Plastics Technology Handbook, Fifth Edition

Plastics Technology Handbook Fifth Edition Plastics Technology Handbook Fifth Edition Manas Chanda CRC Press Taylor & Francis Group 6000 Broken Sound Parkway NW, Suite 300 Boca Raton, FL 33487-2742 © 2018 by Taylor & Francis Group, LLC CRC Press is an imprint of Taylor & Francis Group, an Informa business No claim to original U.S. Government works Printed on acid-free paper International Standard Book Number-13: 978-1-4987-8621-8 (Hardback) This book contains information obtained from authentic and highly regarded sources. Reasonable efforts have been made to publish reliable data and information, but the author and publisher cannot assume responsibility for the validity of all materials or the consequences of their use. The authors and publishers have attempted to trace the copyright holders of all material reproduced in this publication and apologize to copyright holders if permission to publish in this form has not been obtained. If any copyright material has not been acknowledged, please write and let us know so we may rectify in any future reprint. Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced, transmitted, or utilized in any form by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying, microﬁlming, and recording, or in any information storage or retrieval system, without written permission from the publishers. For permission to photocopy or use material electronically from this work, please access www.copyright.com (http://www.copyright .com/) or contact the Copyright Clearance Center, Inc. (CCC), 222 Rosewood Drive, Danvers, MA 01923, 978-750-8400. CCC is a not-for-proﬁt organization that provides licenses and registration for a variety of users. For organizations that have been granted a photocopy license by the CCC, a separate system of payment has been arranged. Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identiﬁcation and explanation without intent to infringe. Library of Congress Cataloging-in-Publication Data Names: Chanda, Manas, 1940- author. Title: Plastics technology handbook / Manas Chanda. Description: Fifth edition. | Boca Raton, FL : CRC Press, Taylor & Francis, 2018. | Series: Plastics engineering series | Includes bibliographical references. Identiﬁers: LCCN 2017035500| ISBN 9781498786218 (hardback : acid-free paper) | ISBN 9781315155876 (ebook) Subjects: LCSH: Plastics. Classiﬁcation: LCC TA455.P5 C46 2018 | DDC 668.4–dc23 LC record available at https://lccn.loc.gov/2017035500 Visit the Taylor & Francis Web site at http://www.taylorandfrancis.com and the CRC Press Web site at http://www.crcpress.com Dedicated to the hallowed memory of my father and mentor Narayan Chanda (1911–2005), a renowned author and a decorated teacher, who showed by example the value of dedication and perseverance in academic pursuits. http://taylorandfrancis.com Contents Preface ............................................................................................................................................. xxvii Author .............................................................................................................................................. xxxi 1 Characteristics of Polymers and Polymerization Processes 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 What Is a Polymer?.................................................................................................................................... 1 Molecular Weight of Polymers................................................................................................................ 3 n ) .............................................................................. 3 1.2.1 Number-Average Molecular Weight (M 1.2.1.1 End-Group Analysis .....................................................................................................3 1.2.1.2 Ebulliometry (Boiling-Point Elevation)....................................................................4 1.2.1.3 Cryoscopy (Freezing-Point Depression) ..................................................................4 1.2.1.4 Membrane Osmometry ................................................................................................4 1.2.1.5 Vapor-Phase Osmometry ............................................................................................5 w ) ............................................................................... 7 1.2.2 Weight-Average Molecular Weight (M 1.2.2.1 Light-Scattering Method ..............................................................................................7 1.2.2.2 Low-Angle Laser Light Scattering (LALLS) ............................................................9 v )............................................................................. 9 1.2.3 Viscosity-Average Molecular Weight (M 1.2.3.1 Dilute Solution Viscometry...................................................................................... 10 1.2.4 Polydispersity Index ...................................................................................................................12 1.2.4.1 Gel Permeation Chromatography........................................................................... 14 Polymerization Reactions........................................................................................................................ 20 1.3.1 Addition or Chain Polymerization .........................................................................................20 1.3.2 Coordination Addition Polymerization .................................................................................24 1.3.3 Step Polymerization....................................................................................................................28 1.3.4 Supramolecular Polymerization ...............................................................................................34 1.3.5 Copolymerization ........................................................................................................................35 Polymerization Processes ........................................................................................................................ 37 1.4.1 Process Characteristics...............................................................................................................37 1.4.1.1 Bulk, Solution, and Suspension Polymerization.................................................. 38 1.4.1.2 Emulsion Polymerization.......................................................................................... 39 1.4.1.3 Microemulsion Polymerization ............................................................................... 43 1.4.2 Industrial Polymerization..........................................................................................................49 1.4.2.1 Heat Removal .............................................................................................................. 49 1.4.2.2 Reactor Agitation........................................................................................................ 50 1.4.2.3 Residence Time ........................................................................................................... 50 1.4.2.4 Industrial Reactors ..................................................................................................... 51 Conﬁgurations of Polymer Molecules.................................................................................................. 58 Conformations of a Polymer Molecule................................................................................................ 58 Polymer Crystallinity ............................................................................................................................... 60 1.7.1 Determinants of Polymer Crystallinity ..................................................................................60 The Amorphous State.............................................................................................................................. 62 Structural Shape of Polymer Molecules............................................................................................... 63 ix x Contents 1.10 Thermal Transitions in Polymers ......................................................................................................... 65 1.10.1 Tg and Tm ....................................................................................................................................65 1.10.2 Regions of Viscoelastic Behavior ...........................................................................................68 1.10.3 Factors Affecting Tg ..................................................................................................................70 1.10.3.1 Chain Flexibility ...................................................................................................... 70 1.10.3.2 Steric Effects.............................................................................................................. 71 1.10.3.3 Conﬁgurational Effects ........................................................................................... 71 1.10.3.4 Effect of Cross-Linking .......................................................................................... 71 1.10.4 Factors Affecting Tm .................................................................................................................71 1.10.4.1 Symmetry .................................................................................................................. 72 1.10.4.2 Intermolecular Bonding......................................................................................... 72 1.10.4.3 Tacticity ..................................................................................................................... 73 1.10.4.4 Branching, Chain Flexibility, and Molecular Weight ..................................... 74 1.10.5 Relation between Tm and Tg ...................................................................................................74 1.11 Designing a Polymer Structure for Improved Properties................................................................ 74 1.12 Cross-Linking of Polymer Chains......................................................................................................... 76 1.12.1 Reactions of Functional Groups.............................................................................................76 1.12.2 Vulcanization..............................................................................................................................78 1.12.3 Radiation Cross-Linking ..........................................................................................................83 1.12.4 Photochemical Cross-Linking.................................................................................................84 1.12.5 Ionic Cross-Linking ..................................................................................................................85 1.13 Solubility Behavior of Polymers ............................................................................................................ 86 1.13.1 Solubility Parameter..................................................................................................................86 1.14 Effects of Corrosives on Polymers ........................................................................................................ 94 1.15 Thermal Stability and Flame Retardation ........................................................................................... 94 1.15.1 Thermal Degradation ...............................................................................................................98 1.15.2 Ablation .....................................................................................................................................102 1.15.3 Flame Retardation ...................................................................................................................103 1.16 Deterioration of Polymers .................................................................................................................... 103 1.16.1 Chemical Deterioration..........................................................................................................104 1.16.2 Degradation by Radiation......................................................................................................106 1.16.3 Microbiological Deterioration...............................................................................................106 1.17 Stabilization of Polymers ...................................................................................................................... 107 1.17.1 Antioxidants and Related Compounds ..............................................................................109 1.17.2 Chemical Structures of Antioxidants ..................................................................................110 1.17.3 Stabilization of Selected Polymers .......................................................................................110 1.17.3.1 Polypropylene.........................................................................................................112 1.17.3.2 Polyethylene............................................................................................................114 1.17.3.3 Polystyrene..............................................................................................................115 1.17.3.4 Acrylonitrile–Butadiene–Styrene Copolymers................................................115 1.17.3.5 Polycarbonate .........................................................................................................115 1.17.3.6 Nylons......................................................................................................................116 1.17.3.7 Thermoplastic Elastomers ...................................................................................116 1.17.3.8 Polyacetal.................................................................................................................117 1.17.3.9 Poly(Vinyl Chloride) ............................................................................................117 1.17.3.10 Rubber......................................................................................................................118 1.18 Metal Deactivators.................................................................................................................................. 119 Contents xi 1.19 Light Stabilizers ....................................................................................................................................... 120 1.19.1 Light Stabilizer Classes ...........................................................................................................121 1.19.1.1 UV Absorbers.........................................................................................................122 1.19.1.2 Quenchers ...............................................................................................................123 1.19.1.3 Hydroperoxide Decomposers .............................................................................124 1.19.1.4 Free-Radical Scavengers.......................................................................................124 1.20 Light Stabilizers for Selected Plastics ................................................................................................. 128 1.20.1 Polypropylene...........................................................................................................................128 1.20.2 Polyethylene..............................................................................................................................128 1.20.3 Styrenic Polymers ....................................................................................................................129 1.20.4 Poly(Vinyl Chloride) ..............................................................................................................130 1.20.5 Polycarbonate ...........................................................................................................................130 1.20.6 Polyacrylates .............................................................................................................................131 1.20.7 Polyacetal...................................................................................................................................131 1.20.8 Polyurethanes ...........................................................................................................................131 1.20.9 Polyamides ................................................................................................................................131 1.21 Diffusion and Permeability................................................................................................................... 131 1.21.1 Diffusion ....................................................................................................................................132 1.21.2 Permeability ..............................................................................................................................132 1.22 Polymer Compounding ......................................................................................................................... 133 1.22.1 Fillers ..........................................................................................................................................135 1.23 Plasticizers ................................................................................................................................................ 137 1.23.1 Phthalic Acid Esters................................................................................................................139 1.23.2 Phosphoric Acid Esters ..........................................................................................................140 1.23.3 Fatty Acid Esters......................................................................................................................140 1.23.4 Polymeric Plasticizers .............................................................................................................141 1.23.5 Miscellaneous Plasticizers......................................................................................................142 1.24 Antistatic Agents..................................................................................................................................... 142 1.24.1 External Antistatic Agents.....................................................................................................143 1.24.2 Internal Antistatic Agents .....................................................................................................144 1.24.3 Chemical Composition of Antistatic Agents.....................................................................144 1.24.3.1 Antistatic Agents Containing Nitrogen............................................................144 1.24.3.2 Antistatic Agents Containing Phosphorus ......................................................146 1.24.3.3 Antistatic Agents Containing Sulfur.................................................................146 1.24.3.4 Betaine-Type Antistatic Agents..........................................................................147 1.24.3.5 Nonionic Antistatic Agents.................................................................................147 1.25 Flame Retardants .................................................................................................................................... 147 1.25.1 Halogen Compounds..............................................................................................................148 1.25.2 Phosphorus Compounds .......................................................................................................149 1.25.3 Halogen–Antimony Synergetic Mixtures...........................................................................149 1.25.4 Intumescent Flame Retardants.............................................................................................150 1.26 Smoke Suppressants ............................................................................................................................... 151 1.27 Colorants .................................................................................................................................................. 151 1.28 Antimicrobials ......................................................................................................................................... 152 1.29 Toxicity of Plastics.................................................................................................................................. 152 1.29.1 Plastic Devices in Pharmacy and Medicine.......................................................................153 1.29.1.1 Packing ....................................................................................................................153 1.29.1.2 Tubings and Blood Bag Assemblies..................................................................153 1.29.1.3 Implants...................................................................................................................154 1.29.1.4 Adhesives.................................................................................................................154 xii Contents 1.29.1.5 Dental Materials ....................................................................................................154 1.29.1.6 Nanomedicines and Drug Delivery...................................................................155 1.29.2 Biodegradable Plastics and Bioplastics ...............................................................................155 1.29.3 Oxo-Biodegradable Plastics ...................................................................................................156 1.29.4 Toxicity of Plastic Combustion Products ..........................................................................157 1.29.5 Toxicity Testing .......................................................................................................................157 References........................................................................................................................................................... 157 2 Fabrication Processes 2.1 2.2 2.3 2.4 2.5 2.6 2.7 Types of Processes ..................................................................................................................................161 Tooling for Plastics Processing............................................................................................................161 2.2.1 Types of Molds.........................................................................................................................162 2.2.2 Types of Dies ............................................................................................................................162 2.2.3 Tool Design ...............................................................................................................................163 Compression Molding ..........................................................................................................................164 2.3.1 Open Flash ...............................................................................................................................165 2.3.2 Fully Positive ............................................................................................................................165 2.3.3 Semipositive .............................................................................................................................165 2.3.4 Process Applicability ...............................................................................................................165 Transfer Molding ...................................................................................................................................166 2.4.1 Ejection of Molding.................................................................................................................168 2.4.2 Heating System.........................................................................................................................168 2.4.3 Types of Presses .......................................................................................................................169 2.4.4 Preheating ..................................................................................................................................169 2.4.5 Preforming.................................................................................................................................170 2.4.6 Flash Removal...........................................................................................................................170 Injection Molding of Thermoplastics.................................................................................................170 2.5.1 Types of Injection Units.........................................................................................................171 2.5.2 Clamping Units ........................................................................................................................172 2.5.3 Molds ..........................................................................................................................................173 2.5.3.1 Mold Designs............................................................................................................173 2.5.3.2 Number of Mold Cavities......................................................................................175 2.5.3.3 Runners......................................................................................................................175 2.5.3.4 Gating.........................................................................................................................176 2.5.3.5 Valve Gates ...............................................................................................................178 2.5.3.6 Venting ......................................................................................................................179 2.5.3.7 Parting Line ..............................................................................................................179 2.5.3.8 Cooling.......................................................................................................................179 2.5.3.9 Ejection ......................................................................................................................180 2.5.3.10 Standard Mold Bases ..............................................................................................180 2.5.4 Structural Foam Injection Molding .....................................................................................180 2.5.5 Co-Injection (Sandwich) Molding .......................................................................................180 2.5.6 Gas-Assisted Injection Molding ...........................................................................................181 Injection Molding of Thermosetting Resins.....................................................................................182 2.6.1 Screw-Injection Molding of Thermosetting Resins..........................................................182 Extrusion...................................................................................................................................................185 2.7.1 Extruder Capacity ....................................................................................................................186 2.7.2 Extruder Design and Operation ...........................................................................................186 2.7.2.1 Typical Screw Construction ..................................................................................186 2.7.2.2 Screw Zones ..............................................................................................................187 Contents 2.8 2.9 2.10 2.11 2.12 2.13 2.14 xiii 2.7.2.3 Motor Drive............................................................................................................ 187 2.7.2.4 Heating..................................................................................................................... 187 2.7.2.5 Screw Design .......................................................................................................... 188 2.7.3 Multiple-Screw Extruders..................................................................................................... 189 2.7.4 Blown-Film Extrusion........................................................................................................... 190 2.7.5 Flat Film or Sheet Extrusion ............................................................................................... 192 2.7.6 Pipe or Tube Extrusion ........................................................................................................ 194 2.7.7 Wire and Cable Coverings................................................................................................... 195 2.7.8 Extrusion Coating .................................................................................................................. 195 2.7.9 Proﬁle Extrusion..................................................................................................................... 196 Blow Molding .........................................................................................................................................196 2.8.1 Extrusion Blow Molding ...................................................................................................... 197 2.8.2 Injection Blow Molding........................................................................................................ 198 2.8.3 Blow Molds.............................................................................................................................. 198 Calendering .............................................................................................................................................199 Spinning of Fibers .................................................................................................................................200 2.10.1 Melt Spinning.........................................................................................................................202 2.10.2 Dry Spinning ..........................................................................................................................202 2.10.3 Wet Spinning..........................................................................................................................202 2.10.4 Cold Drawing of Fibers........................................................................................................204 Electrospinning of Polymer Nanoﬁbers...........................................................................................204 Thermoforming .....................................................................................................................................208 2.12.1 Vacuum Forming ..................................................................................................................208 2.12.2 Pressure Forming...................................................................................................................209 2.12.3 Mechanical Forming .............................................................................................................210 Casting Processes ..................................................................................................................................210 2.13.1 Simple Casting........................................................................................................................210 2.13.2 Plastisol Casting.....................................................................................................................210 2.13.2.1 Dip Casting.......................................................................................................... 212 2.13.2.2 Slush Casting ....................................................................................................... 212 2.13.2.3 Rotational Casting.............................................................................................. 213 Reinforcing Processes...........................................................................................................................213 2.14.1 Molding Methods .................................................................................................................. 214 2.14.1.1 Hand Lay-Up or Contact Molding ................................................................ 214 2.14.1.2 Spray-Up .............................................................................................................. 215 2.14.1.3 Matched Metal Molding ................................................................................... 215 2.14.1.4 Vacuum-Bag Molding ....................................................................................... 216 2.14.1.5 Pressure-Bag Molding ....................................................................................... 216 2.14.1.6 Filament Winding .............................................................................................. 217 2.14.1.7 Pultrusion............................................................................................................. 218 2.14.1.8 Prepreg Molding................................................................................................. 220 2.14.2 Fibrous Reinforcements .......................................................................................................223 2.14.2.1 Glass Fibers.......................................................................................................... 223 2.14.2.2 Graphite/Carbon Fibers, the Beginning ........................................................ 224 2.14.2.3 Manufacture of Graphite (Carbon) Fibers ................................................... 225 2.14.2.4 Graphite/Carbon Fibers and Fabrics ............................................................. 227 2.14.2.5 Graphite/Carbon Fiber-Reinforced Plastics ................................................. 229 2.14.2.6 Manufacture of CFRP Parts............................................................................. 230 2.14.2.7 Applications of CFRP Products ...................................................................... 231 2.14.2.8 Aramid Fibers ..................................................................................................... 234 xiv 2.15 2.16 2.17 2.18 2.19 2.20 2.21 Contents 2.14.2.9 Applications.........................................................................................................235 2.14.2.10 Extended-Chain Polyethylene Fibers .............................................................235 Reaction Injection Molding.................................................................................................................. 237 2.15.1 Machinery ...............................................................................................................................238 2.15.2 Polyurethanes.........................................................................................................................239 2.15.3 Nylons......................................................................................................................................239 Structural Reaction Injection Molding .............................................................................................. 240 2.16.1 Applications............................................................................................................................241 Resin Transfer Molding ........................................................................................................................ 241 Foaming Processes.................................................................................................................................. 242 2.18.1 Rigid Foam Blowing Agents ...............................................................................................244 2.18.2 Polystyrene Foams ................................................................................................................244 2.18.2.1 Extruded Polystyrene Foam.............................................................................244 2.18.2.2 Expandable Polystyrene ....................................................................................245 2.18.2.3 Structural Foams.................................................................................................246 2.18.3 Polyoleﬁn Foams ...................................................................................................................246 2.18.4 Polyurethane Foams .............................................................................................................251 2.18.4.1 Flexible Polyurethane Foams...........................................................................251 2.18.4.2 Rigid and Semirigid Foams..............................................................................254 2.18.5 Foamed Rubber .....................................................................................................................255 2.18.6 Epoxy Resins ..........................................................................................................................255 2.18.7 Urea-Formaldehyde Foams.................................................................................................256 2.18.8 Silicone Foams .......................................................................................................................257 2.18.9 Phenolic Foams .....................................................................................................................258 2.18.10 Poly(Vinyl Chloride) Foams ..............................................................................................258 2.18.11 Special Foams ........................................................................................................................261 Rapid Prototyping/3D Printing ........................................................................................................... 263 Rubber Compounding and Processing Technology ....................................................................... 265 2.20.1 Compounding Ingredients ..................................................................................................265 2.20.1.1 Processing Aids.................................................................................................. 268 2.20.1.2 Fillers.................................................................................................................... 269 2.20.2 Mastication and Mixing ......................................................................................................270 2.20.2.1 Open Mill............................................................................................................ 270 2.20.2.2 Internal Batch Mixers....................................................................................... 271 2.20.3 Reclaimed Rubber .................................................................................................................272 2.20.4 Some Major Rubber Products............................................................................................274 2.20.4.1 Tires.......................................................................................................................274 2.20.4.2 Belting and Hoses...............................................................................................276 2.20.4.3 Cellular Rubber Products .................................................................................278 Miscellaneous Processing Techniques................................................................................................ 278 2.21.1 Coating Processes..................................................................................................................278 2.21.1.1 Fluidized Bed Coating .......................................................................................279 2.21.1.2 Spray Coating......................................................................................................280 2.21.1.3 Electrostatic Spraying ........................................................................................280 2.21.1.4 Smart Coatings....................................................................................................281 2.21.1.5 Electrografted Coatings.....................................................................................285 2.21.2 Powder Molding of Thermoplastics .................................................................................287 2.21.2.1 Static (Sinter) Molding......................................................................................287 2.21.2.2 Rotational Molding ............................................................................................287 2.21.2.3 Centrifugal Casting ............................................................................................288 xv Contents 2.21.3 Adhesive Bonding of Plastics............................................................................................288 2.21.3.1 Solvent Cementing ............................................................................................ 289 2.21.3.2 Adhesive Bonding ............................................................................................. 289 2.21.3.3 Joining of Speciﬁc Plastics............................................................................... 292 2.21.4 Welding..................................................................................................................................294 2.21.4.1 Hot-Gas Welding .............................................................................................. 295 2.21.4.2 Fusion Welding.................................................................................................. 295 2.21.4.3 Friction Welding................................................................................................ 295 2.21.4.4 High-Frequency Welding ................................................................................ 295 2.21.4.5 Ultrasonic Welding........................................................................................... 296 2.21.5 Joining Polymer–Metal Hybrids ......................................................................................296 2.22 Decoration of Plastics.........................................................................................................................299 2.22.1 Painting..................................................................................................................................299 2.22.2 Printing ..................................................................................................................................300 2.22.2.1 Gravure Printing................................................................................................ 300 2.22.2.2 Flexography ........................................................................................................ 300 2.22.2.3 Screen Process Printing.................................................................................... 300 2.22.2.4 Pad Printing........................................................................................................ 301 2.22.2.5 Flex Printing ....................................................................................................... 301 2.22.3 Hot Stamping .......................................................................................................................301 2.22.4 In-Mold Decorating ............................................................................................................302 2.22.5 Embossing .............................................................................................................................302 2.22.6 Electroplating........................................................................................................................303 2.22.7 Vacuum Metallizing............................................................................................................303 References........................................................................................................................................................... 304 3 Plastics Properties and Testing 3.1 3.2 Introduction.............................................................................................................................................307 Mechanical Properties ..........................................................................................................................307 3.2.1 Stress and Strain .......................................................................................................................308 3.2.2 Stress–Strain Behavior.............................................................................................................310 3.2.3 Viscoelastic Behavior of Plastics ...........................................................................................313 3.2.3.1 Modulus and Compliance .....................................................................................313 3.2.4 Stress–Strain–Time Behavior.................................................................................................314 3.2.4.1 The WLF Equations................................................................................................316 3.2.5 Creep Behavior .........................................................................................................................317 3.2.6 Maxwell Model .........................................................................................................................318 3.2.6.1 Stress–Strain Relation .............................................................................................318 3.2.6.2 Generalized Maxwell Model .................................................................................320 3.2.7 Kelvin or Voigt Model ............................................................................................................322 3.2.7.1 Stress–Strain Relation .............................................................................................322 3.2.8 Four-Element Model ...............................................................................................................324 3.2.9 Zener Model ..............................................................................................................................325 3.2.10 Superposition Principle ...........................................................................................................326 3.2.11 Isometric and Isochronous Curves.......................................................................................327 3.2.12 Pseudoelastic Design Method................................................................................................328 3.2.13 Effect of Temperature..............................................................................................................330 3.2.14 Time–Temperature Superposition........................................................................................331 3.2.15 Dynamic Mechanical Properties ...........................................................................................333 xvi Contents 3.2.16 3.3 3.4 Rheological Behavior ..............................................................................................................334 3.2.16.1 Classiﬁcation of Fluid Behavior .........................................................................335 3.2.16.2 Effect of Shear Rate on Viscosity.......................................................................338 3.2.16.3 Effect of Molecular Weight on Viscosity .........................................................338 3.2.16.4 Effect of Temperature on Polymer Viscosity..................................................339 3.2.16.5 Effect of Pressure on Viscosity ...........................................................................340 3.2.16.6 Weissenberg Effects...............................................................................................340 3.2.16.7 Irregular Flow or Melt Fracture.........................................................................340 3.2.17 Measurement of Viscosity .....................................................................................................341 3.2.17.1 Rotational Viscometers ........................................................................................342 3.2.17.2 Capillary Rheometers ...........................................................................................343 3.2.18 Plastics Fractures .....................................................................................................................344 3.2.19 Impact Behavior of Plastics...................................................................................................345 3.2.20 Fatigue of Plastics ....................................................................................................................348 3.2.21 Hardness ....................................................................................................................................351 3.2.22 Indentation Hardness .............................................................................................................351 3.2.22.1 Brinell Hardness Number ...................................................................................351 3.2.22.2 Vickers Hardness Number..................................................................................351 3.2.22.3 Knoop Hardness Number ...................................................................................352 3.2.22.4 Rockwell Hardness Number ...............................................................................352 3.2.22.5 Barcol Hardness.....................................................................................................352 3.2.22.6 Durometer Hardness ............................................................................................353 3.2.23 Rebound Hardness ..................................................................................................................354 3.2.24 Scratch Hardness .....................................................................................................................355 3.2.25 Stress Corrosion Cracking of Polymers .............................................................................356 Reinforced Plastics..................................................................................................................................356 3.3.1 Types of Reinforcement..........................................................................................................356 3.3.2 Types of Matrix ........................................................................................................................357 3.3.3 Analysis of Reinforced Plastics .............................................................................................357 3.3.3.1 Continuous Fibers ...................................................................................................357 3.3.3.2 Discontinuous Fibers ..............................................................................................360 3.3.3.3 Fiber Length Less than lc .......................................................................................362 3.3.3.4 Fiber Length Equal to lc ........................................................................................363 3.3.3.5 Fiber Length Greater than lc ................................................................................363 3.3.4 Deformation Behavior of Fiber-Reinforced Plastic ..........................................................364 3.3.5 Fracture of Fiber-Reinforced Plastics ..................................................................................365 3.3.5.1 Tension ......................................................................................................................365 3.3.5.2 Compression.............................................................................................................365 3.3.5.3 Flexure or Shear.......................................................................................................366 3.3.6 Fatigue Behavior of Reinforced Plastics..............................................................................366 3.3.7 Impact Behavior of Reinforced Plastics ..............................................................................366 Electrical Properties................................................................................................................................366 3.4.1 Dielectric Strength ...................................................................................................................367 3.4.2 Insulation Resistance...............................................................................................................368 3.4.3 Arc Resistance...........................................................................................................................369 3.4.4 Dielectric Constant ..................................................................................................................370 3.4.4.1 Polarization and Dipole Moment ........................................................................372 3.4.4.2 Dielectric Constant versus Frequency ................................................................374 3.4.4.3 Dielectric Constant versus Temperature............................................................374 xvii Contents 3.4.4.4 Dielectric Losses.......................................................................................................375 3.4.4.5 Dielectric Losses of Polar Polymers ....................................................................376 3.5 Optical Properties ...................................................................................................................................377 3.5.1 Optical Clarity ..........................................................................................................................377 3.5.2 Index of Refraction ..................................................................................................................378 3.5.3 Piped Lighting Effect ...............................................................................................................380 3.5.4 Stress-Optical Characteristics ................................................................................................380 3.6 Thermal Properties ................................................................................................................................381 3.6.1 Speciﬁc Heat..............................................................................................................................381 3.6.2 Thermal Expansion .................................................................................................................382 3.6.3 Thermal Conductivity.............................................................................................................382 3.6.4 Transition Temperatures and Temperature Limitations ................................................383 3.6.5 Standard Properties of Plastics .............................................................................................384 3.7 Identiﬁcation of Common Plastics .....................................................................................................384 3.7.1 Behaviors on Heating and Ignition......................................................................................394 3.7.2 Tests for Characteristic Elements.........................................................................................395 3.7.3 Speciﬁc Tests .............................................................................................................................397 3.8 Plastics Analysis by Instrumental Methods ......................................................................................402 3.8.1 IR Spectroscopy ........................................................................................................................402 3.8.1.1 Methods of Measurement......................................................................................403 3.8.1.2 Instruments ...............................................................................................................404 3.8.1.3 Sample Preparation .................................................................................................404 3.8.1.4 Fourier Transform IR Spectroscopy....................................................................406 3.8.1.5 Qualitative Analysis ................................................................................................407 3.8.1.6 Quantitative Analysis..............................................................................................415 3.8.2 NMR Spectroscopy ..................................................................................................................417 3.8.2.1 General Principles ...................................................................................................418 3.8.2.2 Chemical Shift ..........................................................................................................421 3.8.2.3 Shielding Mechanisms............................................................................................423 3.8.2.4 Spin–Spin Coupling ................................................................................................426 3.8.2.5 Applications in Polymer Analysis .......................................................................426 References........................................................................................................................................................... 