resonance, 233-237, 253, 263-264, 321 Nucleating agent, 193 Nucleation, 280, 282-283 Nucleic acids, 218, 264 Nucleophilic substitution reactions, 3 1 Number-average molecular weight, 17, 187-197,219-220,223 Nylon, 231, 351 Nylons, 407411, 413, 460,467, 472, 486491, 495,499,501 crystal structure, 269 history, 408-409 N-methyl, 335, 355 numbering system, 407-408 price and production volume, 408 see also Polyamides Nylon-6, 1I, 47,269,305, 355,407-41 1,431, 467,502 See also Polycapsolactam Nylon-1 1, 269, 410, 431 Nylon-12, 410 Nylon-66, 11,46, 152-153, 158, 233, 269, 355, 407-411, 431, 477, 501-502 See also Poly(hexamethy1ene adipamide) Nylon-610, 46, 269, 289, 408, 410, 431. See also Poly(hexarnethy1ene sebacamide) Nylon-612, 410 Octet, 11 Oil extension, 374,5 17-51 8,521 resistance, 370, 376, 378-379,418, 519-520 Olefinic groups, infrared spectra, 233 0-Olefms, 83, 91, 96, 101 One-stage phenolic resin, 439 Optically-active polymers, 264-266 Optical properties of polymers, 249-250, 344-345, 351, 385, ,388, ,394 Order-disorder transitions, 271-273, 323-324, 399 Organisols, 397 Orientation, 240,252,291-295,350,399-400, 415,421-422,459,475,488,494 Osmometers, 193-1 94 Osmometry, membrane, 188-1 90, 193-1 97, 219-221
Molecular-weight distribution (Continued): early recognition, 5, 10 gamma, 70 graphical representation, 5, 33-39, 174-1 75 integral, 178-180 of "living" polymers, 70, 90 logarithmic normal, 70 and mixing in polymerization, 136-1 37, 145 most probable, 33-39, 69 of' multichain polymer, 68 of natural rubbex; 372 Poisson, 70, 90 of polyethylene, 363-364, 380 in radical polymerization, 68-70 in stepreaction polymerization, ,3340, 42-46 Zimm, 69 Molecular weights of polymers: cellulose, 421 natural rubber, 372 polyethylene, 195, 197, 220-221 polyisobutylene, 377 Monomers, 4 mode of'polymerization, 83 reactivity in copolymerization, 114-1 15, 119 304 Mooney viscosimete~; Morphology, 241,273-281,289-295 of melt-crystallized polymers, 276-281, 283-284 of single crystals, 273-276 of spherulites, 279-281 str ain-induced, 289-295 and tensile strength, 342 Multichain polymers, 39-40, 68 Multicomponent polymerization, 109 Multilayer films, 470, 481 Multiple-screw extruder, 462 Multipolymer systems, 469-470 "Mutual" recipe, 129, 374 Natural fibers, see Cotton; Silk; Wool polymers, 8 rubber, see Rubber, natural N- carboxyanhydrides,97 Neoprene, 378-379,514, 520 Network elastic, 309-3 11 polymers, 5,6, 40-46 see also Crosslinking Vulcanization Neutron scattering, 159, 204, 303 Newtonian flow, 302-303, 306, 318, 326,341 Nitrile rubber, 375-376, 396, 508, 514, 516-517,520
SUBlECT INDEX
vapor-phase, I88 Osmotic pressure, 160, 163, 167, 187-190, 193-198, 218-221. See also Osmometry, membrane Ostwald-Fenske viscometer, 209 Overcure, 5 12-5 13, 52 1 Oxidation-reduction polymers, 138 Oxidative aging: coupling, 428 elastomers, 371, 375--377, 379, 389, 517-518 plastics, 365, 369, 385, 396, 403, 418 Ozone resistance, 371, 375, 377, 379, 418, 517,519-520 Ozonolysis, 141 Yaracrystals, 240, 278, 287, 292, 337, 515 Parison, 459460 Partial molar quantities, 163- 163 specific volume, 205 Particle light-scattering functions, 200--201 Partition factor, 174 Pearl polymerization, see Suspension polymerization Pentane interference, 157 Penultimateunit effect, 109 Peptide linkage, 231,407,411 Perfluoroalkyl-tetrafluoroethylene copolymer, 401 Permanent press, 498, 500, 504 set, 307, 371, 423, 513, 516, 519-521 Permeability, 195, 197, 252, 344, 347, 364, 366,401402,418,422,519-520 Peroxides, 50,63, 131, 144, 362, 767, 371, 384, 388, 398, 443, 450, 452, 467, I 509, 517 Phase contrast microscopy, 240 diagrams, 169, 171, 173, 176 ePuilibrium separation, 168-1 76, 470 452 Phenol-aldehyde polymers, 29, 43-40, Phenol-formaldehyde polymers, 10, 480 Phenolic resins, 436440,453,471, 473, 480, 509 pPhenylene group, 334,413 Photoinitiation, photopolymerization, 50, 65-67, 73, 76, 141 Photometer, 202-203 Physical testing, 245-252 Piezoelectricity, 25 1 Pigmentation, 473-474, 499 Pilling, 491 Pipe, 364,366-367, 370, 397, 403 Plasticity, 7
Plasticization, plasticizers, 339, 341, 3-14, 348, 365, 395-197,404,410,422.424, 472473,481,499,518 Plasticizing resins, 351 Plastics: molecular structure requirements for, 351-354 prices and production volumes, 361, ,366, 368. 370. 383. 387, 389, 395, 408, 447, 451 processing, 457--48 1 property tables, 475-480 role among materials, 354 Plastisol, 397, 467 Pliability, 364 Poisson distribution, 70, 90 Polarity, 15, 83, 117-1 18, 153, 346347, 351 Polaizability, 13, 25 1 Polarized radiation, 199, 232,240,295 Polyacenaphthalene, 339 Polyacetals, 28 See also Acetal resins Polyacetylene, 25 1,429 Polyacrylamide, 131, 144-145, 390 Polyacrylates, 137, 142, 144, 389 Polylacrylic acid), 60, 131, 218, 389-390 Polylacrylic acid co styrene), 348 Polyacrylonitrile,390-391,427,431,471,489, 495 "black orlon," 427 cohesive energy density, 16, 339 crystal structure, 287 degradation, 142, 404 glass transition, 3 39 polymerization, 72, 132-1 33, 390 solubility parametel, 153 spinning, 495--496 Polylacrylonitrile co butadiene), 122,375-376 Polylacrylonitrile co methyl methacrylate), 142 PolyIacrylonitrile co styrene), 122, 385-386, 396,470 Polylacrylonitrile co vinyl acetate), 390 Polylacrylonitrile co vinyl chloride), 390, 397 Polylacrylonitrile co vinylidene chloride), 390 Polylacrylonitrile co vinyl pynolidone), 390 Polylalkylene glycols), 498 Polyfalkyl methacrylates), 389 Polyamic acid, 428 Polyamide-imide, 428 Polyamides, 28, 30, 290, 407-41 1, 413, 473, 477,499,501-502 aromatic, 11, 41C-411