430 4 Industrial Polymers 4.1 4.2 Introduction .............................................................................................................................................433 Part I: Addition Polymers.....................................................................................................................434 4.2.1 Polyoleﬁns..................................................................................................................................435 4.2.1.1 Polyethylene..............................................................................................................435 4.2.1.2 Polypropylene...........................................................................................................440 4.2.1.3 Polyallomer ...............................................................................................................447 4.2.1.4 Poly(Vinyl Chloride) ..............................................................................................448 4.2.1.5 Poly(Vinylidene Chloride).....................................................................................456 4.2.1.6 Polytetraﬂuoroethylene and Other Fluoropolymers........................................457 4.2.1.7 Polyisobutylene ........................................................................................................461 4.2.1.8 Polystyrene ................................................................................................................462 4.2.1.9 Polybutadiene (Butadiene Rubber)......................................................................463 4.2.1.10 Polyisoprene..............................................................................................................463 4.2.1.11 Polychloroprene .......................................................................................................464 xviii Contents 4.2.2 4.3 Oleﬁn Copolymers...................................................................................................................464 4.2.2.1 Styrene–Butadiene Rubber ....................................................................................464 4.2.2.2 Nitrile Rubber...........................................................................................................465 4.2.2.3 Ethylene–Propylene Elastomer.............................................................................465 4.2.2.4 Butyl Rubber.............................................................................................................467 4.2.2.5 Thermoplastic Elastomers .....................................................................................467 4.2.2.6 Fluoroelastomers......................................................................................................470 4.2.2.7 Styrene–Acrylonitrile Copolymer ........................................................................472 4.2.2.8 Acrylonitrile–Butadiene–Styrene Terpolymer...................................................472 4.2.2.9 Ethylene–Methacrylic Acid Copolymers (Ionomers)......................................474 4.2.3 Acrylics .......................................................................................................................................475 4.2.3.1 Polyacrylonitrile .......................................................................................................475 4.2.3.2 Polyacrylates .............................................................................................................476 4.2.3.3 Polymethacrylates....................................................................................................477 4.2.3.4 Polyacrylamide .........................................................................................................479 4.2.3.5 Poly(Acrylic Acid) and Poly(Methacrylic Acid) ..............................................480 4.2.3.6 Acrylic Adhesives ....................................................................................................481 4.2.4 Vinyl Polymers .........................................................................................................................482 4.2.4.1 Poly(Vinyl Acetate).................................................................................................482 4.2.4.2 Poly(Vinyl Alcohol) ................................................................................................483 4.2.4.3 Poly(Vinyl Acetals) .................................................................................................484 4.2.4.4 Poly(Vinyl Cinnamate) ..........................................................................................485 4.2.4.5 Poly(Vinyl Ethers)...................................................................................................485 4.2.4.6 Poly(Vinyl Pyrrolidone).........................................................................................486 4.2.4.7 Poly(Vinyl Carbazole) ............................................................................................486 Part II: Condensation Polymers ..........................................................................................................486 4.3.1 Polyesters ...................................................................................................................................487 4.3.1.1 Poly(Ethylene Terephthalate) ...............................................................................487 4.3.1.2 Poly(Butylene Terephthalate) ...............................................................................488 4.3.1.3 Poly(Dihydroxymethylcyclohexyl Terephthalate)............................................490 4.3.1.4 Unsaturated Polyesters...........................................................................................491 4.3.1.5 Aromatic Polyesters ................................................................................................497 4.3.1.6 Wholly Aromatic Copolyester..............................................................................499 4.3.1.7 Polycarbonates .........................................................................................................500 4.3.2 Polyamides.................................................................................................................................503 4.3.2.1 Aliphatic Polyamides ..............................................................................................503 4.3.2.2 Aromatic Polyamides..............................................................................................511 4.3.2.3 Polyimides .................................................................................................................513 4.3.3 Formaldehyde Resins ..............................................................................................................519 4.3.3.1 Phenol–Formaldehyde Resins...............................................................................519 4.3.3.2 Urea–Formaldehyde Resins...................................................................................524 4.3.3.3 Melamine–Formaldehyde Resins .........................................................................527 4.3.4 Polyurethanes............................................................................................................................529 4.3.4.1 Polyurethane Rubbers and Spandex Fibers .......................................................529 4.3.4.2 Flexible Polyurethane Foam..................................................................................534 4.3.4.3 Rigid and Semirigid Polyurethane Foams .........................................................535 4.3.4.4 Polyurethane Coatings ...........................................................................................536 4.3.5 Ether Polymers .........................................................................................................................537 4.3.5.1 Polyacetal...................................................................................................................538 4.3.5.2 Poly(Ethylene Oxide)..............................................................................................540 xix Contents 4.3.5.3 Applications ..............................................................................................................541 4.3.5.4 Poly(Propylene Oxide) ...........................................................................................544 4.3.5.5 Epoxy Resins.............................................................................................................545 4.3.5.6 Poly(Phenylene Oxide)...........................................................................................557 4.3.6 Cellulosic Polymers .................................................................................................................559 4.3.6.1 Regenerated Cellulose.............................................................................................560 4.3.6.2 Cellulose Nitrate ......................................................................................................560 4.3.6.3 Cellulose Acetate......................................................................................................561 4.3.6.4 Other Cellulose Esters ............................................................................................562 4.3.6.5 Cellulose Ethers........................................................................................................562 4.3.7 Sulﬁde Polymers.......................................................................................................................563 4.3.7.1 Polysulﬁdes ...............................................................................................................563 4.3.7.2 Poly(Phenylene Sulﬁde) .........................................................................................564 4.3.8 Polysulfones...............................................................................................................................565 4.3.8.1 Properties...................................................................................................................566 4.3.9 Polyether Ketones ....................................................................................................................567 4.3.10 Polybenzimidazole ...................................................................................................................569 4.3.11 Silicones and Other Inorganic Polymers ...........................................................................570 4.3.11.1 Silicones ...................................................................................................................570 4.3.11.2 Polyphosphazenes .................................................................................................576 4.3.11.3 Polythiazyl...............................................................................................................577 4.3.12 Polyblends..................................................................................................................................577 4.3.12.1 Prediction of Polyblend Properties ...................................................................581 4.3.12.2 Selection of Blend Components.........................................................................582 4.3.12.3 Compatibilization of Polymers...........................................................................584 4.3.12.4 Industrial Polyblends............................................................................................586 4.3.12.5 Nanoblends .............................................................................................................586 4.3.13 Interpenetrating Polymer Networks ....................................................................................587 4.3.13.1 Industrial IPNs.......................................................................................................589 References........................................................................................................................................................... 590 5 Polymers in Special Uses 5.1 5.2 5.3 5.4 Introduction .............................................................................................................................................593 High-Temperature and Fire-Resistant Polymers.............................................................................593 5.2.1 Temperature-Resistant Polymers .........................................................................................594 5.2.2 Fire-Resistant Polymers ..........................................................................................................596 Liquid Crystal Polymers........................................................................................................................597 5.3.1 Thermotropic Main-Chain Liquid Crystal Polymers ......................................................603 5.3.2 Side-Chain Liquid Crystal Polymers ...................................................................................605 5.3.3 Chiral Nematic Liquid Crystal Polymers ...........................................................................606 5.3.4 Properties of Commercial LCPs ...........................................................................................609 5.3.5 Applications...............................................................................................................................611 Conductive Polymers .............................................................................................................................611 5.4.1 Filled Polymers .........................................................................................................................612 5.4.1.1 EMI Shielding...........................................................................................................614 5.4.1.2 Conductive Coating ................................................................................................616 5.4.1.3 Signature Materials..................................................................................................617 xx Contents 5.4.2 5.5 5.6 5.7 5.8 5.9 5.10 5.11 5.12 5.13 5.14 5.15 Inherently Conductive Polymers..........................................................................................618 5.4.2.1 Doping .......................................................................................................................620 5.4.2.2 Conducting Mechanisms .......................................................................................625 5.4.2.3 Applications ..............................................................................................................627 5.4.3 Photoconductive Polymers ....................................................................................................635 Electroactive Polymers...........................................................................................................................636 5.5.1 Ionic EAPs .................................................................................................................................636 5.5.1.1 Polymer–Metal Composites ..................................................................................636 5.5.1.2 Ionic Polymer Gels..................................................................................................637 5.5.1.3 Carbon Nanotubes ..................................................................................................637 5.5.1.4 Conductive Polymers..............................................................................................639 5.5.2 Electronic EAPs........................................................................................................................640 5.5.2.1 Ferroelectric Polymers............................................................................................640 5.5.2.2 Polymer Electrets.....................................................................................................640 5.5.2.3 Electrostrictive Polymers........................................................................................641 5.5.2.4 Dielectric Elastomers ..............................................................................................642 Polymers in Fiber Optics ......................................................................................................................644 Polymers in Nonlinear Optics .............................................................................................................648 Langmuir–Blodgett Films......................................................................................................................648 Piezo- and Pyroelectric Polymers .......................................................................................................650 5.9.1 Applications...............................................................................................................................651 Polymeric Electrolytes............................................................................................................................653 Polymers in Photoresist Applications ................................................................................................656 5.11.1 Negative Photoresists..............................................................................................................659 5.11.2 Positive Photoresists ...............................................................................................................661 5.11.2.1 Near-UV Application .......................................................................................... 661 5.11.2.2 Mid- and Deep-UV Photoresists ...................................................................... 663 5.11.3 Electron Beam Resists.............................................................................................................667 5.11.4 Plasma-Developable Photoresists.........................................................................................667 Photoresist Applications for Printing.................................................................................................669 5.12.1 Printing Plates ..........................................................................................................................669 5.12.1.1 Relief or Raised-Image Plates ............................................................................ 669 5.12.1.2 Photolithography/Planographic Plates ............................................................ 670 5.12.1.3 Photogravure ......................................................................................................... 670 5.12.2 Photoengraving ........................................................................................................................671 5.12.3 Printed Circuits........................................................................................................................671 5.12.4 Collotype and Prooﬁng Systems ..........................................................................................671 Optical Information Storage.................................................................................................................672 Polymers in Adhesives...........................................................................................................................673 5.14.1 Solvent-Based Adhesives........................................................................................................674 5.14.2 Water-Based Adhesives..........................................................................................................675 5.14.3 Hot Melt Adhesives ................................................................................................................676 5.14.4 Radiation-Curable Adhesives................................................................................................677 Degradable Polymers .............................................................................................................................678 5.15.1 Packaging Applications ..........................................................................................................679 5.15.2 Medical and Related Applications .......................................................................................680 5.15.2.1 Synthetic Polymers............................................................................................... 680 5.15.2.2 Controlled Release Agents.................................................................................. 682 5.15.2.3 Tissue Engineering ............................................................................................... 683 Contents xxi 5.16 Ionic Polymers......................................................................................................................................... 685 5.16.1 Physical Properties and Applications..................................................................................686 5.16.1.1 Ionic Cross-Linking ..............................................................................................686 5.16.1.2 Ion-Exchange..........................................................................................................687 5.16.1.3 Hydrophilicity ........................................................................................................689 5.16.2 Ionomers ....................................................................................................................................691 5.16.2.1 Polyethylene Ionomers.........................................................................................691 5.16.2.2 Elastomeric Ionomers...........................................................................................694 5.16.2.3 Ionomers Based on Polytetraﬂuoroethylene ...................................................695 5.16.2.4 Ionomers Based on Polysulfones .......................................................................696 5.16.3 Polyelectrolytes .........................................................................................................................697 5.16.3.1 Ion-Exchangers ......................................................................................................697 5.16.3.2 Polycarboxylates ....................................................................................................705 5.16.3.3 Integral Polyelectrolytes.......................................................................................706 5.17 Synthetic Polymer Membranes............................................................................................................ 707 5.17.1 Membrane Preparation...........................................................................................................707 5.17.1.1 Wet-Extrusion Process.........................................................................................708 5.17.1.2 Hollow Fiber Membranes....................................................................................709 5.17.2 Membrane Modules ................................................................................................................712 5.17.3 Applications...............................................................................................................................713 5.18 Hydrogels and Smart Polymers........................................................................................................... 714 5.18.1 Smart Polymers........................................................................................................................716 5.19 Dendritic Polymers................................................................................................................................. 722 5.19.1 Applications ..............................................................................................................................724 5.20 Shape Memory Polymers ...................................................................................................................... 725 5.21 Microencapsulation ................................................................................................................................ 727 5.21.1 Processes for Microencapsulation ........................................................................................728 5.21.1.1 Complex Coacervation.........................................................................................728 5.21.1.2 Polymer–Polymer Incompatibility ....................................................................729 5.21.1.3 Interfacial and In Situ Polymerization .............................................................729 5.21.1.4 Spray Drying ..........................................................................................................733 5.21.1.5 Fluidized-Bed Coating..........................................................................................733 5.21.1.6 Co-Extrusion Capsule Formation......................................................................734 5.21.1.7 Other Processes......................................................................................................734 5.21.2 Applications...............................................................................................................................735 5.22 Nanosize Polymers ................................................................................................................................. 737 5.22.1 Polymer Nanoparticles ...........................................................................................................737 5.22.2 Polymer Nanospheres.............................................................................................................738 5.22.3 Polymer Nanoﬁbers ................................................................................................................739 5.22.4 Polymer Nanowires, Nanotubes, and Nanorods..............................................................740 5.23 Polymer Nanocomposites ..................................................................................................................... 742 5.24 Polymer–Clay Nanocomposites........................................................................................................... 743 5.25 Polymer–Carbon Nanocomposites ..................................................................................................... 746 5.25.1 Graphite-Based PNCs .............................................................................................................747 5.25.2 CNT-Based PNCs ....................................................................................................................747 5.25.2.1 Nanotube Functionalization ...............................................................................747 5.25.2.2 Nanocomposite Fabrication Methods...............................................................748 5.25.2.3 Nanotube Alignment in Composites ................................................................750 5.25.2.4 Properties of Nanocomposites ...........................................................................751 xxii Contents 5.25.3 Graphene-Based PNCs ........................................................................................................753 5.25.3.1 Technical Production of Graphene ............................................................... 754 5.25.3.2 Preparation of Nanocomposites..................................................................... 760 5.25.3.3 Properties of Graphene–PNCs ....................................................................... 762 5.26 Microﬁbrillar/Nanoﬁbrillar Polymer Composites.........................................................................764 5.26.1 Microﬁbrillar/Nanoﬁbrillar Polymer–Polymer Composites .......................................765 5.26.1.1 Manufacturing Steps......................................................................................... 766 5.26.1.2 Properties and Applications............................................................................ 768 5.26.2 Microﬁbrillar/Nanoﬁbrillar SPCs......................................................................................770 5.26.2.1 Manufacturing Steps......................................................................................... 770 5.26.2.2 Properties and Applications............................................................................ 771 5.27 Wood–Polymer Composites..............................................................................................................774 5.27.1 WPC Feedstocks....................................................................................................................775 5.27.1.1 Wood.................................................................................................................... 775 5.27.1.2 Plastics ................................................................................................................. 775 5.27.1.3 Compounded Pellets......................................................................................... 776 5.27.1.4 Additives.............................................................................................................. 776 5.27.2 Manufacture of WPC Products .........................................................................................776 5.27.2.1 Compounding .................................................................................................... 777 5.27.2.2 Extrusion ............................................................................................................. 777 5.27.3 Properties of WPC Products..............................................................................................779 5.27.4 Applications of WPC Products .........................................................................................780 References........................................................................................................................................................... 781 6 Recycling and Disposal of Waste Plastics 6.1 6.2 6.3 6.4 6.5 Introduction.............................................................................................................................................795 Outline of Recycling Methods.............................................................................................................799 Recycling of Poly (Ethylene Terephthalate) .....................................................................................804 6.3.1 Direct Reuse ..............................................................................................................................804 6.3.2 Reuse after Modiﬁcation.........................................................................................................806 6.3.2.1 Glycolysis...................................................................................................................806 6.3.2.2 Methanolysis.............................................................................................................808 6.3.2.3 Ammonolysis............................................................................................................808 6.3.2.4 Hydrolysis .................................................................................................................809 6.3.2.5 Depolymerization in Supercritical Fluids...........................................................809 6.3.2.6 Enzymatic Depolymerization................................................................................810 6.3.3 Incineration ...............................................................................................................................810 Recycling of Polyurethanes ..................................................................................................................811 6.4.1 Thermopressing Process .........................................................................................................811 6.4.2 Kneader Process .......................................................................................................................812 6.4.3 Hydrolysis ..................................................................................................................................812 6.4.3.1 Glycolysis...................................................................................................................813 6.4.3.2 Ammonolysis............................................................................................................813 Recycling of Poly (Vinyl Chloride) ....................................................................................................815 6.5.1 Characterization of Used PVC..............................................................................................816 6.5.2 In-Line PVC Scrap...................................................................................................................816 6.5.3 PVC Floor Coverings ..............................................................................................................818 6.5.4 PVC Rooﬁng Sheets.................................................................................................................818 6.5.5 Post-Consumer PVC ...............................................................................................................819 6.5.6 Vinyloop and Texyloop Processes........................................................................................819 xxiii Contents 6.6 6.7 Recycling of Cured Epoxies................................................................................................................820 Recycling of Mixed Plastics Waste ...................................................................................................821 6.7.1 Direct Reuse ............................................................................................................................822 6.7.2 Homogeneous Fractions.......................................................................................................824 6.7.3 Liquefaction of Mixed Plastics............................................................................................825 6.8 Post-Consumer Polyethylene Films..................................................................................................825 6.9 Recycling of Ground Rubber Tires ...................................................................................................826 6.10 Recycling of Car Batteries ...................................................................................................................828 6.11 Plastic Recycling Equipment and Machinery .................................................................................828 6.11.1 Plastocompactor.....................................................................................................................829 6.11.2 Debaling and Initial Size Reduction..................................................................................829 6.11.2.1 Shredder ...............................................................................................................830 6.11.2.2 Cutter or Guillotine ...........................................................................................830 6.11.2.3 Screw Shredder ...................................................................................................830 6.11.2.4 Granulators ..........................................................................................................830 6.11.2.5 Fine Grinding ......................................................................................................831 6.11.3 Cleaning and Selection .........................................................................................................831 6.11.3.1 Dry Separation ....................................................................................................832 6.11.3.2 Wet Separation ...................................................................................................834 6.11.3.3 Other Methods....................................................................................................834 6.11.4 Resin Detectors: Type and Conﬁguration........................................................................834 6.11.5 Automatic Sortation..............................................................................................................836 6.11.5.1 PVC/PET and Commingled Plastics Sortation ...........................................836 6.12 Upcycling of Waste Plastics ...............................................................................................................838 6.13 Plastics Waste Disposal in Landﬁlls .................................................................................................839 6.14 Energy Recovery from Waste Plastics .............................................................................................841 6.15 Disposal of E-Waste Plastics ..............................................................................................................842 References........................................................................................................................................................... 843 7 Trends in Polymer Applications 7.1 7.2 7.3 Introduction.............................................................................................................................................847 Polymers in Packaging ..........................................................................................................................848 7.2.1 Retorting.....................................................................................................................................850 7.2.2 Asceptic Packaging ..................................................................................................................850 7.2.3 Hot-Filling .................................................................................................................................851 7.2.4 Controlled-Atmosphere Packaging ......................................................................................851 7.2.5 High-Barrier Films...................................................................................................................851 7.2.6 Oxygen Scavenger-Based Packaging ....................................................................................852 7.2.7 Plastic Bottles ............................................................................................................................852 7.2.8 Chemical Containers ...............................................................................................................853 7.2.9 Dual Ovenables.........................................................................................................................853 7.2.10 Closures ......................................................................................................................................854 7.2.11 Biodegradable Packaging ........................................................................................................854 7.2.12 Pharmaceutics Packaging and Nanomedicines .................................................................855 7.2.13 Wood–Plastic Composites in Packaging.............................................................................856 Polymers in Building and Construction...........................................................................................856 7.3.1 Rooﬁng .......................................................................................................................................857 7.3.2 Flooring ......................................................................................................................................857 7.3.3 Windows ....................................................................................................................................858 7.3.4 Pipes ............................................................................................................................................859 xxiv Contents 7.3.5 7.3.6 7.4 7.5 7.6 7.7 7.8 7.9 7.10 Insulation ...................................................................................................................................859 Polymer–Concrete Composites.............................................................................................860 7.3.6.1 Fiber-Reinforced Polymer in Concrete...............................................................862 7.3.7 Wood–Plastic Composites .....................................................................................................863 7.3.8 Smart Healable Polymer Composites..................................................................................863 7.3.9 Biodegradable Composites.....................................................................................................864 Polymers in Corrosion Prevention and Control .............................................................................864 7.4.1 Flue Gas Desulfurization ........................................................................................................865 7.4.2 Chemical Resistant Masonry.................................................................................................865 7.4.3 Piping Systems..........................................................................................................................866 7.4.4 Boiler and Cooling Water Treatment .................................................................................866 7.4.5 Biodegradable Scale Inhibitor ...............................................................................................866 7.4.6 Reinforcing Steel in Concrete................................................................................................867 Plastics in Automotive Applications ..................................................................................................868 7.5.1 Exterior Body Parts .................................................................................................................868 7.5.2 Interior Components...............................................................................................................869 7.5.3 Load-Bearing Parts ..................................................................................................................870 7.5.4 Under-the-Bonnet (Hood).....................................................................................................870 7.5.5 Future Trends ...........................................................................................................................871 7.5.6 Polymer Nanocomposites ......................................................................................................871 7.5.7 “Green” Composites ................................................................................................................872 Polymers in Aerospace Applications..................................................................................................873 7.6.1 Carbon Fibers ...........................................................................................................................873 7.6.2 Resins ..........................................................................................................................................874 Polymers in Electrical and Electronic Applications........................................................................875 7.7.1 Wire and Cable Insulation.....................................................................................................876 7.7.2 Polymer Insulators...................................................................................................................877 7.7.3 Printed Circuit Boards ............................................................................................................877 7.7.4 Connectors.................................................................................................................................877 7.7.5 Enclosures ..................................................................................................................................878 7.7.6 Optical Fibers............................................................................................................................878 7.7.7 Information Storage Discs .....................................................................................................879 7.7.8 Polymeric FET and LED ........................................................................................................880 7.7.9 Polymer-Based Solar Cells .....................................................................................................881 Polymers in Agriculture and Horticulture........................................................................................883 7.8.1 Plastic Film ................................................................................................................................884 7.8.2 Plastic Crates.............................................................................................................................885 7.8.3 Building ......................................................................................................................................885 7.8.4 Pipes and Hoses .......................................................................................................................885 7.8.5 Greenhouses ..............................................................................................................................885 Polymers in Domestic Appliances and Business Machines ..........................................................886 7.9.1 Large Appliances ......................................................................................................................886 7.9.2 Small Appliances ......................................................................................................................888 7.9.3 Business Equipment ................................................................................................................888 7.9.4 Air Filters ...................................................................................................................................889 7.9.5 Solar Systems ............................................................................................................................889 Polymers in Medical and Biomedical Applications ........................................................................890 7.10.1 Medical Packaging ...................................................................................................................890 7.10.2 Nontoxic Sterilizable Items....................................................................................................890 7.10.3 Biodegradable Polymers .........................................................................................................892 Contents xxv 7.10.4 Conducting Polymer Nanotubes.......................................................................................... 892 7.10.5 Biomimetic Actuators.............................................................................................................893 7.10.6 Dental Resin Composites.......................................................................................................894 7.10.7 Appliances .................................................................................................................................894 7.10.8 Disposables................................................................................................................................894 7.11 Polymers in Marine and Offshore Applications.............................................................................. 895 7.11.1 Cables .........................................................................................................................................895 7.11.2 Coatings .....................................................................................................................................895 7.11.3 Other Applications..................................................................................................................896 7.12 Polymers in Sport ................................................................................................................................... 896 7.12.1 Synthetic Surfaces....................................................................................................................896 7.12.2 Footwear ....................................................................................................................................897 7.12.3 Equipment.................................................................................................................................897 7.13 Renewable Synthetic Polymers ............................................................................................................ 898 References ............................................................................................................................................................ 900 Appendices A1 Trade Names for Some Industrial Polymers ................................................................ 903 A2 Commonly Used Abbreviations for Industrial Polymers .......................................... 915 A3 Formulations of Flame-Retarded Selected Thermoplastics ....................................... 919 A4 Formulations of Selected Rubber Compounds............................................................. 923 A5 Formulations of Selected PVC Compounds ................................................................. 927 A6 Formulations of Polyurethane Foams ............................................................................ 931 A7 Conversion of Units ........................................................................................................... 935 A8 Typical Properties of Polymers Used for Molding and Extrusion .......................... 937 A9 Typical Properties of Cross-Linked Rubber Compounds.......................................... 943 A10 Typical Properties of Representative Textile Fibers .................................................... 947 Index.................................................................................................................................................. 949 http://taylorandfrancis.com Preface In answer to a question about what he thinks to be chemistry’s biggest contribution to science and to society, Lord Todd, president of the Royal Society of London, said (as quoted in Chem. Eng. News 50, 40, 1940): I am inclined to think that the development of polymerization is, perhaps, the biggest thing chemistry has done, where it has had the biggest effect on everyday life. The world would be a totally different place without artiﬁcial ﬁbers, plastics, elastomers etc. Even in electronics what would you do without insulation? And you come back to polymers again. Polymers are made of macromolecules, which are large molecules made by joining together many smaller ones while processable compositions based on polymers are, by common usage, called plastics. The term plastics thus includes all of the many thousands of grades of commercial materials, ranging in application from squeeze bottles, bread wrappers, fabrics, paints, adhesives, rubbers, wire insulators, foams, greases, oils, and ﬁlms to automobiles and aircraft components, and missile and spacecraft bodies. As plastics thus touch everybody’s life in many different ways, in many different forms, and for many different needs, under diverse conditions of use and applications, a comprehensive source of information about different aspects of plastics, such as synthesis and manufacture, processing and fabrication, properties and testing, stability and degradation, or recycling and disposal is needed. Plastics Technology Handbook was thus conceived and published, in its ﬁrst edition, way back in 1987, in recognition of this need. As the plastics world has progressed, expanded, and changed rapidly with time, this handbook also has been constantly revised, expanded, and updated, in tune with this change, to reach the present stage of ﬁfth edition, in which, however, the textbook-style approach and objective presentation of the earlier editions have been retained, while enlarging, upgrading, and updating several areas of greater interest and adding many new topics of current focus and potential importance. Starting with basic concepts, deﬁnitions, and terminologies of polymer science, the opening chapter of the book describes different types of average molecular weights of polymers, theories behind them and experimental methods of determination, different types of polymerization processes (addition, condensation, coordination, and supramolecular), reaction mechanisms and kinetics, industrial methods of polymerization (bulk, solution, suspension, emulsion, and microemulsion), and polymerization reactors (batch-type, plug ﬂow, cascading, etc). Molecular characteristics and structural shapes of polymer molecules are analyzed, focusing on polymer conﬁgurations and conformations with effects on polymer crystallinity and solid state morphology that greatly inﬂuence physical and chemical properties of polymer materials. Thermal transitions (glass transition and melting) of polymers and their relation to molecular structure and morphology are rationalized. Viscoelastic behavior, cross-linking effects of polymer chains, polymer solubility and its predictability, as well as effects of corrosives are discussed. Other facets of polymer characteristics including thermal stability, ﬂame retardation and ablation, and thermal, chemical, and radiation degradation are explored. Chemical methods of stabilization of polymers are explained, focusing on all common polymers, individually. Since polymers are not useful without the addition of additives (i.e., polymer compounding), various additives, namely, ﬁllers, plasticizers, antistatic agents, ﬂame retardants, smoke suppressants, colorants, and antimicrobials, are highlighted. Toxicity of plastics is also discussed with focus on plastic devices used in pharmacy, medicines, nanomedicines, drug delivery, and toxicity testing. Beginning with a general discussion of tooling for plastics processing—especially molds, dies, and tool design—the second chapter of the book presents, in a simple yet elaborate manner, essential features of the most common methods of processing thermosetting plastics (namely, compression and transfer molding) and thermoplastics (namely, extrusion, injection, blowing, and calendering), besides other processes, such as thermoforming, slush molding, and spinning with persistent focus on functioning and methodology of the processes. The signiﬁcance of new developments in mold design in respect to runners and gates is discussed. All important fabrication processes are fully illustrated and lucidly explained. These include plunger xxvii xxviii Preface injection molding, screw injection molding, foam injection molding, blown ﬁlm extrusion, ﬂat ﬁlm extrusion, pipe extrusion, extrusion blow molding, injection blow molding, calendering, conventional ﬁber spinning, electrospinning (of nanoﬁbers), thermoforming, and casting processes. All aspects of reinforced plastics—both glass-reinforced and graphite/carbon-reinforced plastics—and their molding techniques are described and illustrated with diagrams. The processes of pultrusion, prepreg molding, reaction injection molding, resin transfer molding, conventional foaming, and syntactic foaming are presented. Rubber compounding and processing technology, including rubber reclamation and fabrication of rubber-based products, are discussed, followed by a host of other processing techniques, which include spray coating, ﬂuidized bed coating, electrostatic spraying, stimuli-responsive coating, powder molding, centrifugal casting, adhesion bonding, high-frequency welding, ultrasonic welding, and plastics decorative processes (e.g., gravure printing, ﬂexography, screen printing, in-mold decorating, embossing, electroplating, and vacuum metallizing). More recent developments in polymer fabrication and processing, such as joining polymer– metal hybrids, smart coatings, rapid prototyping/3D printing, and ﬂex printing, are also presented. Besides the remarkable ease and scope of plastics fabrication, it is the wide range of properties inherent in plastics or imparted by various means that gives plastics the dominant place among all materials. Chapter 3 discusses plastics properties under four main headings—mechanical, electrical, optical, and thermal—giving theoretical derivations where necessary and rationalizing them on molecular and structural basis. Mechanical properties considered include elastic stress–strain, viscoelastic stress–straintime, creep (with mathematical modeling), pseudoelastic design, time–temperature superposition principle, dynamic mechanical properties, hardness, impact strength, and stress–corrosion cracking behavior. The mechanical behavior of ﬁber-reinforced plastics is separately discussed and mathematically modeled. The electrical behavior of plastics is evaluated in terms of the speciﬁc properties, such as dielectric strength, dielectric constant, insulation resistance or resistivity, and arc resistance. Characteristic aspects of light transmitting ability and optical clarity, refractive index, and light piping, as well as stress-optical behavior of speciﬁc plastics, are discussed. Thermal properties are highlighted by comparing different plastics in respect of speciﬁc heat, thermal expansion, thermal conductivity, and thermal transitions with speciﬁc focus on application aspects. An important facet of materials development and proper materials selection is testing and standardization for various properties. The latter part of Chapter 3 is devoted to this aspect. A number of standard test methods for plastics are presented schematically (in simpliﬁed form), highlighting the principles of the tests and methods of measurements. The last part of the chapter is devoted to chemical identiﬁcation and analysis of common plastics. A layman yet systematic procedure, involving heating and ignition, detection of hetero-elements (N, S, halogens), solubility observation, and other tests, is presented followed by exact identiﬁcation via speciﬁc chemical tests. Instrumental methods of plastics analysis, both qualitative and quantitative, are also presented, an elaborate and illustrative coverage being given to the most commonly used methods, namely, IR and NMR spectroscopy. Industrial polymers, that is, those that are produced on very large, large, and relatively large scales— including the so-called engineering polymers that possess superior mechanical properties for engineering applications—are discussed in Chapter 4. These polymers have been classiﬁed into three broad categories— addition polymers (i.e., chains consisting entirely of C–C bonds), condensation polymers (i.e., chains with hetero-atoms, e.g., O, N, S, Si, in the backbone chain), and special polymers (i.e., products with special properties, such as temperature and ﬁre resistance, photosensitivity, electrical conductivity, and piezoelectric properties). Further classiﬁcation of addition, condensation, and special polymers has been made on the basis of monomer composition, and all important polymers in each class have been discussed in detail with focus on production methods, characteristic properties, and principal applications. The main classes thus considered are polyoleﬁns, acrylics, vinyl polymers, polyesters, polyamides, formaldehyde resins, polyurethanes, ether polymers, cellulosic polymers, and sulﬁde polymers. Polysulfones, polyether ketones, polybenzimidazole, silicones, and other inorganic polymers are included under the special polymers category. Blend polymers, including industrial polyblends, nanoblends, and interpenetrating network polymers, are also given due coverage. Preface xxix Besides a handful of high-volume polymers like PE, PP, PVC, PS, PC, nylon, polyesters, and so on, which are very visible in everyday life, there are hundreds of other polymers, polymer derivatives, and polymeric combinations that play special and often critical roles in diverse ﬁelds of human activity. Chapter 5 is devoted to such polymers. For a systematic discussion, these polymers have been placed in different groups according to their properties and/or areas of uses, namely, high-temperature and ﬁreresistant polymers, liquid crystal polymers, electroactive polymers, photoresist polymers, shape-memory polymers, conducting polymers, photoconductive polymers, polymers in ﬁber optics, optical information storage, and nonlinear optics, piezo- and pyro-electric polymers, ionic and ion-exchange polymers, packaging polymers, adhesive polymers, and biodegradable polymers. A wide range of novel applications of polymers are highlighted. These include electrostatic discharge (ESD) protection, rechargeable batteries, polymer solar cells, actuation and robotics, sensors, “smart” coatings, fuel cells, micro- and nanolithography, micro- and nano-electronics, controlled release and drug delivery, tissue engineering, reverse osmosis, solar desalination (by the “Sirotherm” process), oil–surfactant–water separation, hemodialysis, microencapsulation, plasmapheresis, oxygenator, carbonless copying, and light-sensitive color imaging. Nanotechnology being one of the most promising technologies of the present century, an elaborate coverage is given to polymer nanomaterials (nanoparticles, nanospheres, nanoﬁbers, nanowires, nanotubes, and nanorods), polymer–clay nanocomposites, polymer–carbon nanocomposites and the exciting new domain of graphene-based polymer nanocomposites. Microﬁbrillar/nanoﬁbrillar polymer– polymer and single-polymer composites—a fascinating development of recent years based on the novel concept of “converting instead of adding”—are also highlighted in Chapter 5, at the end of which the socalled green composites or wood–polymer composites are discussed with focus on recent developments. While plastics recycling has been practiced for many years with focus on homogeneous industrial scraps and homogeneous post-consumer plastics, the industry also accepted the challenge of recycling heterogeneous plastics waste based on new technologies of separation and processing. Chapter 6 deals with both these aspects of recycling and discusses various methods of waste plastics disposal, used in practice, with their merits and demerits. Different recycling methods currently available for predominant polyethylene terephthalate (PET) wastes are discussed, as well as those of polyurethanes, poly(vinyl chloride), and cured epoxies. For mixed wastes of plastics, methods considered are various direct use technologies, separation into single materials for reuse, and liquefaction for converting into oil. Considering its magnitude, importance, and several unique problems that it presents, rubber tire recycling is treated in detail. Recycling methods of car batteries are reviewed. Equipment and machinery commonly used for various steps in waste plastics recycling operation, including automatic sortation, are described. The new concept of upcycling (as opposed to “downcycling” or recycling) of waste plastics is reviewed. Focus is also given on issues associated with plastics waste disposal, advantages and disadvantages of landﬁll, gas recovery prospects, impact of biodegradable plastics, composting, and energy generation by incineration. While there are continuous improvements in the established uses of polymers, new applications are being developed in diverse areas of human activity. Chapter 7, which is the last chapter of the book, presents a comprehensive overview of new developments in polymer applications in recent years. The areas considered are packaging, building and construction, corrosion prevention and control, automotive, aerospace applications, electrical and electronic applications, agriculture and horticulture, domestic appliances and business machines, medical and biomedical applications, marine and offshore applications, and sports. Furthermore, a state-of-the-art account of emerging new areas such as single-polymer composites, polymer solar cells, and renewable synthetic polymers, which are believed to hold much promise for the future, is given. In writing a book of this kind, I accumulated indebtedness to a wide range of people, not the least to the authors of earlier publications in the ﬁeld, which include such classic reference books as Modern Plastics Handbook (Charles A. Harper, ed., McGraw-Hill, New York, 2000) and Plastics Materials (J. A. Brydson, Butterworth-Heinemann, Oxford, UK, 1999). My faculty colleagues in the Department of Chemical Engineering at the Indian Institute of Science (IISc), Bangalore, and innumerable students and academic xxx Preface associates in other universities and colleges have provided much welcome stimulation and direct help. I have also received much support and encouragement from Professor Ganapathy Ayappa, the departmental chairman. I am greatly indebted to all of them. I also acknowledge with gratitude the painstaking work of Ms. B. G. Girija in preparing computer graphics of a large body of ﬁgures, diagrams, and illustrations that adorn the pages of the book. My thanks are due to Mr. Ravichandra, Manager of the Campus Xeroxing and Services Center at IISc, for providing very valuable and crucial help without which preparation of this new edition would not have been possible. While it is not possible to mention the names of all students who have been part of my academic career and a source of strength, special mention must be made of two among them—Dr. Amitava Sarkar and Dr. Ajay Karmarkar—who have provided me invaluable help and assistance in many ways in my long journey of authorship. Among my academic friends abroad, I would especially mention Prof. G. L. Rempel (University of Waterloo, Waterloo, Canada), Prof. Stoyko Fakirov (Auckland University, Auckland, New Zealand), and Prof. Kenneth J. Wynne (Virginia Commonwealth University, Richmond, Virginia) for the valuable help and support that I received from them. At the end, I would record with gratitude the signiﬁcant contribution of Dr. Salil K. Roy, my erstwhile college-mate and formerly a professor in the postgraduate program in civil engineering of the Petra Christian University, Surabaya, Indonesia, who motivated me to write this book on plastics and provided me with a good amount of valuable literature in the initial phase of the project. I would like to especially thank Mrs. Allison Shatkin, Materials Science and Chemical Engineering Editor at CRC Press (Taylor & Francis), for her deep sensitivity, understanding, and stimulating encouragement in my work. The excellent cooperation that I have received from Ms. Florence Kizza, Engineering and Environmental Sciences Editor, and Ms. Teresita Munoz, the Editorial Assistant, is highly appreciated. Finally, I should thank Ms. Adel Rosario of Manila Typesetting Company (Philippines) and her highly efﬁcient team for contributing so well to the shaping of the book. I am grateful to my wife Mridula, daughter Amrita, and granddaughter Mallika for their extraordinary tolerance and ungrudging support without which this voluminous manuscript could not have been prepared. Manas Chanda Author Manas Chanda was formerly a professor in the Department of Chemical Engineering, Indian Institute of Science, Bangalore, India. He also worked as a summer-term visiting professor at the University of Waterloo, Ontario, Canada, with regular summer visits to the university from 1978 to 2000. He guided the PhD research of 15 students and about 20 Masters dissertations. A ﬁve-time recipient of the International Scientiﬁc Exchange Award from the Natural Sciences and Engineering Council, Canada, Dr. Chanda is the author and coauthor of more than 100 scientiﬁc papers, articles, and books. The books he authored include Engineering Materials Science (3 volumes), 1979, published by Macmillan (New Delhi, India); Introduction to Polymer Science and Chemistry: A Problem Solving Approach, 2nd Edition, 2013; and Plastics Technology Handbook, 5th Edition, 2018, the last two published by CRC Press (Boca Raton, Florida). His biographical sketch is listed in Marquis’ Who’s Who in the World Millennium Edition (2000) by the American Biographical Society. A Fellow of the Indian National Academy of Engineers (New Delhi) and a member of the Indian Institute of Chemical Engineers (Calcutta) and Indian Plastics Institute (Mumbai), he received his BS (1959) and M. Tech (1962) degrees from Calcutta University and his PhD degree (1966) from the Indian Institute of Science, Bangalore, India. xxxi http://taylorandfrancis.com 1 Characteristics of Polymers and Polymerization Processes 1.1 What Is a Polymer? A molecule has a group of atoms which have strong bonds among themselves but relatively weak bonds to adjacent molecules. Examples of small molecules are water (H2O), methanol (CH3OH), carbon dioxide, and so on. Polymers contain thousands to millions of atoms in a molecule which is large; they are also called macromolecules. Polymers are prepared by joining a large number of small molecules called monomers. Polymers can be thought of as big buildings, and monomers as the bricks that go into them. Monomers are generally simple organic molecules containing a double bond or a minimum of two active functional groups. The presence of the double bond or active functional groups acts as the driving force to add one monomer molecule upon the other repeatedly to make a polymer molecule. This process of transformation of monomer molecules to a polymer molecule is known as polymerization. For example, ethylene, the prototype monomer molecule, is very reactive because it has a double bond. Under the inﬂuence of heat, light, or chemical agents this bond becomes so activated that a chain reaction of selfaddition of ethylene molecules is generated, resulting in the production of a high-molecular-weight material, almost identical in chemical composition to ethylene, known as polyethylene, the polymer of ethylene (Figure 1.1). The difference in behavior between ordinary organic compounds and polymeric materials is due mainly to the large size and shape of polymer molecules. Common organic materials such as alcohol, ether, chloroform, sugar, and so on, consist of small molecules having molecular weights usually less than 1,000. The molecular weights of polymers, on the other hand, vary from 20,000 to hundreds of thousands. The name polymer is derived from the Greek poly for many and meros for parts. A polymer molecule consists of a repetition of the unit called a mer. Mers are derived from monomers, which, as we have seen for ethylene, can link up or polymerize under certain conditions to form the polymer molecule. The number of mers, or more precisely the number of repetitions of the mer, in a polymer chain is called the degree of polymerization (DP). Since the minimum length or size of the molecule is not speciﬁed, a relatively small molecule composed of only, say, 3 mers might also be called a polymer. However, the term polymer is generally accepted to imply a molecule of large size (macromolecule). Accordingly, the lowermolecular-weight products with low DP should preferably be called oligomers (oligo = few) to distinguish them from polymers. Often the term high polymer is also used to emphasize that the polymer under consideration is of very high molecular weight. Because of their large molecular size, polymers possess unique chemical and physical properties. These properties begin to appear when the polymer chain is of sufﬁcient length—i.e., when the molecular weight exceeds a threshold value—and becomes more prominent as the size of the molecule increases. The 1 2 Plastics Technology Handbook H H H Heat/light Catalyst C — — C H H * C — — C H H H Activated ethylene H H H H H H H H H H + C C C C C C* H H H H H H H * C — — C H H H H H H H C C C H H H C C C C* + n C H H H H H n+2 Polyethylene Intermediate steps during formation of polyethylene. FIGURE 1.1 dependence of the softening temperature of polyethylene on the degree of polymerization is shown in Figure 1.2a. The dimer of ethylene is a gas, but oligomers with a DP of 3 or more (that is, C6 − or higher parafﬁns) are liquids, with the liquid viscosity increasing with the chain length. Polyethylenes with DPs of about 30 are greaselike, and those with DPs around 50 are waxes. As the DP value exceeds 400 or the molecular weight exceeds about 10,000, polyethylenes become hard resins with softening points about 100°C. The increase in softening point with chain length in the higher-molecular-weight range is small. The relationship of such polymer properties as tensile strength, impact strength, and melt viscosity with molecular weight is indicated in Figure 1.2b. Note that the strength properties increase rapidly ﬁrst as the chain length increases and then level off, but the melt viscosity continues to increase rapidly. Polymers with very high molecular weights have superior mechanical properties but are difﬁcult to process and fabricate due to their high melt viscosities. The range of molecular weights chosen for commercial polymers represents a compromise between maximum properties and processability. Tensile strength, melt viscosity Softening temperature (°C) 100 75 50 25 Tensile strength Melt viscosity Commercial range 0 500 (a) 1000 1500 Degree of polymerization (DP) Molecular weight (b) FIGURE 1.2 Polymer properties versus polymer size. (a) Softening temperature of polyethylene. (b) Tensile strength, and melt viscosity. (Adapted from Seymour, R. B. and Carraher, C. E. Jr., 1992. Polymer Chemistry. An Introduction. Marcel Dekker, New York.) 3 Characteristics of Polymers and Polymerization Processes 1.2 Molecular Weight of Polymers In ordinary chemical compounds such as sucrose, all molecules are of the same size and therefore have identical molecular weights (M). Such compounds are said to be monodisperse. In contrast, most polymers are polydisperse. Thus a polymer does not contain molecules of the same size and, therefore, does not have a single molecular weight. In fact, a polymer contains a large number of molecules—some big, some small. Thus there exists a variation in molecular size and weight, known as molecular-weight distribution (MWD), in every polymeric system, and this MWD determines to a certain extent the general behavior of polymers. Since a polymer consists of molecules of different sizes and weights, it is necessary or an average degree of polymerization (DP). to calculate an average molecular weight (M) The molecular weights commonly used in the characterization of a polydisperse polymer are the number average, weight average, and viscosity average molecular weights. Consider a sample of a polydisperse polymer of total weight W in which N = total number of moles; Ni = number of moles of species i (comprising molecules of the same size); ni = mole fraction of species i; Wi = weight of species i; wi = weight fraction of species i; Mi = molecular weight of species i; xi = degree of polymerzation of species i. n) 1.2.1 Number-Average Molecular Weight (M From the deﬁnition of molecular weight as the weight of sample per mole, we obtain X n = W = M N X =X Ni Mi N X Wi Wi =Mi =X = wi wi =Mi X ni Mi =X 1 wi =Mi (1.1) (1.2) n by the mer weight M0, we obtain a number-average degree of polymerization, DPn , where Dividing M X n Ni x i M DPn = = X M0 Ni (1.3) n is obtained by end-group analysis or by measuring a colligative property such as The quantity M elevation of boiling point, depression of freezing point, or osmotic pressure [1,2]. 1.2.1.1 End-Group Analysis End-group analysis can be used to determine Mn of polymer samples if the substance contains detectable end groups, and the number of such end groups per molecule is known beforehand. End-group analysis has been applied mainly to condensation polymers, since these polymers, by their very nature, have reactive functional end groups. The end groups are often acidic or basic in nature, as exempliﬁed by the carboxylic groups of polyesters or the amine groups of polyamides; such groups are conveniently esti n is derived according to mated by titration. From the experimental data, M n = f w e M a (1.4) where f is the functionality or number of reactive groups per molecule in the polymer sample, w is the weight of the polymer, a is the amount of reagent used in the titration, and e is the equivalent weight of the reagent. 4 Plastics Technology Handbook 1.2.1.2 Ebulliometry (Boiling-Point Elevation) In applying this method, the boiling point of a solution of known concentration is compared to that of the solvent at the same pressure. For ideally dilute solutions, the elevation of the boiling point, T – Tb, is related to the normal boiling temperature of the solvent Tb, its molar latent heat of evaporation Le, and molecular weight M1, as well as to the molecular weight of the solute M2, and relative weights of solvent and solute W1 and W2, respectively, by DTb = T – Tb = RT 2b W2 M1 Le W1 M2 (1.5) For convenience, Equation 1.5 is rewritten as DTb = RT b2 M1 1000W2 = k e m2 1000Le W1 M2 (1.6) where ke is the molal boiling-point elevation constant of the solvent given by ke = (RT 2b M1)/(1000Le) and m2 is the solute molality (in units of moles per kilogram), given by m2 = ð1000W2 Þ=ðW1 M2 Þ To determine a molecular weight, one measures DTb for a dilute solution of solute in solvent and calculates m2 from Equation 1.6. The molecular weight M2 of the solute is then equal to the value of (1000W2)/(W1m2). Ebulliometry, like end-group analysis, is limited to low-molecular-weight polymers. 1.2.1.3 Cryoscopy (Freezing-Point Depression) Calculation of the freezing-point depression of the solvent and hence the molecular weight of the solute by this method proceeds exactly the same way as for the boiling-point elevation. For cryoscopy of ideal solutions, equations corresponding to those for DTb and ke are DTf = −kfm2 and kf = (RT 2f M1)/(1000Lf), where DTf ≡ T − Tf is the freezing-point depression, Tf is the freezing point of pure solvent, and Lf is the molar latent heat of fusion. The solvent’s molal freezing-point depression constant kf is calculated in the same way as ke is calculated in ebulliometry. Some kf values so obtained are as follows: water, 1.8; acetic acid, 3.8; benzene, 5.1; succinonitrile, 20.3; camphor, 40. The large kf of camphor makes it especially useful in molecular weight determinations. Like ebulliometry, the cryoscopic method is also limited to relatively n up to 50,000. low-molecular-weight polymers with M 1.2.1.4 Membrane Osmometry Osmotic pressure is the most important among all colligative properties for the determination of molecular weights of synthetic polymers. To explain osmotic pressure, let us imagine a box (Figure 1.3) that is divided into two chambers, one containing a polymer solution and the other containing pure solvent, separated by a semipermeable membrane (typically cellophane) that allows solvent to pass through but not polymer, because the diffusion rate of the much larger polymer molecules through the pores of the membrane is negligibly small. As the solvent enters the solution side to establish equilibrium, the pressure on the solution side must be greater. Either by waiting till equilibrium is reached or by measuring and compensating for pressures automatically, osmotic pressure p can be measured at several concentrations of the polymer solution. According to thermodynamic theory, p c c!0 = RT M (1.7) where p = osmotic pressure (g/cm2) = hr; h = difference in liquid levels at equilibrium (cm); r = density of solvent (g/cm3); c = concentration (g/cm3); T = absolute temperature (K); M = molecular weight (g/mole); and R = gas constant, 8.48 × 104 (g cm/mole K). 5 Characteristics of Polymers and Polymerization Processes h3 h2 Osmotic head (h) h1 h1 h2 h3 Solution of B in A Pure A Semipermeable membrane FIGURE 1.3 Schematic diagram showing the development of osmotic head as a function of time, where h1 represents the initial liquid levels, h2 denotes the levels after some time, and h3 represents the levels when equilibrium is attained. Equation 1.7 holds only at inﬁnite dilution, where the osmotic pressure is best represented by an attenuated power series: p M = 1 + A2 Mc + A3 M 2 c c RT (1.8) where A2 and A3 are the second and third virial coefﬁcients, respectively. Often, A3 can be taken as equal to (A2/2)2, so that Equation 1.8 can be rewritten as p 1=2 c = RT M 1=2 A Mc 1+ 2 2 (1.9) Plots of (p/c)1/2 versus concentration are usually linear and can be extrapolated to inﬁnite dilution (c = 0). The second virial coefﬁcient is obtained from such plots by dividing the slope by the intercept and by M/2. In static osmometers, the heights of liquid in capillary tubes attached to the solvent and solution compartments (Figure 1.3) are measured. At equilibrium, the hydrostatic pressure corresponding to the difference in liquid heights is the osmotic pressure. The main disadvantage of this static procedure is the considerable length of time required for attainment of equilibrium. However, this can be overcome by using commercially available automatic osmometers (Figure 1.4) that operate on the null-point principle (i.e., solvent pressure is adjusted by a servo mechanism until a sensor can detect no tendency for solvent to ﬂow through the membrane in either direction) and equilibrium can be reached in 5 to 10 min. With care, molecular weight of 10,000 to 500,000 can be measured with about 1% accuracy. 1.2.1.5 Vapor-Phase Osmometry Vapor-phase osmometry is based on vapor pressure lowering, which is a colligative property. The method n . There is no membrane in a vapor-pressure osmometer. Instead, there are two matched therefore gives M thermistors in a thermostated chamber that is saturated with solvent vapor (Figure 1.5). With a hypodermic syringe, a drop of solution is placed on one thermistor and similarly a drop of solvent of equal size on the other thermistor. The solution has a lower vapor pressure than the solvent at the same temperature, 6 Plastics Technology Handbook Servo motor Ampliﬁer Solution Membrane h Elevator Solvent Solvent reservoir Optical detector Light source Bubble FIGURE 1.4 Schematic diagram of essential components of a high-speed membrane osmometer (Hewlett-Packard Corp., Avondale, Pennsylvania). Wheatstone bridge circuit Galvanometer Syringe (solution) Syringe (solvent) Thermistor beads Chamber Solvent cup Thermal block FIGURE 1.5 Schematic diagram of a vapor-phase osmometer. and so the solvent vapor condenses on the solution droplet. The solution droplet, therefore, starts getting diluted as well as heated up by the latent heat of condensation of the solvent condensing on it. In a steady state, the total rise in temperature DT can be related by an equation, DT 1 = ks + Bc + Cc 2 + … c Mn (1.10) where c is the solute concentration and ks is an instrument constant, which is normally determined for a given solvent, temperature, and thermistor pair, by using solutes of known molecular weight. Characteristics of Polymers and Polymerization Processes 7 The temperature difference between the two thermistors can be measured very accurately as a function of the bridge imbalance output voltage, DV. The operating equation is DV K = + KBc c M (1.11) where K is the calibration constant. A plot of DV/c versus c (where c is the solution concentration) is made and extrapolated to zero concentration to obtain the ordinate intercept (DV/c)c!0. The calibration constant K can be computed using the equation K = M ðDV=cÞc!0 (1.12) where M is the molecular weight of the known standard sample. To determine the molecular weight of an unknown sample, solutions of the sample are made in different concentrations in the same solvent used for the standard sample and the whole procedure is repeated to obtain the ordinate intercept. The molecular weight of the unknown sample is then given by n = K=(DV=cÞc!0 M (1.13) The upper limit of molecular weights for vapor-phase osmometry is considered to be 20,000. Development of more sensitive machines has extended this limit to 50,000 and higher. w) 1.2.2 Weight-Average Molecular Weight (M n , the molecular weight of each species is weighted by Equation 1.1 indicates that in the computation of M the mole fraction of that species. Similarly, in the computation of weight-average molecular weight the molecular weight of each species is weighted by the weight fraction of that species: X X Wi Mi w = M wi Mi = X (1.14) Wi X = X Ni Mi2 Ni M i (1.15) w , by the mer weight: The weight-average degree of polymerization, DPw , is obtained by dividing M X w Wi x i M DPw = = X (1.16) M0 Wi w can be determined by measuring light scattering of dilute polymer solution [3,4]. M w is always M n . Thus for a polymer sample containing 50 mol% of a species of molecular weight 10,000 higher than M n = and 50 mol% of species of molecular weight 20,000, Equation 1.1 and Equation 1.15 give M w = ½(10,000)2 + (20,000)2 =½10,000 + 20,000 = 17,000. 0:5(10,000 + 20,000) = 15,000 and M 1.2.2.1 Light-Scattering Method The measurement of light scattering by polymer solutions is an important technique for the determination of weight-average molecular weight, Mw. It is an absolute method of molecular weight measurement. It also can furnish information about the size and shape of polymer molecules in solution and about parameters that characterize the interaction between solvent and polymer molecules. The experimental 8 Plastics Technology Handbook technique is, however, exacting, mainly because of the large difference in intensity of the incident beam and light scattered by the polymer solution. If a polymer is dissolved in a solvent, the light scattered by the polymer molecules far exceeds that by the solvent, and this provides an absolute measure of molecular weight (Figure 1.6). The Debye relationship, which provides the basis of determining polymer molecular weight from solution scattering is [5] Hc 1 = + 2A2 c k MP(q) (1.17) where H is a lumped constant including the refractive index of the solvent and the change in refractive index of the solution with polymer concentration (determined separately using a differential refractometer). The intensity of light is measured at angle q and concentration c. The second virial coefﬁcient A2 is determined from the data, while P(q), called the particle scattering factor (a complex function of molecular shape), is simply the ratio of the scattering intensity to the intensity in the absence of interference, measured at the same angle q. The light intensity factor k is derived from the “raw” galvanometer readings Ig and Igs when the sample cell contains the solution and the solvent, respectively, with the photocell of Figure 1.6 positioned at an angle q in both cases. The equation used is [5] k= (Ig − Igs ) sin q : 1 + cos2 q (1.18) Since it is known that P(q) = 1 at q = 0, it is customary to extrapolate both q and c to 0 so that one can then obtain the molecular weight M from Equation 1.17. This can be done by plotting Hc/k versus concentration at constant values of q and then plotting the intercept 1/MP(q) versus sin2(q/2) to give the intercept 1/M. However, this double extrapolation to zero q and zero c can be effectively done on the same plot by the Zimm method. Zimm plots (Figure 1.7) consist of graphs in which Hc/k is plotted against [sin2 (q/2) + bc], where b is a constant arbitrarily chosen to give an open display of the experimental data and to enable the two extrapolations to c = 0 and q = 0 to be carried out with comparable accuracy. (It is often convenient to take b = 100.) In practice, intensities of scattered light are measured at a series of concentrations and at several angles for each concentration. The Hc/k values are plotted, as shown in Figure 1.7. The extrapolated points on the q = 0 line, for example, are the intersections of the lines through the Hc/k values for a ﬁxed c and various q values with the ordinates at the corresponding bc values. Similarly, the c = 0 line is drawn through the intersections of the lines through the Hc/k values, for a ﬁxed q and various c values, with the corresponding sin2 (q/2) ordinates. The q = 0 and c = 0 lines intersect on the ordinate and the intercept is equal to 1/M, that is, the reciprocal of the molecular weight. It can be shown Focused monochromatic light beam (λ) Sample cell Transmitted beam Io θ = 0° θ r θ = 135° θ = 45° Movable photo cell θ = 90° iθ To galvanometer FIGURE 1.6 Arrangement of the apparatus required to measure light scattered from a solution at different angles with respect to the incident beam. 9 Characteristics of Polymers and Polymerization Processes c = 0.002 c = 0.002 θ = 135° c = 0.002 θ = 120° c = 0.002 θ = 105° c = 0.002 c = 0.002 θ = 90° θ = 150° c = 0.002 θ = 75° c = 0.002 θ = 60° c = 0.0015 c = 0.002 c = 0.002 θ = 30° θ = 45° θ = 150° θ = 0° c = 0.0010 θ = 150° Hc 6 –1 κ × 10 (mol g ) 4.0 3.0 θ=0 line 2.0 c = 0 line Constant c line c=0 θ = 0° Constant θ line c = 0.00075 θ = 150° c = 0.0005 θ = 150° c=0 θ = 150° c in g/cm3 1.0 0 0.2 0.4 0.6 0.8 1.0 1.2 sin2 (θ/2) + 100c FIGURE 1.7 A typical Zimm plot. The concentration units employed are g/cm3. The symbols o represent extrapolated points. (From Chanda, M. Introduction to Polymer Science and Chemistry. A Problem Solving Approach 2013. CRC Press, Boca Raton, FL.) that this molecular weight obtained by the light scattering method is the weight average molecular weight w ), as deﬁned by Equation 1.14. of the polymer (M 1.2.2.2 Low-Angle Laser Light Scattering (LALLS) In some commercial light-scattering instruments, conventional light sources have been replaced by helium-neon (He-Ne) lasers (l = 6328 Å). Because of the high intensity of these light sources, it is possible to make scattering measurements at much smaller angles (2°–10°) than with conventional light sources and also for smaller samples at lower concentrations. Since, at low angles, the particle scattering factor P (q) approaches unity, Equation 1.17 effectively reduces to Hc 1 = + 2A2 c k M (1.19) Therefore, the intercept of a linear plot of Hc/k versus c at a single (small) angle gives 1/M (and hence w ), while one-half of the slope gives the second virial coefﬁcient A2. The method is thus much simpler M and it avoids the laborious double extrapolation of the aforesaid Zimm plot of the conventional light scattering method. One disadvantage of the LALLS method, however, is that chain dimensions cannot be obtained since scattering is measured only at a single angle (q). v) 1.2.3 Viscosity-Average Molecular Weight (M The viscosity-average molecular weight is deﬁned by the equation v = M hX wi Mia i1=a = hX .X i1=a Ni Mi1+a Ni Mi (1.20) v = M w and for a = −1, M v = M n . Thus, M v falls between M w and M n , and for many polymers For a = 1, M it is 10%–20% below Mw . Mv is calculated from the intrinsic viscosity [h] by the empirical relation va ½h = K M (1.21) 10 Plastics Technology Handbook where K and a are constants. [h] is derived from viscosity measurements by extrapolation to “zero” concentration [5,6]. In correlating polymer properties (such as reactivity) which depend more on the number of molecules n is a more useful parameter than M w or M v . Conin the sample than on the sizes of the molecules, M versely, for correlating polymer properties (such as viscosity) which are more sensitive to the size of the w or M v is more useful. polymer molecules, M Because it is easy to determine, the melt index often is used instead of molecular weight in routine characterization of polymers. It is deﬁned as the mass rate of polymer ﬂow through a speciﬁed capillary under controlled conditions of temperature and pressure. The index can often be related empirically to some average molecular weight, depending on the speciﬁc polymer. A lower melt index indicates a higher molecular weight, and vice versa. 1.2.3.1 Dilute Solution Viscometry V is known as the Mark–Houwink–Sakurada (MHS) equation. Equation 1.20 used for the calculation of M The classical method for determining K and a values of this equation involves fractionation of a whole polymer into subspecies, or fractions, with narrow molecular weight distributions. An average molecular n ) or light scattering (M w ) and, if the weight can be determined on each such fraction, by osmometry (M V of monodisperse polymer. fractions are narrow enough, the measured average can be approximated to M V The intrinsic viscosities measured at a constant temperature for a number of such fractions of known M are ﬁtted to the equation [5] V) ln½h = ln K + a ln (M (1.22) to obtain the MHS constants K and a for the particular polymer/solvent system at the temperature w V is closer to M of viscosity measurement. (Since actual fractions are not really monodisperse and M n , it is a better practice to determine the molecular weight by light scattering than by than to M osmometry.) The two constants K and a are derived from the intercept and slope of a linear least-squares ﬁt to [h] − M data for a series of fractionated polymers. The method assumes that K and a are ﬁxed for a given polymer type and solvent and do not vary with polymer molecular weight. (This is not strictly true, however, and the MHS constants determined for higher-molecular-weight species may depend on the molecular weight range. Tabulations of such constants therefore usually list the molecular weights of fractions for which the particular K and a values were determined.) The determination of intrinsic viscosity is performed very easily with simple glass viscometers. Since the viscosity of a liquid depends markedly on temperature, viscosity measurements must be made at a carefully controlled temperature (within ±1°C). Before a measurement, the viscometer is therefore equilibrated in a carefully controlled thermostatic bath at the required temperature. Two popular viscometers are of the Ostwald and Ubbelohde types, shown in Figure 1.8. To operate the Ostwald viscometer, a given volume of liquid is introduced into bulb B through stem A and is drawn up by suction till it ﬁlls the bulb C and moves beyond the ﬁducial mark a. The suction is then released and the time taken by the liquid meniscus to pass between the ﬁducial marks a and b across the bulb C is measured. The liquid obviously ﬂows under a varying driving force proportional to the changing difference (h) in the levels of the liquids in the two tubes. To ensure that this driving force is the same in all cases, the same amount of liquid must always be taken in bulb B. The above condition of always using the same volume of liquid does not apply, however, in the case of the Ubbelohde suspended-level viscometer, shown in Figure 1.8b. A modiﬁed design of the Ubbelohde viscometer is shown in Figure 1.8c. For measurement with the Ubbelohde viscometer, a measured volume of polymer solution with a known concentration is pipetted into bulb B through stem A. This solution is transferred into bulb C by applying a pressure on A with compressed air while column D is kept closed. When the pressure is released, the solution in bulb E and column D drains back into bulb B and the end of the capillary remains free of 11 Characteristics of Polymers and Polymerization Processes A A D D a a a b C C A b Capillary b C Capillary Capillary E h E B B B (a) (b) (c) FIGURE 1.8 Common types of glass viscometers. (a) Ostwald viscometer. (b) Ubbelohde suspended-level viscometer. (c) A modiﬁed Ubbelohde suspended-level viscometer (see text for description). liquid. The solution ﬂows from bulb C through the capillary and around the sides of the bulb E into bulb B. The volume of liquid in B has no effect on the rate of ﬂow through the capillary because there is no back pressure on the liquid emerging from the capillary as the bulb E is open to atmosphere. The ﬂow time t for the solution meniscus to pass between the ﬁducial marks a and b on bulb C above the capillary is noted. Since the volume of solution in B has no effect on the ﬂow time t, the solution in B can be diluted in situ by adding a measured amount of solvent through A. The diluted solution, whose concentration is easily calculated from the solvent added, is then raised up into C, as before, and the new ﬂow time is measured. In this way, the concentration of the solution in B can be changed by successive dilution with measured volumes of solvent and the corresponding ﬂow times can be determined. Denoting the terms related to solvent with subscript zero, a ratio of viscosities of solution (h) and solvent (ho) in terms of respective ﬂow times (t and to) for the same volume and through the same capillary is given by h r t = ho ro to (1.23) Equation 1.23 can be derived [5] from the Hagen–Poiseuille equation relating liquid viscosity to volumetric ﬂow rate for ﬂow through a capillary tube of given length and radius and under a given pressure difference between the ends of the tube. The ratio h/ho is known as the relative viscosity (hr). Other terminologies commonly used for solution viscosity are as follows: speciﬁc viscosity (hsp) = hr − 1; reduced viscosity (hsp/c) = (hr − 1)/c; inherent 1 h −1 . viscosity (hinh) = (ln hr)/c; intrinsic viscosity (½h) = limc!0 c ho For dilute solutions, r is very close to ro and Equation. 1.23 simpliﬁes to h t = ho to (1.24) Thus, the ratio of viscosities needed for the determination of intrinsic viscosity [h], deﬁned as above, can be obtained simply from ﬂow times without measuring absolute viscosities. Since, however, the h/ho ratios are obtained from Equation 1.24 with measurements made at ﬁnite concentrations of the solution, it becomes necessary to extrapolate the data to zero concentration in order to satisfy the deﬁnition of [h]. 12 Plastics Technology Handbook ηinh × 10–2 (cm3 g–1) (ηsp/c) × 10–2 (cm3 g–1) 1.6 1.4 1.2 1.0 0 0.4 0.1 0.2 0.3 Concentration, c × 102 (g cm–3) FIGURE 1.9 A typical plot of hsp/c and hinh against c. (From Chanda, M. 2013. Introduction to Polymer Science and Chemistry. A Problem Solving Approach, CRC Press, Boca Raton, FL.) There are a variety of ways to carry out this extrapolation. The variation in solution viscosity (h) with increasing concentration (c) can be expressed as a power series in c. The equations usually used are the Huggins equation [5] hsp 1 h = − 1 = ½h + kH ½h2 c + k0H ½h3 c2 + … c c ho (1.25) and the Kraemer equation [5] hinh = ln (h=ho ) = ½h – k1 ½h2 c – k1′ ½h3 c – … c (1.26) It is easy to show that both equations should extrapolate to a common intercept equal to [h]. The usual calculation procedure thus involves a double extrapolation of Equations 1.25 and 1.26 on the same plot V from the MHS equation. (see Figure 1.9) to determine [h] and hence M 1.2.4 Polydispersity Index The ratio of weight-average molecular weight to number-average molecular weight is called the dispersion or polydispersity index (I). It is a measure of the width of the molecular-weight distribution curve (Figure 1.10) and is used as such for characterization purposes. Normally I is between 1.5 and 2.5, but it may range to 15 or greater. The higher the value of I is, the greater is the spread of the molecular-weight distribution of the polymer. For a monodisperse system (e.g., pure chemicals), I = 1. There is usually a molecular size for which a given polymer property will be optimum for a particular application. So a polymer sample containing the greatest number of molecules of that size will have the optimum property. Since samples with the same average molecular weight may possess different molecular-weight distributions, information regarding molecular-weight distribution is necessary for a proper choice of polymer for optimum performance. A variety of fractionation techniques, such as fractional precipitation, precipitation chromatography, and gel permeation chromatography (GPC), based on properties such as solubility and permeability, which vary with molecular weight, may be used for separating polymers of narrow size ranges. 13 Characteristics of Polymers and Polymerization Processes 0.5 Weight fraction 0.4 0.3 0.2 0.1 0 FIGURE 1.10 4 12 16 20 24 28 32 Mean molecular weight 8 36 40 × 103 Molecular-weight distribution of a polymer. Example 1: A sample of poly(vinyl chloride) is composed according to the following fractional distribution (Figure 1.10). Wt fraction 0.04 0.23 0.31 0.25 0.13 0.04 Mean mol. wt × 10−3 7 11 16 23 31 39 n, M w , DPn , and DPw . (a) Compute M (b) How many molecules per gram are there in the polymer? Answer: (a) Mean mol. wt (Mi) Wt fraction (wi) wi × Mi wi/Mi 0.04 7,000 280 0.57 × 10−5 0.23 11,000 2,530 2.09 × 10−5 0.31 0.25 16,000 23,000 4,960 5,750 1.94 × 10−5 1.90 × 10−5 0.13 31,000 4,030 0.42 × 10−5 0.04 S 39,000 1,560 19,110 0.10 × 10−5 6.21 × 10−5 From Equation 1.2 n = M 1 = 16,100 g=mole 6:21 10−5 From Equation 1.14 w = 19,110 g=mole M 1 mer weight of vinyl chloride (C2H3Cl) = (2)(12)+(3)(1)+35.5 = 62.5 g/mer DPn = 16,000 g=mole = 258 mers=mole 62:5 g=mer DPw = 19,110 g=mole = 306 mers=mole 62:5 g=mer 14 Plastics Technology Handbook (b) Number of molecules per gram = X wi (Avogadro number) Mi = (6:21 10−5 )(6:02 1023 ) = 3:74 1019 molecules=g 1.2.4.1 Gel Permeation Chromatography Gel permeation chromatography (GPC) is an extremely powerful method for determining the complete molecular weight distribution (MWD) and average molecular weights [5]. It is essentially a process for the separation of polymer molecules according to their size. The separation occurs as a dilute polymer solution is injected into a solvent stream, which then passes through a column packed with porous gel particles, with the porosity being typically in the range 50–106 Å. GPC is also known as gel ﬁltration, gel exclusion chromatography, size-exclusion chromatography, and molecular sieve chromatography. The principle of GPC is simple. It is shown schematically in Figure 1.11. A schematic layout of a typical GPC system is shown in Figure 1.12. Consider a stationary column packed with ﬁnely divided solid particles, all having the same pore size. The smaller polymer molecules that are able to enter the pores (tunnels) of the gel particles will have longer effective paths than larger molecules and will hence be “delayed” in their passage (elution) through the column. On the other hand, larger polymer molecules with a coil size greater than the pore diameter will be unable to enter the pores and will thus be swept along with the solvent front to appear in the exit from the GPC column (Figure 1.11) and reach the detector (Figure 1.12) ahead of the smaller molecules. The volume of solvent thus required to elute a particular polymer species from the point of injection to the detector is known as its elution volume. Molecular weight can be calculated from the GPC data only after calibration of the GPC system in terms of elution volume or retention time with polymer standards of known molecular weights (see below). The most common type of column packing used for analysis of synthetic polymers consists of polystyrene gels, In Substrate particle with pores (tunnels) Out FIGURE 1.11 A schematic of the principle of separation by gel permeation chromatography. Black circles represent molecules of coil sizes ≤ pore diameter, while crosses represent molecules of coil sizes > pore diameter. If a sample with molecular size distribution enters the column at the same time, the molecules will emerge from the column sequentially, separated according to molecular size, from larger to smaller. 15 Characteristics of Polymers and Polymerization Processes Pump with pressure gauge Eﬄuent Detector Recorder Collector Solvent reservoir Filter Columns Sample injection FIGURE 1.12 Schematic layout of a typical gel permeation chromatography apparatus. It is, however, normal practice to use a set of several columns each packed with porous gel particles having a different porosity, depending on the range of molecular sizes to be analyzed. (Adapted from Allcock, H. R. and Lampe, F. W., Contemporary Polymer Chemistry, 2nd ed., 1990. Prentice Hall, Englewood Cliffs, NJ.) called styragel particles (hence the term gel permeation). These are highly porous polystyrene beads and are highly cross-linked. For GPC in aqueous systems, the common packings are cross-linked dextran (Sephadex) and polyacrylamide (Biogel). Porous glass beads can be used with both aqueous and organic solvents. After passage through the column(s), the solvent stream (eluant) carrying the size-separated polymer molecules passes through a detector, which responds to the weight concentration of polymer in the eluant. The most commonly used detector is a differential refractometer. It measures the difference in refractive index between the eluted solution and the pure solvent. This difference is proportional to the amount of polymer in solution. Spectrophotometers are also used as alternative or auxiliary detectors. The elution volume (also called the retention volume) is the volume of solvent that has passed through the GPC column from the time of injection of the sample. It is conveniently monitored by means of a small siphon, which actuates a marker every time it ﬁlls with eluant and dumps its contents. The raw GPC data are thus available as a trace of detector response proportional to the amount of polymer in solution and the corresponding elution volumes. A typical GPC record (gel permeation chromatogram) is shown in Figure 1.13. It can yield a plot of the MWD, since the ordinate corresponding to the detector response can be transformed into a weight fraction of total polymer while the elution volume axis can be transformed into a logarithmic molecular weight scale by suitable calibration (explained below). Using a baseline drawn through the recorder trace, the chromatogram heights are measured for equal small increments of elution volume. The weight fraction corresponding to a particular elution volume is taken as the height of the ordinate divided by the sum of the heights of all the ordinates under the trace. This process normalizes the chromatogram (see Example 3). Let us now consider how the elution volume axis of a chromatogram, such as shown in Figure 1.13, can be translated into a molecular weight scale. This necessitates calibration of the particular GPC column using monodisperse polymer samples. The main problem encountered in this task is that monodisperse or very narrow distribution samples of most polymers are not generally available. However, such samples are available for a few speciﬁc polymers. For polystyrene, for example, anionically polymerized samples of narrow MWDs with polydispersity index less than 1.15 are commercially available in a wide range of molecular weights (103 to 106). Using such narrow MWD samples, a polystyrene calibration of molecular 16 Plastics Technology Handbook Elution volume FIGURE 1.13 A typical gel permeation chromatogram. The lower trace with short vertical lines is the differential refractive index while the upper curve is an absorption plot at a ﬁxed ultraviolet frequency. The short vertical lines are syphon dumps counted from the time of injection of the sample. The units of the ordinate depend on the detector, while those of the abscissa can be in terms of syphon volumes (counts) or volume of solvent. weight versus elution volume can be easily obtained for the given GPC column (or columns) and the given GPC solvent. A problem that would then remain is to establish a relationship for the particular GPC column between the elution volume and molecular weight of some other chemically different polymer. An approach to solving this problem is described below. A series of narrow MWD polystyrene samples used with the particular GPC column and the GPC solvent yield a set of GPC chromatograms, as shown in Figure 1.14. The peak elution volumes and the corresponding molecular weights thus provide a polystyrene calibration curve (Figure 1.15) for the particular GPC column and solvent used. In the next step, known as universal calibration, the polystyrene calibration curve is translated to one that will be effective for another given polymer in the same apparatus and solvent. To extend the calibration to other polymers, a calibration parameter that is independent of the chemical nature of the polymer, that is, a universal calibration parameter, is required. Experimentally, it has been found that such a parameter could be the product of the intrinsic viscosity and molecular weight (i.e., [h]M). Thus, as shown in Figure 1.16, the logarithm of the product [h]M plotted against elution volume, with tetrahydrofuran used as the solvent, provides a single curve for a wide variety of polymers, which thus suggests that a universal calibration procedure may be possible. (Such a single curve for different polymers is not obtained, however, by simply plotting logM against elution volume.) A theoretical validity of the aforesaid experimental observation can also be obtained from a consideration of the hydrodynamic volume of the polymer, as shown below. Amount of polymer in eluant Increasing molecular weight Elution volume FIGURE 1.14 Gel permeation chromatography elution curves for polymer standards having very narrow molecular weight distribution. 