568
SUBJECT INDEX Poly(decamethy1ene adipate), 305 Poly(dibuty1 furnarate co vinyl acetate), 392 Polydichlorophosphazene, 430 Poly(2,5-dichlorostyrene co styrene), 385 Poly(diethy1 fumarate co isobutylene), 101 Poly(diethy1 fumarate co styrene), 112-1 13 Poly(diethy1 fumarate co vinyl chloride), 396 Poly(diethy1 maleate co vinyl chloride), 396 Pol,y(2,3-dimethyl butadiene), 506 Poly(dimethy1 siloxane), 153, 305, 449-452 Poly(N:N-diphenylacrylamide co styrene), 385 Polydispersity, 200, 204 Poly(diviny1 benzene co styrene), 386, 470 Polyelectrolytes, 21 8, 390 Polyester: fibers, 414, 472, 48.7-492, 495, 499-502 resins. 353,442-445, 453, 463464, 466467,470-471,480 Polyesters, 290, 41 3-416, 45 1, 497 crystal structure, 269 en&group analysis, 187 heats and entropies of fusion, 336 history, 413 hydrolysis, 137 melting points, 332-333, 355 polymerization, 25,27,28, 33-35,413 reactions, 137 unsaturated, 11 Polyetherimide, 428 Polyethers, 445-446, 497 Polyethersulfone, 4 19 Poly(ethy1 acrylate co ethylene), 369-370 Poly(ethy1 acrylate co methyl methacrylate), 389 Poly(ethy1 acrylate co vinyl acetate), 392 Polyethylene, 4, 11, 361-368, 467-468, 472, 474,478,501,503 branched, 11,195,245,253,325,361-366, 478 chain folding, 274-278 characteristic ratio, 158 chlorosulfonated, 371, 380, 5 19 cohesive energy density, 16 crosslinked, 144, 367-368 crystallinity, 286-289, 364 crystal structure, 239-240, 266-268 and properties, 324 degradation, 142-1 43 density, 342-344, 362, 364 drawing, 290-29 1 electrical properties, 347, 351, 365 electronic properties, 25 1 fibrils, 292 glass transition, 323
Polyamides (Contrnued): crystal structures, 267, 269 elastic, 324 end-group analysis, 187 glass transition, 338 heat and entropy of fusion, 332-2 j3 infrared spectra, 289 melting points, 332-333, 336, ZZS. 355 polymerization, 27, 29, 409 single crystals, 273-274 f see also Nylon; speclfic chernrccl names o the polymers Poly(u-aminocaproic acid), see Polycaprolactam Polyanhydrides, 28 Polyarylate, 429 Polyaryletherketone, 429 Polyarylethersulfone, 429 Polybenzimidazole, 429, 431 Poly(bis-p aminocyclohexylmethane dodecanediamide), 410 Polybutadiene, 376, 386, 506, 520 crystal structure, 267, 269 degradation, 143 glass t~ansition,339 GPC, 216 history, 506-507 * polymerization, 72, 88, 91, 122, 131 solubility par meter, 153 stereoregular, 91, 376 tacticity, 263 Poly(butadiene co styrene), 122, 142, 180, 374-375, 379, 385-386 glass transition, 339 history, 374, 507 plastic copolymers, 385-386 polymerization, 129 see also SBR Poly(1-butene), 270-271, 369-370, 477 Poly(1-butene co ethylene), 369 Poly(ptert-butyl styrene), 386 Poly(buty1ene terephthalate), 4 15 Polycaprolactam, 4, 269, 274, 305,407-41 1 Polycarbonate, 11, 417-418, 431, 477 Poly(carbon monoxide co ethylene), 267 Poly(2-chloroethyl vinyl ethe~ ethyl co acrylate), 389 Polychloroprene, 11, 269-270, 378-370 See also Neoprene Polychlorostyrenes, 347, 386 Polychlorotrifluoroethvlene, 195, 402 ~oly(chlorotrifluoroethy1ene co ethylene), 403 Poly(chlorotrifluoroethylene co vinylidene fluoride), 403
SUBJECT INDEX
heat -3 entropy of hsion, 286-287, 296 electronic p~operties,251 md f~ee energy of polyrnenzation, 72, 75 hydrolysis, 137 infrard spectrum, 231, 289, 363-364 infrared spectra, 23 3 irraL&fd, 144 polymerization, 27, 46, 414 line&-, J 1, 366-368,478 quenching, 326 briz&mess, 325 solubility parameter, 153 c h b folding, 274, 284 specific beat, 243 cw;ridjnation polymerization, 91, 366 Poly(2-ethylhexyl acrylate co vinyl acetate), wy-~al structure, 23 9-240 392 end-goup analysis, 187 Polyformaldehyde, see Polyoxymethylene ex~cded-chaincrystals, 279, 285 Poly(forma1dehyde co melamine, urea), see rings3 spherulites, 280-281 Amino acids, resins sinde crystals, 273-276, 284 Poly(formaldehyde co phenol), see Phenolic sohkility, 152, 355 resins stre-is relaxation, 325 Poly(fumaronitxile co styrene), 109, 385 mechanical properties, 324, 364 Polyglycols, 416 Poly(hexafluor opropylene co melt index, 304-305, 342 melting point, 331-333, 335, 355, 362, 365 tetrafluoroethylene), 401, 476 , melt ~ i i m s i t y 340, 364 Poly(hexafluoropropylene co vinylidene molecular motion, 323 fluoride), 403 weight, 195, 197,220-221 Poly(hexamethy1ene adipamide), 27, 233, osmometry, 195, 197 407-41 1 polymmization, 97, 128, 135, 362 characteristic ratio, 158 polymorphism, 268 cohesive energy density, 16 single crystals, 273--276 crystal structule, 269 softening temperature, 345-346, 364 solubility, 152 solubility parameter, 153 parameter, 153 solution viscosity, 21 3 Poly(hexamethy1ene adipamide co specific volume, 288 hexamethylene terephthalamide), stiffness, 343-344, 355, 364 336-337 stress-strain curve, 246 Poly(hexamethy1ene sebacamide), 408,410 structure, 362-364 crystal structure, 269 tinsile strength, 342, 355, 364 degree of crystallinity, 289 ultra-high molecular weight, 367 Poly(hexamethy1ene sebacamide co see alro Polyethylene, branched; hexamethylene terephthalamide), Polyethylene, linear 336-337 Poly(ethy1ene co isobutylene), 369 Polyimide, 11, 245, 427-428, 431, 478 Poly(ethy1ene co rnaleic anhydride), 369 Polyisobutyfene, 4, 266, 295, 376-378 Poly(ethy1ene co methyl 'acr~late),370 cohesive-energy density, 16 Poly(ethy1ene co propylene), 369 copolymers, crystallinity of, 266 Poly(ethy1ene co tetrafluoroethylene), 267,403 crystal structure, 270-271 Poly(ethy1ene co vinyl acetate), 369-370 degradation, 143 Poly(ethy1ene co vinyl alcohol), 267 melt viscosity, 305 Poly(ethy1ene glycol), 9, 499 phase separation, 172 Poly(ethy1ene glycol dimethacrylate co methyl polymerization, 72, 84-86, 97 methacrylate), 138 radiation chemistry, 144 Poly(ethy1ene oxide), 89, 158, 416 solution viscosity, 2 11 Poly(ethy1ene terephthalate), 11, 27, 41 3-415, stress relaxation, 313-314, 31 9, 325 43 1,478,489 see also Butyl rubber calorimetry, 243 Poly(isobuty1ene co isoprene), 266, 376.-378 cohesive-energy density, 16 See also Butyl lubber crystallization, 326 Polyisoprene, 4, 372-373, 380, 506, 520 crystal structure, 269 cohesive-energy density, 16 DTA, 244 269-270 crystal structu~e,
SUBJECT INDEX
Polyisobutylene (Continued): degradation, 143 polymerization, 72, 91, 13 1 stereoregular forms, 264, 373 see also Rubber; natural Polyfcis-1,4-isoprene), see Rubber, natural Poly(trans-1,4-isoprene), see Gutta percha Poly(isoprene co styrene), 379 Poly(laury1 methacrylate), 389 Polfimaleic anhydride co stilbene), 101 Poly(ma1eic anhydride co styrene), 109 Polymer: blends, 175--176 definition, 3 melts, 301-307, 340-341 properties: a determinants, 330-33 1 and structure, :330-354 tables, 475-480, 501-503, 518-520 and utilization, 349-354 see also names ofpolymers reactions, 126, 136-144 Polymeric reagents, 138 Polymerization, 25-144, 465 bulk, 127-128, 144,362, 384, 388, 392, 395 classification of methods, 4-5, 25-26, 127 cycle, 139-1 40 diene-divinyl, 138-140 emulsion, 127-132, 135, 374-375, 378, 389, 392, 395 gas-phase, 128 heterogeneous, 127-1 33, 366 homogeneous, 127-128 monomers and mechanisms, 83 with polymer precipitating, 132-1 33, 390, 395,415 r.eaction engineering 134-136, 144-145 redox, 131-132, 390, 398, 402 solid-phase, 133 solution, 127-128, 144, 362, 389-390, 392, 395 suspension, 132-1 33, 135, 144-145, ,384, 386--388, 392, 395 see also Anionic polymerization; Cationic polymerization; Copolymerization; Radical polymerization; Rate of polymerization; Stepreaction polymerization Polymer-solvent interactions, 15 1-1 81, 472 Polymethacrylates, 137, 142, 389 Poly(methacry1ic acid), 389-390 Polymethacrylonitrile, 88, 133, 390 Poly(methy1 acrylate), 60, 13 1, 390
Poly(methy1 acrylate co vinyl chloride), 122 Poly(rnethy1 cyanoacrylate), 76, 390 Polfimethyl methacrylate), 11, 181, 266, 355, 387-390,452,476 autoacceleration, 60 ceiling temperature, 72 chaacteristic ratio, 158 cohesiveenergy density, 16 crystal structure, 270-271 degradation, 142-143 glass transition, 323, 329 GPC, 216,223 graft copolymer with natural rubber, 347-348 infrared spectra, 23 1 melt viscosity, 305 molecular-weight distribution, 69 NMR spectra, 236,263 polymerization, 52, 72, 127, 131-132, 388 radiation chemistry, 144 solubility parameter, 153 stereoregularity, 263 tacticity, 263 TGA, 245 Poly(methy1methacrylate co styrene), 107-108, 120,216 Poly(4-methyl pentenel), 96,369-370, 380 Poly(cu-methyl styrene), 72, 143, 339, 386 Poly(pmethy1 styrene), 386 Poly(methy1 vinyl ketone), 427 Polymorphism, 268, 370 Polynorbornene, 429 Poly(cu-olefins), 352, 369 end-group analysis, 187 melting points, 335 polymerization, 9 1, 96 radiation chemistry, 144 single crystals, 273 tacticity, 261, 263 Poly(o1efins co SO2), 142 Polyoxymethylene, 11,415416,476 heat and entropy of fusion, 287,296 single crystals, 273 solid-phase polymerization, 133 x-ray diffraction pattern, 294 Polypeptides, 27C273, 411413. See also Silli; Wool Poly@phenylene), 32, 251, 429 Polfiphenylene oxide), 428, 470, 477 Polfim-phenylene ~phthalamide), 410-41 1 Poly(pheny1ene sulfide), 419, 428-429 Poly(ppheny1ene terephthalamide), 41 1 Poly(pheny1 esters), 30 Poly(pheny1 siloxane), 2 16 Polyphosphates, 2 18
SUBJECT INDEX
5 71