17 Characteristics of Polymers and Polymerization Processes 100 Molecular weight, Mx × 10–4 40 10 4 1.0 0.4 0.1 100 120 140 160 180 Peak elution volume, Ve (cm3) FIGURE 1.15 A typical polystyrene standard calibration curve for GPC. 109 (η) M 108 107 106 105 Polystyrene (“linear”) Polystyrene (“comb”) Polystyrene (“star”) Poly(methyl methacrylate) (linear) Poly(vinyl chloride) Poly(butadiene) Poly(phenyl siloxane) Poly(styrene-co-methyl methacrylate) Poly(styrene-g-methyl methacrylate) Poly(styrene-g-methyl methacrylate) (comb) 18 20 22 24 26 28 Elution volume (5-mL counts, THF solvent) 30 FIGURE 1.16 A universal calibration curve for several polymers in tetrahydrofuran. (Drawn with data of Grubisic, Z., Rempp, P., and Benoit, H. 1967. Polymer Lett. 5, 753.) 18 Plastics Technology Handbook As long ago as 1906, it was shown by Einstein that the viscosity of a dilute suspension of spherical particles relative to that of the suspending medium is given by the expression [5] h=ho − 1 = 2:5j (1.27) where f is the volume fraction of the suspended material. If all polymer molecules exist in solution as discrete entities, without overlap and each solvated molecule of a monodisperse polymer has an equivalent volume (or hydrodynamic volume) V and molecular weight M, then the volume fraction f of solvent-swollen polymer coils at a concentration c (mass/volume) is j = c V NAV =M (1.28) where NAV is Avogadro’s number. Combination of Equations 1.27 and 1.28 yields 1 h − ho 2:5VNAV = ho M c (1.29) Using this equation and applying the deﬁnition of intrinsic viscosity [h] given earlier, 1 ½h = limc!0 c h − ho ho = 2:5NAV limc!0 V M (1.30) Multiplying both sides by M and taking logarithm gives logð½hM Þ = log (2:5NAV ) + logðlimc!0 V Þ (1.31) The product [h]M is thus seen to be a direct function of the hydrodynamic volume of the solute at inﬁnite dilution. Since studies of GPC separations have shown that polymers appear in the eluate in inverse order of their hydrodynamic volumes in the particular solvent, it may thus be stated that two different polymers that appear at the same elution volume in a given solvent and particular GPC column at a given temperature have the same hydrodynamic volumes and hence the same [h]M characteristics [5]; that is, logð½hx Mx Þ = logð½hs Ms Þ (1.32) where the subscripts x and s indicate the unknown polymer X (i.e., polymer with unknown molecular weight) and the standard polymer S, respectively. If each intrinsic viscosity term in Equation 1.27 is replaced by its MHS expression (Equation 1.21), one obtains for the two polymers at equal elution volumes log Kx Mx ax+1 = log Ks Ms as+1 (1.33) Solving for log Mx gives log Mx = 1 1 + ax log Ks 1 + as + log Ms Kx 1 + ax (1.34) The elution volume (Ve) that corresponds to a GPC peak in the unknown polymer is used to obtain a value of log Ms from the polystyrene standard curve (Figure 1.15) that has been obtained in the same column and solvent, and Mx is then calculated from Equation 1.34. Alternatively, a number of values of Ve can be chosen and a new calibration curve for the polymer X can be constructed using a standard curve 19 Characteristics of Polymers and Polymerization Processes such as Figure 1.15 and Equation 1.34. The procedure is applicable only if the MHS constants, Ks, as, Kx, and ax, are known. Values of Ks and as are available in the literature for the standard polymer in various solvents, and in many cases, values of Kx and ax may also be available. However, if the desired MHS constants are not available for the polymer under study in the GPC solvent or for the standard in the same solvent, they can be determined by measuring the intrinsic viscosities, as described earlier. Example 2: A series of narrow distribution polystyrene standards dissolved in chloroform were injected into a GPC column at 35°C yielding a set of chromatograms. The following data of peak elution volumes and corresponding sample molecular weights were reported (Dawkins, J. V. and Hemming, M. 1975. Makromol. Chem. 176, 1777): s 10−3 (g/mol) M Ve (cm3) 867 670 411 160 98.2 51 19.8 10.3 3.7 122.7 126.0 129.0 136.5 141.0 147.0 156.5 162.5 170.0 Using the above data for polystyrene standards, construct a calibration curve for the molecular weight elution volume of polymer X in chloroform at 35°C. The MHS constants in chloroform at 35°C may be taken as K = 4.9 × 10−3 cm3/g, a = 0.79 for polystyrene and K = 5.4 × 10−3 cm3/g, a = 0.77 for polymer X. s versus Ve gives the polystyrene calibration curve (cf. Answer: A semilogarithmic plot of M Figure 1.15) for the given GPC column, solvent, and temperature. Substitution of the MHS constants in Equation 1.33 gives the expression log Mx = −0.0238 + 1.0113 log Ms. To construct an elution calibration curve (Mx vs. Ve) for polymer X, various values of Ve are assumed and corresponding to each of them the Ms value is ﬁrst obtained from the polystyrene calibration curve (Figure 1.15) and then Mx from the above expression. A semilog plot of Mx versus Ve gives the required calibration curve (Figure 1.17). Standard molecular weight, Ms × 10–4 Example 3: The GPC column of Example 2 was used for the determination of molecular weight of a sample of the same polymer X. After injecting a chloroform solution of the polymer into the GPC 100 40 10 4 1 0.4 0.1 100 140 180 Peak elution volume, Ve (cm3) FIGURE 1.17 Elution calibration curve (Mx vs. Ve) for polymer X (Example 2) derived from polystyrene calibration curve and MHS constants. 20 Plastics Technology Handbook column, the refractive index difference (Dň) between the eluted solution and pure solvent was measured as a function of elution volume (Ve), which yielded the following data [5]: Dň × 105 Ve (4-mL count) 0.6 40 3.4 39 12.4 38 15.0 37 9.9 36 3.0 35 0.4 34 n and M w and the polydispersity Using the calibration curve obtained in Example 2, calculate M index of the sample. Answer: The molecular weight corresponding to each elution volume is determined from the elution calibration curve for this polymer in Figure 1.17. The corresponding weight fraction wi is computed from the refractive index difference by the following relation based on the assumption that Dň is proportional to concentration and the proportionality factor is independent of molecular weight: wi = Dň/SDň. The results are tabulated below. wi = Dňi/SDň Mi × 10−3 (from Figure 1.17) 34 0.009 182 1620 0.049 35 36 0.067 0.221 117 76 7851 16,834 0.573 2.914 37 0.336 50 16,780 6.712 38 39 0.277 0.076 33 21 9154 1598 8.406 3.624 40 0.013 14 187 0.957 S = 54 × 103 S = 23.236 Elution volume (4-mL count) wiMi (wi/Mi) × 106 n = 1=S(wi =Mi ) = 43 103 g mol−1; polydispersity w = Swi Mi = 54 103 g mol−1M Therefore, M index = Mw =Mn = 1:25. 1.3 Polymerization Reactions There are two fundamental polymerization reactions. Classically, they have been differentiated as addition polymerization and condensation polymerization. In the addition process, no by-product is evolved, as in the polymerization of vinyl chloride (see below); whereas in the condensation process, just as in various condensation reactions (e.g., esteriﬁcation, etheriﬁcation, amidation, etc.) of organic chemistry, a lowmolecular-weight by-product (e.g., H2O, HCl, etc.) is evolved. Polymers formed by addition polymerization do so by the successive addition of unsaturated monomer units in a chain reaction promoted by the active center. Therefore, addition polymerization is called chain polymerization. Similarly, condensation polymerization is referred to as step polymerization since the polymers in this case are formed by stepwise, intermolecular condensation of reactive groups. Another polymerization process that has now appeared as a new research area of considerable interest is supramolecular polymerization (see later). 1.3.1 Addition or Chain Polymerization In chain polymerization, a simple, low-molecular-weight molecule possessing a double bond, referred to in this context as a monomer, is treated so that the double bond opens up and the resulting free valences 21 Characteristics of Polymers and Polymerization Processes join with those of other molecules to form a polymer chain. For example, vinyl chloride polymerizes to poly(vinyl chloride): H2C Polymerization CH CH CH2 Cl n (1.35) Cl Poly(vinyl chloride) Vinyl chloride It is evident that no side products are formed; consequently the composition of the mer or repeating unit of the polymer (–CH2–CHCl–) is identical to that of the monomer (CH2═CHCl). The identical composition of the repeating unit of a polymer and its monomer(s) is, in most cases, an indication that the polymer is an addition polymer formed by chain polymerization process. The common addition polymers and the monomers from which they are produced are shown in Table 1.1. Chain polymerization involves three processes: chain initiation, chain propagation, and chain termination. (A fourth process, chain transfer, may also be involved, but it may be regarded as a combination of chain termination and chain initiation.) Chain initiation occurs by an attack on the monomer molecule by a free radical, a cation, or an anion; accordingly, the chain polymerization processes are called free-radical polymerization, cationic polymerization, or anionic polymerization. (In coordination addition or chain polymerization, described below separately, the chain initiation step is, however, assumed to be the insertion of the ﬁrst monomer molecule into a transition metal–carbon bond.) A free radical is a reactive substance having an unpaired electron and is usually formed by the decomposition of a relatively unstable material called an initiator. Benzoyl peroxide is a common free-radical initiator and can produce free radicals by thermal decomposition as O O O R C O O C R R C . . O + R + CO2 (1.36) (R = Phenyl group for benzoyl peroxide initiator) Free radicals are, in general, very active because of the presence of unpaired electrons (denoted by dot). A free-radical species can thus react to open the double bond of a vinyl monomer and add to one side of the broken bond, with the reactive center (unpaired electron) being transferred to the other side of the broken bond: O H + H2C R C O O H C R C O CH2 C X X (1.37) (X = CH3, C6H5, Cl, etc.) The new species, which is also a free radical, is able to attack a second monomer molecule in a similar way, transferring its reactive center to the attacked molecule. The process is repeated, and the chain continues to grow as a large number of monomer molecules are successively added to propagate the reactive center: H H R C O CH2 C O X Successive addition of monomer R C O O ( CH2 C )m (1.38) X This process of propagation continues until another process intervenes and destroys the reactive center, resulting in the termination of the polymer growth. There may be several termination reactions depending on the type of the reactive center and the reaction conditions. For example, two growing Vinylacetate CH2 CH CH CH3 CH CH2 CH 5. CH2 CH2 Acrylonitrile H2C Styrene Propylene Ethylene CH2 4. 3. 2. 1. Monomer O O C CN CH3 Poly(vinyl acetate) Polyacrylonitrile Polystyrene (PS) Polypropylene (PP) Polyethylene (PE) TABLE 1.1 Typical Addition Polymers (Homopolymers) CH2 CH2 CH2 O CH3 C n n CH CN CH O n n CH3 CH CH2 CH CH2 CH2 Polymer n (Continued) Emulsion paints, adhesives, sizing, chewing gum, e.g., Flovic, Mowilith, Mowicoll. Widely used as ﬁbers; best alternative to wool for sweaters, e.g., Orlon, Acrilan. Transparent and brittle; used for cheap molded objects, e.g., Styron, Carinex, Hostyren, Lustrex. Modiﬁed with rubber to improve toughness, e.g., High impact Polystyrene (HIPS) and acrylonitrile–butadiene– styrene copolymer (ABS). Expanded by volatilization of a blended blowing agent (e.g., pentane) to make polystyrene foam, e.g., Styrocell, Styrofoam. High density polyethylene (HDPE) and low density polyethylene (LDPE); molded objects, tubing, ﬁlm, electrical insulation, used for household products, insulators, pipes, toys, bottles, e.g., Alkathene, Lupolan, Hostalen, Marlex. Lower density, stiffer, and higher temperature resistance than PE; used for water pipes, integral hinges, sterilizable hospital equipment, e.g., Propathene, Novolen, Moplen, Hostalen, Marlex. Comments 22 Plastics Technology Handbook 11. 10. 9. 8. CH2 Butadiene CH2 Isoprene Isobutylene CH2 CH CH CH3 C CH3 C CH2 CH3 Methyl methacrylate Tetraﬂuroethylene CF2 7. O CH CH2 CH2 OCH3 C C CH3 CF2 CH Cl Vinyl chloride CH2 6. CF2 cis-1,4-Polybutadiene cis-1,4-Polyisoprene Polyisobutylene (PIB) CH2 CH2 CH n CH2 n Cl CH CH CH2 CH CH2 CH3 C CH3 CF2 CH3 C CH2 Poly(methyl methacrylate) (PMMA) Polytetraﬂuroethylene CH2 Polymer Poly(vinyl chloride) (PVC) Typical Addition Polymers (Homopolymers) Monomer TABLE 1.1 (CONTINUED) O n n OCH3 C C CH3 n Tires and tire products, e.g., Cis-4, Ameripol-CB, Diene. Tires, mechanical goods, footwear, sealants, caulking compounds, e.g., Coral, Natsyn, Clariﬂex I. Lubricating oils, sealants, copolymerized with 0.5– 2.5 mol% isoprene to produce Butyl rubber for tire inner tubes and inner liners of tubeless tires. Transparent sheets and moldings; more expensive than PS; known as organic glass, used for aeroplane windows; e.g., Perspex, Plexiglass, Lucite, Diakon, Vedril. High temperature resistance, chemically inert, excellent electrical insulator, very low coefﬁcient of friction, expensive; moldings, ﬁlms, coatings, used for non-stick surfaces, insulation, gaskets; e.g. Teﬂon, Fluon. Water pipes, bottles, gramophone records, plasticized to make PVC ﬁlm, leather cloth, raincoats, ﬂexible pipe, tube, hose, toys, electrical cable sheathing, e.g., Benvic, Darvic, Geon, Hostalit, Solvic, Vinoﬂex, Welvic. Comments Characteristics of Polymers and Polymerization Processes 23 24 Plastics Technology Handbook radicals may combine to annihilate each other’s growth activity and form an inactive polymer molecule; this is called termination by combination or coupling: H H R C O C .m + . C CH2 O X X R C O CH2 n O C CH2 O R O (1.39) H H C m C CH2 n O C X X O R A second termination mechanism is disproportionation, shown by the following equation: H R C O O R C O O H . .C CH2 X X O C n O H H H CH2 C m + CH2 C m–1 CH2 C H + C X X X R (1.40) H CH C X CH2 n–1 O C R O In chain polymerization initiated by free radicals, as in the previous example, the reactive center, located at the growing end of the molecule, is a free radical. Similarly, in chain polymerizations initiated by ionic systems, the reactive center is ionic, i.e., a carbonium ion (in cationic initiation) or a carbanion (in anionic initiation). Regardless of the chain initiation mechanism—free radical, cationic, or anionic—once a reactive center is produced it adds many more molecules in a chain reaction and grows quite large extremely rapidly, usually within a few seconds or less. (However, the relative slowness of the initiation stage causes the overall rate of reaction to be slow and the conversion of all monomers to polymers in most polymerizations requires at least 30 min, sometimes hours.) Evidently, at any time during a chain polymerization process the reaction mixture will consist only of unreacted monomers, high polymers and unreacted initiator species, but no intermediate sized molecules. The chain polymerization will thus show the presence of high-molecular-weight polymer molecules at all extents of conversion (see Figure 1.18). In certain ionic chain polymerizations, which feature a fast initiation process coupled with the absence of reactions that terminate the propagating reactive centers, molecular weight increases linearly with conversion. This is known as “living” ionic chain polymerization. 1.3.2 Coordination Addition Polymerization Many polymers are now manufactured on a commercial scale using Ziegler–Natta catalysts, an outstanding example being polypropylene of high molecular weight which cannot be made by commercial processes of free-radical or ionic chain polymerization. Perhaps the best known Ziegler–Natta systems are those derived from TiCl4 or TiCl3 and an aluminum alkyl. The catalyst systems appear to function by formation of a coordination complex between the catalyst, growing chain, and incoming monomer. Hence the process is referred to as coordination addition polymerization and the catalysts as coordination catalysts. Polymers with stereoregular structures (see later) can be produced with these catalysts. The efﬁciency or activity of the early Ziegler–Natta catalyst systems was low. The term activity usually refers to the rate of polymerization, expressed in terms of kilograms of polymer formed per gram of catalyst. Thus a low activity meant that large amounts of catalyst were needed to obtain reasonably high yields of polymer, and the spent catalyst had then to be removed from the product to avoid contamination. This problem effectively disappeared with the advent of subsequent generations of catalysts leading to large increases in activity without loss of stereospeciﬁcity. This was achieved by increasing the 25 Characteristics of Polymers and Polymerization Processes Molecular weight (b) (c) (a) 0 20 40 60 80 100 % Conversion FIGURE 1.18 Variation of molecular weight with conversion in (a) step polymerization, (b) free-radical polymerization, and (c) ionic chain polymerization. effective surface area of the active component by more than two orders of magnitude through impregnation of the catalyst on a solid support such as MgCl2 or MgO. For example, in contrast to a typical TiCl3–AlR3 catalyst which yields about 50–200 g of polyethylene per gram of catalyst per hour per atmosphere of ethylene, as much as 200,000 g of polyethylene and over 40,000 g of polypropylene per gram titanium per hour may be produced using a MgCl2-supported catalyst, thus obviating the need to remove the spent catalyst (a costly step) from the product. Such catalyst systems are often referred to as high-mileage catalysts. Stereospeciﬁcity of the catalyst is kept high (>90%–98% isotactic dyads) by adding electron-donor additives such as ethyl benzoate. The catalyst complex of the TiCl3/AlR3 system essentially acts as a template for the successive orientation and isotactic placement of the incoming monomer units. Though a number of structures have been proposed for the active species, they fall into either of two general categories: monometallic and bimetallic, depending on the number of metal centers. The two types can be illustrated by the structures (I) and (II) for the active species from titanium chloride (TiCl4 or TiCl3) and alkylaluminum (AlR3 or AlR2Cl). R (R)CI TI (R)CI AI CI (I) R (R)CI CI CI (R)CI CI TI CI (II) Structure (I), representing a bimetallic species, is the coordination complex that arises from the interaction of the original catalyst components (titanium and aluminum compounds) with exchange of R and Cl groups. The placing of R and Cl groups in parentheses signiﬁes that the exact speciﬁcation of the ligands on Ti and Al cannot be made. Structure (II), representing a typical monometallic species, constitutes an active titanium site at the surface of a TiCl3 crystal. Besides the four chloride ligands that the central Ti atom shares with its neighboring Ti atoms, it has an alkyl ligand (received through exchange reactions with alkyl aluminum) and a vacant orbital (□). 26 Plastics Technology Handbook The truly active bimetallic catalysts are complexes that have an electron-deﬁcient bond, e.g., Ti⋯C⋯Al in (I). Chain propagation by the bimetallic mechanism [7] occurs at two metal centers of the bridge complex as shown in Figure 1.19. With the chain growth taking place always from the metal end, the incoming monomer is oriented, for steric reasons, with the ═CH2 group pointing into the lattice and the CH3 group to one side, with the result that the process leads to the formation of an isotactic polymer. While a limited amount of experimental evidence does lend support to the bimetallic concept, majority opinion, however, favors the second and simpler alternative, the monometallic mechanism (described next). It is generally accepted that the d-orbital in the transition element is the main source of catalytic activity and that it is the Ti-alkyl bond that acts as the polymerization center where chain growth occurs. For aTiCl3 catalyst the active center [8] is formed by the interaction of aluminum alkyl with an octahedral vacancy around Ti, as shown in Figure 1.20. To elaborate, the ﬁve-coordinated Ti3+ on the surface has a vacant d-orbital, represented by -□, which facilitates chemisorption of the aluminum alkyl and this is followed by alkylation of the Ti3+ ion by an exchange mechanism to form the active center TiRCl4-□. The vacant site at the active center can accommodate the incoming monomer unit, which forms a p-complex with the titanium at the vacant d-orbital and is then inserted into the Ti-alkyl bond. The sequence of steps is shown in Figure 1.21 using propylene as the monomer. CH CH (a) CH3 Incoming monomer CH3 CH CH3 2 CH CH2 Al Ti Cl CH2 Ti Al Partial delocalization of alkyl CH 3 bridge CH2 CH CH CH2 Monomer insertion CH2 (d) Al Ti CH2 Al Ti Cl CH3 Another sequence of monomer addition CH3 CH2 CH Six-membered ring transition state CH Cl Al Ti Sequences CH3 CH2 Repeated Al Ti (c) Polymer chain attached to metal end CH3 CH2 (e) CH3 Cl CH2 CH (b) Cl Bridge complex CH3 CH3 2 (f) Cl FIGURE 1.19 Bimetallic mechanism for stereospeciﬁc polymerization. (Adapted from Patat, F. and Sinn, H. 1958. Angew. Chem., 70, 496.) 27 Characteristics of Polymers and Polymerization Processes Chemisorbed C H C H 2 5 2 5 Al-alkyl Vacant d-orbital Cl Cl Cl Al Cl Ti Cl Al(C2H5)3 Cl Cl Cl Cl Alkylation of Ti3 ion Ti Cl C2H5 TiCl3 C2H5 Cl ClAl(C2H5)2 Ti Cl Cl Cl Vacant site FIGURE 1.20 Interaction of aluminum alkyl with an octahedral vacancy around Ti in the ﬁrst stage of monometallic mechanism. (After Cossee, P. 1967. The Stereochemistry of Macromolecules, A. D. Ketley, ed., Vol. 6. Marcel Dekker, New York.) C2H5 CI C2H5 CI CH Ti CH3 CI Vacant site CI C Monomer CI CH CI CI H H C2H5 CI CH3 CI Migration of alkyl chain CI CI π-complex of monomer Ti CI C CI Active center at Ti CH2 H Ti CI CH2 CH3 New active center CI CI CI C H CH3 CH2 2 CH CI CH 3 CH2 CH CH3 C2H5 CI H CH3 C2H5 Momomer insertion between Ti-alkyl bond Another sequence of steps CH CI Ti C2H5 C Ti CI C CI H H Transition state before insertion of monomer Growing polymer chain Ti Vacant site at original position CI CI FIGURE 1.21 Monometallic mechanism for stereospeciﬁc polymerization. (After Cossee, P. 1967. The Stereochemistry of Macromolecules, A. D. Ketley, ed., Vol. 6. Marcel Dekker, New York.) After the monomer is inserted into the Ti-alkyl bond, the polymer chain migrates back to its initial position, while the vacant site migrates to its original position to accept another monomer molecule. This migration is necessary, as otherwise an alternating position would be offered to the monomer leading to the formation of a syndiotactic polymer instead of an isotactic polymer. The termination of a polymer chain growing at an active center may occur by various reactions, as shown below with propylene as the example. 28 Plastics Technology Handbook 1. Chain Transfer to Monomer Ti CH2 CH(CH3) + CH3 CH CH2 ktr,M Ti CH2CH2CH3 (1.41) + CH2 C(CH3) Ti CH2 CH(CH3) + CH3 CH CH2 ktr,M Ti CH + CH3 CH CH3 (1.42) CH(CH3) where •–Ti represents the transition metal active center on the catalyst site at which chain propagation takes place. Note that it is the methylene carbon atom from the monomer that is bonded to the transition metal atom (cf. Figure 1.49). 2. Chain transfer to the Group I–III metal alkyl: Ti CH2 CH(CH3) + Al(C2H5)3 Ti CH2CH3 + (C2H5)2Al – CH2 (1.43) CH(CH3) 3. Spontaneous intramolecular b-hydride transfer: Ti CH2 ks CH(CH3) Ti H + CH2 C(CH3) (1.44) 4. Chain transfer to an active hydrogen compound such as molecular hydrogen (external agent): Ti CH2 CH(CH3) + H2 ktr,H2 Ti H + CH3 CH(CH3) (1.45) The above reactions terminating the growth of polymer chains are indeed chain transfer reactions since in each case a new propagating chain is initiated. The relative extents of these reactions depend on various factors such as the monomer, the initiator components, temperature, concentrations, and other reaction conditions. Under normal conditions of polymerization, intramolecular hydride transfer is negligible and termination of propagating chains occurs mostly by chain transfer processes. Being a highly effective chain transfer agent, molecular hydrogen is often used for polymer molecular weight control. 1.3.3 Step Polymerization Step polymerization occurs by stepwise reaction between functional groups of reactants. The reaction leads successively from monomer to dimer, trimer, tetramer, pentamer, and so on, until ﬁnally a polymer molecule with large DP is formed. Note, however, that reactions occur at random between the intermediates (e.g., dimers, trimers, etc.) and the monomer as well as among the intermediates themselves. In other words, reactions of both types, namely, n mer + monomer ! (n + 1) mer and n mer + m mer ! (n + m) mer occur equally. Thus, at any stage the product consists of molecules of varying sizes, giving a range of molecular weights. The average molecular weight builds up slowly in the step polymerization process, and a high-molecular-weight product is formed only after a sufﬁciently long reaction time when the conversion is more than 98% (see Figure 1.18a). Since most (though not all) of the step polymerization processes involve poly-condensation (repeated condensation) reactions, the terms step polymerization and condensation polymerization are often used synonymously. Consider, for example, the synthesis of a polyamide, i.e., a polymer with amide 29 Characteristics of Polymers and Polymerization Processes (–CONH–) as the characteristic linkage. If we start with, say, hexamethylenediamine and adipic acid as reactants, the ﬁrst step in the formation of the polymer (nylon) is the following reaction producing a monoamide: = = H2N – (CH2)6 – NH2 + HO – C – (CH2)4 – C – OH O O (1.46) = = H2N – (CH2)6 – NH – C – (CH2)4 – C – OH + H2O O O The reaction continues step-by-step to give the polyamide nylon-6,6. The overall reaction may thus be represented as n H2N CH2 NH2 + n HO C 6 CH2 4 H NH CH2 6 NH C CH2 C OH O O 4 C (1.47) + n OH H2O O O Poly(hexamethylene adipamide) We see that the composition of the repeating unit (enclosed in square brackets) equals that of two monomer molecules minus two molecules of water. Thus a condensation polymer may be deﬁned as one whose synthesis involves elimination of small molecules or whose repeating unit lacks certain atoms present in the monomer(s). With the development of polymer science and the synthesis of new polymers, the previous deﬁnition of condensation polymer is inadequate. For example, in polyurethanes (Table 1.2), which are classiﬁed as condensation polymers, the repeating unit has the same net composition as the two monomers (i.e., a diol and a diisocyanate), which react without eliminating any small molecule. To overcome such problems, chemists have introduced a deﬁnition which describes condensation polymers as consisting of structural units joined by internal functional groups such as = = = = = – = – CO– ester –C–O– , amide – C –NH – , imide – N , urethane – O –C – NH – , O O CO– O O sulfide –S– , ether –O– , carbonate – O – C – O – , and sulfone – S – linkages. O O A polymer satisfying either or both of the above deﬁnitions is classiﬁed as a condensation polymer. Phenol–formaldehyde, for example, satisﬁes the ﬁrst deﬁnition but not the second. Some condensation polymers along with their repeating units and condensation reactions by which they can be synthesized are shown in Table 1.2. Some high-performance polymers prepared by polycondensation are listed in Table 1.3. The ring-opening polymerizations of cyclic monomers, such as propylene oxide, H3C O CH CH2 CH2 CH O n CH3 (1.48) or –caprolactam O C CH2 5 NH NH CH2 (1.49) 5 CO n Polyurethane (PU) Polyester Polyamide (PA) R R n HO n HO R n HO R n H2N R R n H2N n HO R n H2N O COH OH + n OC N O OH + n R"O C O O + n Cl C O n HOC OH + n HO C O COH NH2 NH2 + TABLE 1.2 Typical Condensation Polymers NCO O C OR" O Cl R NH R O C O COH C OH HO R' R' R' H R' R' O CO O C H n H HO O + OH HO n H H (n–1) H2O O OC O OC O R' R' R' O n H O R" n R' R' NHC CO O CO O NHC O NHC (n–1) H2O R R R OC NH R R + NH NH Polymerization Reactiona Cl + (2n–1) HCl OH + (2n–1) H2O R O OC NH R' + (2n–1) R"OH + (2n–1) H2O n n O (n–1) O C O C NCO (Continued) Rubbers, foams, coatings; e.g., Vulkollan, Adiprene C, Chemigum SL, Desmophen A, Moltopren. Textile ﬁbers, ﬁlm, bottles; poly(ethylene terephthalate) (PET) e.g., Terylene, Dacron, Melinex, Mylar. Moldings, ﬁbers, tirecord; poly(hexamethylene adipamide) (Nylon 6,6) e.g., Ultramid A; polycaprolactam (nylon-6), e.g., Ultramid B, Akulon, Perlenka, poly (hexamethylene sebacamide) (Nylon-6,10), e.g., Ultramid S, Zytel. Comments 30 Plastics Technology Handbook a H2N n n H2N n N R Si R OH N + O n CH2 NH2 + n CH2 O C NH2 C NH2 N O C O OH CH2 HO Cl + n Na2 Sx + n CH2 R C OH n HO n Cl R Si R O OH Typical Condensation Polymers Sx n N NH C C N C NH2 N O NH C (n–1) OH NHCH2 n NH CH2 + (n–1) H2O + n H2O + n H2O n + 2n NaCl H + (n–1) H2O CH2 n R Polymerization Reactiona R, R′, R″ represent aliphatic or aromatic ring. The repeating unit of the polymer chain is enclosed in parentheses. Melamine– formaldehyde (MF) Urea–formaldehyde (UF) Phenol–formaldehyde (PF) Polysiloxane Polysulphide TABLE 1.2 (CONTINUED) Particle-board binder resin, paper and textile treatment, molding compounds, coatings, e.g., Beetle, Resolite, Cibanoid. Dinnerware, table tops, coatings, e.g., Formica, Melalam, Cymel. Plywood adhesives, glass– ﬁber insulation, molding compound, e.g., Hitanol, Sirfen, Trolitan. Elastomers, sealants, ﬂuids, e.g., Silastic, Silastomer, Silopren. Adhesives, sealants, binders, hose, e.g., Thiokol. Comments Characteristics of Polymers and Polymerization Processes 31 Polyetheretherketone (PEEK) Polyethersulfone (PES) Polycarbonate (PC) n KO n KO n HO OK + n F O S O CH 3 C CH 3 Polymer Type and Polycondensation Reaction Cl O C O O OH + n Cl C TABLE 1.3 Some High-Performance Condensation Polymers F Cl O S O O O O (n–1) KCl n + CH 3 C CH 3 C O O C O (Continued) Moldings and sheets; transparent, tough and physiologically inert: HCl + (2n–1) used for safety glasses, lenses, n screens and glazings, electrical and electronics, appliances, compact discs, e.g. Merlon, Baylon, Jupilon. Moldings, coatings, membranes; rigid, transparent, self-extinguishing, resistant to heat deformation: used for electrical components, molded circuit boards, appliances operating at high temperatures, e.g., Victrex PES. Moldings, composites, bearings, coatings; very high continuous use + (2n–1) KF temperature (260°C): used in n coatings and insulation for high performance wiring, composite prepregs with carbon ﬁbers, e.g., Victrex PEEK. Comments 32 Plastics Technology Handbook Polyimide Poly(p-phenylene terephthalamide) Poly(phenylene sulphide) (PPS) n O n Cl n Cl O O C O O O + n H2N NH2 N O O C O O NH2 C Cl + n H2N n + (2n–1) NaCl O Cl + n Na2S S Some High-Performance Condensation Polymers Polymer Type and Polycondensation Reaction TABLE 1.3 (CONTINUED) N H H O O N C N O + (2n–1) NaCl Films, coatings, adhesives, laminates; outstanding in heat resistance, ﬂame resistance, abrasion (2n–1) H O 2 + resistance, electrical insulation resistance, resistance to oxidative n degradation, high energy radiation and most chemicals (except strong bases): used in specialist applications, e.g., Kapton, Vespel. n Moldings, composites, coatings; outstanding in heat resistance, ﬂame resistance, chemical resistance and electrical insulation resistance: used for electrical components, mechanical parts, e.g., Ryton, Tedur, Fortron. High modulus ﬁbers; as strong as steel but have one-ﬁfth of weight, ideally suited as tire cord materials and for ballistic vests, e.g., Kevlar, Twaron. Comments Characteristics of Polymers and Polymerization Processes 33 34 Plastics Technology Handbook proceed either by chain or step mechanisms, depending on the particular monomer, reaction conditions, and initiator employed. However, the polymers produced in Equation 1.48 and Equation 1.49 will be structurally classiﬁed as condensation polymers, since they contain functional groups (e.g., ether, amide) in the polymer chain. Such polymerizations thus point out very clearly that one must distinguish between the classiﬁcation based on polymerization mechanism and that based on polymer structure. The two classiﬁcations cannot always be used interchangeably. Both structure and mechanism are usually needed in order to clearly classify a polymer. 1.3.4 Supramolecular Polymerization Supramolecular polymers are a relatively new class of polymers in which monomeric repeating units are held together with directional and reversible (noncovalent) secondary interactions, unlike conventional macromolecular species in which repetition of monomeric units is mainly governed by covalent bonding. A schematic comparison of a covalent polymer and a supramolecular polymer is shown in Figure 1.22. The directionality and strength of the supramolecular bonding, such as hydrogen bonding, metal coordination, and p–p interactions, are important features resulting in polymer properties in dilute and concentrated solutions, as well as in the bulk. It should be noted that supramolecular interactions are not new to polymer science, where hydrogen bonding and other weak reversible interactions are important in determining polymer properties and architectures. However, for linear supramolecular polymers to form, it is a prerequisite to have strong and highly directional interactions as a reversible alternative for the covalent bond. Hydrogen bonds between neutral organic molecules, though they hold a prominent place in supramolecular chemistry because of their directionality and versatility, are not among the strongest noncovalent interactions. Hence, either multiple hydrogen bonds with cooperativity must be used or hydrogen bonds should be supported by additional forces like excluded volume interactions [9]. Though the concept has been known for years, it was not known how to incorporate such sufﬁciently strong but still reversible interactions. However, in the past two decades following the development of strong hydrogen-bonding dimers, several research groups have applied these dimers for the formation of hydrogen-bonded supramolecular polymers. Thus the ﬁnding by Sijbesma et al. [10] that derivatives of 2-ureido-4[1H]-pyrimidinone (UPy, 1 in Figure 1.23) are easy to synthesize and they dimerize strongly (dimerization constant>106 M−1 in CHCl3) by self-complementary quadrupole (array of four) hydrogen bonding (2 in Figure 1.23) prompted them to use this functionality as the associating end group in reversible self-assembling polymer systems. A difunctional UPy compound, 4 in Figure 1.24, possessing two UPy units can be easily made in a one step procedure, from commercially available compounds, methylisocytosine (R═CH3) and hexyldiisocyanate (R═C6H12). The compound forms very stable and long polymer chains (5 in Figure 1.24) in solution as well as in the bulk [9,11]. Dissolving a small amount of the compound in (a) (b) FIGURE 1.22 Schematic representation of (a) a covalent polymer and (b) a supramolecular polymer. (After Brunsveld, L., Folmer, B. J. B., Meijer, E. W., and Sijbesma, R. P. 2001. Chem. Rev., 101, 4071. With permission.) 35 Characteristics of Polymers and Polymerization Processes O N N NH2 O H H N N R' O = C = N - R' N N H O H R R Solvent 1 UPy R H N O R' N N N O H H H H O N N N R' O N H R 2 FIGURE 1.23 Synthesis of a monofunctional 2-ureido-4[1H]–pyrimidinone (UPy) (1) and dimerization of 1 in solution forming a quadrupole hydrogen-bonded unit. (After Sijbesma, R. P., Beijer, F. H., Brunsveld, L., Folmer, B. J. B., Hirschberg, J. H. K., Lange, R. F. M., Lowe, J. K. L., and Meijer, E. W. 1997. Science, 278, 1601. With permission.) chloroform gives solutions with high viscosities, while calculations show that polymers with molecular weights of the order of 106 can be formed. Deliberate addition of small amounts of monofunctional compounds (1 in Figure 1.23) results in a sharp drop in viscosity, proving that linkages between the building blocks are reversible and unidirectional and that the monofunctional compounds act as chain stoppers. For the same reason, the supramolecular polymers show polymer-like viscoelastic behavior in bulk and solution, whereas at elevated temperatures they exhibit liquid-like properties [9]. The quadrupole hydrogen-bonded unit can be employed in the chain extension of telechelic oligomers such as polysiloxanes, polyethers, polyesters, and polycarbonates [11]. Thus the electrophilic isocyanate group (–NCO) of “synthon” (3 in Figure 1.24) can be reacted with common nucleophilic end groups (–OH or –NH2) of telechelic oligomers, resulting in supramolecular polymers by chain extension (Figure 1.25). Thus the material properties of telechelic polymers have been shown to improve dramatically upon functionalization with synthon, and materials have been obtained that combine many of the mechanical properties of conventional macromolecules with the low melt viscosity of oligomers [9]. In contrast to conventional high-molecular-weight polymers, supramolecular (reversible) polymers with a high “virtual” molecular weight show excellent processability due to the strong temperature dependency of the melt viscosity [11]. Moreover, hybrids between blocks of covalent macromolecules and supramolecular polymers can be easily made. 1.3.5 Copolymerization All the addition polymers we have considered so far (Table 1.1) contain only one type of repeating unit or mer in the chain. Polymers can also be synthesized by the aforesaid processes with more than one type of 36 Plastics Technology Handbook N O NH2 N O H H N N R"–NCO + OCN – R"–NCO N N H O H R R 3 (a) UPy O N H H N N R" H H N N O N N O NCO O N H H R R 4 UPy n 5 (b) FIGURE 1.24 Preparation of (a) UPy possessing an isocyanate functional group (3) and (b) a difunctional UPy compound (4) which forms a supramolecular polymer (5) by hydrogen bonding (cf. Figure 1.23). (After Brunsveld, L., Folmer, B. J. B., Meijer, E. W., and Sijbesma, R. P. 2001. Chem. Rev., 101, 4071 and Folmer, B. J. B., Sijbesma, R. P., Versteegen, R. M., van der Rijt, J. A. J., and Meijer, E. W. 2000. Adv. Mater., 12, 12, 874. With permission.) (UPy -NCO) 6 7 FIGURE 1.25 Schematic representation of the formation of supramolecular polymer (7) by chain extension of reactive telechelic oligomer with UPy. (After Folmer, B. J. B., Sijbesma, R. P., Versteegen, R. M., van der Rijt, J. A. J., and Meijer, E. W. 2000. Adv. Mater., 12, 12, 874. With permission.) mer in the chain. Such polymers are called copolymers. They are produced by polymerizing a mixture of monomers (copolymerization) [12] or by special methods. Copolymers can be of different types, depending on the monomers used and the speciﬁc method of synthesis. The copolymer with a relatively random distribution of the different mers in its structure is referred to as a random copolymer. Representing, say, two different mers by A and B, a random copolymer can be depicted as ABBABBBAABBAABAAABBA Characteristics of Polymers and Polymerization Processes (a) 37 (b) (c) (d) FIGURE 1.26 Copolymer arrangements. (a) Two different types of mers (denoted by open and ﬁlled circles) are randomly placed. (b) The mers are alternately arranged. (c) A block copolymer. (d) A graft copolymer. Other copolymer structures [13] are known: alternating, block, and graft copolymer structures (Figure 1.26). In the alternating copolymer the two mers alternate in a regular fashion along the polymer chain: ABABABABABABABABABAB A block copolymer is a linear copolymer with one or more long uninterrupted sequences of each mer in the chain: AAAAAAAAAABBBBBBBBBB A graft copolymer, on the other hand, is a branched copolymer with a backbone of one type of mer to which are attached one or more side chains of another mer. AAAAAAAAAAAAAAAA B B B B B Copolymerization, which may be compared to alloying in metallurgy, is very useful for synthesizing polymers with the required combination of properties. For example, polystyrene is brittle and polybutadiene is ﬂexible; therefore copolymers of styrene and butadiene should be more ﬂexible than polystyrene but tougher than polybutadiene. The general-purpose rubber GRS (or SBR), the ﬁrst practical synthetic rubber, is a copolymer of styrene and butadiene. 1.4 Polymerization Processes 1.4.1 Process Characteristics Free-radical chain or addition polymerizations are commonly carried out by four different processes: (a) bulk or mass polymerization, (b) solution polymerization, (c) suspension polymerization, and (d) emulsion polymerization. Processes (c) and (d) are essentially of the heterogeneous type containing a large proportion of nonsolvent (usually water) acting as a dispersion medium for the immiscible liquid monomer. Bulk and solution polymerizations are homogeneous processes, but some of these homogeneous systems may become heterogeneous with progress of polymerization owing to the polymer formed being insoluble 38 Plastics Technology Handbook in its monomer (for bulk polymerization) or in the solvent used to dilute the monomer (for solution polymerization). Step polymerizations such as polyesteriﬁcation and polyamidation are carried out in bulk or solution phase with provision to remove the by-products of condensation reactions. 1.4.1.1 Bulk, Solution, and Suspension Polymerization The kinetic schemes for free-radical polymerizations in bulk monomer, solution, or suspension (but not in emulsion) are the same, the rate of polymerization (Rp) being given by [5] − d ½M=dt = Rp = kp =kt 1=2 ð fkd ½IÞ1=2 ½M (1.50) where [M] and [I] are the concentrations of monomer and initiator, respectively; kd, kp, and kt are the rate constants for initiator decomposition, chain propagation, and chain termination, respectively; f is the initiator efﬁciency or efﬁciency of initiation. The general ranges of values of these quantities are [14] as follows: [M]: 10 to 10−1 mol L−1, [I]: 10−2 to 10−4 mol L−1, Rp: 10−4 to 10−6 mol L−1 s−1, kp: 104 to 102 L mol−1 s−1, kd: 10−4 to 10−6 s−1, kt: 108 to 106 L mol−1 s−1, and f: 0.3 to 0.8. The above equation shows that the rate of polymerization depends directly on the monomer concentration and on the square root of the rate of initiation. Thus, doubling the monomer concentration doubles the polymerization rate, while doubling pﬃﬃﬃ the rate of initiation or initiator concentration increases the polymerization rate only by the factor 2. The overall extent of polymerization or (fractional) monomer conversion (p) over a period of time (t) is obtained by integration of Equation 1.50. This gives [5] − lnð 1 − pÞ = 2 kp =kt 1=2 ð f ½Io =kd Þ1=2 1 − e−kd t=2 (1.51) where [I]o is the initial concentration of the initiator. This equation can be used to calculate the amount of polymer produced (i.e., the moles of monomer converted to polymer) in time t at a given temperature or for determining the time needed to reach different extents of conversion for actual polymerization systems where both [M] and [I] decrease with time. Polymerization in bulk, that is, of undiluted monomer, minimizes any contamination of the product. But bulk polymerization is difﬁcult to control because of the high exothermicity and high activation energies of free-radical polymerization and the tendency toward the gel effect (in some cases). By carrying out the polymerization of a monomer in a solvent (solution polymerization), these disadvantages of the bulk process can be avoided. The solvent acting as a diluent reduces the viscosity gain with conversion and allows more efﬁcient agitation or stirring of the medium, thereby enabling better transfer and dissipation of heat. Solution polymerization is, however, advantageous only if the polymer formed is to be applied in solution (avoiding the need for solvent removal at the end), such as for making coating (lacquer) grade poly(methyl methacrylate) resins from methyl methacrylate and related monomers. Suspension polymerization combines the advantages of both the bulk and solution polymerization techniques. It is used extensively in the mass production of vinyl and related polymers. In suspension polymerization (also referred to as bead or pearl polymerization), the monomer is suspended as droplets by efﬁcient agitation in a large volume (continuous phase) of nonsolvent, commonly referred to as the dispersion or suspension medium. Water is used as the suspension medium for water insoluble monomers because of its obvious advantages. Styrene, methyl methacrylate, vinyl chloride, and vinyl acetate are polymerized by this suspension process. The size of the monomer droplets in suspension polymerization usually ranges between 0.1 and 5 mm in diameter. Suspension is maintained by mechanical agitation and addition of stabilizers. Low concentrations of suitable water-soluble polymers, such as carboxymethyl cellulose or methyl cellulose, poly (vinyl alcohol), gelatin, and so on, are used as suspension stabilizers. They raise the medium viscosity and effect stabilization by forming a thin layer on the monomer/polymer droplets. Water-insoluble inorganic compounds such as bentonite, kaolin, magnesium silicate, and aluminum hydroxide, in ﬁnely divided Characteristics of Polymers and Polymerization Processes 39 state, are sometimes used to prevent agglomeration of the monomer droplets. Initiators soluble in monomer, such as organic peroxides, hydroperoxides, or azocompounds—often referred to as oil-soluble initiators—are used. Each monomer droplet in a suspension polymerization thus behaves as a miniature bulk polymerization system and the kinetics of polymerization within each droplet are the same as those for the corresponding bulk polymerization (Equations 1.50 and 1.51). At the end of the polymerization process, the monomer droplets appear in the form of tiny polymer beads or pearls (hence the term bead or pearl polymerization). As mentioned previously, the two main termination processes for propagating chain radicals in all freeradical polymerizations are combination or coupling (Equation 1.39) and disproportionation (Equation 1.40). The degree of polymerization (DP) of polymer chain, and hence polymer molecular weight, depends on which termination process is prevalent. It can be shown [5] that the number average degree of polymerization (DPn ) is related by DPn = kp 2 ½M2 1 (2 − etc ) kt Rp (1.52) where etc is the fraction of propagating chains terminating by coupling, while kp, kt, Rp, and [M] are as in Equation 1.50. It is evident from Equation 1.52 that the degree of polymerization, and hence polymer molecular weight, decreases with the increase in the polymerization rate in free-radical chain polymerization (not when the reaction is carried out in emulsion [see below]). 1.4.1.2 Emulsion Polymerization In emulsion polymerization, monomers are polymerized in the form of emulsions and the polymerization in most cases involves free-radical reactions. Like suspension polymerization, the emulsion process uses water as the medium. Polymerization is much easier to control in both these processes than in bulk systems because stirring of the reactor charge is easier owing to lower viscosity, and removal of the exothermic heat of polymerization is greatly facilitated with water acting as the heat sink. Emulsion polymerization, however, differs from suspension polymerization in the nature and size of particles in which polymerization occurs, in the type of substances used as initiators, and also in mechanism and reaction characteristics. Emulsion polymerization normally produces polymer particles with diameters of 0.1–3 mm. On the other hand, polymer nanoparticles of sizes 20–30 nm are produced by microemulsion polymerization (see later). The original theory of emulsion polymerization is based on the qualitative picture of Harkins [15] and the quantitative treatment of Smith and Ewart [16]. The essential ingredients in an emulsion polymerization system are water, a monomer (not miscible with water), an emulsiﬁer, and an initiator that produces free radicals in the aqueous phase. Monomers for emulsion polymerization should be nearly insoluble in the dispersing medium but not completely insoluble. A slight solubility is necessary as this will allow the transport of monomer from the emulsiﬁed monomer reservoirs to the reaction loci (explained later). Emulsiﬁers are soaps or detergents, and they play an important role in the emulsion polymerization process. A detergent molecule is typically composed of an ionic hydrophilic end and a long hydrophobic chain. Some examples are as follows: Anionic detergent: Sodium laurate: CH3(CH2)10COO− Na+ Sodium alkyl aryl sulfonate: CnH2n+1C6H4SO3− Na+ Cationic detergent: Cetyl trimethyl ammonium chloride: C16H33N+(CH3)3 Cl− Anionic and cationic detergent molecules may thus be represented by —•− and —•+, respectively, indicating hydrocarbon (hydrophobic) chains with ionic (polar) end groups. Let us now consider the locations of the various components in an emulsion polymerization system. A micelle of an anionic detergent can be depicted as a cluster of detergent molecules (—•−) with the 40 Plastics Technology Handbook hydrocarbon chains directed toward the interior and their polar heads in water (see Figure 1.27). In the same way, detergent molecules get adsorbed on the surface of an oil droplet suspended in water. Such materials are therefore said to be surface active and are also called surfactants. When a relatively water-insoluble vinyl monomer, such as styrene, is emulsiﬁed in water with the aid of an anionic surfactant and adequate agitation, three phases result (see Figure 1.27): (1) an aqueous phase in which small amounts of both monomer and surfactant are dissolved (i.e., they exist in molecular dispersed state); (2) emulsiﬁed monomer droplets that are supercolloidal in size (>10,000 Å), stability being imparted by the reduction of surface tension and the presence of repulsive forces between the droplets since a negative charge overcoats each monomer droplet; and (3) submicroscopic (colloidal) micelles that are saturated with monomer. This three-phase emulsion represents the initial state for emulsion polymerization (Figure 1.27). Stage I (see Figure 1.27) begins when a free radical–producing water-soluble initiator is added to the three-phase emulsion described above. The commonly used initiator is potassium persulfate, which decomposes thermally to form water-soluble sulfate radical ions: 50−60° C S2 O8 2− ! 2SO4 − • (1.53) The rate of radical generation by an initiator is greatly accelerated in the presence of a reducing agent. Thus, an equimolar mixture of FeSO4 and K2S2O8 at 10°C produces radicals by the reaction S2 O8 2− + Fe2+ ! Fe3+ + SO4 2 – + SO4 − • (1.54) about 100 times as fast as an equal concentration of the persulfate alone at 50°C. (Redox systems generally ﬁnd use for polymerizations only at lower temperatures.) Initial stage Stage I M I Monomer accumulation inside micelle M MM M MM A Water-soluble initiator I In M nM M A M I A added M Out I 2 A M MM Soap molecule out M MM M MM Out Micelle converts into monomer-polymer particle Out A In Monomer droplet (emulsiﬁed) A Soap-like free radical in In M M M : Dissolved monomer molecule; Dissolved soap molecule; A Monomer droplet (emulsiﬁed) Soap-like free radical FIGURE 1.27 Schematic of emulsion polymerization showing three phases present. (After Williams, D. J. 1971. Polymer Science and Engineering, Prentice Hall, Englewood Cliffs, NJ.) Characteristics of Polymers and Polymerization Processes 41 The sulfate radical ions generated from persulfate react with the dissolved monomer molecules in the aqueous phase to form ionic free radicals (–•): SO4 − • + ðn + 1ÞM!− SO4 ðMÞn M• 1.55 These ionic free radicals can be viewed as soap-like anionic free radicals as they are essentially made up of a long hydrocarbon chain carrying an ionic charge at one end and a free-radical center at the other (represented as −A—· in Figure 1.27). They thus behave like emulsiﬁer molecules and because of the existence of a dynamic equilibrium between micellar emulsiﬁer and dissolved emulsiﬁer, they also can at some stage be implanted in some of the micelles. Once implanted this way in a micelle, a soap-like anionic free radical initiates polymerization of the solubilized monomer in the micelle. The micelle, thus “stung,” grows in size by reacting with the solubilized monomer, and to replenish it, more monomer enters the micelle from monomer droplets via the aqueous dispersion phase. The “stung” micelle is in this way transformed into a monomer–polymer (M/P) particle (see Figure 1.27). Thus, in Stage I, the system will consist of an aqueous phase containing dissolved monomer, dissolved soap-like free radicals, micelles, “stung” micelles, M/P particles, and monomer droplets. The rate of overall polymerization increases continuously since nucleation of new particles (i.e., conversion of micelles into M/P particles) and particle growth occur simultaneously. For the same reason, a particle size distribution occurs during Stage I. However, at 13%–20% monomer conversion, nearly all the emulsiﬁer will be adsorbed on the M/P particles and the micelles will disappear. Since new particles mostly originate in micelles, with the disappearance of micelles, the nucleation of new M/P particles essentially ceases. This marks the end of Stage I. In Stage II (Figure 1.28) that follows, there occurs a continued growth of the existing M/P particles in the absence of any new particle nucleation. Free radicals enter only the M/P particles where polymerization takes place as the particles are supplied with monomer from the emulsiﬁed monomer droplets via the aqueous phase. Stage II thus features a constant overall rate of polymerization (Figure 1.28). It is followed by Stage III, which begins when the overall rate of polymerization begins to deviate from linearity and nonlinear growth rate is observed. The nonlinearity may appear (a) in the form of a decrease in rate as a result of dwindling monomer concentration inside M/P particles, or (b) in the form of an increase, if the gel effect (generally attributed to a greater decrease of the termination rate constant kt, as Conversion (%) Trommsdorﬀ eﬀect Stage III: Nonlinear growth rate (diﬀusioncontrolled regime) 60% Stage II: • No new particle nucleation • Constant rate of growth }25% to 30% Emulsiﬁed monomer droplets disappear ~15% Stage I: Simultaneous particle nucleation and growth Time FIGURE 1.28 Schematic conversion–time curve for a typical emulsion polymerization showing three main stages of the polymerization process. 