Polyphthalocyanines, 430 tacticity, 263, 324, 384 Polypropiolactone, 133 termination, 52 Polypropylene, 295, 355, 368-369, 380, 478, thermal polymerization, 58-59, 73 Polfistyrene co vinyl acetate), 106 487,501,503 chain stiffness, 333 Polfistyrene co vinyl alcohol), 348 characteristic ratio, 158 Polfistyrene co N-vinyl carbazole), 385 cohesive-energy density, 339 Polfistyrene co vinyl thiophene), 111-1 13 coordination polymerization, 91 Polysulfides, 29, 31, 41 8, 421, 514, 520 crystal structure, 270-271 Polysulfonamides, 30 Polysulfones, 41 8, 478 glass transition, 323, 339 GPC, 217 Polytetrafluoroethylene, 11, 152, 355, heat and entropy of fusion, 333 398401,404,476,496-497,501, of polymerization, 72 503 melting point, 33 chain stiffness, 333 melt properties, 341 crystal phase changes, 271-273,296, 399 molecular motion, 323, 369 structure, 271-273, 296 molecular-.weightdistribution, 2 17 degradation, 142-143, 399 NMR spectra, 263 electrical proper-ties, 347, 351, 400 single crystals, 273 EPR, 237-238 solubility parameter, 153 flex life, 345-346 glass transition, 323 stress relaxation, 325 tacticity, 263 heat and entropy of fusion, 287, 296, 333 Polyfpropylene glycol), 397, 499 of polymerization, 72 PolyfGpropylene oxide), 266 infrared spectrum, 23 1 Polypyr~ole, 429 melting point, 333, 399 Polysiloxanes, 29, 340 melt viscosity, 341, 399 Polfisodium styrene sulfonate), 386 molecular motion, 323 Polystyrene, 4,10,11,18Q-181,202,222,266, radiation chemistry, 143 295, 355, 383-387,428,470,477, 153 solubility par amete~, 479 spinning, 401 anionic polymerization, 84-85, 89 TGA, 245 Polfitrimethylene glutarate), 280 cationic polymerization, 86 ceiling temperature, 72 Polyureas, 28, 30, 332, 355 chain transfer; 62-43 Polyurethanes, 28, 30, 138, 195, 332-333, characteristic ratio, 158 351, 3 5 5 , 4 1 6 4 1 7 , 4 3 1 , 4 4 6 4 4 8 , cohesive-energy density, 16 453, 460, 467, 470, 480,497, 503, coUigative properties, 189 520 crystal structure, 270-271 Polfivinyl acetals), 136, 394- 395 degradation, 142-143, 385, 404 Poly(viny1 acetate), 11,267,355,391-392,404 glass transition, 339, 384 amorphous nature, 267 GPC, 214-217 chain branching 340, 392, 404 heat and entropy of' polymerization, 72 cohesive-energy density, 16 infiared spectrum, 23 1 degradation, 142 "living" polymer, 90 melt viscosity, 340-341 melt viscosity, 305 polymerization, 72, 128, 131, 392 orientation in, 295 reactions, 136, 392-393 phase separation, 169, 171-,172 solubility parameter, 153 polymerization, 52, 72, 74, 77, 97, 122, Polfivinyl acetate co vinyl chloride), 101, 123, 131-132, 384 396-397,472 polymer-solvent interaction constant, 164 Polyfvinyl acetate co vinyl stearate), 392 radiation chemistry, 144 Poly(viny1 alcohol), 128, 133, 194-195, 355, reactions, 137 392-394,404,472-473,498 solubility parameter, 153 crystallizability, 267 solution viscosity, 210 crystal structure, 268
572
SUBJECT INDEX
Protein fibers, proteins, 204, 218, 264, 270-273,351,420,473,499 Proton NMR, 234-237 Pseudoplasticity, 302 Pultrusion, 444, 463-464, 470 Pure-gum vulcanizates, 5 13, 5 15 Pyramidal crystals, 275-276 Pyroelectricity, 25 1 Q-e scheme, 117-1 18, 123 Quenching, 326,402 Quiana, 41 1 Radiation: chemical effects of, 143-144 crosslinking, 144, 368, 371 degradation, 143 initiation by, 51 Radical coupling, 32 Radical ions, 84 Radical polymerization, 49-74, 442 activation energy, 72-74, 76 autoacceleration, 5 9-60 batch, 127 bulk, 127-128 configurations in, 54-55 continuous, 127 dead-end, 67 diene-divinyl, 138-140 emulsion, 67, 128-132 end-group analysis, 187 free energy, 71 -72 heat, 71-72 induction period, 54 inhibition, inhibitors, 53-54, 64-65, 75,77 initiation, initiators, 50-51, 56, 73, 76 efficiency, 51, 58 vs ionic polymerization, 83 kinetic chain length, 60-63 kinetics, 56-68 laboratory methods, 55-56 mechanism, 49-55 molecular weight and distribution, 60-63 monomers, 49-50 non-steady state, 65-67 photoinitiation, photopolymerization, 50, 65-67,76, 141 pressure effects, 74 propagation, 52,57-59,68, 72-73 radical lifetime, 66-67 radicals, 50-51, 65-67 See also Radicals rate, 55, 68, 72-74
Poly(viny1 alcohol) (Contznued): fiber properties, 351, 393-394 reactions, 136, 393-394,404 Poly(viny1 arnine), 218 Poly(viny1 butyral), 195, 394-395, 404 Poly(viny1 carbazole), 133 Poly(viny1 chloride), 4, 1I, 295, 395-398,464, 466, 472473,479 cohesive-energy density, 16 crystallinity, degree of, 324 crystal structure, 267-269 degradation, 142-143,404 electrical uses, 35 1 glass transition, 339 GPC, 216 infrared spectra, 23 1 mechanical properties, 324 NMR spectra, 263 polymerization, 72, I31 -1 32, 395 radiation chemistry, 143 ~eactions,137 solubility par meter, 153 tacticity, 263 TGA, 245 Poly(viny1 chloride co vinylidene chloride), 396-397 Poly(viny1 fluoride), 402-403 Poly(viny1 formal), 394 Poly(viny1idene chloride), 132, 143, 241, 339, 479 Poly(viny1idene fluoride), 264, 403, 476 Poly(viny1 isobutyl ether), 263 Poly(4-vinyl pyridine), 218 Poly(viny1 pyr~olidone),133, 390 Poly(viny1 stearate), 133 "Popit" bead model, 159 Postforming 467-468 Power factor, 250-251 Precipitation, 132-1 33 Preform, premix, prepreg, 400,444 Preparative GPC, 215 ultracentrifugation, 207 Pressure, effects in radical polymerization, 74 Prices, see Fiber; Plastics; Rubber Primary bonds, 11-1 2 Processabllity, 374, 385, 388, 400-401, 470, 472,475 Processing, 457-521 Production volumes, see Fiber; Plastics; Rubber Promoters, 5 10 See also Vulcanization Propagation, 52,57-59,68, 72-73, 85-87, 89-90, 95,512, 518 Propylene, 75, 83, 117