42 Plastics Technology Handbook compared to the propagation rate constant kp, owing to increased viscosity at higher conversions) becomes important with the monomer reservoirs having already disappeared in Stage II. A simple analysis of emulsion polymerization kinetics is based on the idealized emulsion system as depicted in Figure 1.27. However, the treatment centers only around Stage I and Stage II, as no general theory for Stage III is available. From theoretical considerations, it can be shown [5,16] that the number (Np) of M/P particles at the end of Stage I is Np = 0:53ðas ws Þ0:6 ðRr =uÞ0:4 (1.56) where as is the area occupied by the unit weight of surfactant, ws is the weight concentration of surfactant, Rr is the rate of radical generation, and u is the rate of volume growth of a M/P particle. Equation 1.56 indicates that the particle number depends on the 0.6 power of the surfactant concentration and on the 0.4 power of the initiator concentration, since the rate of radical generation, Rr, by thermal dissociation of initiator is given by Rr = 2NAv kd ½I (1.57) where NAv is Avogadro’s number (6.023 × 1023 molecules mol−1), [I] is the initiator concentration (mol/L) in aqueous phase, and kd is the thermal dissociation rate constant (s−1) of the initiator, with two radicals being produced from each initiator molecule. (Typical values [17] of the quantities in Equation 1.56 are as follows: Np ~ 1015–1016 per cm3 of aqueous phase, Rr ∼ 1012–1014 radicals (number) cm−3 s−1, asws ∼ 105 cm2/cm3 aqueous phase, u ~ 10−20 cm3 s−1.) For estimating the rate of polymerization in Stage II, one may generally assume that free radicals enter the particles singly, that is, one at a time. For example, considering typical values pertinent to emulsion polymerization, if the rate of generation of free radicals (Rr) in the aqueous phase is 1014 per second per milliliter and the value of the number of M/P particles is 1015 per milliliter, then assuming that all the radicals generated eventually enter M/P particles (since there are no micelles), the rate of radical entry into a particle will average out to about 0.1 radical per second, that is, one radical every 10 s. When a (soap-like) free radical enters a M/P particle, it initiates the polymerization of monomer in the particle, but the polymerization is terminated when another free radical enters the same particle (since, as a simple calculation using known kt values shows, two radicals cannot coexist in the same particle and they would terminate mutually within a few thousandths of a second). The particle thereafter remains inactive till another free radical enters and initiates the polymerization afresh. Thus, if a radical enters a M/P particle every 10 s as calculated above, the particle will experience alternate periods of activity (growth) and inactivity (no growth), each of 10 s duration. In other words, each M/P particle will remain active for half of the total time and inactive for the other half. This situation will remain unchanged even if there is a change in the rate of radical entry into the particle. The average number of active radical per particle can thus be considered to be ½ and independent of the rate at which radicals enter the particle. (These concepts are known collectively as the Smith–Ewart theory). The rate of polymerization in a M/P particle, Rpp (mol s−1), is thus given by [5] Rpp = kp ½M 2NAv (1.58) where kp is the propagation rate constant (L mol−1 s−1), [M] denotes the monomer concentration (mol/L) in a M/P particle, and NAv is Avogadro’s number. If Rpp is constant and the number of particles per unit volume, Np (L−1), is constant, then the overall rate of emulsion polymerization per unit volume, Rp (mol L−1 s−1), is simply given by − Np kp ½M d½M = Rp = dt 2NAv (1.59) Characteristics of Polymers and Polymerization Processes 43 Substituting for Np from Equation 1.56 gives Rp = 0:53ðas ws Þ0:6 0:4 kp ½M Rr u 2NAv (1.60) According to the simple Smith–Ewart model described above, the rate of polymerization in Stage II will thus depend on the 0.6 power of the surfactant concentration (ws) and the 0.4 power of the rate of radical generation in aqueous phase, Rr, which, in turn, is related to the aqueous phase (Stage I) initiator concentration, [I], through Equation 1.57. Since chain termination takes place as soon as a radical enters an active M/P particle, the rate of chain termination can be equated to the rate of radical entry into the particles (Rr/Np) and the degree of polymerization, which is given by the ratio of the rate of chain propagation to the rate of chain termination, can be expressed as [5] DPn = kp ½M Np kp ½M = Rr Rr =Np (1.61) Equation 1.61 shows that the degree of polymerization DPn , like the rate of polymerization Rp (Equation 1.56), is directly dependent on the number of particles. Thus, unlike polymerization by the bulk, solution, and suspension techniques, polymerization by the emulsion technique permits simultaneous increase in rate and degree of polymerization by increasing the number of polymer particles (Np), that is, by increasing the surfactant concentration, at a ﬁxed rate of initiation. This possibility of combining high molecular weight with high polymerization rate is one reason for the popularity of the emulsion technique. The Smith–Ewart kinetic scheme, described above, is highly idealized, though valuable for its simplicity. It explains adequately only a small part of the vast literature on emulsion polymerization. Thus, it works well for monomers such as styrene, butadiene, and isoprene, which have very low water solubility (<0.1%), but it fails to apply quantitatively for many monomers with higher solubility (1%–10%), such as vinyl chloride, methyl methacrylate, vinyl acetate, and methyl acrylate, where initiation in the aqueous phase followed by precipitation of polymer assumes importance. It also fails where the emulsion system has larger particles (>0.1–0.15 mm in diameter), which can accommodate more than one growing chain simultaneously. 1.4.1.3 Microemulsion Polymerization According to IUPAC deﬁnition, microemulsions are dispersions (made of water, oil, and surfactants) that are optically isotropic and thermodynamically stable and have dispersed droplet diameters varying approximately from 1 to 100 nm (usually 10 to 50 nm, i.e., 100 to 500 Å). The small particle size leads to a translucent or even to a transparent system if the particle size is a few hundred angstroms. In comparison, the average diameter of droplets in an ordinary emulsion (macroemulsion), discussed above, is in the micron range. Microemulsions, like micellar dispersions, are liquid dispersions containing surfactant aggregates. However, whereas in micellar dispersions the aggregates are made of surfactant only and are usually dispersed in water, in microemulsions, the aggregates are much larger and have large liquid cores (oil or water) surrounded by a surfactant monolayer that stabilizes the dispersion. In many cases, the micellar aggregates are spherical, but they can also be tubular and, in a few cases, they can also grow very long and entangle like polymers [18]. While the aqueous component of microemulsions may contain salt(s) and/or other ingredients, the “oil” component may actually be a mixture of different hydrocarbons and oleﬁns. Microemulsions form upon simple mixing of the components and do not require the high-shear conditions generally used to make ordinary (macro) emulsions. Besides the optical clarity (or translucency) of a microemulsion and the small (10–50 nm) droplet size of the dispersed phase, an additional feature that distinguishes it from ordinary emulsions is that the average drop size does not grow with time owing to thermodynamic stability. 44 Plastics Technology Handbook There are three basic types of microemulsions, namely, direct (oil dispersed in water, denoted by “o/w”), reversed or inverse (water dispersed in oil, denoted by “w/o”), and bicontinuous (i.e., regions of water and oil). The domains of the dispersed phase are either globular or interconnected (giving a bicontinuous microemulsion), as shown schematically in Figure 1.29. These are stabilized by an interfacial ﬁlm of surfactant (usually in combination with a cosurfactant), its molecules being oriented at the interface such that the hydrophilic ends are in the aqueous phase and the hydrophobic ends are in the oil phase. To prepare a microemulsion, in a simple procedure, milky emulsions can be ﬁrst prepared using water, oil, and surfactant and then lower alkanols (butanol, pentanol, and hexanol) can be added in controlled amounts so as to obtain transparent or translucent solutions comprising dispersions of either water-in-oil (w/o) or oil-in-water (o/w) in nanometer or colloidal dispersions. The lower alcohols added are called cosurfactants. They lower the interfacial tension between oil and water sufﬁciently for almost spontaneous formation of the aforesaid microheterogeneous systems. Various surfactant-to-cosurfactant ratios can be used in the preparation. The miscibility of oil, water, and amphiphile (i.e., surfactant plus cosurfactant) depends on the overall composition, which, in turn, depends on the system. Ternary (water/surfactant/oil), pseudo-ternary (water/amphiphile/oil), or explicitly quaternary (water/surfactant/cosurfactant/oil) phase diagrams are usually employed to describe the phase manifestation that is essential in the study of microemulsions. These phase diagrams help deﬁne the microemulsion areas. Samples from the best combinations, that is, those that produce the largest area of microemulsion, can be subjected to further studies. The knowledge of phase manifestations of the pseudo-ternary (water/amphiphile/oil) or explicitly quaternary (water/surfactant/cosurfactant/oil) mixtures has been systematized. According to Winsor [19], four types of microemulsion phases exist in equilibrium. These phases are commonly referred to as Winsor phases; they are Winsor I: with two phases, the lower o/w microemulsion phase in equilibrium with the upper excess oil; Winsor II: with two phases, the upper w/o microemulsion phase in equilibrium with excess water; Winsor III: with three phases, middle microemulsion phase (o/w plus w/o, called bicontinuous) in equilibrium with upper excess oil and lower excess water; Winsor IV: in single phase, with oil, water, and surfactant homogeneously mixed. Interconversion among these phases can be achieved by varying the proportions of the components. Figure 1.30 gives a composite representation of the aforesaid features of microemulsion forming systems. Being thermodynamically stable, “nano-dispersions” of water-in-oil or oil-in-water, microemulsions can be considered as microreactors to carry out chemical reactions and, in particular, to synthesize nanomaterials. Microemulsions thus have received much recent attention as media for synthesis of nanoparticles like Pt, Pd, Rh, and Ir (by reducing the corresponding salts in the water micropools of w/o microemulsions with hydrazine or hydrogen gas) and as media for polymerization to produce thermodynamically stable latexes in the nanosize range (<50 nm), not attainable with classical emulsion polymerization processes. Water droplet Oil droplet (a) (b) (c) FIGURE 1.29 Idealized microemulsion structures: (a) oil-in-water (o/w) microemulsion; (b) water-in-oil (w/o) microemulsion; (c) bicontinuous microemulsion. 45 Characteristics of Polymers and Polymerization Processes Oil O WI W W μe W III O L2 W IV W O W II D L1 Water Surfactant O W W W O FIGURE 1.30 Ternary phase diagram (schematic) of water-oil-surfactant mixtures showing Winsor classiﬁcation and probable internal structures: L1, one-phase region of normal micelles or oil-in-water (o/w) microemulsion; L2, reverse micelles or water-in-oil (w/o) microemulsions; D, anisotropic lamellar liquid crystalline phase. Other symbols: me, microemulsion; O, oil; W, water. (After Paul, B. K. and Moulik, S. P. 2001. Curr. Sci., 80, 990.) Though polymers with large molecular weights (>106) can be produced from monomers at fast reaction rates in both emulsion and microemulsion processes, it is only in the latter that stable latexes with polymer particles smaller than 50 nm can be easily obtained, since in this case polymerization occurs in the monomer reservoir encapsulated in a nanosize space [20]. While microemulsions exhibit a wide variety of microstructures, it is the spherical oil-in-water (o/w) or water-in-oil (w/o) microstructures that have provoked the greatest interest to carry out polymerization in practice. The phase diagram concept provides a reliable basis for the use of microemulsion in polymerization systems. The phase diagram maps the thermodynamically stable regions of o/w and w/o microemulsions. Experimental determination of single-phase (microemulsion) regions thus precedes microemulsion polymerization. In fact, a thorough study involving composition and characterization of phase diagrams of various systems, including those that contain monomer, must be done before performing any microemulsion polymerization. To give an example, for polymerization of methyl acrylate in microemulsions using sodium dodecyl sulfate (SDS) (C12H25SO4Na) as surfactant and pentanol (C5H11OH) as cosurfactant, Stoffer and Bone [21], in their pioneering work on microemulsion polymerization, determined boundaries of single-phase region by titrating with water various surfactant/cosurfactant/monomer mixtures, ﬁrst to the point of dissolution of surfactant (which marked one limit of the solubility region) and then to the appearance of turbidity (which deﬁned another limit). Complementary information was obtained by titration of water/surfactant mixtures with a cosurfactant/monomer solution. Figure 1.31 thus shows a microemulsion region that contains 25% methyl acrylate. While Stoffer and Bone used w/o microemulsions stabilized by SDS and pentanol to carry out polymerization of methyl acrylate and methyl methacrylate at 50°C using oil-soluble initiator benzoyl peroxide or AIBN and hence observed kinetic similarity to solution polymerization, Atik and Thomas [22] carried out polymerization of styrene in o/w microemulsion and provided the ﬁrst account of a microemulsion polymerization that produces nanosize spherical latex particles (Figure 1.32). They, however, found that the stability of the microemulsion was limited by the solubility of the polymer formed. 46 Plastics Technology Handbook D (H2O) 10 90 30 70 50 50 30 70 10 90 B 10 (Surfactant) 30 50 70 C 90 (Cosurfactant) FIGURE 1.31 A microemulsion region (schematic) containing water, surfactant (sodium dodecyl sulfate), cosurfactant (pentanol), and 25% monomer (methyl methacrylate). (After Stoffer, J. O. and Bone T. 1980. J. Polym. Sci. Polym. Chem. Ed., 18, 2641.) 200 nm FIGURE 1.32 Electron micrograph of radiation-polymerized styrene microemulsion stabilized by surfactant cetyl trimethyl ammonium bromide and cosurfactant hexanol. (Reprinted with permission from Atik, S. S. and Thomas, K. J. 1981. J. Am. Chem. Soc., 103, 4279. Copyright 1981 American Chemical Society.) Guo et al. [23] also carried out microemulsion polymerization of styrene in an o/w system with SDS surfactant and 1-pentanol cosurfactant using a water-soluble K2S2O8 or an oil-soluble 2,2′-azobis-(2methylbutyronitrile) (AMBN) initiator at 70°C. The reaction yielded stable latex of small size (20–30 nm) and high molecular weight (1.2 × 105), which implied that each latex particle consisted of two or three polystyrene molecules. The maximum polymerization rate and number of particles varied with the 0.47 and 0.40 powers of K2S2O8 concentration and 0.39 and 0.38 powers of AMBN, respectively, in agreement with the 0.4 power predicted by Smith–Ewart theory, Case II (see above). This consistency was attributed to the comparable size of microemulsion droplets and micelles. Since the work of Atik and Thomas in 1981, most work was done in four- or ﬁve-component (including cosurfactant) microemulsion. However, the presence of a fourth component, such as alcohol cosurfactant, substantially limits the utility of microemulsion, mainly for two reasons. First, the cosurfactant complicates the phase behavior of the microemulsion system, whereas the ternary microemulsion formulations 47 Characteristics of Polymers and Polymerization Processes are considerably easier to deal with. Second, alcohol present as a cosurfactant can act as a chain transfer agent, interfering with the desired polymerization and reducing the polymer molecular weight. In this context, the ﬁrst report on polymerization in ternary microemulsions (without cosurfactant alcohol) by P’erez-Luna et al. [24] using cationic surfactant dodecyl trimethylammonium bromide (DTAB) and styrene monomer assumes signiﬁcance. The one-phase microemulsion region was determined visually by styrene titration of aqueous micellar solution of DTAB at a ﬁxed temperature to obtain phase boundaries that were also checked by preparing samples with compositions below and above the titration determined phase boundaries. In the one-phase microemulsion region of styrene/water/DTAB mixtures thus determined at 60°C (using a few parts per million of hydroquinone to inhibit thermal polymerization), styrene polymerization was carried out with K2S2O8 (1 wt% with respect to the monomer) as initiator. In unpolymerized microemulsions, two apparent particle sizes were always observed— a small size of ca. 0.6–0.8 nm in hydrodynamic radius and a larger composition-dependent size (6–15 nm), believed to be DTAB micelles and styrene-swollen droplets, respectively. The microlatexes produced by polymerization remained stable with respect to coagulation for months and the particles were spherical with radii in the range 20–30 nm and apparently monodisperse when observed with TEM (see Figure 1.33). The anionic surfactant sodium bis(2-ethylhexyl sulfosuccinate) (AOT, shown below) produces microemulsion without the addition of any cosurfactant. CH3 CH2 O CH3 (CH2)3 CH CH2 O C CH3 (CH2)3 CH CH2 O C CH3 CH2 CH SO3Na CH2 O (AOT) In the ﬁrst account of inverse microemulsion polymerization reported in 1982, Leong and Candau [25] used inverse (w/o) microemulsion consisting of water–acrylamide mixture dispersed in toluene and stabilized by AOT surfactant without requiring the presence of any cosurfactant. (The microemulsion, however, needed the presence of a large amount of the surfactant.) Photopolymerization of the inverse 100 nm FIGURE 1.33 Transmission electron micrograph of a microemulsion containing 13.8 wt% DTAB, 8 wt% styrene, and 78.2 wt% water after polymerization at 60°C with K2S2O8 initiator (1 wt% with respect to monomer). (Reprinted with permission from P’erez-Luna, V. H., Puig, J. E., Castano, V. M., Rodriguez, B. E., Murthy, A. K., and Kaler, E. W. 1990. Langmuir, 6, 1040. Copyright 1990 American Chemical Society.) 48 Plastics Technology Handbook microemulsion was rapid with total conversion to polymer in less than 30 min, producing lowpolydispersity polyacrylamide of high molecular weight (∼3 × 106) conﬁned in small (<50 nm in diameter) micellar particles. The microemulsions remained perfectly transparent and stable and no phase separation took place during the polymerization process. Cross-linked polyacrylamide latexes, or microgels, were also prepared by using a 100:1 mixture of acrylamide and methylenebis(acrylamide). Candau et al. [26] later carried out the polymerization of AOT-stabilized acrylamide inverse microemulsion by a thermal process using either oil-soluble AIBN or water-soluble K2S2O8 as the initiator. The polymerization displayed a novel feature that each ﬁnal latex particle consisted of one single polymer molecule in a collapsed state, suggesting that the kinetics of the reaction did not follow the Smith–Ewart theory but were characterized by continuous particle nucleation. The polymerization of water-soluble monomers and more particularly acrylamide (AM) in w/o microemulsions has been investigated extensively because of the numerous applications of polyacrylamide. In short, polymerization reactions can be carried out in microemulsions of all types of structures— droplet-type (o/w or w/o) microemulsions or non–droplet-type (bicontinuous) microemulsions. A monomer (or monomers) can be incorporated in any of the water and oil phases of microemulsions of any of the above structural types and polymerized by any of the normal free-radical methods (thermal, photochemical, or high-energy radiation). Such polymerizations can be used to obtain very small polymer particles on the order of the primary micelle size. Thus, when a hydrophobic monomer (or monomers) is the dispersed oil phase, the o/w microemulsion can be polymerized, typically forming a spherical stable latex composed of polymer particles as small as 15 nm in diameter. On the other hand, an inverse microemulsion polymerization can be performed when an aqueous solution of a hydrophilic monomer, such as acrylamide, is the dispersed phase in w/o microemulsion, producing the corresponding polymer microlatex. Microgels result from both o/w and w/o microemulsions by cross-linking copolymerization in oil and water dispersed phase, respectively. If, however, a monomer occupies the continuous phase, polymerization may produce a solid material with the dispersed phase entrapped in its matrix. On the other hand, by replacing both the oil and water phases in a microemulsion with hydrophobic and hydrophilic monomers, respectively, copolymerization may also be achieved, opening a new way to polymer synthesis with interesting possibilities. Microlatexes obtained from o/w microemulsions are often used in the same form to utilize their properties, while w/o systems are commonly used to produce water-soluble polymers of high molecular weight. As a general rule, the higher the surfactant/monomer ratio in the initial formulation, the smaller the particle size of the ﬁnal latex product. Therefore, both high solid contents and small-size particles can hardly be achieved simultaneously. Moreover, the size of microlatex particles (usually determined by quasi-elastic light scattering, QELS, and transmission electron microscopy techniques) signiﬁcantly exceeds those of the precursor microemulsion droplets. Thus, the latex particles are typically 20–60 nm compared to 4- to 5-nm-diameter globular (o/w or w/o) microemulsions. However, the ﬁnal latex particles are still bigger (around 50–150 nm) if the polymerizing microemulsions are bicontinuous, simply because of a larger monomer content of the latter. As a general rule, the particle size increases on increasing the monomer content or decreasing the surfactant content and/or the initiator concentration. The latexes formed by microemulsion polymerization of monomers in the dispersed state can be precipitated in a large excess of non-solvent and dried under vacuum. The molecular weights of the polymers thus obtained are high, usually ranging from 106 to 107, and in some cases, exceptionally high molecular weights (Mw ≈ 2.5 to 3.3 × 107) have been reported [27]. However, when alcohols are used as cosurfactants in the formulation, chain transfer reactions can occur, which reduce the molecular weight. As a few polymer chains of high molecular weight are conﬁned in the microlatex particle, these must be highly collapsed in order to ﬁll the nanosize space of the particle (d ∼ 40 nm). The use of microemulsions as polymerization media offers new prospects for producing nanosize polymeric materials with novel and interesting properties. The main limitation, however, arises from the necessity of using rather high levels of surfactant (about 10% of the total mass) in order to stabilize a large Characteristics of Polymers and Polymerization Processes 49 internal surface area. This prompted extensive investigations on new surfactants and there are reports of microemulsion formulations with surfactant concentrations less than 2 wt%. Industrial polymers made from microemulsions are now commercially available. Microemulsion-based polymers play an important role in the application of enhanced oil recovery (EOR) technology, as water-soluble polymers are used to raise the viscosity of the injection water, the most common being polyacrylamide. The latex obtained from w/o microemulsion polymerization of water-soluble monomers is used in EOR as drive ﬂuids to achieve better displacement and volumetric sweep efﬁciency in pushing micellar ﬂood through underground reservoirs. Stable microlatexes from microemulsion polymerization are also preferred for application as particulate carriers in pharmaceutical and medical ﬁelds because of their subcellular size, sustained release properties, and biocompatibility with tissue and cells [28]. Thus, procedures have been proposed based on inverse microemulsion polymerization of water-soluble polymers for the preparation of nanoparticles (∼50 nm) and nanocapsules. 1.4.2 Industrial Polymerization The heart of any polymer plant is the reactor section where the monomer is converted to polymer. In the case of polymerization, the reactor assumes an additional importance because it is here that the ultimate properties of the polymer such as molecular weight, molecular structure, molecular weight distribution, copolymer composition, and so on are largely set. To achieve these properties in the desired degree, the reactor must satisfy the following: (a) efﬁcient removal of heat, (b) provide the necessary residence time, (c) provide uniform mixing for good temperature control and reactant homogeneity, and (d) control the degree of backmixing in a continuous polymerization. 1.4.2.1 Heat Removal Polymerization processes are accompanied by liberation of heat. All common vinyl monomers generate about the same amount of heat during polymerization on a molar basis. For poly(vinyl chloride) PVC, it is approximately 23 kcal/mole or 650 Btu/lb, and for styrene, it is 17 kcal/mole (300 Btu/lb). The heat is released over a short period of 5 to 10 h and, moreover, the peak heat release may be several times larger than this average. To remove this released heat, polymerization reactors are generally jacketed and water cooled. Polymerization reactors are usually constructed either of stainless steel or glass-lined carbon steel. The heat transfer rate in the former is much higher than that in the latter. Thus, the overall heat transfer coefﬁcient for a stainless steel vessel can be as high as 125 Btu/(h)(°F)(ft2), compared to 55–70 for a glasslined vessel. The difference arises from the additional heat transfer resistance offered by the glass layer. Therefore, wherever glass-lining is required to reduce reactor fouling, only minimum glass thickness should be speciﬁed. Removal of heat is relatively easy in small reactors, but becomes a problem in larger ones, since the ratio of cooling jacket surface to volume of vessel becomes smaller with larger size. When a cylindrical vessel is scaled up maintaining dimensional similarity, the increase in heat transfer area is proportional to the volume increase raised to the power 0.67. Large kettles require, in addition to cooling jackets, provision for additional heat removal aids. With suspension polymerization, this is commonly achieved by means of cooling bafﬂes. In this case, the bafﬂes provided for agitation are also made to serve as cooling aids. Two common types of cooling bafﬂes are (a) bottom-entering pipe-type cooling bafﬂe, mounted through the bottom and (b) plate cooling bafﬂe, mounted through jacketed side wall and removable through manhole. Another frequently used method of supplemental cooling is by the use of reﬂux condensers. The presence of a diluent or solvent, which either boils lower than the monomer or forms an azeotropic mixture, aids in heat removal through reﬂuxing. Heat removal is easier in semicontinuous operation, where monomer and initiator are added gradually, and is even easier in continuous operation. 50 Plastics Technology Handbook 1.4.2.2 Reactor Agitation An important aspect of polymer reactor design and scale-up is agitation. Various types of agitators are used, ranging from marine-type propellers, ﬂat-blade, and pitched turbines to high-shear types of impellers (Figure 1.34). Because of erosion, there are practical top-speed limitations that should not be exceeded. These correspond to 1200 ft/min for ﬂat-bladed or pitched turbines and 1800 ft/min for glassed, retreating-curve turbines. 1.4.2.3 Residence Time The duration of polymerization has signiﬁcant inﬂuence on polymer properties such as molecular weight distribution, particle size, particle surface, conﬁguration, or composition. Excessive stay in the polymerization reactor results in “hard spots.” The phenomenon of hard spots is familiar to all who are concerned with quality control. In the manufacture of PVC, hard spots mean particles of closed surface unable to absorb plasticizer. In other cases, such as polystyrene or polyethylene, hard spots are either high-molecularweight or cross-linked polymer particles that will not melt as easily as the main portion of the batch. To reduce the formation of hard spots, the kettles have to be cleaned frequently with pressure water, steam, or solvent. In this respect, batch polymerization has advantage over continuous polymerization. With batch reactor, moreover, the polymerization time may be kept exactly uniform for the entire batch, resulting in a product of uniform speciﬁcations. In continuous processes, the polymerization time depends on the residence time in the reaction vessel. Assuming that a single agitated vessel is employed for continuous suspension or emulsion polymerization and that “perfect” mixing takes place, the normalized exit age distribution, E, is given by [29] E = (1=t )e−t=t FIGURE 1.34 (1.62) Open curved blade Open pitched blade Open ﬂat blade Three-blade marine propeller Center-disk straight blade Center-disk curved blade Helix Anchor Agitator types for polymerization reactors. 51 Characteristics of Polymers and Polymerization Processes Input A Output C Input B A Output Output C B Time FIGURE 1.35 Pulse responses of perfectly mixed vessels and plug ﬂow reactor with small extents of axial dispersion. where t is the mean residence time given by the ratio of the volume of the vessel and volumetric ﬂow rate of liquid. Equation 1.62 is plotted as graph (a) in Figure 1.35. Assuming that the mean residence time (t ) in the single vessel is, say, 4 h, integration shows that 10% of the fed monomer would remain less than 25 min, 63% for less than 4 h, and 0.7% for longer than 21 h, which is caused by backmixing. This wide variation of monomer residence time in the reactor could lead to a wide distribution of product quality. Thus, in contrast to batch emulsion polymerization of a monomer like vinyl chloride that yields a single particle-size polymer, continuous operation using a single mixed vessel would result in a broad particle size distribution, dependent on the difference of residence time. For the ﬂow through a cascade of two equal-sized completely mixed vessels, instead of one vessel, the exit age distribution is much narrower (curve B in Figure 1.35). Integration shows that for a mean residence time of 4 h in each of the vessels, only 0.5% of the fed monomer is in the system less than 25 min and only 26.4% of the monomer is less than 4 h. Further increase in the number of mixed vessels in cascade produces more and more narrow age distribution, leading to close approximation to a plug ﬂow (curve C in Figure 1.35). Figures 1.36 and 1.37 show continuous cascading mixed vessels and plug ﬂow tubular reactors, both designed to prevent backmixing and broadening of residence time distribution in the system. Cascading reactors are widely used in the manufacture of rubber latex by emulsion polymerization and plug ﬂow tubular reactors in the manufacture of low-density polyethylene. 1.4.2.4 Industrial Reactors The reactor type used for large-scale production of polymers depends on many factors, such as physical state of the monomer (gas, liquid, or solid), polymerization mechanism (free radical, coordination, ionic, or condensation), polymerization method (bulk, solution, suspension, emulsion, or ﬂuidization), polymer solubility in monomer, heat transfer issues, viscosity changes, and ﬁnal form of the product (solution, emulsion, bead, or powder). 1.4.2.4.1 Polymers Soluble in Their Monomers Some common polymers, which are produced by bulk (or mass) polymerization technique and are soluble in their own monomers, are polystyrene, polyamides, and polyesters. A tremendous rise in melt viscosity 52 FIGURE 1.36 Plastics Technology Handbook Continuous cascading reactors. Monomer + initiator Heating Cooling Polymer plus unreacted monomer FIGURE 1.37 Plug ﬂow tubular reactor. is experienced during bulk polymerization for these polymers. The reactor form is dictated strongly by the reaction mass viscosity. Vertical agitated vessels with turbine agitators may be selected for viscosity as high as 2000–5000 cP. Beyond this, anchor-type agitators are used for viscosities in the range of 50,000–100,000 cP. Ribbon or helical agitators may be used for viscosities as high as 500,000 cP, but beyond this, agitated vessels are generally not practical. Instead, a variety of double-screw devices are commonly used. Double-screw devices (discussed in Chapter 2) can act as reactors for materials with viscosities as high as 5 million cP. Although such devices resemble conventional extruders, the screws are specially designed for long residence time with vents for handling high vapor loads. These reactors are not applicable to polymer processes requiring a signiﬁcant amount of heat transfer because at such high viscosities, heat transfer is an extremely difﬁcult problem. Double-screw–type reactors have found application in polycondensation reactions, where heat transfer requirements are low. They are excellent plug ﬂow machines, lending themselves well to continuous polymerization processes. The use of extruder as the main polymerizer can have a number of advantages since the extruder by its nature is a high-rate, continuous conveyor and mixer of high-viscosity products. Increase in viscosity is, in fact, advantageous in a screw extruder since it improves the conveying capacity. Heat exchange through the barrel and screw surfaces also allows for a good temperature control. It should thus be possible to carry polymerizations to high conversions in an extruder reactor. 53 Characteristics of Polymers and Polymerization Processes Since screw extruders are commonly used for polymer processing operations such as palletizing, compounding, mixing, and devolatilizing, a signiﬁcant advantage of the extruder reactor may arise from the combination of several processing steps in an integrated operation. For example, the preparation of the feedstocks, addition of catalysts, the polymerization reaction, followed by devolatilizing, online compounding, and extrusion can all be done in one single piece of equipment, thereby eliminating several solidiﬁcation and remelting stages. It also results in space savings, compact plant design, and ease of control. These advantages can far outweigh the initial high cost of the extruder reactor. Apart from its signiﬁcant economic savings, the one-step integrated operation, by eliminating polymer degradation in remelting, may make ultrahigh-molecular-weight grades of polymer processable. In the case of highly exothermic free-radical polymerizations, such as that of styrene, polymerization cannot be completed in an agitated vessel because of very high viscosities in the ﬁnal stages. Therefore, when one polymerizes styrene in bulk in an agitated vessel, one stops at 30%–40% styrene conversion and transfers the viscous syrup to another type of reactor. In the 1930s, Otto Rohem used plate-and-frame press for the second stage reactor. The resulting polymer blocks are cooled, ground, and granulated. This batch plate-and-frame process is still used. The capacity of one standard press is 5000–6000 tons/year. The most widely used process for polymerization of styrene today, however, is the continuous mass process using a tower as a reactor (Figure 1.38). (The continuous tower process was also adopted for vinyl acetate polymerization.) According to the procedure of I. G. Farbenindustrie, the pre-polymerized styrene is fed at the top of the polymerization tower. The polymerization is completed to a conversion of 92%– 98% in the tower, with temperatures of 140°C at the top and 200°C at the bottom, within 3–10 h. Monomeric styrene may be removed as vapor at the top of the tower and recycled to the prepolymerization vessels. After 96%–98% conversion, the polymer melt that is discharged at the bottom is devolatilized in a vented vacuum extruder and blended with lubricants, colorants, and other additives. Monomer Monomer 30%–40% conversion Batch prepolymer kettles 80°C–100°C 100°C–100°C 110°C–140°C 140°C–160°C Continuous tower reactor 160°C–180°C 180°C–200°C Short barrel extruder Cooling belt Pellets FIGURE 1.38 Bulk polymerization of styrene (tower process). 54 Plastics Technology Handbook 1.4.2.4.2 Polymers Insoluble in Their Monomers Unlike polystyrene, PVC is insoluble in its monomer and so precipitates out during bulk polymerization. Bulk polymerization has an advantage over the standard suspension polymerization in that the product is of higher purity, being free of foreign materials, such as dispersing agents or salts that inﬂuence moisture absorption, transparency, and electrical properties unfavorably. In the original one-step Pechiney–Saint Gobain (PSG) process, a 12-m3 ball mill was used as reactor (Figure 1.39). The ball densiﬁes the precipitated ﬂuffy PVC to a ﬁne powder, suitable for extrusion or calendaring. PSG further developed a two-step bulk polymerization process (Figure 1.40). One line consists of one 8-m3 fast agitated pre-polymerization kettle followed by three or four 16-m3 batch ribbon blender-type autoclaves. Half of the vinyl chloride is fed to the pre-polymerization kettle ﬁtted with a reﬂux condenser and polymerized to 7% conversion in less than 1 h by the addition of a fast initiator. The speed of agitation determines the size of the polymer particles. These polymer particles act as seed in the second step of polymerization performed in the ribbon blender-type autoclave to which the second half of vinyl chloride is fed. Polymerization is completed in the autoclave in 5–9 h. The unreacted monomer is removed by evaporation, condensed, and reused in the following batch. After applying vacuum, the resin is discharged to screens for removing oversized particles. This two-stage bulk polymerization process gives dry-blend PVC resin. 1.4.2.4.3 Solution Polymerization Solution polymerization is frequently employed in copolymerization, where the copolymer formed is soluble in the solvent, and the polymer solution is used directly in surface coating or as adhesive. Most stereospeciﬁc elastomers are produced by solution polymerization. The method involves a generally inert hydrocarbon that reduces the viscosity of the polymerization mass. Typical examples of Horizontal autoclave Metal balls FIGURE 1.39 PSG reactor (one-step process). 1/2 of feed 1/2 of feed Monomer Vacuum evaporator Prepolymerized fast agitation 7% conversion in 1 h FIGURE 1.40 Ribbon-blender type autoclave 5–9 h Two-step process for bulk polymerization of vinyl chloride. Screen Dry blend PVC resin 55 Characteristics of Polymers and Polymerization Processes elastomers produced by this method are cis-polybutadiene, polyisoprene, SBR solution, and certain types of ethylene-propylene terpolymers. In another method of solution polymerization, the polymer has only limited solubility in solvent, frequently only to the extent of a few percent. This may be called solvent–nonsolvent polymerization. This polymerization system is applied in the production of high-density polyethylene, polypropylene, butyl rubber, and certain types of ethylene–propylene terpolymer. Solvent–nonsolvent polymerizations are all operated continuously, and reactor residence time is normally in the range of 1 to 4 h. Slurry concentration ranges from 15% to 40% by weight and the heat-transfer coefﬁcient generally ranges from 35 to 200 Btu/(h)(ft2)(°F). For such high levels of heat transfer, the agitation in the reactor is in the range of 15 to 40 hp/1000 gallons of reactor capacity. Polymer buildup on the reactor wall depends on the turbulence at the wall, the quality of the surface ﬁnish, and the surface temperature. For solution polymerization, various means of heat removal may be applicable, such as cold-solvent feed, reﬂux cooling, or an external cooling loop. For this last one, the preferable type of cooler is a doublepipe exchanger or a single cell. 1.4.2.4.4 Suspension Polymerization Resins are obtained in the form of beads. Most PVC, dry blend, and general-purpose resin are produced by batch suspension polymerization. Goodrich and Wacker in the 1930s developed the batch process and equipment design, which have changed little. A ﬂow sheet of the process is shown in Figure 1.41. Innovations in the process centered on (a) a faster initiator, (b) a larger reactor, and (c) continuous ﬂash dryers combined with batch-operated ﬂuidized bed dryers. Several designs for continuous suspension polymerization are shown in Figure 1.42. These consist of towers with multiple-blade agitators. Suspension polymerization is frequently combined with bulk polymerization such as in the manufacture of high-impact polystyrene and ABS (see Figure 1.43). Polybutadiene, GRS, or another elastomer is dissolved in styrene or a mixture of styrene and acrylonitrile. Thermal polymerization is carried out at 100°C for 18 h to a conversion of 30%. The syrup is then suspended in water containing 0.3% benzoyl peroxide as initiator plus a mixture of suspending agents. The polymerization is completed at 100°C within 6 h. The product is of uniform bead size and is centrifuged, washed, and dried. 1.4.2.4.5 Emulsion Polymerization Emulsion polymerization has found its largest application in the manufacture of synthetic rubber, such as polybutadiene, GRS, and other copolymer rubbers. Cascading kettles (Figure 1.36) are used in their continuous polymerization process. Agitated jacketed kettle Let down tank FIGURE 1.41 Centrifuge Rotary dryer Suspension polymerization of vinyl chloride: batch process (Goodrich–Wacker). 56 Plastics Technology Handbook Reﬂux condenser Feed Feed Perforated plates H/D = 6-15/1 Agitated chamber Distance between turbines increases downward Baﬄe (a) FIGURE 1.42 (b) Continuous suspension polymerization reactor: (a) Union carbide. (b) BASF. Rubber Monomer Preliminary dissolution Bulk polymerization Water Suspension polymerization P To centrifuge P FIGURE 1.43 Flow sheet for bulk suspension polymerization (intermittent). High-pressure continuous emulsion polymerization of ethylene is known for many years. A ﬂow diagram of a continuous process is illustrated in Figure 1.44. High-impact polystyrene, ABS, and similar ter- and tetra-polymers are produced by graft copolymerization using the emulsion method. To produce ABS graft polymers, styrene and acrylonitrile are grafted upon polybutadiene latex and blended with SAN copolymer latex in the desired proportion (see Figure 1.45). The main application of emulsion PVC is as paste resin. The resin particles are extremely ﬁne and relatively nonabsorptive toward plasticizers with which it forms a paste at room temperature. In 1935, I. G. Farbenindustrie started the ﬁrst production by employing a batch process. PVC resin is also made by continuous emulsion polymerization process. It started in 1937. 57 Characteristics of Polymers and Polymerization Processes pH adjustor K2S2O8 Ethylene Emulsiﬁer Water Compressor Feed tank Emulsion reactor P Defoam tank Screen FIGURE 1.44 Vacuum wiped ﬁlm evaporator Flow sheet for high-pressure continuous emulsion polymerization of ethylene. Initiator emulsiﬁer water Acrylonitrile Butadiene Latex blend tank Graft copolymer latex SAN latex Styrene Coagulant (water) Filter Emulsion polymerization reactor FIGURE 1.45 Dryer Polybutadiene latex Plant for ABS graft polyblend. Product 58 Plastics Technology Handbook 1.5 Conﬁgurations of Polymer Molecules The long threadlike shape of polymer molecules induces generally a random arrangement leading to inter- and intramolecular entanglement, somewhat like a bowl of cooked spaghetti. A typical molecule of polyethylene, for example, might be represented by a cylindrical chain with a length of 50,000 Å and a diameter of less than 5 Å. This is similar to a rope that is 45 m long and 4.5 mm in diameter. A molecule such as this can easily get knotted and entangled with surrounding molecules. In the structure some parts of the molecular chains can be more ordered than others. The ordered regions are termed micelles or crystallites. These regions are embedded in the unordered or amorphous matrix. It is quite logical that if we want a molecule to go into some kind of ordered, repetitive pattern, then its structure must also have a regularly repeating pattern. The degree of crystallinity of the polymer will thus increase with the linearity and steric regularity of the molecules and also with interchain attractive forces. Linear polymers are found in nature, or they may be formed by polymerization of simple monomers. When monosubstituted ethylene monomers (CH2═CHR) polymerize, the addition reaction may be headto-tail, head-to-head/tail-to-tail, or a random mixture of the two: CH2 CH CH2 R CH CH2 CH R CH2 R CH R Head-to-tail CH CH CH2 R R CH2 CH CH R R CH2 CH CH CH2 CH2 R R Head-to-head/tail-to-tail CH2 CH R Random CH2 CH CH2 CH2 R The head-to-tail conﬁguration is preferred almost to the exclusion of the other two. An important reason for this is steric hindrance, which favors head-to-tail reaction, especially if R is bulky. Another aspect of stereoregularity is tacticity. Figure 1.46 show a polymer chain in which all of the chain carbons are in the same plane. Three conﬁgurations can be obtained: A polymer molecule is isotactic if all the substituted groups lie on the same side of the main chain. In a syndiotactic polymer molecule the substituted groups regularly alternate from one side to the other. The molecule is atactic if the positioning of substituted groups is random. The relative arrangement of groups and atoms in successive monomer units in a polymer chain not only affects the crystallinity but also induces completely different properties in polymers. One example of this effect is found in polypropylene steroisomers. (The three stereoisomers of polypropylene can be obtained by replacing R by CH3 in Figure 1.46.) The structural difference results in profound variations in the properties of polypropylene isomers. As is evident from Table 1.4, the three polypropylene isomers appear to be three altogether different materials. 1.6 Conformations of a Polymer Molecule Just as a rope can be stretched, folded back on itself, curled up into a ball, entangled, knotted, and so forth, a polymer molecule can take on many conformations. Consider a molecule as a chain of N links, each link of length l0 and attached to the preceding link by a rotating joint. 59 Characteristics of Polymers and Polymerization Processes R H R C C R H R H R H R H C C C C C H H H H H H H H R H H H R H H H C C C C H H C C C H H H R C C H H H C C C C C H H R H H H R H H R H R C C C (c) H H H H H H H H H R (b) H R C C C (a) H R R H H H C C C C C H H H H H R H C C C C C C H R H H H R H H H R H R H H H C C C C C C R H H H H H R H C C C C FIGURE 1.46 Diagrams of (a) isotactic, (b) syndiotactic, and (c) atactic conﬁguration in a vinyl polymer. The corresponding Fischer projections are shown on the right. TABLE 1.4 Properties of Polypropylene Stereoisomers Stereoisomers Property Isotactic Syndiotactic Atactic Appearance Hard solid Hard solid Melting temperature (°C) 175 131 Soft rubbery <100 Density (g/cc) Tensile strength [psi (N/m2)] 0.90–0.92 5,000 (3.4 × 107) 0.89–0.91 – 0.86–0.89 – Solubility Insoluble in most organic solvents Soluble in ether and aliphatic hydrocarbon Soluble in common organic solvents Crystallinity (%) <70 – – Glass transition temperature (°C) 0 to −35 – −11 to −35 For a free rotating polymer chain the average conformation is characterized by the mean square distance 〈 r 2 〉 between the ends of the chain and is given by [5] 〈 r2 〉 = Nl02 (1.63) If we assume that adjacent links in the chain form ﬁxed angles q with each other but rotate freely about that angle, then 〈 r2 〉 = 1 − cos q 2 Nl = 2Nl02 1 + cos q 0 (1.64) since bonds in a tetrahedral carbon unit are at 109.5° to each other [5]. Thus, for a polyethylene molecule p comprising, say, 40,000 freely rotating –CH2– units (l0=1.54 Å), the ﬃﬃﬃﬃﬃﬃ end-to-end distance would only be 2N l0 or approximately 435 Å. In contrast, if fully extended, the 60 Plastics Technology Handbook molecule will be in all-trans conformation with a linear zigzag structure (like corrugated sheets) shown in (III). (The dotted lines denote bonds and the wedges signify bonds above the plane of the page.) H H H C H H C C H H H H H H C C H H C C H H C C H H C H H (III) H The end-to-end distance of this fully extended polyethylene molecule will be Nl0 sin(109.5°/2), or approximately 50,000 Å, for a molecule comprising 40,000 –CH2– units, and the contour length is Nl0 or 60,000 Å. 1.7 Polymer Crystallinity X-ray scattering and electron microscopy have shown that the crystallites are made up of lamellae which, in turn, are built-up of folded polymer chains as explained below. Lamellae are thin, ﬂat platelets on the order of 100–200 Å (0.01–0.02 mm) thick and several microns in lateral dimensions, while polymer molecules are generally on the order of 1,000–10,000 Å long. Since the polymer chain axis is perpendicular to the plane of the lamellae, as revealed by electron diffraction, the polymer molecules must therefore be folded back and forth within the crystal. This arrangement has been shown to be sterically possible. In polyethylene, for example, the molecules can fold in such a way that only about ﬁve chain carbon atoms are required for the fold, that is, for the chain to reverse its direction. Each molecule folds up and down in a regular fashion to establish a fold plane. As illustrated in Figure 1.47a, a single fold plane may contain many polymer chains. The height of the fold plane is known as the fold period. It corresponds to the thickness of the lamellae. Figure 1.47b shows an idealized model of lamellae structure with ideal stacking of lamellar crystals. A more useful model, however, is that of stacks of lamellae interspersed with and connected by amorphous regions that consist of disordered chain segments of polymer molecules. Such a model, referred to as interlamellar amorphous model (Figure 1.47c), helps explain the ductility and strength of polymers as a direct consequence of the molecular links between the lamellae forming interlamellar ties. For semicrystalline polymers with amorphous regions to the tune of 20%–50%, it is often more advantageous to adopt a fringed micelle or fringed crystalline model (Figure 1.47d). It pictures polymers as two-phase systems in which the amorphous regions are interspersed between the randomly distributed crystallites. 1.7.1 Determinants of Polymer Crystallinity The extent to which polymer molecules will crystallize depends on their structures and on the magnitudes of the secondary bonds forces among the polymer chains: the greater the structural regularity and symmetry of the polymer molecule and the stronger the secondary forces, the greater the tendency toward crystallization. We give a few examples: 1. Linear polyethylene has essentially the best structure for chain packing. Its molecular structure is very simple and perfectly regular, and the small methylene groups ﬁt easily into a crystal lattice. Linear polyethylene (high density) therefore crystallizes easily and to a high degree (over 90%) even though its secondary forces are small. Branching impairs the regularity of the structure and makes chain packing difﬁcult. Branched polyethylene (low density) is thus only partially (50%–60%) crystalline. Most of the differences in properties between low-density and high-density polyethylenes can be attributed to the higher crystallinity of the latter. Thus, linear polyethylenes have higher density than the branched material (density range of 0.95–0.97 versus 0.91–0.94 g/cm3), higher Characteristics of Polymers and Polymerization Processes 61 Fold surface Chain ends Fold plane (side view) (a) Loose chain end (b) Amorphous region Interlamellar ties Lamellae (d) (c) FIGURE 1.47 Schematic representation of (a) fold plane showing regular chain folding, (b) ideal stacking of lamellar crystals, (c) interlamellar amorphous model, and (d) fringed micelle model of randomly distributed crystallites. 2. 3. 4. 5. melting point (typically 135 versus 115°C), greater stiffness (modulus of 100,000 versus 20,000 psi), greater tensile strength, greater hardness, and less permeability to gases and vapors. Substituents hanging off polymer chains lead to difﬁculties in packing and generally decrease the tendency toward crystallization. Moreover, crystallization does not take place easily when polymer molecules have a low degree of symmetry. Thus, polymers such as polystyrene, poly(methyl methacrylate), poly(vinyl acetate), etc., all of which have bulky side groups oriented at random with respect to the main carbon chain (in atactic polymers), show very poor crystallization tendencies and tend to have amorphous structures. However, crystallinity would result if the side groups could be arranged in a regular orientation. Indeed, this can be done by controlled polymerization with properly chosen catalysts. Copolymerization reduces the structural symmetry of a polymer. Thus it is a very effective method of decreasing the crystallization tendency of a polymer. Chain ﬂexibility also affects the crystallizability of a polymer. Excessive ﬂexibility in a polymer chain, as in natural rubber and polysiloxanes, gives rise to difﬁculty in chain packing, with the result that such polymers remain almost completely in the amorphous state. In the other extreme, excessive rigidity in polymers due to extensive cross-linking, as in thermosetting resins like phenol– formaldehyde and urea–formaldehyde, also results in an inability to crystallize. The presence of polar groups—such as amide, carboxyl, hydroxyl, chlorine, ﬂuorine, and nitrile— along the polymer chains greatly increases the intermolecular or secondary attraction forces, which is favorable for crystallization. However, high secondary forces alone may not give rise to high crystallinity unless the chain segments are aligned. Mechanical stretching of the polymer makes this alignment easier. For example, nylon-6,6 (a polyamide) has less than the expected degree of crystallinity in the unstretched condition and is used as a plastic. Highly crystalline strong ﬁbers are produced by stretching (cold-drawing) the polyamide polymer 400%–500%. Mechanical stretching 62 Plastics Technology Handbook also makes it possible to develop a degree of order and crystallinity in several other thermoplastic resins that do not ordinarily crystallize. An unusual example of alignment and crystallization on stretching is rubber. 6. The degree of crystallinity of polymeric materials is reduced by adding plasticizers. Crystallization in many synthetic resins is not always desirable because it makes shaping more difﬁcult and reduces transparency by closely packing neutralization of the intermolecular forces of attraction by coming between polymer molecules, thus enhancing ﬂexibility and plasticity, are often added to the polymeric mass before shaping. The oldest example is celluloid, made by plasticizing nitrocellulose (ordinarily a crystalline material) with camphor. Cellophane (regenerated cellulose ﬁlm produced by a viscose process) is plasticized with glycerine to prevent crystallization and loss of transparency. Polyvinyl chloride (PVC) is made ﬂexible by adding plasticizers, such as dioctyl phthalate, for use as wire coating, upholstery, ﬁlm and tubing. The unplasticized rigid PVC is used for the production of pipe, sheet, and molded parts. The disadvantage of plasticizers is that they reduce the tensile strength and chemical resistance of the material. 1.8 The Amorphous State The amorphous state is the characteristic of all polymers at temperatures above their melting points (except under special circumstances where liquid crystals may form). If a molten polymer retains its amorphous nature on cooling to the solid state, the process is called vitriﬁcation. In the vitriﬁed amorphous state, the polymer resembles a glass. It is characteristic of those polymers in the solid state that, for reasons of structure, exhibit no tendency toward crystallization. The amorphous solid state is characterized by glass transition (Tg), which is described in a later section. We consider below only the behavior of polymer melt. When an amorphous polymer achieves a certain degree of rotational freedom, it can be deformed. If there is sufﬁcient freedom, the molecules begin to move past one another and the polymer ﬂows. The science of deformation and ﬂow is called rheology. It is of fundamental importance in industrial applications since it is usually in the molten state that polymers are molded into useful objects. To cause a polymer to deform or ﬂow requires the application of a force. If a force is applied and then withdrawn quickly, the polymer molecules tend to revert to their previous undeformed conﬁguration, a process called relaxation. In other words, the polymer melt exhibits a certain elastic quality. This elasticity comes about because the molecules were disturbed from what was a thermodynamically favorable arrangement. If, however, the force is applied gradually and consistently, the molecules begin to ﬂow irreversibly. (Silly putty, a siloxane polymer, is ideal for demonstrating this effect. If dropped, it bounces; but it can be shaped by the slow application of pressure.) Because of entanglement of the polymer chains and frictional effects, the ﬂowing liquid will be very viscous. This combination of properties, namely elasticity and viscous ﬂow, is why polymers are referred to as viscoelastic materials. A consequence of viscoelasticity of polymer melt is die swell, which refers to the fact that the thickness of the melt emerging through a narrow oriﬁce or die is greater than the width of the die opening. This is explained as follows: as the molecules ﬂow rapidly through the die opening, they are compressed, and when they emerge, the resultant reduction in pressure causes the molecules to rebound to a degree. This dimensional increase of the extrudate must be taken into account by engineers who design polymer processing machinery (Chapter 2). Chain entanglement that contributes to the high viscosity of polymer melt is clearly going to change as the molecular weight increases. Molecular weight is thus a critical variable in polymer rheology. Studies c ) for entanhave shown that with ﬂexible chain polymers, there exists a critical molecular weight (M c varies c falls in the range 4,000–15,000. Although M glement to begin. For most common polymer, M from one polymer to another, it has been shown that elimination of mass effects arising from substituents on the chains and calculation of chain length leads to a value that is remarkably constant from one Characteristics of Polymers and Polymerization Processes 63 polymer to another and corresponds to a DP of about 600. In other words, a critical chain length, rather than a critical molecular weight per se, is necessary for entanglement. 1.9 Structural Shape of Polymer Molecules Polymers can be classiﬁed, based on the structural shape of polymer molecules, as linear, branched, or cross-linked. Schematic representations are given in Figure 1.48. Linear polymers have repeating units linked together in a continuous length (Figure 1.48a). When branches protrude from the main polymer chain at irregular intervals, the polymer is termed a branched polymer. Branches may be long or short, forming a comblike structure (Figure 1.48b), or divergent (Figure 1.48c), forming a dendritelike structure. [Regularly repeating side groups which are a part of the monomer structure are not considered as branches. Thus polypropylene is a linear polymer, as are polystyrene and poly(methyl methacrylate).] Both linear and branched polymers are thermoplastic; that is, they can be softened and hardened reversibly by changing the temperature. Fabricating processes like injection molding, extrusion molding, casting, and blowing take advantage of this feature to shape thermoplastic resins. The rigidity of thermoplastic resins at low temperatures is attributed to the existence of secondary bond forces between the polymer chains. These bonds are destroyed at higher temperatures, thereby causing ﬂuidity of the resin. Polymers used as textile ﬁbers are linear. However, they must satisfy two additional requirements: (1) high molecular weight and (2) a permanent orientation of the molecules parallel to the ﬁber axis. The molecules must have a high degree of order and/or strong secondary forces to permit orientation and crystallization. The chain orientation necessary to develop sufﬁcient strength by crystallization is achieved by a process known as could drawing, in which the initially formed ﬁlaments (unoriented or only slightly oriented) are drawn at a temperature above the glass transition temperature (discussed later), which is the temperature at which sufﬁcient energy is available to the molecular segments to cause them to begin to rotate. (b) (a) (c) (d) (e) FIGURE 1.48 Schematic representation of (a) linear, (b and c) branched, and (d and e) cross-linked polymers. The branch points and junction points are indicated by heavy dots. 64 Plastics Technology Handbook Elastomeric materials, like thermoplastic resins and ﬁbers, are essentially linear polymers. But certain distinctive features in their molecular structure give rise to rubberlike elasticity. Elastomeric polymers have very long chain molecules occurring in randomly coiled arrangements in the unstressed condition. A large deformation is thus possible merely by reorienting the coiled molecules. When elongated, the molecular coils partially open up and become aligned more or less parallel to the direction of elongation. The aligned conﬁguration represents a less probable state or a state of lower entropy than a random arrangement. The aligned polymer chains therefore have a tendency to return to their original randomly coiled state. The large deformability of elastomeric materials is due to the presence of a certain internal mobility that allows rearranging the chain orientation, the absence of which in linear chain plastic materials (at normal temperatures) constitutes the essential difference between the two groups. Although the aforesaid requirements are necessary conditions for ensuring a large extent of deformability, the remarkable characteristic of the rubbery state—namely, nearly complete recovery—cannot be obtained without a permanent network structure, since permanent deformation rather than elastic recovery will occur. A small amount of cross-linkage is necessary to provide this essential network structure [30,31]. Natural rubber (polyisoprene), for example, simply ﬂows like an extremely viscous liquid at room temperature if it is not cross-linked. Cross-links are introduced into a rubber by heating raw rubber with sulfur (1%–2% by weight) and accelerating agents. Sulfur reacts with the double-bonded carbon atoms to produce a network structure, as shown schematically in Figure1.49. The amount of crosslinkage must be as small as possible to retain the structure; excessive cross-linkages will make the internal structure too stiff to permit even the required rearrangement of chain orientation during both deformation and recovery—in other words, it will destroy the rubbery state. An example of this is best CH3 CH2 C CH CH2 n (a) CH3 SX CH2 C CH CH3 SX CH3 CH2 CH2 C CH CH2 CH2 SX CH2 C CH SX C CH CH3 SX CH2 CH2 C CH3 CH CH2 CH2 C CH CH3 SX (b) (c) FIGURE 1.49 Vulcanization of natural rubber with sulfur. (a) Linear polyisoprene (natural rubber). (b) Vulcanized structure (idealized) showing cross-links of sulﬁde groups. The number, x, of sulfur atoms in a cross-link is 1 or 2 in efﬁcient vulcanization systems but may be as high as 8 under conditions where cyclic and other structures are also formed in the reaction. (c) The effect of cross-linking is to introduce points of linkage or anchor points between chain molecules, restricting their slippage. The cross-links in elastomers are typically a few hundred carbon atoms apart. Characteristics of Polymers and Polymerization Processes 65 furnished by ebonite, which is a rigid plastic made by vulcanizing natural rubber with large quantities of sulfur. Example 4: (a) How much sulfur is required to fully cross-link natural rubber? (b) What is the sulfur content of vulcanized natural rubber that is 50% cross-linked? (Assume that each cross-link contains one sulfur atom). Answer: Mer weight of isoprene (Figure 1.49a): C5 H8 = (5)(12) + (8)(1) = 68 g=mer Assuming x = 1 in Figure 1.49b, one sulfur atom, on the average, is required for cross-linking per mer of isoprene. Therefore, (a) Amount of sulfur = (32/68)100 = 47 g/100 g of raw rubber (b) Sulfur content = 100(0.5)(32)/[0.5(32)+68] = 19% Example 5: A rubber contains 60% butadiene, 30% isoprene, 5% sulfur, and 5% carbon black. What fraction of possible cross-links are joined by vulcanization? (Assume that all the sulfur is used in cross-linking.) Answer: 1 mer weight of butadiene (C4H6) = (4)(12)+(6)(1) = 54 g/mer 1 mer weight of isoprene (C5H8) = (5)(12)+(8)(1) = 68 g/mer 1 atomic weight of sulfur = 32. Assuming that, on the average, one sulfur atom per mer is required for cross-linking, we get Fraction of cross‐links = 5=32 = 0:101 or 10:1 % 60=54 + 30=68 1.10 Thermal Transitions in Polymers The term “transition” refers to a change of state induced by changing the temperatures or pressure. Two major thermal transitions are the glass transition and the melting, the respective temperatures being called Tg and Tm. 1.10.1 Tg and Tm All polymers are hard rigid solids at sufﬁciently low temperatures, but as the temperature rises a thermoplastic polymer eventually acquires sufﬁcient thermal energy to enable its chains to move freely enough for it to behave like a viscous liquid (assuming there is no degradation). There are several ways in which a polymer can pass from the solid to the liquid phase, depending on the structure of the polymer chains and their arrangement in the sample. The different types of thermal response in this transition from a rigid solid to an eventually liquid state can be illustrated in several ways. One of the simplest and most satisfactory is to trace the change in speciﬁc volume, as shown schematically in Figure 1.50. In ﬁrst-order transitions, such as melting, there is a discontinuity in the volume–temperature plot (Figure 1.50) or enthalpy–temperature plot at the transition temperature. In second-order transitions, only a change in slope occurs and thus there is a marked change in the ﬁrst derivative or temperature coefﬁcients, as illustrated in Figure 1.51. The glass transition is not a ﬁrst-order transition, as no discontinuities are observed at Tg when the speciﬁc volume or entropy of the polymer is measured as a 66 Plastics Technology Handbook A Speciﬁc volume (V ) T °m Tm F D C B Glass Melt Tg E G Glass H Crystallitie s Temperature (T) FIGURE 1.50 Schematic representation of the change of speciﬁc volume of a polymer with temperature for a completely amorphous sample (A–B–C), a semicrystalline sample (A–D–E), and a perfectly crystalline material (A–F–G). a V Tg (a) Tg T (c) T Cp H Tg Tg (b) T (d) T FIGURE 1.51 Idealized variations (a, b) in volume (V) and enthalpy (H). Also shown (c, d) are a, the volume coefﬁcient of expansion, and Cp, the heat capacity, which are, respectively, the ﬁrst derivatives of V and H with respect to temperature (T). function of temperature. However, the ﬁrst derivative of the property–temperature curve, i.e., the temperature coefﬁcient of the property, exhibits a marked change in the vicinity of Tg; for this reason it is sometimes called a second-order transition. a= 1 V ∂V ∂T (1.65) P where V is the volume of the material, and a has the units K−1. While this quantity increases rather sharply at Tg, the increase actually occurs over a range (10°C–30°C). Similar changes occur in the heat capacity at constant pressure (Cp), which is the ﬁrst derivative of enthalpy H with respect to temperature. Characteristics of Polymers and Polymerization Processes 67 A thermoplastic polymer may be completely amorphous in the solid state, which means that the polymer molecular chains in the specimen are arranged in a totally random fashion. Referring to Figure 1.50, the volume change in amorphous polymers follows the curve ABC. In the region C–B, the polymer is a glassy solid and has the characteristics of glasses, including hardness, stiffness and brittleness. In the glassy region, the available thermal energy (RT energy units/mol) is insufﬁcient to allow rotation about single bonds in the polymer backbone overcoming intramolecular energy barriers, and movements of large (some 10–50 consecutive chain atoms) segments of the polymer chain cannot take place. But as the sample is heated, it passes through a temperature Tg, called the glass transition temperature, above which it softens and becomes rubberlike. This is an important temperature and marks the onset of extensive molecular motion which is reﬂected in marked changes in properties, such as speciﬁc volume, refractive index, stiffness, and hardness. Above Tg, the material may be more easily deformed or become ductile. A continuing increase in temperature along B–A leads to a change of the rubbery polymers to a viscous liquid without any sharp transition. In a perfectly crystalline polymer, all the chains would be contained in regions of three-dimensional order, called crystallites, and no glass transition would be observed, because of the absence of disordered chains in the sample. A perfectly crystalline polymer, on heating, would follow curve G–F–A, melting at Tm° to become a viscous liquid. Perfectly crystalline polymers are, however, not encountered in practice and real polymers may instead contain varying proportions of ordered and disordered regions in the sample. These semicrystalline polymers usually exhibit both Tg and Tm (not Tm° ) corresponding to the disordered and ordered regions, respectively, and follow curves similar to E–H–D–A. As Tm° is the melting temperature of a perfectly crystalline polymer of high molecular weight Tm is lower than Tm° and more often represents a melting range, because the semicrystalline polymer contains crystallites of various sizes with many defects which act to depress the melting temperature. Both Tg and Tm are important parameters that serve to characterize a given polymer. While Tg sets an upper temperature limit for the use of amorphous thermoplastics like poly(methyl methacrylate) or polystyrene and a lower temperature limit for rubbery behavior of an elastomer like SBR rubber or 1,4-cispolybutadiene, Tm or the onset of the melting range determines the upper service temperature. Between Tm and Tg, semicrystalline polymers tend to behave as a tough and leathery material. As a general rule, however, semicrystalline polymers are used at temperatures between Tg and a practical softening temperature which lies above Tg and below Tm. (The onset of softening is usually measured as the temperature required for a particular polymer to deform a given amount under a speciﬁed load. These values are known as heat deﬂection temperatures. Such data do not have any direct relations with Tm, but they are widely used in designing with plastics.) The Tg and Tm values for some polymers are shown in Table 1.5. In general, both Tg and Tm are affected in the same manner by considerations of polymer structure [32]. Thus, both Tg and Tm increase with higher molecular symmetry, structural rigidity, and secondary forces of polymer chains. The Tg and Tm values of a polymer determine the temperature range in which it can employed. Amorphous elastomeric polymers, for example, must be used at temperatures (region B–D in Figure 1.50) well above Tg to permit the high, local segmental mobility required in such materials. Thus styrene– butadiene (25/75) copolymer (Tg = −57°C), polyisoprene (Tg = −73°C), and polyisobutylene (Tg = −73°C) can be used as rubbers at ambient temperatures. Amorphous structural polymers, such as polystyrene and poly(methyl methacrylate), depend on their glasslike rigidity below Tg for their utility; they should therefore have high Tg values so that under ambient conditions they are well below Tg. Tough, leatherlike polymers are limited for use in the immediate vicinity of their Tg. Such behavior is observed in vinyl chloride-based plastics, which are used as substitutes for leather in automobile seat covers, travel luggage, and ladies’ handbags. Highly crystalline ﬁber-forming polymers must be used at temperatures substantially below Tm (about 100°C), since changes in crystal structure can occur as Tm is approached. The Tm of a ﬁber must therefore be above 200°C to remain unaffected at use temperatures encountered in cleaning and ironing. (Tm should not, however, be excessively high—not more than 68 Plastics Technology Handbook TABLE 1.5 Glass Transition Temperatures (Tg) and Crystalline Melting Temperatures (Tm) of Polymers Polymer Tg (°C) Tm (°C) −115 137 Polyoxymethylene −85 181 Polyisoprene (natural rubber) Polyisobutylene −73 −73 28 44 Polypropylene −20 176 Poly(vinylidene chloride) Poly(chlorotriﬂuoroethylene) (kel-F) −19 45 190 220 Poly(hexamethylene adipamide) (nylon-6,6) 53 265 Poly(ethylene terephthalate) (Terylene, Dacron) Poly(vinyl chloride) 69 81 265 212 Polystyrene 100 240 Poly(methyl methacrylate) (Perspex, Lucite) Cellulose triacetate 105 105 200 306 Polytetraﬂuoroethylene (Teﬂon) 127 327 Polyethylene (high density) 300°C; otherwise spinning of the ﬁber by melt-spinning processes may not be possible.) The Tg of a ﬁber, on the other hand, should have an intermediate value, because too high a value of Tg would interfere with the stretching operation as well as with ironing, and too low a value of Tg would not permit crease retention in fabrics. Nylon and terylene, as may be seen from Table 1.5, therefore have optimal values of Tm and Tg. Semicrystalline polymers with about 50% crystallinity are used at temperatures between Tg and Tm, since in this range the material exhibits moderate rigidity and a high degree of toughness, somewhat analogous to reinforced rubber. Branched polyethylene (low density), with Tg = −120°C and Tm = 115°C, used at ambient temperatures is a typical example. 1.10.2 Regions of Viscoelastic Behavior As we have seen above, the transition that separates the glassy state from the viscous state is known as the glass–rubber transition. This transition attains the properties of a second-order transition at very slow rates of heating or cooling. In order to clearly locate the region of this transition and to provide a broader picture of the temperature dependence of polymer properties, the principal regions of viscoelastic behavior of polymers will be brieﬂy discussed. Broadly, there are ﬁve regions of viscoelastic behavior for linear amorphous polymers as shown in Figure 1.52. In region I (a–b), the polymer is glassy and frequently brittle. Typical examples are polystyrene and poly(methyl methacrylate) at room temperature. Young’s modulus in this region just below the glass transition temperature is approximately 3 × 1010 dyne/cm2 (3 × 109 Pa) and it is nearly the same for a wide range of polymers. Range II (b–c) is the glass transition region. Typically, the modulus drops a factor of about a thousand in a 20°C–30°C range. The behavior of polymers in this region is best described as leathery. For static or quasistatic measurements, such as illustrated in Figure 1.52, the glass transition temperature, Tg, is often taken at the maximum rate of turndown of the modulus at the elbow, i.e., where d2E/dT2 is at a maximum. A few Tg values are shown in Table 1.5. Region III (c–d) in Figure 1.52 is the rubbery plateau region. After a sharp drop that the modulus exhibits in the glass transition region, it again becomes almost constant in the rubbery plateau region, with typical values of 2 × 107 dyne/cm2 (2 × 106 Pa). In this region, polymers exhibit high rubber elasticity, so much so that an elastomer in this region can be stretched perhaps several hundred percent and substantially snap back to its original length on being released. 69 Characteristics of Polymers and Polymerization Processes 11 a 10 I b 9 10 8 II 8 7 7 c 6 III 6 d IV 5 e 5 Log E (Pa) Log E (dyne/cm2) 9 4 V f 3 4 Temperature FIGURE 1.52 Five regions of viscoelastic behavior for a linear, amorphous polymer: I (a–b), II (b–c), III (c–d), IV (d–e), and V (e–f). Also illustrated are effects of crystallinity (dotted line) and cross-linking (dashed line). In Region III (c–d) of Figure 1.52, two cases need be distinguished. First, if the polymer is linear, the solid line is followed on which the modulus drops off slowly with increasing temperature. The width of the plateau is governed primarily by the molecular weight of the polymer; the higher the molecular weight, the longer the plateau. Second, if the polymer is cross-linked, the dashed line in Figure 1.52 is followed, and improved rubber elasticity is observed, with the creep portion suppressed. A common example of a cross-linked polymer above its glass transition temperature is the ordinary rubber band. For cross-linked polymers, region III remains in effect at higher temperatures up to the decomposition temperature, and regions IV and V do not occur. [The discussion above has been limited to amorphous polymers. However, if the polymer is semicrystalline, the dotted line in Figure 1.52 is followed. Since the crystalline regions in the polymer matrix tend to behave as a ﬁller phase and also as a type of physical cross-link between the chains, the height of the plateau (i.e., the modulus) will be governed by the degree of crystallinity.] In the case of linear amorphous polymers, raising the temperature past the rubbery plateau region brings them to the rubbery ﬂow region—region IV. In this region, the polymer exhibits both rubber elasticity and ﬂow properties depending on the time scale of the experiment. For experiments performed in a short time, the physical entanglements of polymer chains are not able to relax and the material still behaves rubbery. For longer-duration experiments, the increased molecular motion imparted by the increased temperature cause chains to move, resulting in visible ﬂow. An example of a material in the rubbery ﬂow region at ambient temperature is Silly Putty which bounces like a ball when thrown (quick experiment) but deforms like a taffy when pulled (a much slower experiment). Region V (e–f) in Figure 1.52 is the liquid ﬂow region which is reached at still higher temperatures where the increased kinetic energy of the chains permits them wriggle out through entanglements rapidly and move as individual molecules, often producing highly viscous ﬂow. This is the melting temperature and it is always above the glass transition temperature. Example 6: A new polymer was reported to soften at 60°C but it is not known for sure whether the softening was a glass transition or a melting point. Describe a simple experiment to distinguish between the two possibilities. It is known however, that the new polymer is essentially a linear polymer. Answer: If 60°C is a glass transition, then heating the polymer slowly past 60°C would take it to the rubbery plateau region (region III in Figure 1.52), where the modulus E, and hence hardness would 70 Plastics Technology Handbook remain fairly constant with increase of temperature. For a melting transition, however, the modulus would drop rapidly and the polymer would become increasingly softer in a similar experiment. Pressing one’s thumb in an object is a simple way to gauze the object’s hardness. Its scientiﬁc analog is the measurement of hardness by indentation. In practice, the point of a weighted needle is allowed to rest on the polymer surface as the temperature is raised. The movement of the needle as it penetrates the surface can be monitored by means of an ampliﬁcation gauge. Though less accurate than other more sophisticated methods, it is useful for the preliminary engineering-oriented examination of systems. 1.10.3 Factors Affecting Tg As Tg marks the onset of molecular motion, a number of factors which affect rotation about links (necessary for movement of polymer chains) will also inﬂuence the Tg of a polymer. These include (a) chain ﬂexibility, (b) molecular structure (steric effects), (c) molecular weight, and (d) branching and cross-linking. 1.10.3.1 Chain Flexibility The ﬂexibility of the chain is undoubtedly the most important factor inﬂuencing the Tg of a polymer. The chain ﬂexibility depends more on the rotation or torsion of skeletal bonds than on changes in bond angles. When a randomly coiled chain is pulled out into an elongated conformation, the skeletal bonds “unwind” rather than undergo angular distortion (see Figure 1.53). Thus, ﬂexibility on a macroscopic scale depends on torsional mobility at the molecular level. If a highly ﬂexible chain is present, Tg will generally be low and if the chain is rigid, the Tg value will be high. For symmetrical polymers, the chemical nature of the backbone chain is the important factor determining the chain ﬂexibility and hence Tg. Chains made up of bond sequences that are able to rotate easily are ﬂexible, and hence polymers containing –(–CH2–CH2–)–,–(–CH2–O–CH2–)–, or –(–Si–O–Si–)– links will have correspondingly low values of Tg. For example, poly(dimethyl siloxane) has one of the lowest Tg values known(−123°C) presumably because the Si–O bonds have considerable torsional mobility. The value of Tg is raised markedly by the insertion of groups that stiffen the chain by impeding rotation, so that more thermal energy is required to set the chain in motion. Particularly effective in this respect is the p-phenylene ring. Thus, a chain consisting entirely of p-phenylene rings, namely, poly(p-phenylene) (IV) n (IV) Si O CH3 FIGURE 1.53 Elasticity of a polymer such as silicone rubber depends on the ease with which a random coil chain can be stretched out. The stretching is a consequence of the unwinding of bonds rather than a marked widening of bond angles. 71 Characteristics of Polymers and Polymerization Processes has a highly intractable, rigid structure with no softening point. This structure can be modiﬁed by introducing ﬂexible groups in the chain to produce tractable polymers with high values of Tg. Some examples are poly(phenylene oxide) (V), Tg = 83°C and poly(p-xylylene) (VI), Tg = 280°C as compared to polyethylene, Tg = −93°C, and poly(ethylene oxide), Tg = −67°C. O (V) n CH2 n (VI) 1.10.3.2 Steric Effects When the polymer chains are unsymmetrical, with repeat units of the type –(–CH2–CHX–)–, an additional restriction is imposed by steric effects depending on the size of the pendant group X. Bulky pendant groups hinder the rotation about the backbone and cause Tg to increase. There is some evidence of a correlation between Tg and the size of the pendant group measured by its molar volume VX. For example, Ti (°C) increases with increasing VX (cm3/mol) in the progressive series, polyethylene (Tg = −93, VX = 3.7), polypropylene (Tg = −20, VX = 25.9) polypropylene (Tg = 100, VX = 92.3), and poly(vinylnaphthalene) (Tg = 135, VX = 143.9). Superimposed on the pendant group size factor are the effects of polarity and the intrinsic ﬂexibility of the pendant group itself. Greater intermolecular interactions due to polar groups hinder molecular motion and increase Tg. Thus polar groups tend to encourage a higher Tg than nonpolar groups of similar size, as seen when comparing Tg (°C) and VX (cm3/mol) of polypropylene (Tg = −20, VX = 25.9), poly(vinyl chloride) (Tg = 81, VX = 22.1), and polyacrylonitirile (Tg = 105, VX = 30.0). On the other hand, greater ﬂexibility of side chain leads to lower Tg as is evident on examination of the polyacrylate series from methyl through butyl: poly(ethylacrylate) (Tg = −24, VX = 82.3), poly(propyl acrylate) (Tg = −48, VX = 104.5), and poly(butyl acrylate) (Tg = −55, VX = 126.7), where Tg is in °C and VX is in cm3/mol. A further increase in steric hindrance is caused by the presence of an a-methyl group, which restricts rotation even further, thus causing Tg to increase. Typical examples are the pair polystyrene (Tg = 100°C)– poly(a-methylstyrene) (Tg = 172°C) and the pair poly(methyl acrylate) (Tg = 6°C)–poly(methyl methacrylate) (Tg = 105°C). 1.10.3.3 Conﬁgurational Effects It should be noted that the steric effects of the pendant groups considered above are simply additional contributions to the main chain effects. Similarly cis–trans isomerism in polydienes and tacticity variations in certain a-methyl substituted polymers alter chain ﬂexibility and hence affect Tg. Well-known examples of cis–trans variations are polybutadiene cis(Tg = −108°C) and trans(Tg = −18°C) or polyisoprene cis(Tg = −73°C) and trans(Tg = −53°C). An example of tacticity variation is poly(methyl methacrylate) for which the isotactic, atactic, and syndiotactic stereostructures are associated with Tg values of 45, 105, and 115°C, respectively. 1.10.3.4 Effect of Cross-Linking When cross-links are introduced into a polymer, the molecular motion in the sample is restricted and Tg rises. This transition is ill-deﬁned for a high cross-link density, but at lower values Tg is found to increase linearly with the number of cross-links. 1.10.4 Factors Affecting Tm The application of macroscopic thermodynamics leads to some useful generalizations of factors affecting Tm. At Tm, the free energy change is zero, i.e., 72 Plastics Technology Handbook DGm = DHm − Tm DSm = 0 where Tm = DHm/DSm. This expression for Tm predicts that a high melting point can be the result of a high value of the enthalpy change DHm and/or a small value of the entropy change DSm in melting. The former corresponds to stronger binding of adjacent but unbonded units in the polymer lattice and thus to higher degree of crystallinity. The factors that affect crystallinity and Tm can be classiﬁed as symmetry, intermolecular bonding, tacticity, branching, and molecular weight. These are discussed below. If DSm is small, melting does not result in a large gain of conformational entropy and, to some degree, the structure of the solid must persist in the melt. For example, molecules of isotactic polypropylene crystallize in the form of helices and these are thought to occur in the melt as well, thus making DSm small and Tm high. 1.10.4.1 Symmetry The formation of stable crystalline regions (crystallites) in a polymer requires (a) that an efﬁcient closepacked arrangement of the polymer chains can be achieved in three dimensions and (b) that a favorable change in internal energy is obtained during this process. This imposes restrictions on the type of chain that can form crystallites easily and one would expect linear chains with high degree of symmetry such as polyethylene, polytetraﬂuoroethylene, various polyesters, and polyamides to crystallize most readily. Linear symmetrical molecules such as polyethylene, polytetraﬂuoroethylene, and other linear molecules with more complex backbones containing –(–O–)–, –(–COO–)–, and –(–CONH–)– groups, such as polyethers, polyesters, and polyamides, all posses a suitable symmetry for crystallite formation and usually assume extended zigzag conformations when aligned in the lattice. On the other hand, chains containing irregular units, which impair the symmetry, reduce the ability of the polymer to crystallize. Thus cis double bonds (VII), o-phenylene groups (VIII), and m-phenylene groups (IX) all encourage bending and twisting in the chains and make close packing very difﬁcult. H H C C (VII) (VIII) (IX) If, however, the phenylene rings are para-oriented, the chains retain their axial symmetry and can crystallize more readily. Similarly, double bonds in trans conﬁguration maintain the chain symmetry thus allowing for crystallite formation. This is highlighted by a comparison of the amorphous elastomeric cis-polyisoprene (Tm = 28°C) with highly crystalline trans-polyisoprene (Tm = 74°C), which is a nonelastomeric rigid polymer, or cis-1,4-polybutadiene (Tm = −11°C) with trans-1,4-polybutadiene (Tm = 148°C). Another aspect of high chain symmetry is the possibility of molecular motion within the crystal lattice contributing to higher Tm. For example, polyethylene and polytetraﬂuoroethylene are both sufﬁciently symmetrical to be considered as smooth, stiff cylindrical rods. In the crystal, these rods tend to roll over each other and change position when thermally agitated. This motion within the crystal lattice, called premelting, effectively stabilizes the lattice. Consequently, more thermal energy is required to break down the crystallite into a disordered melt, ant Tm is raised. Irregularly shaped polymers with bends and bumps in the chain cannot move in this way without disrupting the crystal lattice, and so have lower Tm values. 1.10.4.2 Intermolecular Bonding Any interaction between polymer chains in the crystal lattice serves to hold the structure more ﬁrmly and raise the melting temperature. In polyethylene, the close packing in crystallites achieved due to high chain symmetry and lack of substituents on the chains allows the van der Walls forces to act cooperatively and 73 Characteristics of Polymers and Polymerization Processes O C N δ+ δ– H O C N δ+ δ– H O C δ+ H N (a) CH2 H2C CH2 CH2 H2C C C H O H O H CH2 CH2 H2C CH2 CH2 CH2 H2C CH2 H2C CH2 H2C H2C CH2 H2C H2C N N N H O H O H C C C CH2 CH2 CH2 H 2C H2C CH2 O N N N O C H2C H2C H2C H2C (b) FIGURE 1.54 (a) Hydrogen bonds between neighboring chains of polyamide. (b) Arrangement of chains in hydrogen-bonded sheets in the crystal structure of nylon-6,6. provide additional stability to the lattice. In polymers containing polar substituents, for example, Cl, CN, or OH, the chains can be aligned and held rigidly by the strong dipole–dipole interactions between the substituents. This effect is more obvious in the symmetrical polyamides. These polymers can form intermolecular hydrogen bonds which greatly enhance crystallite stability and raise the Tm signiﬁcantly. This is illustrated in Figure 1.54 for nylon-6,6, where the extended zigzag structure greatly facilitates the formation of regular intermolecular hydrogen bonding. The increased stability is reﬂected in the value of Tm, which for nylon-6,6 is 267°C, compared with 137°C for high-density polyethylene. 1.10.4.3 Tacticity Stiffening of the polymer chain due to the presence of large pendant groups tends to raise Tm, but this will also increase the difﬁculty of close packing to form a crystalline array and so tend to lower Tm. However, 74 Plastics Technology Handbook this latter problem can be overcome if the pendant groups are arranged in a regular fashion along the chain. Thus isotactic polymers tend to form helices to accommodate the substituents in the most stable steric positions, and being of regular forms these helices are capable of regular alignment promoting crystallite formations. Thus isotactic polystyrene is semicrystalline (Tm = 240°C), though atactic polystyrene is amorphous. Syndiotactic polymers are also sufﬁciently regular to crystallize, not necessarily as a helix, but rather in glide planes. 1.10.4.4 Branching, Chain Flexibility, and Molecular Weight Chain ﬂexibility has a direct bearing on the melting point. Insertion of groups that stiffen the chain increases Tm, while introducing ﬂexible groups into the chain lowers the value of Tm (see “Factors Affecting Tm”). Branching in the side group tends to stiffen the chain and raise Tm, as shown in the series poly(but-1-ene) (Tm = 126°C), poly(3-methyl but-1-ene) (Tm = 145°C), poly(3,3′-dimethyl but-1-ene) (Tm > 320°C). However, if the side groups is ﬂexible and nonpolar, Tm is lowered. Also, if the chain is substantially branched to reduce the packing efﬁciency, the crystalline content is lowered, as is the melting point. A good example is low-density polyethylene where extensive branching lowers the density and Tm of the polymer. Molecular weight in relatively lower order can have signiﬁcant effect on Tm. Since chain ends are relatively free to move and the number of chain ends increases with the decrease in molecular weight Tm is lowered because less energy is then required to stimulate chain motion and melting. For example, polypropylene with molecular weight 2,000 has Tm = 114°C, whereas a sample with molecular weight 30,000 has Tm = 170°C. 1.10.5 Relation between Tm and Tg While Tm is a ﬁrst-order transition, Tg is a second-order transition and this precludes the possibility of a simple relation between them. There is, however, a crude relation between Tm and Tg. Thus the ratio Tg/ Tm range from 0.5 to 0.75 when the temperatures are expressed in Kelvin. The relation is represented in Figure 1.55 where a broad band covers most of the results for linear hompolymers and the ratio (Tg/Tm) lies between 0.5 and 0.75 for about 80% of these [33,34]. The ratio is closer to 0.5 for symmetrical polymers such as polyethylene and polybutadiene, but closer to 0.75 for unsymmetrical polymers, such as polystyrene and polyisoprene. The difference in these values may be related to the fact that in unsymmetrical chains with repeat units of the type –(CH2–CHX–)– an additional restriction to rotation is imposed by steric effects causing Tg to increase, and conversely, an increase in symmetry lowers Tg. 1.11 Designing a Polymer Structure for Improved Properties The three principles [35] applied to give strength and resistance to polymers are (1) crystallization, (2) cross-linking, and (3) increasing inherent stiffness of polymer molecules. Combinations of any two or all of the three strengthening principles have proved effective in achieving various properties with polymers (Figure 1.56). For polymers composed of inherently ﬂexible chains, crystallization and crosslinking are the only available means to enhance polymer properties. The factors affecting the crystallinity of polymers have been discussed previously, and the methods of introducing cross-links between polymer molecules are discussed later. The third, and relatively new, strengthening principle is to increase chain stiffness. One possible way of stiffening a polymeric chain is to hang bulky side groups on the chain to restrict chain bending. For example, in polystyrene, benzene rings are attached to the carbon backbone of the chain; this causes stiffening of the molecule sufﬁcient to make polystyrene a hard and rigid plastic with a softening temperature higher than that of polyethylene, even though the polymer is neither cross-linked nor crystalline. The method is advantageous because the absence of crystallinity makes the material completely transparent, and the absence of cross-linking makes it readily moldable. A similar example is poly(methyl methacrylate). 75 Characteristics of Polymers and Polymerization Processes 600 Block copolymers 500 Linear homopolymers 300 .5 T m Tm (°K) 400 0.7 5 T g T m =0 200 = Random copolymers T g 100 0 100 200 300 Tg (°K) 400 500 FIGURE 1.55 Plot of Tm against Tg for linear homopolymers with Tg/Tm in the range 0.5 and 0.75 broadly demarcating Tm−Tg domains covered by homopolymers and copolymers. Crystallization (polyethylene, polyvinyl chloride, polypropylene, nylon, etc.) (Resilient and oil resistant rubber, e.g., neoprene) Cross-linking (cured rubber, thermosetting cesins) (Structural plastics) (Heat-resistant plastics) (Terylence, cellulose acelate ﬁbers and ﬁlms) Chain stiﬀening (polyimides, ladder molecules) FIGURE 1.56 Three basic principles—crystallization, cross-linking, and chain stiffening—for making polymers strong and temperature resistant are represented at the three corners of the triangle. The sides and the center of the triangle indicate various combinations of the principles. (After Mark, H. F. 1967. Sci. Am., 217, 3, Sept. 19, 148, see also p. 156.) 76 Plastics Technology Handbook However, the disadvantage of attaching bulky side groups is that the material dissolves in solvents fairly easily and undergoes swelling, since the bulky side groups allow ready penetration by solvents and swelling agents. This problem can be eliminated by stiffening the backbone of the chain itself. One way to do this is to introduce rigid ring structures in the polymer chain. A classic example of such a polymer is cellulose, which is the structural framework of wood and the most abundant organic material in the world. Its chain molecule consisting of a string of ring-shaped condensed glucose molecules has an intrinsically stiff backbone. Cellulose therefore has a high tensile strength. In poly(ethylene terephthalate) ﬁber the chains are only moderately stiff, but the combination of chain stiffness and crystallization sufﬁces to give the ﬁber high strength and a high melting point (265°C). The new plastic polycarbonate containing benzene rings in the backbone of the polymer chain (Table 1.3) is so tough hat it can withstand the blows of a hammer. Ladder polymers based on aromatic chains consist of double-stranded chains made up of benzene-type rings [36]. These hard polymers are completely unmeltable and insoluble. The combination of all three principles has led to the development of new and interesting products with enhanced properties. Composites based on expoxy and urethane-type polymers may be cited as an example. Thus, the stiff polymeric chains of epoxy and urethane types are cross-linked by curing reactions, and ﬁllers are added to produce the equivalent of crystallization. 1.12 Cross-Linking of Polymer Chains Cold ﬂow can be prevented by cross-links between the individual polymer chains. The structure of polymer chains present in the cross-linked polymers is similar to the wire structure in a bedspring, and chain mobility, which permits one chain to slip by another (which is responsible for cold ﬂow) is prevented. Natural rubber, for example, is a sticky product with no cross-linking, and its polymer chains undergo unrestricted slippage; the product has limited use. However, as we have seen, when natural rubber is heated with sulfur, cross-linking takes place (Figure 1.49). Cross-linking by sulfur at about 5% of the possible sites gives rubber enough mechanical stability to be used in automobile tires but still enables it to retain ﬂexibility. Introducing more sulfur introduces more cross-links and makes rubber inﬂexible and hard. Cross-linking of polymer chains can be brought about by (1) reactions of functional groups; (2) vulcanization, using peroxides, sulfur, or sulfur-containing compounds; (3) free-radical reactions caused by ionizing radiation; (4) photolysis involving photosensitive functional groups; or (5) coulombic interactions of ionic species. 1.12.1 Reactions of Functional Groups Cross-linking by chemical reactions is an important process in polymer technology. A few common examples of cross-linking involving prepolymers are illustrated in Figure 1.57 through Figure 1.63. The cross-links are usually formed between prepolymer molecules by foreign molecules [e.g., styrene in unsaturated polyesters (Figure 1.61)] or by small chain segments, as in phenolic resin (Figure 1.57) and glyptal resin (Figure 1.63). A high degree of cross-linking gives rise to three-dimensional or space network polymers in which all polymer chains are linked to form on giant molecule. Thus, instead of being composed of discrete molecules, a piece of highly cross-linked polymer constitutes, essentially, just one molecule. At high degree of cross-linking, polymers acquire rigidity, dimensional stability, and resistance to heat and chemicals [31]. Because of their network structure such polymers cannot be dissolved in solvents and cannot be softened by heat; strong heating only cause decomposition. Polymers or resins which are transformed into a crosslinked product, and thus take on a set on heating, are said to be of thermosetting type. Quite commonly, these materials are prepared, by intent, in only partially polymerized states (prepolymers), so that they may be deformed in the heated mold and then hardened by curing (cross-linking). The most important resins in commercial applications are phenolic resins (Figure 1.57), amino-resins (Figure 1.58), epoxy resins (Figure 1.59), vinylesters (Figure 1.60) unsaturated polyester resins (Figure 1.61), 77 Characteristics of Polymers and Polymerization Processes OH n n C + OH OH H Acid O OH CH2 CH2 CH2 CH2 catalyst H B-Stage resin (Novolac) OH OH CH2 CH2 H2C CH2 CH2 CH2 OH O H2C OH CH2 CH2 OH CH2 CH2 OH CH2 CH2 CH2 CH2 CH2 OH OH CH2 CH2 OH CH2 O CH2 OH OH OH C-Stage resin Equations (idealized) for the production of phenol-formaldehyde resins. FIGURE 1.57 NH2 O C H N + H2C O NH2 Urea Alkaline catalyst Formaldehyde O C H2N N C NH2 C N N C + CH2 N CH2 C O N CH2 C O N CH2 N N CH2 N C Melamine CH2 CH2OH C H N CH2OH Dimethylol urea N C O N CH2 N N CH2 N C O O CH2 H N O NH2 Monomethylol urea NH2 N CH2OH Network polymer (UF) FIGURE 1.58 The two important classes of amino resins are the products of condensation reactions of urea and melamine with formaldehyde. Reactions for the formation of urea-formaldehyde amino resins (UF) are shown. Preparation of melamine-formaldehyde resins is similar. urethane foams (Figure 1.62), and the alkyds (Figure 1.63). The conversion of an uncross-linked thermosetting resin into a cross-linked network is called curing. For curing, the resin is mixed with an appropriate hardener and heated. However, with some thermosetting systems (e.g., epoxies and unsaturated polyesters), the cross-linked or network structure is formed even with little or no application of heat. Linear polymers having functional groups can be cross-linked using suitable polyfunctional agents. Thus cellulosic ﬁbers (cotton, rayon) are cross-linked by reaction of the hydroxyl groups of cellulose with formaldehyde, diepoxides, diisocyanates, and various methylol compounds such as urea‐formaldehyde prepolymers, N,N′-dimethylol-N,N′-dimethylene urea, and trimethylomelamine. Cross-linking imparts crease and wrinkle resistance and results in crease-resistant fabrics. 78 Plastics Technology Handbook CH3 O CH2 H + CH2Cl CH C CH3 Epichlorohydrin aq. NaOH Bisphenol-A O CH2 OH CH3 CH2 CH O OH C O CH2 CH CH2 O O CH3 Epoxy resin intermediate H3C C CH3 O CH2 CH CH2 O Curing of an epoxy resin O CH2 CH CH2 + H2N Epoxy end group R NH2 Diamine CH2 H2C CH CH CH2 N OH OH R N CH2 CH CH2 CH2 CH2 CH CH2 OH OH Network polymer FIGURE 1.59 Epoxy monomers and polymer and curing of epoxy resins. Polyamines such as diethylenetriamine (H2NC2H4NHC2H4NH2) are widely used for the production of network polymers by room temperature curing. 1.12.2 Vulcanization Vulcanization is a general term applied to cross-linking of polymers, particularly elastomers. Many polymers are cross-linked by compounding with a peroxide such as dicumyl peroxide or di-t-butyl peroxide and then heating the mixture. Peroxide-initiated cross-linking of saturated polymers such as polyethylene proceeds by hydrogen abstraction (Equation 1.67) by radicals resulting from hemolytic cleavage of peroxide (Equation 1.66), followed by radical combination (Equation 1.68): ROOR RO· + CH2CH2 CH CH 2 (1.66) 2RO· CHCH2 (1.67) CHCH 2 + CH CH 2 + ROH (1.68) CHCH 2 79 Characteristics of Polymers and Polymerization Processes O CH2 CH CH Diepoxide C O = CH2 = CH + CH2 = CHCOOH CH2 Catalyst CH2 CH CH OH O Vinyl ester Acrylic acid CH2 O OH C = O CH =CH2 O Styrene and peroxide catalyst C n CH2 O CH2 CH CH OH OH O CH2 O C = CH = CH2 HC O CH2 CH CH OH OH CH2 O C = C O FIGURE 1.60 CH2 CH2 = CH n O HC CH2 CH CH CH2 O Equations for the preparation of a vinyl ester. With unsaturated polymers, hydrogen abstraction probably occurs at the allylic position (Equation 1.69) with subsequent cross-linking again resulting from radical combination (Equation 1.70): CH 2CH = CHCH 2 + RO· CHCH = CHCH 2 CHCH = CHCH 2 + CHCH = CHCH 2 CHCH = CHCH 2 CHCH = CHCH 2 + ROH (1.69) (1.70) Moreover, addition-transfer processes (Equation 1.71 and Equation 1.72) can also cause cross-linking, because in many instances considerably more cross-links are formed than would be expected on the basis of only abstraction–combination reactions described above. CHCH = CHCH 2 + CHCH = CHCH 2 CH 2 CH = CHCH 2 CH 2 CH − CHCH 2 (1.71) CH 2 CH = CHCH 2 CHCH = CHCH 2 • CHCH = CHCH 2 (1.72) + CH 2 CH CH 2 CH 2 80 Plastics Technology Handbook O = O C + HO CH2 CH = CH O C = O C O C OH + = = CH CH3 O Maleic anhydride Propylene glycol O O O C CH CH Unsaturated polyester O O C Acid catalyst Phthalic anhydride C O CH2 CH O O O C C CH3 Styrene and peroxide catalyst O CH CH n CH2 O C O CH2 CH O O O C C O O C C O CH3 HC O C O CH CH C O CH2 O CH O O CH3 Network structure FIGURE 1.61 Equations for preparation and curing of an unsaturated polyester resin. The presence of ethylenic unsaturation provides sites for cross-linking by a chain reaction mechanism in the presence of styrene. Phthalic anhydride increases ﬂexibility by increasing spacing of cross-links. Peroxide cross-linking of diene polymers works with all except butyl rubber, which undergoes chain scission. The cross-links formed via peroxides are more thermally stable than those formed via sulfur vulcanization. However, peroxide cross-linking is not economically competitive with sulfur cross-linking because of the high cost of peroxides. Peroxide cross-linking is primarily used for those polymers that cannot be easily cross-linked by sulfur, such as polyethylene and other polyoleﬁns, ethylenepropylene (no diene) rubbers (EPM), and polysiloxanes. Cross-linking of polyethylene increases its strength properties and the upper limit of use temperature. Uses include electrical wire and cable insulation, pipe, and hose. For EPM and polysiloxanes, cross-linking is essential to their use as elastomers. Cross-linking with elemental sulfur is the oldest method of vulcanization, discovered independently in 1839 by Goodyear in the United States and MacIntosh and Hancock in Great Britain. This results in many reactions, producing a variety of cross-links, as represented schematically in Figure 1.64 The rate of vulcanization with sulfur can be, and normally is, increased by addition of accelerators such as zinc salts of dithiocarbamic acids or organosulfur compounds such as tetramethylthiuram disulﬁde (see Chapter 2, Table 1.4). Other compounds, usually zinc oxide and stearic acid, are also added as activators. Although the mechanism of acceleration is not well understood, acceleration is known to increase the number of monosulﬁde and disulﬁde cross-links, and decrease the number of cyclic monosulﬁde groups (Figure 1.64). Other cross-linking reactions are also used for elastomers. A variety of halogen-containing elastomers are cross-linked by heating with a basic oxide (e.g., MgO or ZnO) and a primary diamine. Such elastomers include poly(epichlorohydrin); various co-ad terpolymers f ﬂuorinated monomers such as vinylidene 81 Characteristics of Polymers and Polymerization Processes (a) Prepolymer formation: O C N R N C O + HO Diisocyanate P OH O C N R NH C O P O C Glycol NH R N C O O Urethane prepolymer (b) Chain extension of prepolymer: (i) With water NCO H2O + O NH + OCN Prepolymer Prepolymer C NH + CO2 Urea link O O (ii) With glycols NCO + HO R OH + OCN NH C O R O C NH Urethane link (iii) With amines NCO + H2N R NH2 + OCN O O NH C NH R NH C Double urea link (c) Crosslinking of chain-extended polyurethane: NHCONH NCONH OCN OCNH NCO HNCO NHCONH FIGURE 1.62 NCONH Equations for preparation, chain extension, and curing of polyurethane. O HOCH2 CH CH2OH + OH Glycerol HO O O C C O CH2 O C C O Phthalic anhydride CH O O CH2 O C C O OH Glyptal resin Curing HO O O C C O CH2 CH O O CH2 O C C O FIGURE 1.63 O C O C Equations for preparation of network glyptal resin. O NH O 82 Plastics Technology Handbook (b) (a) (c) (d) (e) (f ) S S S2 SyZ (i) Sx Sn Sn Sn Sn S S (g) (h) FIGURE 1.64 Typical chemical groupings in a sulfur-vulcanized natural rubber network. (a) Monosulﬁde cross-link; (b) disulﬁde cross-link; (c) polysulﬁde cross-link (x = 3–6); (d) parallel vicinal cross-links (n = 1–6) attached to adjacent main-chain atoms; (e) cross-links attached to common or adjacent carbon atoms; (f) intrachain cyclic monosulﬁde; (g) intrachain cyclic disulﬁde; (h) conjugated diene; (i) pendant sulﬁde group terminated by moiety Z derived from accelerator. ﬂuoride, hexaﬂuoropropylene, perﬂuoro(methyl vinyl ether), and tetraﬂuoroethylene (see “Fluoroelastomers”); and terpolymers of alkyl acrylate, acrylonitrile, and 2-chloroethyl vinyl ether (see “Polyacrylates”). Cross-linking involves dehydrohalogenation followed by addition of the diamine to the double bond (Equation 1.73) with the metal oxide acting as an acid acceptor. Some vulcanizations employa dithiol instead of a diamine. Elastomeric terpolymers of alkyl acrylate, ethylene, and a small amount of alkenoic acid (see “Acrylate Polymers”) are vulcanized by addition of diamine. The alkenoic acid is added to introduce sites (C═C) for subsequent cross-linking via reaction with primary diamines. CF 3 CH 2 CF 2 CH 2 CF 2 CF 2 CF 2 –HF (MgO) CH = CFCH 2 CF 2 CF 2 CF(CF 3 ) H 2N CH 2 CFCH 2 R NH 2 CF 2 CF 2 CF(CF 3 ) NH (1.73) R NH CH 2 CFCH 2 CF 2 CF 2 CF(CF 3 ) The vulcanization of polychloroprene (Neoprene) is carried out in different ways. Conventional sulfur vulcanization is not practiced to a large extent, since best physical properties of neoprene rubber are achieved by vulcanization with metal oxides (without diamine), either alone or in combination with sulfur (sometimes together with an accelerator). Halogenated butyl rubber is cross-linked in a similar manner. The mechanism for cross-linking with metal oxide alone is not established. Aging of polymers is often accompanied by cross-linking due to the effect of the surroundings. Such cross-linking is undesirable because it greatly reduces the elasticity of the polymer, making it more brittle and hard. The well-known phenomenon of aging of polyethylene with loss of ﬂexibility is due to crosslinking by oxygen under the catalytic action of sunlight (Figure 1.65a). Cheap rubber undergoes a similar loss of ﬂexibility with time due to oxidative cross-linking (Figure 1.65b). This action may be discouraged by adding to the polymer an antioxidant, such as a phenolic compound and an opaque ﬁller, such as carbon black, to prevent entry of light. 83 Characteristics of Polymers and Polymerization Processes CH2 CH2 CH2 CH2 CH2 CH2 O2, light CH2 CH2 CH2 CH2 CH2 CH2 CH2 CH O CH2 CH2 CH2 CH2 CH2 CH CH2 CH2 CH2 CH2 (a) CH3 CH3 CH2 C CH CH2 CH2 C H3C CH CH2 CH CH2 CH2 C CH3 CH CH2 CH3 CH2 C CH CH2 CH2 C CH CH2 O CH2 C H3C (b) FIGURE 1.65 CH3 CH2 C CH CH2 O2, light CH2 C O CH CH2 CH3 O Aging of (a) polyethylene and (b) natural rubber by oxidative cross-linking. 