SUBJECT INDEX
constants, 5 7, 59, 65-68, 75-77 retardation, retarders, 53-54, 64-65, 75 rotating sector, 65-66 solution, 127-1 28 steady state, 57 -58, 65 suspension, 132-1 33 termination, 52, 57, 60 thermal, 58-59, 73 transfer, 53, 61-64, 75, 139, 507, 512 Radicals, 5 addition to double bonds, 50 cage effect, 5 1 concentration in polymerization, 50, 65-67 detection by EPR, 132-133,237-238 efficiency, 51, 58, 107 generation, 50-51, 56,75, 130 lifetime, 66-67, 75-76 monomers polymerized by, 83 reactions with molecules, 52-53, 61-64 reactivity in copolyme~ization,116 stability, 50 in vulcanization, 509, 5 11-5 12, 518 Radioactive initiators, 187, 399 Radius of gyration, 155, 180, 200, 202-204, 22 1 Raman spectroscopy, 23@-232 Ram-fed injection molding, 459 Random chain, coil, flighf walk, 154-159 180,200-201 Raoult's law, 160-161 Rate constants: anionic polymerization, 90 cationic polymerization, 87-88 ; coordination polymerization, 95 copolymerization, 102-103, 107-109, 117-1 18 See also Reactivity, ratios radical polymerization, 57, 59, 65-67, 72-74 stepreaction polymerization, 33 Rate of polymerization: anionic, 90 cationic, 85, 87 coordination, 95 copolymerization, 102, 107-109 radical, 55, 57-60, 65-68, 72-74 step-r eaction, 33-35 Rayleigh ratio, 199, 221 Rayon, 10,442,472,487,498 acetate, 10 triacetate, 423 viscose, 10, 489492, 495-496, 501-502 Reaction engineering 1 34-136 injection molding, 417, 447478, 460, 475
spinning, 497 Reactivity: and molecular size, 33, 136 of monomers, 114-115, 119 of radicals, 116 ratios, 103-107, 117-119, 123,252 Reactors, polymerization, 134-1 36 Rebound, 307, 5 15 Reciprocating-screw molding machines, 458--459 Redox initiation, 51, 131-1 32, 390, 398, 402, 507 polymers, 138 Reduced viscosity, 208 Reflectance, 249 Refractionation, 177 Refractive increment, 199 index, 198-199,284, 321, 345, 385,400, 476480 gradient, see Specific gravity, refractive increment Regular solutions, 161 Regulator, 63 resins Reinforced plastics, see Polyeste~, Reinforcing agents, see Fille~s Reinforcementof elastomers, 348, 350, 374, 450, 472, 513-516,518,521 of plastics, 353 Relative viscosity, 208 Relaxation time, 312, 318--319, 324 Remote-unit effect, 109 Repeat distance, unit, 4, 239 Reptation, 306 Resilience, 307, 365, 375, 386, 391,412-413, 423, 515 Resin finishes, 490, 500 Resistivity, 25 1, 5 15 Resole, 438,452 Resonance, 115-1 16, 332, 334 Retardation: retarders, 53-54,64-65,75, 83, 377-378, 510-511 time, 3 17-3 19 Retarded elastic, viscoelastic response, 31 7 Reversion, 5 12, 521 Rheological properties, rheology, 301-327,458 Rheometer, 304 Rheopectic, 303, 326 RZM, 4 1 7 , 4 4 7 4 4 8 , 4 6 0 Ringed spherulites, 280-281 Ring formation, 30--3 1, 46, 82 -opening polymerization, 96-97, 121, 146
SUBJECT INDEX
Rocket propellants, 41 8 Roll mills, rolls, 464, 474, 497 Root-mean-square end-to-end distance, 155-157,212-213, 309 Rotating sector, 67 Rotational molding, 460 viscometers, 304 RTV elastomers, 484 Rubber: compounding, 5 16-5 18 elasticity, 301, 307-31 1, 516 gas permeability, 347 masterbatching, 374, 5 17-5 18, 52 1 oil extension, 374, 517-518, 521 oxidative aging, 371, 375, 377, 379, 418, 517,519-520 prices and production volumes, 37 1,
CIS-1,4-polybutadiene,376, 506, 520 cis-1,4-polyisoprene, 372-373, 506, 520 polysulfide, 418, 5 14, 520 polyurethane, 416, 467, 520 SBR, 374-375,506-508,514519,521 silicone, 449-450, 453, 5 14, 520 thermoplastic elastomers, 379 Safety glass, 394--395 Saran, 397 SBR, 143, 374375,506-508,514-519,521 Scaling concepts, 158,167,175-176,181,306 Scanning electron microscopy, 241-242 Scouring, 498, 504 Screw: extruder, 461-462, 481,494 -fed injection molding, 458-459 Secondary bonds, 12-1 4 , s 16 plasticizer, 473 Second-order transition, see Glass, transition, transition temperature, glassy state Second virial coefficient, 163, 167, 190, 202-203,218,220-221 Sector method, 65-67 Sedimentation: constant, coefficient, 207, 221 equilibrium, 205-206 transport, velocity, 207 see Ultracentrifugation, ultracentrifuge Semipermeable membrane, 193-1 95 Semipolar bond, 12 Shear rate: and melt viscosity, 306, 462 and solution viscosity, 210 Shear testing, 246 thickening, thinning, 302-303, 326 Sheeting, 365, 367,425, 464, 466 Sheet-molding compound, 4 4 4 , 4 6 7 4 6 8 , 4 8 0 Shift factor, 312, 314, 322 "Shiskkebab" morphology, 289-290,297 Short-chain branching 363-364 Side chains, see Branching Silicone polymers, 11, 340,430,448-453,480, 500, 514,520 Silk, 28, 270, 412-413, 431, 489-490, 503 Single crystals, 241, 273-276, 296-297 Sintering, 400-401, 497 Size-exclusion chromatography, see Gel, gel point, gelation, filtration, permeation chomatog aphy Sizes, sizing, 389, 393, 498, 504 "Slush" molding, 467 Sm~tkEwart kinetics, 1 30-1 3 1 Smoked sheet rubber, 372
373-374,376,378-379,506-507