1.12.3 Radiation Cross-Linking When vinyl polymers are subjected to ionizing radiation (whether it be photons, electrons, neutrons, or protons), two main types of reaction occur: cross-linking and degradation. Generally, both occur simultaneously, though degradation predominates with high doses of radiation. With low radiation doses, the polymer structure determines which will be the major reaction. Thus, geminally disubstituted polymers, such as poly(a-methylstyrene), poly(methylmethacrylate), and polyisobutylene tend to undergo chain scission, with monomer being formed as the major degradation product. Such polymers will decrease in molecular weight on exposure to radiation. Halogen-substituted polymers, such as poly(vinyl chloride), degrade with loss of halogen. With most other vinyl monomers, cross-linking predominates. Table 1.6 lists some polymers that cross-link and some that degrade (i.e., liable to chain scission). The mechanism of cross-linking is free-radial in nature. The reaction is essentially the same as in peroxide cross-linking except that polymer radicals are formed by the interaction of ionizing radiation with polymer. The reaction probably involves initial rejection of a proton (Equation 1.74), which in turn removes another proton from an adjacent site on a neighboring chain Equation 1.75. Cross-linking then occurs by radical combination as by Equation 1.76. This assumption is reasonable because hydrogen is a major side product, and random formation of radicals would not give the efﬁciency of cross-linking that is generally observed. CH 2 CH 2 hν CH 2 CH 2 + H (neighboring chain) TABLE 1.6 CHCH 2 +H CHCH 2 + H2 Behavior of Polymers Subjected to High Energy Radiation Polymers that Cross-Link Polyethylene Poly(acrylic acid) Polymers that Degrade Polyisobutylene Poly(a-methyl styrene) Poly(methyl acrylate) Poly(methyl methacrylate) Polyacrylamide Polychloroprene Poly(methacrylic acid) Poly(vinylidene chloride) Polydimethylsiloxanes Polypropylene Styrene–acrylonitrile copolymers Polytetraﬂuoroethylene (1.74) (1.75) 84 Plastics Technology Handbook CH CH 2 + CHCH 2 CH CH 2 CHCH 2 (1.76) Low-density polyethylene gives rise to signiﬁcant amounts of gaseous hydrocarbons on irradiation. This probably involves the ejection of a chain branch following the ejection of a proton: CH 2 CH CH CH R +H CH = CH + RH (1.77) R Fragmentation reactions of this type, as well as ejection of hydrogen, lead to double bonds in the polymer chains (Equation 1.77) in addition to cross-linking. Radiolysis effects on numerous vinyl polymers have been studied, but polyethylene is most important. Irradiated polyethylene ﬁlm is used commercially because of its improved mechanical and thermal properties. Uses of radiation cross-linking include electrical wire and cable insulation and heat-shrinkable products (tubing, packaging ﬁlm, and bags). Curing of coating and adhesives are other applications of radiation cross-linking. Polystyrene is quite resistant to radiation (a characteristic of aromatic polymers, in general), but it can be cross-linked with higher doses. A limitation of radiation cross-linking is that radiation does not penetrate deep into the polymer matrix. The method is therefore used primarily with ﬁlms. 1.12.4 Photochemical Cross-Linking Ultraviolet or visible light induced cross-linking (photocross-linking) has assumed increasing importance in recent years. Among the numerous applications are printed circuits for electronic equipment; printing inks; coatings for optical ﬁbers; varnishes for paper arid carton board; and ﬁnishes for vinyl ﬂooring, wood, paper, and metal. Photocross-linking applied to photoresist technology is described in Chapter 5. There are two basic methods for bringing about photocross-linking of polymers; (1) incorporating photosensitizers which absorb photons and thereby induce formation of free radicals, and (2) incorporating groups into the polymer that undergo either light-initiated cross-linking polymerization or photocycloaddition reactions. When photosensitizers such as benzophenone are added to polymer, absorption of ultraviolet radiation results in excitation of the sensitizer followed by hydrogen abstraction from the polymer to yield radical sites (Equation 1.78) available for cross-linking by combination reactions. Copolymers of vinyl esters and ﬂuorinated monomers that can be cross-linked by ultraviolet radiation have been developed for use as weather-resistant wood coatings. In this application, the vinyl ester constitutes about 10% of the copolymer and benzophenone is added as a photosensitizer. The vinyl ester polymer also undergoes a-cleavage reaction (Equation 1.79) with subsequent cross-linking. O Ph O OH C Ph hν (Ph C Ph) [Ph = phenyl (C6H5) group] CH 2 CH C=O hν RH CH 2 CH Ph C Ph + R• + O =C (1.78) OR (1.79) OR A wide variety of functional groups has been used to effect photocycloaddition or light-induced crosslinking. Among such groups are cinnamate (PhCH═CHCO2R), chalcone (PhCH═CHCOPh), and 85 Characteristics of Polymers and Polymerization Processes stilbene (PhCH═CHPh). The groups may be present as part of the backbone or, more commonly, as pendant groups. In most cases, cycloaddition occurs to give cyclobutane cross-links, as shown for cinnamate ester in O C CH CH Ph O O C Ph O O hν + Ph O (1.80) C O Ph CH CH C O The reactive groups may be incorporated into polymer during the polymerization reaction; for example, b-vinyloxethyl cinnamate undergoes cathionic polymerization through the vinyl ether (Equation 1.81) to yield a linear polymer containing pendant cinnamate ester. Alternatively, the group may be added to preformed polymer, as in the Schotten–Baumann reaction of poly(vinyl alcohol) with cinnamoyl chloride (Equation 1.82). CH 2 CH O OCH 2 CH 2 O CCH CH Ph BF3 etherate/toluene [ CH 2 CH ] (1.81) OCH 2 CH 2 OCCH CH Ph O NaOH [ CH 2 CH ] + Ph CH CH OH COCl [ CH 2 CH ] (1.82) O C CH CH Ph O 1.12.5 Ionic Cross-Linking Ionic cross-links fall in the category of thermally labile cross-links, that is, chemical cross-links that break apart on heating and reform on cooling. Examples of ionic cross-linking are the hydrolysis of chlorosulfonated polyethylene with a divalent metal oxide and a source of water, and the neutralization of poly (ethylene-co-methacrylic acid) to salts of divalent metals. In a typical commercial product of chlorosulfonated polyethylene there is one chlorosulfonyl group for each 200 backbone carbon atoms. Magnesia (MgO) and a hydrogenated wood resin (as a source of H2O) can be used as a cross-linking system to yield magnesium sulfonate cross-links in the polymer: CH CH 2 CH 2 SO 2 – SO 2 Cl MgO, H2 O + Mg 2+ – SO 2 Cl CH 2 CH CH SO 2 CH 2 CH (1.83) 86 Plastics Technology Handbook Neutralization of ethylene copolymers containing up to 5%–10% acrylic or methacrylic acid copolymer with a metal salt such as the acetate or oxide of zinc, magnesium, and barium yields products referred to as ionomers. (Commercial products may contain univalent as well as divalent metal salts.) Ionomers are marked by DuPont under the trade name Surlyn. These have interesting properties compared with the nonionized copolymer. Introduction of ions causes disordering of the semicrystalline structure, which makes the polymer transparent. Ionomers act like reversibly cross-linked thermoplastics as a result of microphase separation between ionic metal carboxylate and nonpolar hydrocarbon segments. The behavior is similar to the physical cross-linking in thermoplastic elastomers (see Chapter 4). Ionomers are discussed more fully in Chapter 5. 1.13 Solubility Behavior of Polymers Knowledge of the solubility of various polymers in different solvents is important in assessing their chemical resistance and their application potentialities in the ﬁelds of paints, spinning ﬁbers, and casting ﬁlms. Important also is the knowledge of the solubility of various materials, such as plasticizers and extenders in the polymer, especially since this has an important bearing on plastics formulation. Because of the size and shape of the polymer molecules and other factors, the solubility relations in polymer systems are complex in comparison to those among low-molecular-weight compounds. Some empirical solutbility rules have, nevertheless, been derived for applying to polymer systems, and it is also possible to make certain predictions about solubility characteristics of such systems. The underlying reason that one material can act as a solvent for another is the compatibility of the materials—i.e., the ability of the molecules of the two materials to coexist without tending to separate. If we denote the force of attraction between the molecules of one material A by FAA, that between the molecules of another material B by FBB, and represent that between one A and one B molecule as FAB, then the system will be compatible and a solution will result if FAB>FBB and FAB>FAA. On the other hand, if FAA or FBB>FAB, the system will be incompatible and the molecules will separate, forming two phases. In the absence of any speciﬁc interaction (e.g., hydrogen bonding) between solvent and solute, we can reasonably assume the intermolecular attraction forces between the dissimilar molecules to be approximately given by the geometric mean of the attraction forces of the corresponding pairs of similar molecules; that is, FAB = (FAAFBB)1/2. Consequently, if FAA and equal, FAB will also be similar and the materials should be soluble. 1.13.1 Solubility Parameter A measure of the intermolecular attraction forces in a material is provided by the cohesive energy. Approximately, this equals the heat of vaporization (for liquids) or sublimation (for solids) per mol. The cohesive energy density in the liquid state is thus (DEv/V, in which DEv is the molar energy of vaporization and V is the molar volume of the liquid. The square root of this cohesive energy density is known as the solubility parameter (d), that is, d = (DEv =V)1=2 (1.84) If the vapor behaves approximately like an ideal gas, Equation 1.84 can be written as d = ½(DHv − RT)=V1=2 = ½(DHv − RT)r=M1=2 (1.85) where DHv is the molar enthalpy of vaporization and r is the density of liquid with molecular weight M. For a volatile liquid cohesive energy density and, hence d, can be determined experimentally by measuring DHv and r. 87 Characteristics of Polymers and Polymerization Processes Example 7: Calculate an estimate of the solubility parameter for water at 25°C from its heat of vaporization at the same temperature, given by H2 O(l) = H2 O(g), DH25°C = 10:514 kcal Answer: From Equation 1.85, d 2 = ½(10, 514 cal mol−1 ) − (1:987 cal mol−1 K−1 )(298 K)(1 g cm−3 )=(18 g mol−1 ) = 551:2 cal cm−3 = 2:3 109 J m−3 d = 23:5 (cal cm−3 )1=2 = 48:0 103 (J m−3 )1=2 = 48 MPa1=2 [Conversion factors: 1 cal cm−3 = 4.184 × 106 m−3 = 4.184 × 106 Pa = 4.184 MPa. Hence, 1 (cal cm−3)1/2 = 2.045 MPa1/2] Hildebrand [37] ﬁrst used the solubility parameter approach for calculating estimates of the enthalpy of mixing, DHmix, for mixtures of liquids. The equation employed is DHmix = Vmix f1 f2 (d 1 − d 2 )2 (1.86) where Vmix is the molar volume of the mixture, and d1 and d2 are the solubility parameters of components 1 and 2, respectively. A necessary requirement for solution and blending compatibility is a negative or zero Gibbs free energy change (DGmix) when the solution or blend components are mixed, that is, DGmix = DHmix − TDSmix ≤ 0 (1.87) Since the ideal entropy of mixing (DSmix) is always positive, the components of a mixture can be assumed to be miscible only if DHmix≤TDSmix. Solubility therefore depends on the existence of a zero or small value of DHmix, only positive (endothermic) heats of mixing being allowed, as in Equation 1.86. Miscibility or solubility will then be predicted if the absolute value of the ((d1−d2) difference is zero or small [38]. Speciﬁc effects such as hydrogen bonding and charge transfer interactions can lead to negative DHmix but these are not taken into account by Equation 1.86, and separate considerations must be applied in order to predict their effect on miscibility and solubility [39]. Solubility parameters of solvents can be correlated with the density, molecular weight, and structure of the solvent molecule. According to the additive method of Small [40], the solubility parameter is calculated from a set of additive constants, F, called molar attraction constants, by the relationship d= r X F M (1.88) X where F is the molar attraction constants summed over the groups present in the compound; r and M are the density and the molar mass of the compound. The same procedure is applied to polymers and EquationX 1.88 is used, wherein r is now the density of the amorphous polymer at the solution temperature, Fi is the sum of all the molar attraction constants for the repeat unit, and M is the molar mass of the repeat unit. Values of molar attraction constants for the most common groups in organic molecules were estimated by Small [40] from the vapor pressure and heat of vaporization data for a number of simple molecules. A modiﬁed version of a compilation of molar attraction constants [41,42] is reproduced in Table 1.7. An example of the use of the tabulated molar attraction constants is given in the problem worked out below. 88 Plastics Technology Handbook TABLE 1.7 Group Molar Attraction Constants Group Molar Attraction, F (cal cm3)1/2 mol−1 –CH3 147.3 –CH2– 131.5 >CH– >C< 85.99 32.03 CH2═(oleﬁn) 126.54 –CH═(oleﬁn) >C═(oleﬁn) 121.53 84.51 –CH═(aromatic) 117.12 –C═(aromatic) –O–(ether, acetal) 98.12 114.98 –O–(epoxide) 176.20 –COO– >C═O 326.58 262.96 –CHO 292.64 (CO)2O –OH– 567.29 225.84 –OH aromatic 170.99 –NHs –NH– 226.56 180.03 –N– –C≡N –N═C═O 61.08 354.56 358.66 –S– 209.42 Cl2 –Cl (primary) 342.67 205.06 –Cl (secondary) 208.27 –Cl (aromatic) –Br 161.0 257.88 –Br (aromatic) 205.60 –F 41.33 Structure Feature Conjugation Cis 23.26 −7.13 Trans −13.50 5-membered ring 6-membered ring 20.99 −23.44 Ortho substitution 9.69 Meta substitution Para substitution 6.6 40.33 Source: Hoy, K. L. 1970. J. Paint Technol., 42, 76 and Brandrup J. and Immergut E. eds., 1975. Polymer Handbook, 2nd Ed. Wiley Interscience, New York. 89 Characteristics of Polymers and Polymerization Processes Example 8: Calculate an estimate of the solubility parameter for the epoxy resin DGEBA (diglycidyl ether of bisphenol A) having the repeat unit structure as shown below and density 1.15 g/cm3. CH3 O O C CH2 CH3 CH CH2 n OH Answer: M (for repeating unit) = 284 g/mol. F (cal cm3)1/2/mol Groups –CH3 147.3 2 294.60 –CH2 131.5 2 263.00 1 1 85.99 32.03 >CH– >C< 85.99 32.03 –O– (ether) 114.98 2 229.96 –OH –CH═(aromatic) 225.84 117.12 1 8 225.84 936.96 –C═(aromatic) 6-membered ring Para substitution d= 98.12 4 392.48 −23.44 40.33 2 2 −46.88 80.66 (1:15 g cm−3 )(2496:64 cal1=2 cm3=2 mol−1 ) (284 g mol−1 ) 2494.64 =10.1(cal cm−3)1/2 = 20.7 MPa1/2 The units of the solubility parameter d are in (energy/volume)1/2. The (d values for some common solvents and polymers [43], listed in Table 1.8 and Table 1.9, have units of cal1/2 cm3/2, called hildebrands. The SI value in MPa1/2 may be obtained by multiplying the d value in hildebrand by 2.045. Most tabulated solubility parameters refer to 25°C. However, over the temperature range normally encountered in industrial practice, the temperature dependence of d can be neglected. While the solubility parameter of a homopolymer can be calculated from the molar attraction constants as illustrated in Example 6, the solubility parameter of random copolymers, dc, may be calculated from X d i wi (1.89) dc = where di is the solubility parameter of the homopolymer that corresponds to the monomer i in the copolymer and wi is the weight fraction of repeating unit i in the copolymer [44]. Solubility would be expected if the absolute value of (d1−d2) is less than about unity and there are no strong polar or hydrogen-bonding interactions in either the polymer or the solvent. To allow for the inﬂuence of hydrogen-bonding interactions, solvents have been characterized qualitatively as poorly, moderately, or strongly hydrogen bonded. The solvents listed in Table 1.8 are grouped according to this scheme. It is a useful practice to match both solubility parameter and hydrogen-bonding tendency for predicting mutual solubility. Hansen [45] developed a three-dimensional solubility parameter system based on the assumption that the energy of evaporation, i.e., the total cohesive energy DEt which holds a liquid together, can be divided into contribution from dispersion (London) forces DEd, polar forces DEp, and hydrogen-bonding forces DEh. Thus, DEt = DEd + DEp + DEh (1.90) 90 Plastics Technology Handbook TABLE 1.8 Solubility Parameters for Some Common Solvents d (cal/cm3)1/2 Solvent (i) Poorly H-bonded (Hydrocarbons and their halo-, nitro-, and cyano-products) n-Hexane 7.3 Carbon tetrachloride Toluene 8.6 8.9 Benzene 9.2 Chloroform Methylene chloride 9.3 9.7 Nitrobenzene 10.0 Acetonitrile (ii) Moderately H-bonded (esters, ethers, ketones) 11.9 Isoamyl acetate 7.8 Dioctyl phthalate Tetrahydrofuran 7.9 9.1 Methyl ethyl ketone 9.3 Acetone 1,4-Dioxane 9.9 10.0 Diethylene glycol Monoethyl ether (iii) Strongly hydrogen-bonded (acids, alcohols, aldehydes, amides, amines) 10.2 Piperidine 8.7 Acetic acid Meta-cresol 10.1 10.2 t-Butanol 10.6 I-Butanol Propylene glycol 11.4 12.6 Methanol 14.5 Ethylene glycol Glycerol 14.6 16.5 Water 23.4 Source: Brandrup, J. and Immergut, E. eds. 1975. Polymer Handbook, 2nd Ed. Interscience, Wiley, New York.SI value of d in MPa1/2 is obtained by multiplying the value in (cal/cm3)1/2 by 2.045. Dividing this equation by the molar volume of a solvent, V, gives DEt DEd DEp DEh = + + V V V V (1.91) d 2t = d 2d + d 2p + d 2h (1.92) or where dd, dp, and dh are solubility parameters due to dispersion forces, dipole forces, and hydrogenbonding (or, in general, due to donor–acceptor interactions), respectively. The three parameters, called Hansen parameters, were determined [45,46] empirically on the basis of many experimental observations for a large number of solvents (Table 1.10). Hansen’s total cohesion parameter, dt, corresponds to the Hildebrand parameter d, although the two quantities may not be identical because they are determined by 91 Characteristics of Polymers and Polymerization Processes TABLE 1.9 Solubility Parameters for Some Common Polymers d, (cal/cm3)1/2 H-Bonding Groupa Polytetraﬂuoroethylene 6.2 Poor Polyethylene 8.0 Poor Polypropylene Polyisobutylene 9.2 8.0 Poor Poor Polybutadiene 8.4 Poor Polyisoprene Polystyrene 8.1 9.1 Poor Poor Poly(methyl methacrylate) 9.5 Medium Poly(vinyl acetate) Poly(vinyl chloride) 9.4 9.7 Medium Medium Polymer Cellulose diacetate 11.0 Strong Poly(vinyl alcohol) Polyacrylonitrile 12.6 12.7 Strong Poor Nylon-6,6 13.7 Strong Source: Burrell, H. 1975. Polymer Handbook, J. Brandrup and E. Immergut, eds., 2nd Ed., pp. IV-337–359. Wiley Interscience, New York. Note: SI value of d, in MPa1/2 is obtained by multiplying the value in (cal/cm3)1/2 by 1.045. a The hydrogen-bonding group of each polymer has been taken as equivalent to that of the parent monomer. TABLE 1.10 Hansen Parameters for Solvents at 25°C dd (cal/cm3)1/2 dp (cal/cm3)1/2 dh (cal/cm3)1/2 Acetic acid 6.8 6.0 9.2 Acetone 6.3 4.8 5.4 Benzene I-Butanol 7.9 7.8 4.2 2.8 2.0 7.7 Chloroform 5.4 6.7 3.1 Cyclohexane 8.0 1.5 0.0 1,4-Dioxane Dioctyl phthalate 8.0 8.1 4.9 3.4 3.9 1.5 Ethyl acetate 6.5 4.2 4.3 Ethylene glycol Glycerol 4.9 4.5 7.4 7.5 14.6 15.3 n-Hexane 7.3 0.0 0.0 Methyl ethyl ketone Methanol 7.8 7.4 4.4 6.0 2.5 10.9 Nitrobenzene 8.6 6.8 0.0 Tetrahydrofuran Toluene 8.2 8.0 2.8 3.9 3.9 0.8 Water 5.9 11.1 19.7 m-Xylene 8.1 3.5 1.2 Liquid Source: From Tables of Solubility Parameters 3rd Ed. Chemicals and Plastics Research and Development Dept., Union Carbide Corporation. Tarrytown. N.Y., 1975. Note: SI value of a parameter in MPa1/2is obtained by multiplying the value in (cal/cm3)1/2 by 2.045. 92 Plastics Technology Handbook different methods. Once the three component parameters for each solvent were evaluated, the set of parameters could then be obtained for each polymer (Table 1.11) from solubility ascertained by visual inspection of polymer–solvent mixtures (at concentrations of 10% w/v). When plotted in three dimensions, the Hansen parameters provide an approximately spherical volume of solubility for each polymer in dd, dp, dh space. The scale on the dispersion axis is usually doubled to improve the spherical nature of this volume. The distance of the coordinates (ddi ,dpi ,dhi ) of any solvent i j j j from the center point (d d , d p , d h ) of the solubility sphere of polymer j is j j d = ½4(d id − d d )2 + (d ip − d jp )2 + (d ih − d h )2 1=2 (1.93) This distance can be compared with the radius R of the solubility sphere [47] of the polymer (Table 1.11), and if d<R, the likelihood of the solvent i dissolving the polymer j is high. This works well, despite the limited theoretical justiﬁcation of the method. Example 9: Using Hansen parameters (Table 1.10 and Table 1.11) determine if polystyrene is expected to dissolve in a solvent mixture of 60/40 v/v methyl ethyl ketone/n-hexane. Answer: Denote the solubility parameter components of MEK, n-hexane, and polystyrene using superscripts i, j, and k, respectively. From Table 1.10, MEK: d id = 7:8, d ip = 4:4, d ih = 2:5 all in (cal cm−3 )1=2 n hexane: d d = 7:3, d jp = 0, d h = 0 all in (cal cm−3 )1=2 j j The Hansen parameters are combined on a 60/40 volume fraction basis: d d = 0:6 7:8 + 0:4 7:3 = 7:6 (cal cm−3 )1=2 ij d ijp = 0:6 4:4 + 0:4 0:0 = 2:6 (cal cm−3 )1=2 d h = 0:6 2:5 + 0:4 0:0 = 1:5 (cal cm−3 )1=2 ij TABLE 1.11 Hansen Parameters and Interaction Radius of Some Polymers and Resins Polymer Acrylonitrile–butadiene elastomer Cellulose acetate Epoxy resin dd (cal/cm3)1/2 dp (cal/cm3)1/2 dh (cal/cm3)1/2 R (cal/cm3)1/2 9.1 4.3 2.0 4.7 9.1 10.0 6.2 5.9 5.4 5.6 3.7 6.2 Nitrocellulose 7.5 7.2 4.3 5.6 Polyamide, thermoplastic Polyisoprene 8.5 8.1 −0.9 0.7 7.3 −0.4 4.7 4.7 Poly(methyl methacrylate) 9.1 5.1 3.7 4.2 10.4 10.2 2.8 5.5 2.1 4.7 6.2 6.7 Poly(vinyl chloride) 8.9 3.7 4.0 1.7 Styrene–butadiene rubber 8.6 1.7 1.3 3.2 Polystyrene Poly(vinyl acetate) Source: Data from Hansen, C. M. and Beerbower, A. 1971. Kirk Othmer Encyclopaedia of Chemical Technology, A. Standen, ed., 2nd Ed., Suppl. 889, p. 910. Wiley Interscience, New York. Note: SI value in MPa1/2 is obtained by multiplying the value in (cal/cm3)1/2 by 2.045. 93 Characteristics of Polymers and Polymerization Processes For polystyrene (Table 1.11), d kh = 2:1, R = 6:2 all in (cal cm−3 )1=2 d kd = 10:4, d kp = 2:8, From Equation 1.93, d = ½4(10:4 − 7:6)2 + (2:8 − 2:6)2 + (2:1 − 1:5)2 1=2 = 5:6 (cal cm−3 )1=2 As this value is less than the radius of the polymer solubility sphere (6.2 cal1/2 cm−3/2), the polymer is expected to be soluble. This is found to be the case for a 10% w/w solution. Three-dimensional presentations of solubility parameters are not easy to use and it is more convenient to transform the Hansen parameters into fractional parameters as deﬁned by [48] fd = d d =(d d + d p + d h ) fp = d p =(d d + d p + d h ) (1.94) fh = d h =(d d + d p + d h ) The fractional parameters represent, in effect, the quantitative contribution of the three types of forces to the dissolving abilities for each solvent and can be represented more conveniently in a triangular diagram (Figure 1.66) to provide a visual presentation of the nature of the solvating powers of liquids, including in one picture such diverse liquids as water, alcohols, organic acids, and hydrocarbons [32]. The triangular solubility chart can be used conveniently for the prediction of solubility of polymers. A chart can be constructed for a given resin, identifying each solvent coordinate point to indicate either complete, partial or lack of solubility. In general, solvents which provide clear solutions are grouped in a reasonably well deﬁned area of the chart thus forming a loop and solvents which tend to swell the resin lie near the border of such loops, as shown in Figure 1.66. The probable solubility parameters of the solute polymer 0 1.0 0.2 0.8 0.4 0.6 fh fp Acetone 0.6 0.4 2-Nitropropane Area of solubility 0.8 0.2 Cyclohexanol Cyclohexane 1.0 0 0.2 0.4 0.6 0.8 1.0 fd FIGURE 1.66 Limiting solubility boundary for chlorinated rubber. The solid circle represents the probable solubility parameter of the resin. (After Teas, J. P. 1968. J. Paint Technol., 40, 519.) 94 Plastics Technology Handbook will be at the heart of the solubility area. Any solvent possessing values of fd, fp, and fh to place its plot inside the solubility area has a high probability of producing a clear solution. Conversely, liquids lying outside of the solubility area are very likely to be poor or nonsolvent. Since dd values do not vary greatly, at least among common solvents, it has been suggested that a plot of dp versus dh should be sufﬁcient for most practical purposes. Thus, another and probably more efﬁcient two-parameter representation is made possible by deﬁning [49] d v ≡ d 2d + d 2p 1=2 (1.95) The procedures outlined above have a practical use, but it should be realized that the parametric models are almost entirely empirical. Experimental uncertainties are also involved since solubility measurements are not very accurate. Solubility loops described by the models only indicate the limits of compatibility and always include doubtful observations. 1.14 Effects of Corrosives on Polymers Polymers are resistant to electrochemical corrosion. When a polymer has a d value similar to that of water, dissolution occurs. However, if the d value of the polymer is lower, it is not attacked by water or other polar solvents. The carbon–carbon bonds in polymer backbones are not cleaved even by boiling water. When amide, ester, or urethane groups are present in the polymer backbone, they may be attacked by hot acids or alkalies, and hydrolysis may occur. When these functional groups are present as pendant groups on the polymer chain, the reaction will be similar. However, the tendency for such attack is reduced when alkyl groups are present on the carbon atom attached to the functional group. Thus, poly(methyl methacrylate) but not poly(methyl acrylate) is resistant to acid or alkaline hydrolysis. Atoms or groups with strong carbon bonds, such as ﬂuorine, chlorine, and ether groups, are resistant to attack by aqueous acids and alkalies. The chemical resistance of various plastics is summarized in Table 1.12. 1.15 Thermal Stability and Flame Retardation Most organic polymers decompose when heated to moderate or high temperatures. It is for this reason that few synthetic polymers can be used for long periods of time above 150°C–200°C. This necessitates the use of metals and ceramics for many applications, even though synthetic polymers may be cheaper and, in some cases, stronger on a weight-for-weight basis. Plastics can be grouped into eight temperature-time zones, depending on the temperature at which they retain 50% mechanical or physical properties when heated for different periods in air 50. These temperature–time zones are shown in Figure 1.67, and the materials falling in the different zones are listed in Table 1.13 [50]. The materials in zone 6 and above can compete with metals in high-performance applications because they perform in the same temperature range. Most polymers in zone 6 and above do not burn, but they may char and may be consumed very slowly in direct ﬂames. Rotation of chain segments is more difﬁcult in polymers with cyclic rings in their chains. As a result, such polymers are stiffer and more resistant to deformation and have higher melting points and higher glass transition temperatures. Thermal stability is improved by the presence of aromatic and heterocyclic rings in polymer chains. This general approach has resulted in a number of new commercially available polymers with improved high-temperature properties. Polyimides and polybenzimidazoles are important examples of such high-temperature polymers. They can be exposed for a short time to temperatures as high as 1,112°F (600°C). Polyimides are synthesized [51] by the reactions of dianhydrides with diamines in a two-stage process (Figure 1.68). In the ﬁrst stage, the two materials form an intermediate poly(amic acid). Processing is 95 Characteristics of Polymers and Polymerization Processes TABLE 1.12 Effects of Corrosive Environments on Plastics Environment Plastic Esters Aliphatic Aromatic Chlorinated and Solvents Solvents Solvents Ketones Weak Bases Strong Bases Strong Strong Acids Oxidants Acetal A B A B A B A B A A A B E E E E Acrylic Acrylonitrile–butadiene–styrene (ABS) Cellulose acetate B A C E E D E E E E E E E E E E A A C C B A E C D B E E E D E E A B A C A D E E A C C E C E C E Cellulose acetate butyrate A C D E D E E E B D C E C E C E Cellulose acetate propionate Epoxy (glass ﬁber ﬁlled) A A C B D A E B D A E C E B E C A A B A C B E C C B E C C D E D Furan (asbestos ﬁlled) A A A A A A A A B B B B A A E E Melamine A A A A A A A B B C B C B C B C Phenolic (asbestos ﬁlled) Polyamide A A A A A A A A A A A B C A C A A A C B D B E C A E A E D E E E Polybenzimidazole A A A A A A A A A A A B A B A C Polycarbonate Poly(chlorotriﬂuoroethylene) A A A A A A A A E C E D E A E A A A E A E A E A A A A A A A A A Polyester (glass ﬁber ﬁlled) A B A C B D C C B C C E B B B C Polyethylene Polypropylene C A E D C B E D D B E D D A E C A A A A A A A A A A A C A A A D Polysulfone A A D D E E C D A A A A A A A A Polystyrene Poly(tetraﬂuoroethylene) D A E A D A E A E A E A D A E A A A E A A A E A D A E A D A E A Polyurethane A D C D D E B C A A C D A D A D Poly(vinylchloride) Silicone A B E C D D E D E D E E D B E D A A E B A D E E A C E D B D E E Urea A C A C A C A B B C B C D E B C Note: A = no effect or inert; B = slight effect; C = mild effect; D = softening or swelling; E = severe deterioration. The effects were measured at 25°C (ﬁrst letter in each column) and at 90°C (second letter in each column). accomplished after this stage, since the polymer at this point is still soluble and fusible. The poly(amic acid) is formed into the desired physical form of the ﬁnal polymer product (e.g., ﬁlm, ﬁber, laminate, coating, etc.) and then the second stage of the reaction is carried out, in which the poly(amic acid) is cyclized in the solid state to the polyimide by heating at moderately high temperatures (above 150°C). The polymer after the second stage of the process is insoluble and infusible. Instead of m-phenylenediamine, shown in Figure 1.68, other diamines have also been used to synthesize polyimides—for example, p-phenylenediamine, p,p′-diaminodiphenyl ether, and p,p′-diaminodiphenyl sulﬁde. Polyimides have been used up to 100 h at 600°F (315.6°C). They have been used as a transparent head cover for ﬁre ﬁghters and as excellent ﬂame- and high-temperature-resistant foams, which can be used as structurally stable insulation materials at high temperatures. Polyimide/glass ﬁber composites have many uses in the electronics and aerospace industries and in other high-performance applications. Polybenzimidazoles are synthesized by the reactions of aromatic diacids and aromatic tetraamines (Figure 1.69). Polybenzimidazoles are mostly used as ﬁbers in parachutes, for reentry vehicles, and so on. Composites of polybenzimidazoles with ﬁbers have excellent basic strength and high-temperature performance. These composites have been extensively used in nose fairings, aircraft leading edges, reentry nose cones, radomes, and deicer ducts. 96 Plastics Technology Handbook 1660 3000 1000 500 Zone 8 800 Temperature (°C) Zone 7 300 600 Temperature (°F) 400 Zone 6 200 400 Zone 5 Zone 4 100 Zone 3 Zone 2 200 Zone 1 0 1 10 100 1000 10,000 100,000 FIGURE 1.67 Time-temperature zones indicating thermal stability of plastics (see Table 1.13). (Adapted from Anon. 1968. Plastics World, 26(3), 30.) A weak link in the polybenzimidazole structure is the imino hydrogen. When this hydrogen is replaced by a phenyl group as in N-phenyl polybenzimidazole, a dramatic increase in high-temperature properties in oxidizing atmospheres is obtained. In polymers such as polybhenylenes there are a few aliphatic linkages and many aromatic rings which account for their improved heat resistance. Several polyphenylenes are shown in Figure 1.70. Poly(phenylene oxide) has excellent dimensional stability at elevated temperatures. Repeated steam autoclaving does not deteriorate its properties. Poly(phenylene sulﬁde) is completely nonﬂammable. It is used in the form of composites with both asbestos and glass ﬁbers. Aliphatic linkage is completely eliminated in ladder polymers such as polybenzimidazopyrrolones, commonly called “pyrrones” (Figure 1.71). Such polymers are highly stable in air. They do not burn or melt when heated but form carbon char without much weight loss. They are potentially the ultimate in heat- and ﬂame-resistant materials. Another way to get good heat resistance is to use inorganic material as backbone chain as n silicone polymers. Here organic radicals are attached to silicone atoms on an inorganic silicon–oxygen structure. Presence of silicon–oxygen links gives such materials outstanding heat resistance. A silicone polymer has the structure R Si O R When n is a small number, the structure is that of a silicone oil, whereas silicon rubbers have high values of n. when the ratio R/Si is lower than 2, cross-linked polymers are obtained. Properties of silicone 97 Characteristics of Polymers and Polymerization Processes TABLE 1.13 Plastics Retaining 50% Mechanical or Physical Properties at Temperature in Air Zone 1 Acrylic Cellulose acetate Cellulose acetate butyrate Cellulose acetate propionate Cellulose nitrate Polyallomer Polyethylene, low density Polystyrene Poly(vinyl acetate) Poly(vinyl alcohol) Poly(vinyl butyral) Poly(vinyl chloride) Styrene–acrylonitrile Styrene–butadiene Urea–formaldehyde Zone 2 Acetal Acrylonitrile–butadiene–styrene Copolymer Ethyl cellulose Ethylene–vinyl acetate copolymer Furan Lonomer Polyamides Polycarbonate Polyethylene, high density Polyethylene, cross-linked Poly(ethylene terephthalate) Polypropylene Poly(vinylidene chloride) Polyurethane Zone 3 Poly(monochlorotriﬂuoroethylene) Zone 4 Alkyd Fluorinated ethylene propylene copolymer Melamine formaldehyde Phenol–furfural Polyphenylene oxide Polysulfone Zone 5 Diallyl phthalate Epoxy Phenol–formaldehyde Polyester Poly(tetraﬂuoroethylene) (Continued) 98 Plastics Technology Handbook TABLE 1.13 (CONTINUED) in Air Plastics Retaining 50% Mechanical or Physical Properties at Temperature Zone 6 Polybenzimidazole Polyphenylene Silicone Zone 7 Polyamide–imide Polyimide Zone 8 High-performance plastics being developed using intrinsically rigid linear macromolecules rather than the conventional crystallization and cross-linking Source: Anon. 1968. Plastics World, 26(3), 30. O O C C C C O O O HO O O O C C C C O O H2N NH2 OH Heat NH NH –H2O n Polyamic acid O O C C C C O O N N n Polyimide FIGURE 1.68 Synthesis of polyimide from pyromellitic anhydride and m-phenylenediamine. polymers are greatly affected by the type of organic radical present. For a given chain length, a methyl silicone can be an oily liquid, but a phenyl silicone is a hard and brittle resin. 1.15.1 Thermal Degradation There are basically three types of thermal degradation reactions for vinyl polymers [52,53]: (1) nonchain scission; (2) random chain scission; and (2) depropagation. In practice, mechanisms 2 and 3 blend into one another, with many polymers showing evidence of both processes. 99 Characteristics of Polymers and Polymerization Processes O O O C C O H2N NH2 H 2N NH2 Tetra-aminobiphenyl Diphenyl isophthalate 275°C–300°C (1–2 h) 375°C–400°C (2–3 h) Heat OH N N 2H2O 2 FIGURE 1.69 N N H H n Synthesis of polybenzimidazole by condensation polymerization. Polyphenylene FIGURE 1.70 O O Poly(phenylene oxide) S S Poly(phenylene sulfide) Structures of some polyphenylenes. O N C C N C O FIGURE 1.71 Typical structure of a polybenzimidazopyrrolone. N C N n 100 Plastics Technology Handbook Nonchain scission refers to reactions involving pendant groups that do not break the backbone. Typical of such reactions are dehydrochlorination of poly(vinyl chloride) (Equation 1.96), elimination of acid from poly(vinyl esters)—for example, poly(vinyl acetate) (Equation 1.97)—and elimination of alkene from poly(alkyl acrylate)s (Equation 1.98). ( CH2 ( CH CH ) CH ) + HCl (1.96) Cl ( CH2 CH ) OC ( CH CH ) + HO CH3 C CH3 (1.97) O O ( CH2 CH ) C ( CH2 CH ) C O OCH2CH2R + CH2 CHR (1.98) O OH The ﬁrst two reactions lead to highly colored residues, indicating that the double bonds formed in the polymer backbone are primarily conjugated. Such elimination reactions are not satisfactory for synthesizing polyacetylene, however, since side reactions also occur. But nonchain scission has been used as one approach to solving problem of polyacetylene’s intractability (see “Inherently Conducting Polymers” in Chapter 5). Random chain scission results from homolytic bond-cleavage reactions at week points in the polymer chains. Complex mixtures of degradation products are formed, the origin of which may be explained in terms of radical transfer reactions such as CH 2 CH 2 + CH 2 CH 2 CH 2 + CH 3 CH 2 CH CH 2 CH 2 CH 2 CH 2 Random chain scission occurs with all vinyl polymers to varying degrees, but it occurs less with increasing substitution on the polymer backbone. Depropagation, or depolymerization (unzipping), to give monomer occurs mainly with polymers prepared from 1,1-disubstituted monomers. The depolymerization is a free-radical process that may be initiated either at a chain end or at a random site on the backbone. For example, poly(methyl methyl methacrylate) appears to begin unzipping primarily at the chain ends, whereas poly(a-methylstyrene) does so at random sites along the chain. In both cases, tertiary radicals are formed with each depropagation step: CH 2 R R CCH 2 C R R R CH 2 C + CH 2 R R C R The nature of active chain ends in poly(methyl methacrylate) is a question for debate. At moderate temperatures (220°C) only half of the chains in this polymer unzip, and higher temperatures (350°C) may be needed to decompose the remaining polymer. Apparently, chains which unzip at 200°C are terminated 101 Characteristics of Polymers and Polymerization Processes by unsaturated groups, whereas those which depolymerize only at higher temperatures have saturated end groups. Like poly(methyl methacrylate) and poly(a-methyl styrene), polytetraﬂuroethylene also undergoes 100% conversion to the monomer at elevated temperatures. However, it does so only at low pressures and high temperatures. At atmospheric pressure, the monomer molecules recombine to form dimer and other species. This polymer is one of the most thermally stable polyoleﬁns known, but even so, it cannot withstand prolonged exposure to temperatures above about 350°C–400°C. Polystyrene represents a case in which monomer is only one of several species formed by thermal degradation at 350°C as monomer, dimer, trimer, and tetramer are formed in the relative proportions of 40:10:8:1. The thermal breakdown process is believed to begin at unsaturated linkages which constitute the weak points along the chain. A cleavage at these sites initiates a free-radical mechanism leading to liberation of monomer and to an intramolecular back-biting process. The process liberates dimer, trimer, and so on, by a transfer mechanism such as the shown in Equation 1.99: Ph CH 2 Ph H CH CH CH CH 2 Ph CH CH Ph (1.99) CH 2 CH 2 CH + CH 2 C Ph Ph CH 2 CH CH 2 Ph CH 2 Ph Polyethylene yields very little monomer by thermal degradation. Above about 300°C, the decomposition products form a continuous spectrum of unsaturated hydrocarbons which contain from 1 to at least 70 carbon atoms. The products are believed to be the results of random chain cleavage accompanied by inter- and intramolecular chain transfer. Oxygenated sites may act as weak links for cleavage and the existence of chain branch points may facilitate the transfer process. Polypropylene behaves in a similar manner to polyethylene. The foregoing discussion applies speciﬁcally to oleﬁn addition polymers. Some comments on the thermal behavior of condensation polymers are also appropriate. Polyamides can decompose during melt spinning or molding operations. Such decomposition, usually sight, is apparently initiated by free radicals formed by the hemolytic cleavage of –NH–CH2– skeletal bonds. Water and carbon dioxide are also liberated. The water serves to hydrolyze amide (–NH–CO–) linkages, further shortening the chains. Branches are also formed by reaction of terminal –NH2 groups with carbonyl units (Equation 1.100), leading ultimately to gelation of the molten polymer. C NH –H 2 O C O N NH 2 CH 2 + NH (1.100) CH 2 Polyesters, such as poly(ethylene terephthalate), are fairly stable at temperatures just above the melting point. However, at temperatures between 300 and 550°C, this polymer decomposes to yield carbon dioxide, acetaldehyde, and terephthalic acid together with smaller amounts of other decomposition products, such as water, methane, and acetylene. 102 Plastics Technology Handbook 1.15.2 Ablation When subjected brieﬂy to very high temperatures, some polymers, such as phenolics, can undergo rapid decomposition to gases and porous char, thereby dissipating the heat and leaving a protective thermal barrier on the substrate. This sacriﬁcial loss of material accompanied by transfer of energy is known as ablation. Interaction of a high-energy environment (2,500°C–5,000°C) with the exposed ablative material results in sacriﬁcial erosion of some amount of the surface material, and the attendant energy absorption controls the surface temperature and greatly restricts the ﬂow of heat into the substrate. Most notable applications of ablative materials are in protecting space vehicles during reentry into the earth’s atmosphere, protecting missile nose cones subjected to aerodynamic heating during hypersonic ﬁght in the atmosphere, insulating sections of rocket motors from hot propulsion gases, resisting the intense radiant hear of thermonuclear blasts, and providing thermal protection for structural materials exposed to very high temperatures. Polymers have been used as ablative materials for a combination of reasons [54]. Some of their advantages are (1) high heat absorption and dissipation per unit mass expended, which may range from several hundred to several thousand calories per gram of ablative material; (2) automatic control of surface temperature by self-regulating ablative degradation: (3) excellent thermal insulation; (4) tailored performance by varying the individual material component and composition of ablative systems; (5) design simplicity and ease of fabrication; (6) light weight and low cost. Polymer ablatives, however, can be used only for transitory periods of a few minutes or less at very high temperatures and heat load. Moreover, the sacriﬁcial loss of surface material during ablation causes dimensional changes which must be predicted and incorporated into the design. An ablative material should have a high heat of ablation, which measures the ability of the material to absorb and dissipate energy per unit mass. It should also possess good strength even after charring, since any sloughing off of chunk of material which have not decomposed or vaporized represent poor usage of the ablative system Figure 1.72 proﬁles the various stages of heating, charring, and melting in two phenolic composites, one reinforced with glass and the other with nylon. The glass-reinforced system appears to be mechanically superior because it produces a molten glass surface, whereas the nylon-reinforced system may have higher thermal efﬁciency. Infact Slight volatilization Incomplete charring Molten glass Infact Nylon decomposing Incomplete charring Complete charring 2500°C 1800°C 1300°C 600°C 440°C 330°C 220°C 40°C Phenolic + glass Phenolic + nylon FIGURE 1.72 Temperature distribution in two plastics during steady-state ablation. (After Schmidt, D. L., Mod. Plastics, 37, 131 (Nov. 1960), 147 (Dec. 1960).) 103 Characteristics of Polymers and Polymerization Processes 1.15.3 Flame Retardation All thermoset plastics are self-extinguishing. Among thermoplastics, nylon, polyphenylene oxide, polysulfone, polycarbonate, poly(vinyl chloride), chlorinated polyether, poly(chlorotriﬂuoroethylene) and ﬂuorocarbon polymers have self-extinguishing properties [55]. The burning characteristics of some polymers are summarized in Table 1.14. Halogenation enhances the ﬂame retardancy of polymers. Thus when the chlorine content of PVC, which usually has 56% chlorine, is increased, as in chlorinated PVC (61% chlorine), the oxygen index increases from 43 to 60. (The oxygen index is a minimum percent of oxygen in a N2/O2 gas mixture necessary to support combustion. A material with an oxygen index greater than 27 usually passes the vertical burn test. The burning characteristics of polymers are discussed more fully in Chapter 5.) A discussion on ﬂame-retardant additives is given in a later section on polymer compounding. 1.16 Deterioration of Polymers Deterioration of polymers is manifested in loss of strength, loss of transparency, warpage, cracks, erosion, and so on [56–58]. Hydrophilic materials such as nylon or cellulose acetate can undergo swelling at high humidity or shrinkage due to low humidity. Degradation can occur due to imposition of energy in the form of heat, mechanical action, ultrasonic and sonic energy, radiation such as gamma rays, x-rays, visible light, ultraviolet light, infrared, and electrical action in the from of dielectric effects. Deterioration can occur by chemical effects such as oxidation, ozone attack, hydrolysis, attacks by solvents and detergents, cracking due to swelling of the plasticizer, hardening due to loss of the plasticizer, migration of plasticizers from layer to layer, crazing and cracking, delamination or debonding, void formation, etc. TABLE 1.14 Burning Characteristics and Burn Rates of Some Polymers Polymer Burning Characteristics Burn Ratea (cm/min) Polyethylene Melts, drips 0.8–3.0 Polypropylene Melts, drips 1.8–4.0 Poly(vinyl chloride) Poly(tetraﬂuoroethylene) Difﬁcult to ignite, while smoke Melts, chars, bubbles Self-extinguishing Nonburning Fluorinated ethylene propylene copolymer Does not ignite Nonburning Polybutylene Burns 2.5 Acetal Cellulose acetate Burns, bluish ﬂame Burns, yellow ﬂame sooty smoke 1.3–2.8 1.3–7.6 Cellulose propionate Burns, drips 1.3–3.0 Cellulose acetate butyrate Acrylonitrile–butadiene–styrene (general purpose) Burns, drips Burns 0.8–4.3 2.5–5.1 Styrene–acrylonitrile Melts, chars, bubbles 1.0–4.0 Polystyrene Acrylic Softens, bubbles, black smoke Burns slowly, drips 1.3–6.3 1.4–4.0 Nylons Burns slowly, froths Self-extinguishing Phenylene oxide Polysulfone Self-extinguishing Self-extinguishing Chlorinated polyether Self-extinguishing Polyimide Nonburning Source: Adapted from Kuryla, W. C. and Papa, A. J., eds. 1973. Flame Retardancy of Polymeric Materials, Vol. 3. Marcel Dekker, New York. a ASTM D-635 test procedure. 104 FIGURE 1.73 Plastics Technology Handbook F F F F F F F F F C C C C C C C C C F F F F F F F F F F F F F F F F F F C C C C F + C C C C C F F F F F F F F Scission of polytetraﬂuoroethylene by irradiation. 1.16.1 Chemical Deterioration Liner or thermoplastic polymers may deteriorate by scission. Ultraviolet light and neutrons can easily break a C–C bond of a vinyl-type polymer, producing smaller molecules (Figure 1.73). Cross-linking is another way that linear or thermoplastic polymers may deteriorate. Two wellknown examples are aging of polyethylene and natural rubber, with loss of ﬂexibility due to crosslinking by oxygen under the catalytic action of sunlight (Figure 1.65). Vulcanized rubber has only 5%–20% of its possible positions anchored by sulfur cross-links. Over time there may be further crosslinks of oxygen by the air, and the polymer may thus gradually lose its deformability and elasticity. Because the oxygen molecule is a biradical, its reaction with a polymer usually results in a chin reaction involving free radicals. Thus, a polymer present in air or an atmosphere rich in oxygen can lose an H atom by reacting with O2 to from a free radical, which then reacts with another oxygen molecule to from a peroxy free radical. The latter, in turn, reacts with another unit of a polymer chain to produce a hydroperoxide and another free radical. H C H H C C H H 00 C H H H H C C C C + HOO (1.101) H H O2 H H H H C C C C H H H H H C C C C H O (1.102) H O H H H H C C C C O O H H + H H H H C C C C H (1.103) 105 Characteristics of Polymers and Polymerization Processes Free radicals can combine together and form a stable molecule. 2 H H H H H C C C C C H H H H C C C C H (1.104) H H C C C C H H H H Termination of the chain reaction can also occur with the formation of a peroxide which may be transitory. H H H H H H H H C C C C C C C C H O H O H O O H C C C C H H H H C C C C H H H H (1.105) Inhibitors or antioxidants can also stop the process. H H H H C C C C + RH H H H H H C C C C H +R (1.106) H In this way, the chain is broken, and an inactive radical (R•) is yielded. Aromatic amines or phenolic compounds can act is inhibitors. However, new radicals can also be formed from the peroxy groups H H H H H H H H C C C C C C C C H O O + HO (1.107) H O H H H H H C C C C O H O H 2 H H H H C C C C O H (1.108) C C C C H H H H 106 Plastics Technology Handbook Thus for each original reaction with oxygen there can be numerous propagation reaction. The peroxides may also cleave to give aldehydes, ketones, acids, or alcohols. Due to this type of molecular cleavage, the product becomes soft with lower average molecular weight. H H H H C C C C O aldehyde, ketone, acid, or alcohol (1.109) H O H Ozone provides a very reactive source of oxygen because it breaks down to an oxygen molecule and a single reactive oxygen. O3 ! O2 + O (1.110) The temperature resistance of various plastics in air has been summarized in Figure 1.67 and Table 1.13. Polymers containing hydrolysable groups or which have hydrolysable groups introduced by oxidation are susceptible to water attack. Hydrolyzable groups such as esters, amides, nitriles, acetals, and certain ketones can react with water and cause deterioration of the polymer. The dielectric constant, power factor, insulation resistance, and water absorption are most affected by hydrolysis. For polyesters, polyamides, cellulose, and cellulose either and esters, the hydrolysable groups are weak links in the chain, and hydrolysis of such polymers can cause serious loss of strength. A summary of water absorption characteristics of common plastic and rubbers is presented in Table 1.15. 1.16.2 Degradation by Radiation It was mentioned earlier that radiation may cause both cross-linking or degradation. Which predominates depends on radiation dosage, polymer structure, and temperature, Ultraviolet (UV) light causes 1,1-disubstituted vinyl polymers to degrade almost exclusively to monomer at elevated temperatures, Whereas cross-linking and chain scission reaction predominate at room temperature. Other vinyl polymers undergo cross-linking primarily, regardless of temperature. Irradiation with a beam of x-rays, g-rays, or electrons leads to much higher yields of monomer from 1,1-disubstituted polymers at room temperature. This phenomenon is used advantageously for microetching of resist coatings with electron beams in the preparation of integrated circuits. The process works on the principle that a silicon chip surface will be exposed if an electron beam depolymerizes the protective resin (resist) coating that covers the surface. The exposed silicon can then be doped. Because a beam of high energy radiation has a shorter wavelength than UV light, it is possible to each ﬁner details in the resist than via the alternative photocross-linking process using UV light. Poly(methyl methacrylate) has been studied extensively as an etch-type resist. This subject is discussed more fully in Chapter 5. 1.16.3 Microbiological Deterioration One of the main advantages of synthetic polymers over naturally occurring polymeric materials such as cellulose or leather is their resistance to bacterial or fungal attack. Hence, the synthetic materials are, in general, more permanent. However, a few synthetic polymers are susceptible to biological breakdown and it is clearly important to know, from an applications point of view, which polymers are the most susceptible in a biological environment. Polyurethanes in particular appear to be susceptible to microbial attack, though polyether polyurethanes (see Chapter 4) are more resistant to biological degradation than are polyester polyurethanes. The 107 Characteristics of Polymers and Polymerization Processes TABLE 1.15 Water Absorption Characteristics of Plastics and Rubbers Material Water abs. (%) 24 h on Sample 3.2 mm Thick Phenol–formaldehyde resin cast (no ﬁller) 0.3–0.4 Phenol–formaldehyde resin molded (wood-four ﬁller) 0.3–1.0 Phenol–formaldehyde resin molded (mineral ﬁller) Phenol–furfural resin (wood-ﬂour ﬁller) 0.01–0.3 0.2–0.6 Phenol–furfural resin (mineral ﬁller) 0.2–1.0 Urea–formaldehyde resin (cellulose ﬁller) Melamine–formaldehyde resin (cellulose ﬁller) 0.4–0.8 0.1–0.6 Melamine–formaldehyde resin (asbestos ﬁller) 0.08–0.14 Ethyl cellulose Cellulose acetate (molding) 0.8–1.8 1.9–6.5 Cellulose acetate (high acetyl) 2.2–3.1 Cellulose acetate–butyrate Cellulose nitrate 1.1–2.2 1.0–2.0 Casein plastics 7–14 Poly(vinyl chloride) (plasticized) Poly(vinylidene chloride) (molding) 0.1–0.6 <0.1 Poly(vinyl chloride acetate) (rigid) 0.07–0.08 Poly(vinyl chloride acetate) (ﬂexible) Poly(vinyl formal) 0.40–0.65 0.6–1.3 Poly(vinyl butyral) 1.0–2.0 Allyl resins (cast) Polyester resins (rigid) 0.3–0.44 0.15–0.60 Polyester resin (ﬂexible) 0.1–2.4 Poly(methyl methacrylate) Polyethylene 0.3–0.4 <0.01 Polypropylene 0.01–0.1 Polystyrene Polytetraﬂuoreoethylene 0.03–0.05 0.00 Nylon (molded) 1.5 Rubbers (extruded) Chlorinated rubber 0.4 0.1–0.3 Source: Data mainly from Halim Hamid, S., Amin, M. B., and Maadhah, A. G., eds. 1992. Handbook of Polymer Degradation. Marcel Dekker, New York. precise mechanisms of these degradations are not fully understood. Polyethylene, polypropylene, polyﬂuorocarbons, polyamides, polycarbonates, and many other polymer systems appear to be resistant to biological attack. Table1.16 provides a qualitative assessment of the resistance of plastic and rubbers to attack by microorganisms. The controversial assessment in certain cases may be due to discrepancies in reported data arising from differences in measurement technique, materials used, average molecular weight of materials, ﬁllers, impurities in polymers, etc. 1.17 Stabilization of Polymers We have seen that polymers tend to undergo degradation such as chain scission, depolymerization, crosslinking, oxidation, and so on [59–63]. These changes can be effected by various environmental factors such as heat, light, radiation, oxygen, and water. Various stabilizing ingredients are added to plastics to 108 Plastics Technology Handbook TABLE 1.16 Resistance of Plastics and Rubbers to Attack by Microorganisms Resistance Plastics Poly(methyl methacrylate) Good Polyacrylonitrlle (orlon) Acrylonitrile-vinylchloride copolymer (Dynel) Good Good Cellulose acetate Good, poor Cellulose acetate butyrate Cellulose acetate propionate Good Good Cellulose nitrate Poor Ethyl cellulose Acetate rayon Good Good Viscose rayon Poor Phenol–formaldehyde Melamine–formaldehyde Good Good, poor Urea–formaldehyde Good Nylon Ethylene glycol terephthalate (Terylene) Good Good Polyethylene Good, questionable Polytetraﬂuoroethylene (Teﬂon) Polymonochlorotriﬂuoroethylene Good Good Polystyrene Good Poly(vinyl chloride) Poly(vinyl acetate) Good, questionable Poor Poly(vinyl butryral) Good Glyptal resins (alkyd resins) Silicone resins Poor, moderate Good Rubbers Pure natural rubber (caoutchouc) Natural rubber vulcanizate Attacked Attacked Crude sheet Attacked Pale crepe, not compounded Pale crepe, compounded Attacked Resistant, attacked Smoked sheet, not compounded Attacked Smoked sheet, compounded Reclaimed rubber Resistant, attacked Attacked Chlorinated rubber Resistant Neoprene, compounded GR-S, butadiene-styrene, compounded Resistant Resistant, attacked Hycar OR, butadiene-acrylonitrile compounded Resistant, attacked Buna N, butadiene-acrylonitrile, compounded GR-I (butyl), isobutylene-isoprene, compounded Attacked Resistant, attacked Thiokol, organic polysulﬁde, uncured Attacked Thiokol, organic polysulﬁde, vulcanized Thiokol, organic polysulﬁde, sheets for gasoline tank lining Resistant Attacked Silicone rubber Resistant Source: Adapted from Halim Hamid, S., Amin, M. B., and Maadhah, A. G., eds. 1992. Handbook of Polymer Degradation. Marcel Dekker, New York. 109 Characteristics of Polymers and Polymerization Processes prevent or minimize the degradative effects. These ingredients act either by interfering with degradative processes or by minimizing the cause of degradation. 