properties, 516-520 tables, 5 18-520 see also Rubbers, synthetic radiation chemistry, 144 reinforcement, 348, 350, 374, 450, 472, 513-516,518,521 stress-strain curves, 309-31 1, 508 vulcanization, 138, 140-141, 507-5 13 See also Vulcanization Rubber-modified plastics, 342-343, 507-508, 514-520 Rubber, natural, 10, 194, 372-373 crystallization, 282-283 on stretching, 326 crystal structure, 269-270 graft copolymer with methyl methacrylate, 348 hard, 512 heat and entropy of fusion, 287, 296 infrared spectra, 23 1, 233 polymer-solvent interaction constant, 163-1 64 stress-strain curves, 309--311 volume expansion coefficient, 320 see also Polyisoprene Rubbers, synthetic: acrylic, 389, 5 19 buna, 82, 374,470 butyl, 376-378, 508, 514, 516, 520 chlorosulfonated polyethylene, 37 1, 520 EPDM, 367, 371,520 fluorocarbon, 403,s 20 history, 506-507 neoprene, 378 -379, 514, 520 nitrile, 375-376, 396, 470, 508, 514, 516-517, 520
SUBJEC,TINDEX
Snap, 307 Soap, 129-131, 510 Soda cellulose, 421 Softening temperature, 249, 321, 345-347, 364, 387, 426,457,486 See also Crystalline melting point; Glass, transition, transition temperature, glassy state Soil, 423, 498 Sol, 40,4546 Solid-phase polymerization, 133 Solubility, 15, 151- 154, 180, 252, 345- 347, 413 elastomers, 375 fibers, 390,408,420, 423, 486,495 plastics, 365, 399, 400,402,404,408, 416, 424 parameter, 152-154, 176, 180 Solution SBR, 374 polymerization, 127-128, 144-145, 392 viscosity, 207-21 3 Solvent: effects in polymerization, 83, 86, 97, 119 gradient elution, 177 resistance, 252 elastomers, 375, 377, 417-418, 450 fibers, 409,412,414,420,490, 502-503 plastics, 362-363, 365, 386, 399, 401, 403, 409,442, 448,476-480 see also Chain branching, transfer, acbvation energy Specific gravity: heat, 243,285-287 refractive increment, 199, 221 viscosity, 208 volume, 281-282,285,288,296 see Density Spectroscopy, infrared, 230--233 Spherulites, 240,275,278-283,292,296,342, 345, 351,402 Spin decoupling NMR, 234, 236 Spimeret, 492, 494 Spinning 408-409,412,486,491-497, 501 Spray-up, 444 Spring 316-3 18 Spun-dyed fibers, 499 Stabilization, resonance, 115- 116, 45 1 Stabilizers, 38--39, 46, 132, 364, 369, 396-397,404,410,446,457, 474 Standard specific gravity, 345-346, 399 Staple, 391,414,487, 5 W 5 0 1 Static electricity, 488, 504 Statistics: of chain conformations, 158
of copolymer composition, 110-1 11 of d e g adation, 141-1 42 of step reaction polymerization, 35-46 of vulcanization, 140- 141 Steady state, 57-58, 65, 75, 77, 87, 95, 102, 107, 109 Stegxeaction polymerization, 5, 25-46, 443 branching, branching coefficient, branch point, unit, 40-46 and chain polymerization, 25-26 control of molecular weight, 38-39, 46-47 copolymerization, 119 end group analysis, 187 equilibrium, 3 3 extent of reaction, 34-35, 46 functionality, 26, 29, 41-46 gel, gelation, gel point, 40-46 interchange, 27, 39,46,413-414 interfacial, 30 kinetics, 33-35 linear, 33-40 mechanism, 26-32 moleculax weight and distribution, control, 38-39,46-47,409 linear, 33-39, 46 multichain, 39-40 polyfunctional, 42-46 multichain, 39 -40 nylon, 27, 408-409 poly(ethy1ene terephthalate), 41 3 polyfunctional, 40-46 polymex types in, 26--32 rate constants, 33-35 reactivity, and molecular size, 33, 136 ratios, copolymerization, 119-1 20 ring formation, 30-31,46 statistics, 35-46 Stereochemical contigurations, 236 Stereoregular, stereospecific polymerization, see Coordination polymerization Stereoregularity, 261-266, 335 Steric hindrance, 71, 116, 309-310 Stiffness, 246,321, 333-335, 343-344, 347-348, 350, 353, 364,366-369, 372, 391,410, 412,414-416,420, 442443,450, 471,476480,486, 488,502-503, 507,509,512-515, 519-521 Stined-tank reactors, 135-136, 144-145 Stock, 512-513,521 Stoichiometric balance, 27, 38, 409 Stokes' law, 207 Strain-induced morphology, 289-295 Stress cracking, 252
SUBJECT INDEX
i
Stress cracking (Contrnued): relaxation, 313-314, 316, 327 -strain curves, properties, elastomers, 309-311,326, 378,508, 516, 521 fibers, 414, 488-489, 501 plastics, 246-247, 253 Structu~e index, 5 15 Structure-property relationships, 6-8, 1 4 16, 330-348,364-366, 368-369, 377-378, 386, 399,426427,486, 516-518 Styrenated polyesters, see Polyester, resins Styrene, 65, 68, 71, 83-85, 103, 114, 116, 118, 123, 129, 142, 153, 384, 442443,470 Sulfur vulcanization, 508-5 13, 5 18 Surnmative fractionation, 178 Supercooling, 193, 282-28 3 Superpolymers, 408 Superposition principle, 3 12, 324 Supported metal oxide catalysts, 96 Suspensionpolymerization, 132-133,144,384, 386, 388, 392 Swelling, 15 1, 252, 375, 412, 420, 470 Syndiotactic polyme~s,9 1 See also Tacticity Synthetic fibers, see Fibers rubbers, elastomers, see Rubbers, synthetic Tacticity, 261-263, 338 Tear resistance, strength, 348, 364-365, 378, 415,422,512-515,519-521 testmg, 248 Tenacity, 408-409, 422, 487, 490, 501-503 See also Tensile properties, strength, fibers Tensile properties, strength, 246-247, 341-342 elastome~s, 348, 350, 372, 374-375, 378-379,418,507,509, 512-516, 519-521 fibers, 350, 391, 394, 398, 400, 408-414, 421422,486,495,502-503 plastics, 353, 364, 366-368, 388, 400, 408409,422,441,447,470-471, 476-480 Te~ephthalicacid, 41 3-4 14 Termination: anionic polymerization, 83, 89-90, 97 cationic polymerization, 83, 86-87 coordination polymerization, 95 copolymerization, 107-1 09 diffusion control of, 5 9 4 0 , 109 emulsion polymerization, 145 radical polymerization, 52, 5 7, 61, 68-69, 75-77, 512 Tetrafluoroethylene, 83, 398, 402 Tetramethyl silane, 234 Textile properties, 487-491 Textured fibers, yarns, 492, 500 Texture of polymers, 6, 349 TGA, 244245,427 Thermal analysis, 242-245 -gradient elution, 177-178 initiation, polyme~ization,58-59, 73 properties, 384, 458, 476-480, 502-503, 5 19-520. See also Heat, resistance, stablity, elastome~s property testing 249 Thermistors, 191-192 T hermodynamics: of crystalline melting, 285-286 of phase separation, 168-176 of polymer solutions, 159-1 68 of radical polymerization, 71-74 of rubbe1 elasticity, 308-309 of simple lrquids, 159-16 1 Thermogravimetric analysis, 244245,427 Thermomechanical analysis, 245 Thermoplastic elastomers, 11, 379, 470, 479 Thermoplastics, 6, 36 1-43 1 Thermosetting resins, 5, 428, 436-45 3, 46 1, 465,467,469,475,480 Theta solvenf 159, 186, 21 3 temperature, 159, 165, 169-171, 175, 181, 106,213 Thickening agent, 390, 393 Thixotropic, 303, 306 Tie-temperature equivalence, 3 12--314, 324-325 Tire cord, 41 1, 414, 422 Tires and tubes, 374376, 378-379,417,506, 516,518 TMA, 245 Topology of polymers, 151-152, 339-340 Torsion pendulum, 3 15 Toughness, 324,342-343, 348,353,364,391, 394,401-403,409,413,415,431, 446,502-503,506,515. See also Brittleness; Impact polystyrene, resistance, strength Transfer, molding, 401, 444, 461, 475 See also Chain branching, transfer, activation energy Transitions, see Crystalline melting poinc Glass, transition, transition temperature, glassy state Translucency, 249, 345, 364 Transmittance, 249 Transparency, 249-250,284, 345,364,
S U B J E C T INDEX
577
369-370, 380, 384, 387, 408, 410, 418 Trapped radicals, 132--133 Triboelectricity, 251 Tmmmsdorf effect, 59-60 Tubular..flowreactors, 135-1 36, 144-145, 362 Tungstosilicic acid, 203 Turbidimetric titration, 179 Twin-screw extruder, 462 Two-stage phenolic resin, 439 Ubbelohdc viscometer, 209, 222 UHMW polyethylene, 367 Ultracentrifugation, ultracentrifuge, 205-207, 218-219,221 Ultrahi&molecular weight polyethylene, 367 Ult~avioletlight absorbers, 369, 385, 396, 474 Unit cell, 239-240, 267-273, 295-296 Unperturbed dimensions, 158, 165, 180 Unsatur ation, analysis for, 187 Upper critical solution temperature, 168--170 Newtonian viscosity, 306 Urea-aldehyde polymers, 29, 440-442 Urea-formaldehyde resins, 440-442, 500 Urethane foams, 4 4 7 4 4 8 , 467 See also Polyurethanes Uses of polymers, 349-354 elastomers, 375-376, 378- 379, 418, 513 fibers, 393, 398, 422-423, 501 plastics, 365-367, 369-370, 385, 388-389, 393-395, 397,401,403, 410, 416, 418, 422,424-425, 439, 441-444, 446 Valence electrons, 11 Van der Waals forces, 12-43, 499 Vapor permeability, 252 -phase osmometry, 188, 191, 195, 197, 2 19-220 -pressure lowering, 160, 187- 188, 191 -pressure osmometer, 191 462 Vented ext~uder, Vibration modes, 230-23 1 Vicat test, 249, 345--346 Vinal fibers, 393-394 Vinyl acetate, 55, 65, 68, 103, 114, 116-1 18, 391 chloride, 68, 75, 83, 103, 114, 117-118, 123, 133, 395 esters, 83 ethers, 83 fibers, 397-398 fluoride, 83 groups, analysis for, 187
monomers, 49-50, 442 polymerization, see Radical polymerization resins, 395-397,404,466467,473-474, 4 79 Vinylidene chloride, 83, 114, 117-1 18 Vinyon fibers, 397 Virial coefficients, 163, 190 Viscoelasticity, 301, 31 1-319, 327, 377, 400, 422,488 Viscometers, melt, 304-307, 462 solution, 209 Viscose, 421-422,431 rayon, 10,489-492,495-496,501-502 Viscosity -average molecula~weight, 21 1 improver, 386, 389 melt, 301-307, 331-337, 340-342, 399, 401-402,473,494,5 18 number, ratio, 208 solution, 208-213, 218-220, 222, 399, 472 Voigt element, 3 16-3 18 Volatility, 14-15 Vulcanization, 7,138,140-141,367, 371,374, 378, 389,417,507-513,518,521 Wash and wear, 414, 500 Water repellents, 448, 497, 499, 504 sensitivity, see Moisture content, resistance, sensitivity, elastomers -soluble polymers, 132, 386, 390, 393, 416 Wear resistance, 416, 491 See also Abrasion resistance, elastomers Weathering, 252,353,365,369,375,379,385, 387-388, 391, 394, 396,401-403, 410, 412, 417,424--426, 450, 476--480,502-503,5 19-520 Weaving, 497 Weight-average molecular weight, 17- 18, ' 199-200, 206,220,222-223 Weissenberg effect, 304 Wet spinning, 391,422,491-492,495-497 st~ength, fibers, 408, 420, 422 Wire and cable insulation, 364-367, 379,401, 403, 410,450,464 WLF equation, 307, 312, 322 "Wolf-and-lamb" chemistry, 138 Wool, 2 70, 351,411-412,486487,489491, 498,503 Work content, 15 Wrinkle resistance, 412, 423, 500 Xanthates, xanthation, 412-422, 497 X-ray diffraction, 238-240, 253, 266, 276-278, 288-289,291-294
SUBJECT INDEX
Yarn, textured, 492, 500 Yield point, stress, value, 246, 302-303, 325, 342-345, 364,416 Zaverage molecular size, 201 molecular weight, 206 Zero-entropy-production melting, 285 -strength temperature, 249
Ziegler; Ziegle~Natta catalysts, 91-94, 97, 373 Zimm distribution, 69 -Myerson osmometeq 193-194 plot, 201-203 Zone broadening, 2 15 Zwitte~ions,85, 412

Textbook of Polymer Science

Uploaded by

Copyright:

Available Formats

Textbook of Polymer Science

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Textbook of Polymer Science

Uploaded by

Copyright:

Available Formats

What is discussed in the preface?

What is discussed in the preface?

What are the two major directions of the revision discussed in the preface?

What are the two major directions of the revision discussed in the preface?

TEXTBOOK OF POLYMER SC

FRED W. BILLMEYER, JR.

John Wiley & Sons

Library of Congress Cataloging in Publication Data:

Printed in the United States of' America

Troy, New York January 1984

PREFACE TO THE SECOND EDITION

P E A E 'TO THE SECOND EDITION RFC

~REFAC.'ETO T H E SEC.OND EDITION

PREFACE TO THE SECOND EDITION

Troy, New York September I970

The Science of Large Molecules

Step-Reaction (Condensation) Polymerization

Radical Chain (Addition) Polymerization

PROPERTIES OF COMMERCIAL POLYMERS

TEXTBOOK OF POLYMER SCIENCE

THE SCIENCE OF LARGE MOLECULES

BASIC CONCEPTS OF POLYMER SCIENCE

THE SCIENCE OF LARGE MOLECULES

Polyethylene Poly(viny1 chlodde) Polyisobutylene

Polycaprolactarn (6-nylon) Polyisoprene (natural rubber)

"By convention, the symbol omitted

throughout to represent the benzene ring, double bonds being

BASIC CONCEPTS O F POLYMER SCrlENCE

THE SCIENCE OF LARGE MOLECULES

BASIC C.UNCEPTS OF POLYMER SC1ENC.E

THE SC.IENCE O F LARGE MOLECULES

THE RISE OF MACROMOLECULAR SCIENCE

THE RISE OF MAC:ROMOLECUl.AR SC1ENC.E

THE RISE Of: MACROMOLECULAR SCIENCE

1964 1965 1970 1974

C . MOL.ECUL.ARFORCES AND CHEMICAL. BONDING I N POLYMERS

THE S 1 N . OF LARGE MOL.EC'ULES C E CE

The covalent bond is the predominant bond in polymers

MO~ECULAR FORCES A N D CHEMICAL. BONDING IN POLYMERS

TABLE 1-3. Typical Primary-Bond lengths and Energies

Bond Length Bond

Dissociation Energy - (M/moie)

THE SCIENCE OF LARGE MOL.EC:ULES

M0LEC:ULAR FORCES A N D CHEM1C:AI. B O N D I N G IN POLYMERS

THE SCIENCE OF LARGE MOLECULES

'TABLE 1-4. Cohesive Energy Densities of Linear Polymersa

Cohesive Energy Density (J/cm3)

D. MOLECULAR WEtGHT AND MOLECULAR-WEIGHT'DIS'T'RIBUTION

MOLEC:ULAR WEIGH7 AND MO1.ECULAR-WEIGHT DISTRIBUTION

THE SCIENCE O F LARGE MOLECULES

Typical Ranges of M,/M,, in Synthetic Polymersd

Peebies 1971; Slade 1975; BilIingham 1977

DISCUSSIONQUES'TIONS AND PROBLEMS

and M,, the presence of high- and lowon

THE SCIENCE OF LARGE MOLECUL.ES

STEP-REACTION (CONDENSATION) POLYMERIZATION

A. CLASSIFICATION O F POLYMERS AND POLYMERIZATION MECHANISMS

STEP-REACTION (CONDENSATION) POLYMERIZATION

MECHANISM OF STEPWISE POLYMERIZATION

Types of Condensation Polymers

NH2(CHz)xNHt HOOC(CHz)x,COOH H[-NH(CH2)xNHCO(CHz)x~CO-]yOH H20 HO(CHZ)~OH OCN(CH2)x,CNO + [-O(CHZ)CONH(CHZ)NHCO-]~