1.17.1 Antioxidants and Related Compounds Oxidation, as we have noted, is a free-radical chain process. The most useful stabilizing agents will therefore be those which combine with free radicals, as shown by Equation 1.106, to give a stable species incapable of further reaction. These stabilizing agents are called antioxidants. They are the most frequently employed ingredients in plastics, ﬁbers, rubbers, and adhesives. Stabilization is also achieved in some polymer systems by the use of additives which moderate the degradation reaction. Two large, basic groups of antioxidants are normally distinguished: (1) chain terminating or primary antioxidants, and (2) hydroperoxide decomposers or secondary antioxidant, frequently called synergists. The majority of primary antioxidants are sterically hindered phenols or secondary aromatic amines. They are capable of undergoing fat reaction with peroxy radicals and so are often called radical scavengers. For hindered phenols, for example, such reactions may be represented by the following scheme: OH O ROO + ROOH + CH3 O (1.111) CH3 O O + ROO (1.112) CH 3 H 3C CH 3 OOR ( = tertbutyl group) Stabilization is achieved by the fact that Reaction 1.111 competes with peroxy radical reactions in polymer degradation, such as Reaction 1.103, transforming the reactive peroxy radical into a much less reactive phenoxy radical, which, in turn, is capable of reacting with a second peroxy radical according to Reaction 1.112. The phenoxy radicals formed in Reaction 1.111 do not initiate new radical chains at the normal temperatures of use and testing, but such propagation reactions become possible at high temperatures. The thermal stability of peroxycyclohexadienones (formed by Reaction 1.112) is also limited, and their decomposition leads to new reaction chains. The effectiveness of sterically hindered phenols thus decreases with increasing temperature. The secondary antioxidants are usually sulfur compounds (mostly thioethers and esters of thiodipropionic acid) or trimesters of phosphorous acid (phosphates). A remedy for many types of discoloration in plastics is often the use of a phosphite or a thioether. Both have the ability to react with hydroperoxides, as those formed in Reaction 1.111, to yield nonradical products, following heterolytic mechanisms. The reaction of phosphites to phosphates as an example: P(OR0 )3 + ROOH ! OP(OR0 )3 + ROH (1.113) Thioethers react with hydroperoxides in a ﬁrst stage to yield sulfoxides and alcohols: S R1 R2 + ROOH OS R1 R2 + ROH (1.114) 110 Plastics Technology Handbook Sulfoxides themselves yield, on further oxidation, even more powerful hydroperoxide decomposers than the original sulﬁdes, in that they are able to destroy several equivalents of hydroperoxides. This catalytic effect is explained by the intermediate occurrence of sulfenic acids and sulfur dioxide. The fact that the phenomenon of synergism, which is deﬁned as a cooperative action such that the total effect is greater than the sum of two or more individual effects taken independently, is often observed when primary and secondary antioxidants are combined ahs been explained with the concept of the simultaneous occurrence of the radical reactions (e.g., Equation 1.111) and the nonradical hydroperoxide decomposition (e.g., Equation 1.113 and Equation 1.114). The preceding synergistic effect may be considered as an example of heterosynergism, which arises from the cooperating effect of two or more antioxidants acting by different mechanisms. Homosynergism, on the other hand, involves two compounds of unequal activity that are operating by the same mechanism, for example, a combination of two different chain-breaking antioxidants that normally function by donation of hydrogen to a peroxy free radical. The most likely mechanism of synegism in this case would involve the transfer of hydrogen from one inhibitor molecule to the radical formed in the reaction of the other inhibitor with a peroxy radical (see Equation 1.111), thus regenerating the latter inhibitor and prolonging its effectiveness. 1.17.2 Chemical Structures of Antioxidants While sterically hindered phenols and secondary aromatic amines are the two main chemical classes of antioxidants for thermoplastics, the greatest diversity is, however, found in the class of the sterically hindered phenols. Most of these phenols are commercially available as relatively pure chemicals. Although aromatic amines are often more powerful antioxidants than phenols, their application is limited to vulcanized elastomers since the staining properties of aromatic amines prohibit their use in most thermoplastics. Sterically hindered phenols of commercial importance may be further classiﬁed, according to their structure, into (1) alkylphenols, (2) alkylidene-bisphenols, (3) thiobisphenols, (4) hydroxybenzyl compounds, (5) acylaminophenols, and (6) hydroxyphenyl propionates. An important goal of antioxidant research has been to provide poly-phenols of high molecular weight and low volatility. Most of the commercial sterically hindered phenols have molecular weights in the range 300–600 and above 600. Figure 1.74 depicts structural formulas [61] of the important sterically hindered phenols used as antioxidants. The majority of secondary antioxidants are esters of thiodipropionic acid with fatty alcohols and trimesters of phosphorus acid. Their chemical structures are shown in Figure 1.75. Phosphites are more sensitive towards hydrolysis than esters of carboxylic acids, and this point has to be taken into account when storing and using phosphates. Aromatic phosphates are often preferred since they are inherently more resistant to hydrolysis than aliphatic phosphates. Detailed lists of trade names, manufacturers, and suppliers of antioxidants for thermoplastics are found in Ref. [61,62]. 1.17.3 Stabilization of Selected Polymers Two important considerations in selecting antioxidants for polymers are toxicity and color formation. Thus, for use in food wrapping, any antioxidant or additive must be approved by the appropriate government agency. An antioxidant widely used in food products and food wrapping is butylated hydroxytoluene (BHT) (see Figure 1.74a: R1 = tert-butyl and R2 = –CH3). It is also added in small amounts to unsaturated raw rubbers before shipping to protect them during storage. Materials that are not effective inhibitors when used alone may nevertheless be able to function as synergists by reacting with an oxidized form of an antioxidant to regenerate it and thus prolong its effectiveness. For example, carbon black forms an effective combination with thiols, disulﬁdes, and 111 Characteristics of Polymers and Polymerization Processes (a) R1 Alkylphenols, e.g.: Tert butyl OH R1 R2 CH3 R1 C9H19 H3C CH R2 C9H19 H3C C CH3 HO OH CH3 (b) CH3 Alkylidene-bisphenols, e.g.: OH R3 OH R1 H R2 R1 R2 Tert butyl Tert butyl Tert butyl Tert butyl Tert butyl Sec butyl H3C R1 Tert butyl Tert butyl R2 R3 H CH3 CH3 CH3 H C2H5 CH3 H H CH3 H H C9H19 CH3 FIGURE 1.74 Structural formulas [61] of important sterically hindered phenols used as antioxidants for thermoplastics. (Continued ) elemental sulfur, even though these substances may be almost completely ineffective alone under comparable conditions. Synergistic combinations from a variety of chain terminators and sulfur compounds have been reported. A widely used combination of stabilizers for polyoleﬁns is 2,6-di-tert-butyl-4methylphenol (BHT) (Figure 1.74a) with dilauryl thiodipropionate (DLTDP): O (H25C12O C CH2 CH2)2S This combination is of particular interest because both components are among the small group of stabilizers approved by the U.S. Food and Drug Administration for use in packaging materials for food products. Other synergistic combinations have been reported involving free-radical chain terminators used with either ultraviolet absorbers or metal deactivators as the preventive antioxidants. The differences in 112 Plastics Technology Handbook OH R C3H7 CH2 CH3 HC CH CH3 CH3 R CH3 OH OH OH H3C C CH2 COOCH2 2 (c) OH Thiobisphenols, e.g.: OH OH H3C S HO CH3 (d) S OH CH3 CH3 Hydroxybenzyl compounds, e.g.: HO CH2 A A CH3 H3C 3 CH3 O N O FIGURE 1.74 (CONTINUED) dants for thermoplastics. N N O Structural formulas [61] of important sterically hindered phenols used as antioxi(Continued ) mechanism of action of these different types of stabilizers permit them to act independently but cooperatively to provide greater protection than would be predicted by the sum of their separate effects. 1.17.3.1 Polypropylene Because of the presence of tertiary carbon atoms occurring alternately on the chain backbone, propylene is particularly susceptible to oxidation at elevated temperatures. Since polypropylene is normally processed at temperatures between 220 and 280°C, it will degrade under these conditions (to form lower-molecularweight products) unless it is sufﬁciently stabilized before it reaches the processor. The antioxidants are added at least partially during the manufacturing process and at the least during palletizing. Antioxidant 113 Characteristics of Polymers and Polymerization Processes O R H3C R N N N O R CH2 O H3C R OH O CH2 HO (e) OC2H5 P Oθ Ca 2 Aminophenols, e.g. : R1 R1 HO R2 H H NHR2 COC11H23 COC17H35 SC8H17 R1 N N tert. butyl N SC8H17 (f ) Hydroxyphenylpropionates, e.g. : n 1 HO CH2CH2Co A 2 2 n 2 A OC18H37 O(CH2)6O NH (CH2)6 NH O(CH2)2 S (CH2)2O CH2CH2O O N O 3 N N OCH2CH2 CH2CH2O O FIGURE 1.74 (CONTINUED) dants for thermoplastics. Structural formulas [61] of important sterically hindered phenols used as antioxi- systems in technical use are composed of processing stabilizers, long-term heat stabilizers, calcium- or zinc-stearate, and synergists if necessary. Typical processing stabilizers for polypropylene and butylated hydroxy-toluene (BHT) as the primary antioxidant and phosphates and phosphonates as secondary antioxidants. Examples of the latter that are commonly used are: tetrakis-(2,4-di-tert-butyl-phenyl)-4-4′-bisphenylylenediphosphonite, distearylpentaerythrityl-diphosphonite, tris-(nonylphenyl)-phosphite, tris-(2,4-di-tert-butyl-phenyl)-phosphite and bis(2,4-di-tert-butyl-phenyl)- pentaerythrityl-diphosphite. In commercial polypropylenes, phosphorous compounds are always used together with a sterically hindered phenol. The compounds are commonly added in concentrations between 0.05 and 0.25%. The most important long-term heat stabilizers for polypropylene are phenols of medium (300–600) and especially high (600–1,200) molecular weight, which are frequently used together with thioethers 114 Plastics Technology Handbook (a) Thioethers, e.g.: CH2CH2 COOR R C12H25 C13H27 C18H37 S CH2CH2 (b) COOR Phosphites and phosphonites, e.g.: R1 R2 phenyl n decyl R1O P OR2 R1O nonylphenyl nonylphenyl R RO O O O O P – C18H37 P OR O O P P O (c) Zinc dibutyldithiocarbamate S S (C4H9)2 N FIGURE 1.75 O C S Zn S C N(C4H9)2 Structural formulas [61] of some commercial secondary antioxidants for thermoplastics. as synergists, e.g., dilauryl thiodipropionate (DLTDP), or distearyl thiodipropionate (DSTDP), or dioctadecyl disulﬁde. 1.17.3.2 Polyethylene High-density polyethylenes (HDPE) are less sensitive to oxidation than polypropylene, so lower stabilizer concentrations are generally sufﬁcient. As in polypropylene, antioxidants can be added during a suitable stage of manufacture or during palletizing. The antioxidants in technical use are the same as for polypropylene. Phenols of medium- and high-molecular weight are also active as long-term heat stabilizers. Concentrations between 0.03 and 0.15% are usual. Low-density polyethylene (LDPE) is extensively used for the manufacture of ﬁlms. During processing, which is carried out at temperatures of approximately 200°C, cross-linking, and thus formation of gel, can occur through oxidation if the polymer is not stabilized. Such gel particles are visible in the ﬁlm as agglomerates, known as ﬁsh eyes or arrow heads. The processing stabilizers used in LDPE consist of systems commonly used for polypropylene, namely, combinations of a phosphite or phosphonite and a long-term heat stabilizer (hindered phenol) in overall concentrations up to 0.1%. Concentrations seldom exceed 0.1%, since the compatibility of any additive in LDPE is considerably lower than in any other polyoleﬁns. For primary cable insulation. Cable jackets and pipes manufactured from LDPE, medium to highmolecular-weight grades are used and are frequently cross-linked after processing. Long-term heat Characteristics of Polymers and Polymerization Processes 115 stabilizer is of primary importance in these applications, since lifetimes of up to 50 years are required (usually at elevated temperatures with short-time peaks up to 100°C for cables). The antioxidants have to be extremely compatible and resistant to extraction. Some typical antioxidants customary for insulation are as follows: For power cable insulation. Pentaerythrityltetrakis-3-(3,5-di-tert-butyl-4-hydroxyphenyl)-propionate, polymeric 2,2,4-trimethyl-1,2-dihydroquino-line, 4,4′-thiobis-(3-methyl-6-tert-butyl-phenol), 2,2′-thiodiethyl-bis-3-(3,5-di-tert-butyl-4-hydroxyphenyl)-propionate, and distearyl thiodipropionate are used as synergists. For communication cable insulation. 2,2′-Thiobis-(4-methyl-6-tert-butyl-phenol), 4,4′-thiobis-(3methyl-6-tert-butyl-phenol), 2,2′-methylene-bis-(4-methyl-6-a-methylcyclohexyl-phenol), 1,1,3-tris-(5tert-butyl-4-hydroxy-2-methylphenyl)-butane, 2,2′-thiodiethyl-bis-3-(3,5-di-tert-butyl-4-hydroxy-phenyl)propionate, pentaerythrityl-tetrakis-3-(3,5-di-tert-butyl-4-hydroxy-phenyl)-propionate, and dilauryl thiodipropionate are used as synergists. The simultaneous use of metal deactivators has become increasingly important for cable insulation. This is discussed later. 1.17.3.3 Polystyrene Unmodiﬁed crystal polystyrene is relatively stable under oxidative conditions, so that for many applications the addition of an antioxidant is not required. Nevertheless, repeated processing may lead to oxidative damage of the material, leading to an increase of melt ﬂow index and to embrittlement of the material. Stabilization is effected by the addition of octadecyl-3-(3,5-di-tert-butyl-4-hydroxy-phenyl)propionate at concentrations of up to 0.15%, if necessary in combination with phosphates or phosphonites to improve color. Compared to unmodiﬁed crystal polystyrene, impact polystyrene consisting of copolymers of styrene and butadiene are more sensitive to oxidation. This sensitivity is a consequence of the double bonds in polybutadiene component and manifests itself in yellowing and the loss of mechanical properties of the polymer. In impact polystyrene, the following antioxidants or their mixture are used in total concentrations of 0.1%–0.25%: BHT, octadecyl-3-(3,5-di-tert-butyl-4-hydroxyphenyl)-propionate, 1,1,3-tris-(5tert-butyl-4-hydroxy-2-methylphenyl)-butane, and dilauryl thiodipropionate. 1.17.3.4 Acrylonitrile–Butadiene–Styrene Copolymers Like impact polystyrene, acrylonitrile–butadiene–styrene copolymers (ABS) are sensitive to oxidation caused by the unsaturation of the elastomeric component. The processes for the manufacture of ABS require the drying (at 100°C–150°C) of powdery polymers that are extremely sensitive to oxidation. Thus, antioxidants have to be added before the coagulation step, normally in emulsiﬁed form, although sometimes in solution. The primary antioxidants are frequently sued together with a synergist. Primary antioxidants commonly used for ABS are BHT, 2,2′-methylenebis-(4-ethyl or methyl-6-tert-butyl-phenol), 2,2′methylenebis-(4-methyl-6-cyclohexyl-phenol), 2,2′-methylenebis-(4-methyl-6-nonyl-phenol), octadecyl-3(3,5-di-tert-butyl-4-hydroxyphenyl)-propionate, and 1,1,3-tris-(5-tert-butyl-4-hydroxy-2-methylphenyl)butane. Important synergists are tris-(nonyl-phenyl)-phosphite and dilauryl thiodipropionate. These antioxidants are either liquids or show comparatively low melting points, which is an important prerequisite for the formation of stable emulsions. 1.17.3.5 Polycarbonate Thermoxidative degradation of polycarbonate manifests itself in yellowing that is readily seen because of the transparency of polycarbonate. For this reason, stabilization against discoloration is considered to be important. Severe requirements are also imposed on the nonvolatility and thermostability of the stabilizers for polycarbonate, because the processing temperatures are extraordinarily high (about 320°C). The stabilizers are generally added during the palletizing step. 116 Plastics Technology Handbook Yellowing of polycarbonate during processing is retarded by the addition of phosphates or phosphonites. They are used in concentrations of 0.05%–0.15%, possibly in combination with an epoxy compound as acid acceptor. The addition of these stabilizers not only decreases the yellowness index but also inhibits the rise of the melt ﬂow index and negative inﬂuence on impact strength of processing. The effective processing stabilizers are not suitable, however, for preventing the aging effect of longterm use. For long-term heat stabilization, a sterically hindered phenolic antioxidant is added. An effective antioxidant is octadecyl-3-(3,5-di-tert-butyl-4-hydroxyphenyl)-propionate. 1.17.3.6 Nylons Degradation of nylons due to processing and aging causes discoloration and loss of mechanical properties, although not at the same rate. For example, yellowing is observed already after short periods of oven aging at 165°C, but tensile strength and elongation are hardly affected during the same time period. Discoloration of polyamides during processing can be suppressed to a certain extent by the addition of phosphates, e.g., tris-(2,4-di-tert-butyl-phenyl)-phosphite, in concentrations of 0.2%–0.4%. The stabilization of nylons is mainly a matter of long-term stabilization. Three main groups of stabilizers have become known: (1) copper salts, especially in combination with halogen and/or phosphorus compounds (e.g., copper acetate ? potassium iodide/phosphoric acid), (2) aromatic amines (e.g., N,N′dinaphthyl-p-phenylenediamine or N-phenyl-N′-cyclohexyl-p-phenylenediamine), and (3) hindered phenols. Copper-halogen systems are effective in very low concentrations (10–50 ppm of copper; approximately 1000 ppm of halogen), although they are extracted quite easily with water and cause discoloration of the substrate. Next to the copper-halogen systems, aromatic amines are the most effective stabilizers. They are used in rather high concentrations, from 0.5 to 2%. However, since they have strongly discoloring properties they are used mainly for technical articles, which tolerate discoloration. Hindered phenols do not show the above-mentioned disadvantages. They are the stabilizers of choice whenever good oxidation stability has to be coupled with good color stability and, possibly, food approval of the end article. The most important hindered phenols in use are the following: N,N′-hexamethylenebis-3(3,5-di-tert-butyl-4-hydroxyphenyl)-propionamide, 1,1,3-tris-(5-tert-butyl-4-hydroxy-2-methylphenylbutane, 1,3,5-tris-(3,5-di-tert-butyl-4-hydroxybenzyl)-mesitylene, and BHT. The antioxidants can be added already during polycondensation, the normal concentration ranging from 0.3 to 0.7%. 1.17.3.7 Thermoplastic Elastomers Styrene-based thermoplastic elastomers (see Chapter 4) are sensitive to oxidation since they contain unsaturated soft segments. These elastomers are manufactured by solution polymerization process in aliphatic hydrocarbons. In order to prevent autoxidation during the ﬁnishing steps (stripping, drying), which manifests itself by a rise in melt ﬂow index and discoloration of the raw polymer, antioxidant is added to the polymer solution before ﬁnishing. Hence the antioxidant has to be soluble in the polymerization solvent. A number of hindered phenols are used in practice in a total concentration of approximately 0.5%. Examples of primary antioxidants used are BHT, 1,3,5-tris-(3,5-di-tert-butyl-4-hydroxybenzyl)-mesitylene, and octadecyl-3-(3,5-di-tert-butyl-4-hydroxyphenyl)-propionate. Tris-(nonylphenyl)-phosphite is used as a synergist. BHT may, however, be partially lost during ﬁnishing and drying. The phenols of higher molecular weight are sued as they have low volatility and have the further advantage of protecting the material also during processing and end use. Thermoplastic polyester elastomers (see Chapter 4) contain readily oxidizable polyether soft segments that make stabilization necessary. Essentially two antioxidants, namely, 4,4′-di(a,a-dimethylbenzyl)diphenylamine and N,N′-hexamethylenebis-3-(3,5-di-tert-butyl-4-hydroxyphenyl)-propionamide are in use in concentrations of up to 1%. The antioxidants can be added during pelletizing, or even better, already during polycondensation. When added during polycondensation, N,N′-hexamethylenebis-3-(3,5- 117 Characteristics of Polymers and Polymerization Processes di-tert-butyl-4-hydroxyphenyl)-propionamide is partly chemically bound to the polymer because of its amide structure. The antioxidant thus becomes highly stable to extraction. The oxidative stability of thermoplastic polyurethane elastomers is determined by the length and the structure of the linear polyether or polyester soft segments. The stability of polyether urethanes against autoxidation is distinctly lower compared to that of polyester urethanes, but the latter are less stable to hydrolysis. Antioxidants may be used in polyurethanes for stabilization against loss of mechanical properties and against discoloration in injection molding grades and as gasfading inhibitors in elastomeric ﬁbers. The antioxidant used in thermoplastic polyurethane elastomers are hindered phenols, e.g., BHT, octadecyl-3-(3,5-di-tert-butyl-4-hydroxyphenyl)-propionate, and pentaerythrityl-tetrakis-3-(3,5-di-tertbutyl-4-hydroxyphenyl)-propionate, and aromatic amines, e.g., 4,4′-di-tert-octyl-diphenylamine, as well as their combinations. Aromatic amines can, however, be employed only in very limited concentrations (250–550 ppm at most) because of their discoloring properties. 1.17.3.8 Polyacetal Polyoxymethylenes have a marked tendency to undergo thermal depolymerization with loss of formaldehyde. To prevent thermal depolymerization, polyoxymethylenes are structurally modiﬁed, the two possibilities being acetylation to block the reactivity of the end groups of co-polymerization with cyclic ethers, e.g., ethylene oxide. Polyacetals are also sensitive towards autoxidation, which invariably leads to depolymerization as a result of chain scission. The formaldehyde released by depolymerization is very likely to be oxidized to formic acid, which can catalyze further depolymerization. The stabilizer systems for polyacetals are invariably composed of a hindered phenol with a costabilizer. The hindered phenols in use are 2,2′-methylenebis-(4-methyl-6-tert-butyl-phenol), 1,6hexamethylenebis-3-(3,5-di-tert-butyl-4-hydroxyphenyl)-propionate, and pentaerythrityl-tetrakis-3(3,5-di-tert-butyl-4-hydroxyphenyl)-propionate. A large number of nitrogen-containing organic compounds have been described as costabilizers for polyacetals, e.g., dicyandiamide, melamine, terpolyamides, urea, and hydrazine derivatives. The effectiveness of these compounds is based on their ability to react with formaldehyde and to neutralize acids, especially formic acid, formed by oxidation. In addition to nitrogen compounds, salts of long-chain fatty acids (e.g., calcium stearate, calcium ricinoleate, or calcium citrate) are also used as acid acceptors. The practical concentrations are 0.1–0.5% for the phenolic antioxidant and 0.1–1.0% for the costabilizer. 1.17.3.9 Poly(Vinyl Chloride) Poly(vinyl chloride) (PVC) is relatively unstable under heat and light. The ﬁrst physical manifestation of degradation is a change in the PVC color, which on heating, changes from the initial water-white to pale yellow, orange, brown, and ﬁnally black. Further degradation causes adverse changes in mechanical and electrical properties. The most widely accepted mechanism for PVC degradation is one based on a free-radical chain. Thermal initiation probably involves loss of a chlorine atom adjacent to some structural abnormality, such as terminal unsaturation, which reduces the stability of the C–Cl bond [60]. The chlorine radical thus formed abstracts a hydrogen to form HCl, and the resulting chain radical then reacts to form a chain unsaturation with regeneration of another chlorine radical. Cl + –HCI CH2 CH CH CH Cl CH + Cl CH Cl Thus, as hydrogen chloride is removed, polyene structures are formed: CH2 CH Cl CH2 CH Cl CH2 CH –HCI Cl CH CH CH CH CH CH 118 Plastics Technology Handbook The reaction can also be initiated by ultraviolet light. In the presence of oxygen the reactions are accelerated (as evidenced by the acceleration of color formation), and ketonic structures are formed in the chain. Stabilizers are almost invariably added to PVC to improve its heat and light stability. The species found effective in stabilizing PVC are those that are able to absorb or neutralize HCl, react with free radicals, react with double bonds, or neutralize other species that might accelerate degradation. Lead compounds, such as basic lead carbonate and tribasic lead sulfate, and metal soaps of barium, cadmium, lead, zinc, and calcium are used as stabilizers. Obviously, they can react with HCl. Epoxy plasticizers aid in stabilizing the resin. Another group of stabilizers are the organotin compounds, which ﬁnd application because of their resistance to sulfur and because they can yield crystal-clear compounds. 1.17.3.10 Rubber In stabilizing rubber, antioxidants are classiﬁed as staining or nonstaining, depending on whether they develop color in use. In carbon-black-loaded rubber tire a staining material is not harmful. For white rubber goods, however, it is important that the additives used be colorless and that they stay colorless while protecting against oxidation. Some antioxidants that ﬁnd use in rubber are phenyl-b-naphthylamine (PBNA), which is staining, and 2,2′-thiobis (6-tert-butyl-para-cresol) (see Figure 1.74c), which is nonstaining. It has been established that ozone formation is a general phenomenon that occurs when free radicals can combine with oxygen. Rubber products made form natural rubber, styrene–butadiene rubber (SBR), or nitrile rubber, when stretched moderately and exposed to low concentrations of ozone, crack rapidly and sometimes disastrously. The cracking is caused by cuts that appear on the surface and may penetrate deeply, causing serious damage. These cuts appear only when the rubber is under tension during exposure to ozone and if they cross the lines of tension at an angle of 90°. Since the cracking problem is associated only with elastomers containing some degree of residual unsaturation, the process is considered to be related to the attack of ozone on the unsaturation. Unlike molecular oxygen (O2), ozone appears to add directly to the double bond, which is often followed by chain scission. Use of diene rubbers in an environment containing ozone, such as an automobile tire in some urban locations, therefore calls for stabilization against ozone. The effect of ozone is both greatly delayed and reduced when various antiozonants, such as N,N′-dialkyl-p-phenylene-diamines (X) or similar compounds, are incorporated in the rubber. They may react directly with ozone or with the ozone-oleﬁn reaction products in such a way as to prevent chain scission. SBR is much more receptive to protection than either the nitrile or natural rubbers. Of the latter two the nitrile is more readily inhibited. If a wax is also added in small quantities with the antiozonant, the retardation of the attack of ozone is in certain instances enhanced several fold. H H N R1,R2 = Alkyl N R1 R2 (X) It is believed that waxes function by providing an unreactive layer on the rubber surface. Since, unlike oxygen, ozone reacts only at the surface and does not diffuse into the sample, a surface layer of relatively unreactive wax presents an impervious surface to ozone. The wax thus needs to bloom (i.e., exude to the surface) in order to afford protection. A great deal of attention has therefore been paid to the factors affecting migratory aptitude of wax. Waxes with good protective power have been found to have melting points between 65 and 72°C, refractive indices sin the range of 1.432–1.438, and branching to the extent of 30%–50% side chains [60]. Of the two general categories of waxes—parafﬁn and micro-crystalline—the latter are more strongly held to the surface. However, the use of waxes alone to provide protection against ozone attack is rather well restricted to static conditions of service. Whenever constant ﬂexing is present, even the more strongly held microcrystalline waxes ﬂake off and protection is lost. Combinations of waxes and chemical antiozonants are therefore used to provide protection under both static and dynamic conditions of service. In fact, it is felt that waxes can aid in the diffusion of chemical antiozonant to the rubber surface. Characteristics of Polymers and Polymerization Processes 119 1.18 Metal Deactivators As described previously, thermooxidative degradation of polyoleﬁns proceeds by a typical free-radical chain mechanism in which hydroperoxides are key intermediates because of their thermally-induced hemolytic decomposition to free radicals, which in turn initiate new oxidation chains. However, since the monomolecular hemolytic decomposition of hydroperoxides into free radicals require relatively high activation energies, this process becomes effective only at temperatures in the range of 120°C and higher. But in the presence of certain metal ions, hydroperoxides can undergo catalytic decomposition even at room temperature by a redox reaction to radical products. Two oxidation-reduction reactions can be involved depending on the metal and its state of oxidation: ROOH + Mn+ ! RO• + M(n+1)+ + OH− (1.115) ROOH + M(n+1)+ ! ROO• + Mn+ + H+ (1.116) Mn+ =M(n+1)+ 2ROOH ⟶ RO• + ROO• + H2 O (1.117) The electron transfer producing free radicals as shown above is preceded by the formation of unstable coordination complexes of the metal ions with alkyl hydroperoxides. The relative importance of Reaction 1.115 and Reaction 1.116 depends upon the relative strength of the metal ion as an oxidizing or reducing agent. When the metal ion has two valence states of comparable stability, both Reaction 1.115 and Reaction 1.116 will occur, and a trace amount of the metal can convert a large amount of peroxide to free radicals according to the sum of the two reactions (Reaction 1.117). This is true of compounds of metals such as Fe, Co, Mn, Cu, Ce, and V, commonly called transition metals. The presence of the above-mentioned metal ions increases the decomposition rate of hydroperoxides and the overall oxidation rate in the autoxidation of a hydrocarbon to such an extent that even in the presence of antioxidants, the induction period of oxygen uptake is drastically shortened. In such a case, sterically hindered phenols or aromatic amines even at rather high concentrations, do not retard the oxidation rate satisfactorily. A much more efﬁcient inhibition is then achieved by using metal deactivators, together with antioxidants. Metal deactivators are also known as copper inhibitors, because, in practice, the copper-catalyzed oxidation of polyoleﬁns is by far of greatest importance. This is due to the fact that polyoleﬁns are the preferred insulation material for communication wire and power cables, which generally contain copper conductors. The function of a metal deactivator is to form an inactive complex with the catalytically active metal species. Specially suited for this purpose are chelating agents, which can form metal complexes of high thermal stability. The general feature of chelating agents is that they may contain several ligand atoms like N, O, S, P, alone or in combination with functional groups such as hydroxyl, carboxyl, or carbamide groups. The chelating agent N,N′-bis-(o-hydroxybenzal) oxalyldihydrazide is a good example for the above mentioned structural possibilities. It forms a soluble complex (XI) with the ﬁrst mole of a copper salt by binding at the phenolic group, and a second mole of copper salt binds at amide nitrogens to form an insoluble complex. O O C C HN NH CH N N CH Cu O O (XI) 120 Plastics Technology Handbook Besides their main function to retard efﬁciently the metal-catalyzed oxidation of polyoleﬁns, metal deactivators have to possess a number of ancillary properties to be useful in actual service. These include sufﬁcient solubility or ease of dispersion, high extraction resistance, and low volatility and sufﬁcient thermal stability under processing and service conditions. Within the last two decades, a number of chemical structures have been proposed as metal deactivators for polyoleﬁns. These include carboxylic acid amides of aromatic mono- and di-carboxylic acids and Nsubstituted derivatives such as N,N′-diphenyloxamide, cyclic amides such as barbituric acid, hydrazones and bishydrazones of aromatic aldehydes such as benzaldehyde and salicylaldehyde or of o-hydroxyarylketones, hydrazides of aliphatic and aromatic mono- and di-carboxylic acids as well as N-acylated derivatives thereof, bisacylated hydrazine derivatives, polyhydrazides, and phosphorus acid ester of a thiobisphenol. Though there are metals other than copper (such as iron, manganese and cobalt) that can accelerate thermal oxidation of polyoleﬁns and related polymers such as EPDM, in practice, however, the inhibition of copper-catalyzed degradation of polyoleﬁns is of paramount importance because of the steadily increasing use of polyoleﬁn insulation over copper conductors. Among polyoleﬁns, polyethylene is still the most common primary insulation material for wire and cable. In the United States, high-density polyethylene and ethylenepropylene copolymers are used in substantial amounts for communications wire insulation. For stabilization of polyoleﬁns in contact with copper, it is often mandatory to combine a metal deactivator with an antioxidant. Metal deactivators in actual use are essentially N,N′-bis-[3-(3′,5′-di-tertbutyl-4′-hydroxy-phenyl)propionyl]-hydrazine and N,N′-dibenzaloxalyldihydrazide. The latter compound requires predispersion in a masterbatch because of its insolubility in polyoleﬁns. This is not needed, however, for the former compound, which at commonly used concentrations is molecularly dispersed in polyoleﬁns after processing. The required additive concentrations range from 0.05 to 0.5% depending on the polymer, the nature of the insulation (solid, cellular), whether the cable is petrolatum ﬁlled, and on service conditions. Increasingly, communication cables are ﬁlled with petrolatum to improve waterprooﬁng, whereas the insulation may be solid or cellular. Both trends, namely, petrolatum ﬁlling and cellular insulation, exert an inﬂuence on the oxidative stability of such cables. Thermal stability of high density polyethylene decreases by about 35% in changing from solid to cellular insulation. Moreover, in contact with petrolatum the stability of solid poly-ethylene decreases by 35% and that of cellular polyethylene decreases by 10–40%. Even in the most adverse instance, i.e., cellular insulation in contact with petrolatum, elevated concentrations of antioxidant and metal deactivator make it possible to achieve a high level of stability. 1.19 Light Stabilizers Most polymers are affected by exposure to light, particularly the ultraviolet (UV) portion of sun’s spectrum between 300 and 400 nm. (Fortunately, the earth’s atmosphere ﬁlters out most of the light waves shorter than 300 nm.) Light contributes very actively to polymer degradation, especially when oxygen is present, which is a normal situation for most plastic materials. For example, when rubber is exposed to UV radiation at 45°C, it oxidizes three times as fast as in the dark at 70°C. Light and oxygen induce degradation reaction sin plastics that may not only discolor them but also exert a detrimental inﬂuence on numerous mechanical and physical properties. The inhibition of these degradation reactions is essential or else, the applications of many plastics would be drastically reduced. The inhibition can be achieved through addition of special chemicals, light stabilizers, or UV stabilizers, which are capable of interfering with the physical and chemical processes of light-induced degradation. Carbon black is used as a stabilizer in a limited number of formulations where color is not a criterion. It not only absorbs light, but it can also react with free-radical species that might be formed. The weathering properties of polyethylene are improved by the incorporation of carbon blacks (at 2%–3% concentration). Characteristics of Polymers and Polymerization Processes 121 Weather-resistant wire and cable insulation, pipe for outdoor applications, ﬁlms for mulching, and water conservation in ponds and canals may be made from polyethylene containing carbon black. In spite of the fact that carbon black and other pigments may also protect the plastics from the effects of light, we shall consider in the following only those organic and organo-matallic compounds that impart only slight discoloration or no discoloration at all to the plastics to be stabilized. The most important classes of light stabilizers from a practical point of view are 2-hydroxybenzophenones, 2-hydroxyphenylbenzotriazoles, hindered amines, and organic nickel compounds. In addition, salicylates, p-hydroxy-benzoates, resorcinol monobenzoates, cinnamate derivatives, and oxanilides are also used. Light stabilizers are generally used in concentrations between 0.05 and 2%, the upper limit of this range being employed only exceptionally. The sterically hindered amines represent the latest development in the ﬁeld. They have outperformed the previously available light stabilizer classes in numerous synthetic resins. 1.19.1 Light Stabilizer Classes The designation of the light stabilizer classes is based on mechanisms of UV stabilization. Free radicals are formed in polymers exposed to light, as a consequence of the excitation of absorbing functionalities in the polymer. This is a function of the energy of the light and of the structure of the polymer molecule. Since in the presence of oxygen, the polymer will simultaneously oxidize (photooxidation), it is often difﬁcult to distinguish the pure photochemical processes from the thermal processes (oxidation), which are then superimposed. Some of the fundamental mechanisms involved have been evaluated [63]. The mechanism developed initially for the thermal oxidation of rubber can be applied to other substrates and also to photooxidation. The scheme shown below illustrates a possible reaction sequence to photooxidation: Chain initiation Hydroperoxides ROOH Free radicals Carbonyl compounds > C = O Δ,hν (R , RO , Catalyst residues (Ti, ...) Mn+/M(n+1)+ HO ,HO2 ,...) Charge transfer complexes (RH, O2) (1.118) Chain propagation R• + O2 ! RO•2 (1.119) RO•2 + RH ! RO2 H + R• (1.120) Chain branching ROOH 2ROOH Δ or hν RO + HO Mn+/M(n+1)+ Δ or hν RO2 + RO + H2O Mn+/M(n+1)+ (1.121) (1.122) RO• + RH ! ROH + R• (1.123) HO• + RH ! H2 O + R• (1.124) 122 Plastics Technology Handbook Chain termination 9 > > = R• + R• ! R• + RO•2 ! RO•2 + RO•2 ! > > ; nonradical products (1.125) In the light of the above photooxidation scheme there are four possibilities of protection against UV light. These are based on (1) prevention of UV light absorption or reduction of the amount of light absorbed by the chromophores; (2) reduction of the initiation rate through deactivation of the excited states of the chromophoric groups: (3) intervention in the photooxidative degradation process by transformation of hydroperoxides into more stable compounds, without generation of free radicals, before hydroperoxides undergo photolytic cleavage; and (4) scavenging of the free radicals as soon as possible after their formation, either as alkyl radicals or as peroxy radicals. According to the four possibilities of UV protection mechanism described above, the light stabilizer classes can be designated as (1) UV absorbers, (2) quenchers of excited states, (3) hydroperoxide decomposers, and (4) free radical scavengers. It must be mentioned, however, that this classiﬁcation is a simpliﬁcation and that some compounds may be active in more than one way and often do so. 1.19.1.1 UV Absorbers The protection mechanism of UV absorbers is based essentially on absorption of harmful UV radiation and its dissipation in a manner that does not lead to photosensitization, i.e., conversion to energy corresponding to high wavelengths or dissipation as heat. Besides having a very high absorption themselves, these compounds must be stable to light, i.e., capable of absorbing radiative energy without undergoing decomposition. Hydroxybenzophenones and hydroxyphenyl benzotriazoles are the most extensively studied UV absorbers. Though the main absorptions of 2-hydroxybenzophenone are situated in the uninteresting wavelength domain around 260 nm, substituents such as hydroxy and alkoxy groups push this absorption towards longer wavelengths, between 300 and 400 nm, and at the same time total absorption in the UV absorbers (XII) are essentially derived from 2,4-dihyrdoxybenzophenone (X, R═H). Through choice of adequate alkyl group R in the alkoxy groups it is possible to optimize the protective power and the compatibility with the plastics to be stabilized. O OH C R = H, CH3 to C12H25 OR (XII) Derivatives of 2-hydroxybenzophenone have highly conjugated structures and a capacity to form intramolecular hydrogen bonds that exert a decisive inﬂuence on the spectroscopic and photochemical properties of these compounds. It has been shown with 2-hydroxybenzophenone (XIII) that on exposure to light (XIII) is transformed into enol (XIV), which turns back into its initial form (XIII) on losing thermal energy to the medium (Reaction 1.126): O H C O OH hν C (1.126) (–heat) (XIII) O (XIV ) 123 Characteristics of Polymers and Polymerization Processes The light energy consumed by the UV absorber corresponds to the quantity of energy needed to break the hydrogen bond. This explanation is supported by the fact that compounds that cannot lead to the formation of intramolecular hydrogen bonds (benzophenone or 2-methoxybenzophenone) do not absorb in the UV wavelength range. Hydroxyphenyl benzotriazoles have the structure (XV) where X is H or Cl (chlorine shifts the absorption to longer wavelengths), R1 is H or branched alkyl, and R2 is CH3 to C8H17 linear and branched alkyl (R1 and R2 increase the afﬁnity to polymers). Some technically important materials in this class are 2-(2′-hydroxy-5′-methyl-phenyl)-benzotriazole, 2-(2′-hydroxy-3′-5′-di-tert-butyl-phenyl)-benzotriazole, and 2-(2′-hydroxy-3′,5′-di-tert-butyl-phenyl)-5-chlorobenzotriazole. In comparison with 2-hydroxyben zophenones, the 2-(2′-hydroxyphenyl) benzotriazoles have higher molar extinction coefﬁcients and steeper absorption edges towards 400 nm. HO R1 N N N X R2 (XV) The exact mechanism of light absorption by hydrobenzotriazoles is not known. However, the formation of intramolecular hydrogen bond, as in (XVI), and of zwitter ions having a quinoid structure, as in (XVII), may be responsible for the transformation of light radiation energy into chemical modiﬁcations: H O H R1 N N N X N X O R1 + N (1.127) NR2 R2 (XVII) (XVI) It may be noted that the tendency to form chelated rings by the creation of hydrogen bonds between hydroxide and carbonyl groups [as in (XIII)] or groups containing nitrogen [as in (XVI)] is a characteristic property of all UV absorbers. A fundamental disadvantage of UV absorbers is the fact that they need a certain absorption depth (sample thickness) for good protection of a plastic. Therefore, the protection of thin section articles, such as ﬁlms and ﬁbers, with UV absorbers alone is only moderate. 1.19.1.2 Quenchers Quenchers (Q) are light stabilizers that are able to take over energy absorbed from light radiation by the chromophores (K) present in a plastic material and thus prevent polymer degradation. The energy absorbed by quenchers can be dissipated either as heat (Reaction 1.130) or as ﬂuorescent or phosphorescent radiation (Reaction 1.131): K + hn ! K* (1.128) K + Q ! K + Q* (1.129) Q* ! Q + heat (1.130) Q* ! Q + hn0 (1.131) 124 Plastics Technology Handbook For energy transfer to occur from the excited chromophore K* (donor) to the quencher Q (acceptor), the latter must have lower energy states than the donor. The transfer can take place by two processes: (1) longrange energy transfer or Föster mechanism and (2) contact, or collisional, or exchange energy transfer. The Förster mechanism is based on a dipole-dipole interaction and is usually observed in the quenching of excited states. It has been considered as a possible stabilization mechanism for typical UV absorbers with extinction coefﬁcients greater than 10,000. The distance between chromophore and quencher in this process may be as large as 5 or even 10 nm, provided there is a strong overlap between the emission spectrum of the chromophore and the absorption spectrum of the quencher. However, for an efﬁcient transfer to take place in the contact or exchange energy transfer process, the distance between quencher and chromophore must not exceed 1.5 nm. From calculations based on the assumption of random distribution of both stabilizer and sensitizer in the polymer, it is thus concluded that exchange energy transfer cannot contribute signiﬁcantly to stabilization. This would not apply, however, if some kind of association between sensitizer and stabilizer takes place (for example, through hydrogen bonding). Considering the dominant role of hydroperoxides in polyoleﬁn photooxidation [cf. Equation 1.121 and Equation 1.122], quenching of excited –OOH groups would contribute signiﬁcantly to stabilization. However, since the –OOH excited state is dissociative, i.e., its lifetime is limited to one vibration of the O–O bond, contact energy transfer during this very short time (about 10–3 s) appears highly unlikely if the –OOH group is not already associated with the quencher. The quenching action being thus independent of the thickness of the samples, quenchers are speciﬁcally useful for the stabilization of thin section articles such as ﬁlms and ﬁbers. Metallic complexes that act as excited state quenchers are used to stabilize polymers, mainly polyoleﬁns. They are nickel and cobalt compounds corresponding to the following structure: O H C O N Ni O N C H O (XVIII) Metallic complexes based on Ni, Co, and substituted phenols, thiophenols, dithiocarbamates, or phosphates are used. Typical representatives are nickel-di-butyldithiocarbamate, n-butylamin-nickel2,2′-thio-bis-(4-tert-octyl-phenolate), nickel-bis-[2,2′-thio-bis-(4-tert-octyl-phenolate)] and nickel-(Oethyl-3,5-di-tert-butyl-4-hydroxy-benzyl)-phosphonate. But their use is not as widespread as for other UV absorbers because they tend to be green. 1.19.1.3 Hydroperoxide Decomposers Since hydroperoxides play a determining role in the photooxidative degradation of polymers, decomposition of hydroperoxides into more stable compounds, before the hydroperoxides undergo photolytic cleavage, would be expected to provide an effective means of UV protection. Metal complexes of sulfurcontaining compounds such as dialkyldithiocarbamates (XIX), dialkyldithiophosphates (XX) and thiobisphenolates (XXI) are very efﬁcient hydroperoxide decomposers even if used in almost catalytic quantities. Besides reducing the hydroperoxide content of pre-oxidized polymer ﬁlms, they also can act as very efﬁcient UV stabilizers. This explains the fact that an improvement in UV stability is often observed on combining UV absorbers with phosphite or nickel compounds. 1.19.1.4 Free-Radical Scavengers Besides the absorption of harmful radiation by UV absorbers, the deactivation of excited states by quenchers, and the decomposition of hydroperoxides by some phosphorus and/or sulfur containing 125 Characteristics of Polymers and Polymerization Processes compounds, the scavenging of free-radical intermediates is another possibility of photostabilization, analogous to that used for stability against thermooxidative degradation. It has been shown that compounds (XXI), (XXII), (XXIII), and (XXIV) are effective radical scavengers. The radicals generated by hemolytic cleavage of hydroperoxide (Equation 1.132): S S M 2+ (C 4H 9) 2 NC CN(C 4H 9 ) 2 M = Zn, Ni S S (XIX) (i–C 3H 7) 2 N S Ni P (i–C 3H 7) 2 N N(i–C 3 H 7) 2 S S P S N(i–C 3 H 7) 2 (XX) NH 2C 4H 9 Ni O O S R = tert. octyl R R (XXI) PPOOH ! PPO• +• OH (1.132) HO• + InH ! ½InH…• OH (1.133) PPO• + InH ! PPOH + In• (1.134) may be removed by reactions, such as with a radical scavenger InH. The latest development in the ﬁeld of light stabilizers for plastics is represented by sterically hindered amine-type light stabilizers (HALS). A typical such compound is bis-(2,2,6,6-tetramethyl-4-piperidyl)sebacate (XXV). Since it does not absorb any light above 250 nm, it cannot be considered a UV absorber or a quencher of excited states. This has been conﬁrmed in polypropylene through luminescence measurements. 126 Plastics Technology Handbook C4H9O O CH2 P O OC4H9 P CH2 Ni O O HO ( OH = tert butyl) (XXII) O OH O C C O OC12H25 HO (XXIII) (XXIV) O O O C (CH2)8 NH C (XXV) O NH A low-molecular weight HALS such as (XXV), denoted henceforth as HALS-I, has the disadvantage of relative volatility and limited migration and extraction resistance, which are undesirable in special plastics applications (for example, in ﬁne ﬁbers and tapes). For such applications, it is advantageous to use polymeric sterically hindered amines such as poly-(N-b-hydroxyethyl-2,2,6,6-tetramethyl-4hydroxypiperidyl succinate) represented by (XXVI) and a more complex polymeric hindered amine represented by (XXVII). In later discussions, they will be designated as HALS-II and HALS-III, respectively. Though they do not reach completely the performance of the low-molecular weight HALS-I, they are nevertheless superior to the other common light stabilizers used at several-fold higher concentrations. N O CH 2 CH 2 OCCH 2 CH 2 C O O n (XXVI) N ( CH2 )6 N N N N NH NH n NH tert octyl (XXVII) The protection mechanisms of HALS, known so far mostly from studies with model systems, can be summarized as follows: From ESR measurements it is concluded that, under photooxidative conditions, 127 Characteristics of Polymers and Polymerization Processes HALS are converted, at least in part, to the corresponding nitroxyl radicals (XXVIII). The latter, through Reaction 1.135, are thought to be the true radical trapping species. R R + R N N O (XXVIII) OR (XXIX) (1.135) Another explanation of the UV protection mechanism of HALS involves the hydroxylamine ethers (XXIX) formed in Reaction 1.135. There is indirect evidence that (XXIX) can react very quickly with peroxy radicals, thereby regenerating nitroxyl radicals (Reaction 1.136). Reaction 1.135 and Reaction 1.136, which constitute the “Denisov cycle,” result in an overall slowdown of the usual chain oxidation Reaction 1.119 and Reaction 1.120. R R + + RO 2R RO 2 N N OR (XXIX) O (XXVIII) (1.136) Nitroxyl radicals may also react with polymer radical to form hydroxylamine (Reaction 1.137), and the latter can react with peroxy radicals and hydroperoxides according to Reaction 1.138 and Reaction 1.139: NO +R NOH + RO 2 NOH +ROOH NOH + C = C R (1.137) NO + RO 2H (1.138) NO + H 2O + RO (1.139) The formation of associations between HALS and hydroperoxides followed by reaction of these associations with peroxy radicals (Reaction 1.140 and Reaction 1.141) represents another possibility of retarding photooxidation. NH + HOOR [ NH... HOOR]+ RO 2 [ NH... HOOR] ROOR+ H 2O + NO (1.140) (1.141) 128 Plastics Technology Handbook The hydroperoxides associated with HALS may undergo photolysis producing hydroxy and alkoxy radicals in close proximity to the amines. The radicals may then abstract a hydrogen atom from the amine and form hydroxylamine and hydroxylamine ethers (Reaction 1.142): OH NH ... HOOR NOH + ROH (1.142) N H OR NOR + H 2O Still other mechanisms have been postulated, e.g., the interaction of HALS with a,b,-unsaturated carbonyl compounds and the formation of charge-transfer complex between HALS and peroxy radicals. However, despite extensive publications in the ﬁeld, a complete knowledge of the process occurring in polymer photooxidation in the presence of HALS is still not available. 1.20 Light Stabilizers for Selected Plastics In choosing a light stabilizer for a given plastic, several factors, in addition to its protective power, play an important role. In this respect, one may cite physical form (liquid/solid, melting point), thermal stability, possible interaction with other additives and ﬁllers that may eventually lead to discoloration of the substrate, volatility, toxicity (food packaging), and above all, compatibility with the plastics material considered. An additive can be considered as compatible if, during a long period of time, no blooming or turbidity is observed at room temperature and at the elevated temperatures that may be encountered during projected use of the plastic material. Among the light stabilizer classes available commercially, only a few may be used in a broad range of plastics. Thus, nickel compounds are used almost exclusively in polyoleﬁns, whereas poly(vinyl chloride) is stabilized with UV absorbers only. 1.20.1 Polypropylene For the light stabilization of polypropylene, representatives of the following stabilizer classes are mainly used: 2-(2′-hydroxyphenyl)-benzotriazoles, 2-hydroxy-4-alkoxybenzophenones, nickel-containing light stabilizers, 3,5-di-tert-butyl-4-hydroxybenzoates, as well as sterically hindered amines (HALS). Nickelcontaining light stabilizers are used exclusively in thin sections, such as ﬁlms and tapes, whereas all other classes may be used in thin as well as thick sections, though UV absorbers have only limited effectiveness in thin section. Nickel-containing additives are also used as “dyesites” because they allow the dying and printing of polypropylene ﬁbers with dyestuffs susceptible to complexation with metals. 1.20.2 Polyethylene As all polyoleﬁns, polyethylene is sensitive to UV radiation, although less than polypropylene. For outdoor use polyethylene needs special stabilization against UV light. The light stabilizers for polyethylene are in principle the same as for polypropylene. On accelerated weathering, HALS show much better performance in HDPE tapes than UV absorbers, despite the latter being used in much higher concentrations. The comparison between HALS is, however, in favor of the polymeric HALS-III, which has the same performance when added at a concentration of 0.05% as HALS-I and HALS-II at 0.1%. Among the numerous commercial light stabilizers only a few are suitable for low density polyethylene (LDPE). This is mainly due to the fact that most light stabilizers are not sufﬁciently compatible at levels of Characteristics of Polymers and Polymerization Processes 129 concentration necessary for the required protection and so they bloom more or less rapidly. Initially, UV absorbers of the benzophenone and benzotriazole types were used to protect LDPE materials. With the development of nickel-quenchers, signiﬁcant improvement in the light stability of LDPE ﬁlms has been achieved. For reasons of economy, however, combinations of nickel-quenchers with UV absorbers are mainly used. The service life of LDPE ﬁlms may be increased by raising the concentrations of light stabilizers. Additive contents close to 2% may be found in greenhouse ﬁlms thought to last up to 3 years outdoors. A further improvement in UV stability of LDPE was expected with the development of HALS. However, LDPE compatibility of HALS available in the early years was insufﬁcient, resulting in relatively poor performance on outdoor weathering. It was only with the development of polymeric HALS that these difﬁculties were overcome. Tests have shown that HALS-II is signiﬁcantly superior to UV absorbers and Ni-stabilizers so that the same performance can be achieved with much smaller concentrations. However, use of combinations of the polymeric HALS-II with a UV absorber leads to a signiﬁcant improvement of the efﬁciency in comparison with the HALS used alone at the same concentration as the combination. A further boost of the performance can be achieved through use of the polymeric HALS-III. For example, the performance of HALS-II can be reached by using HALS-III at about half the concentration. The superiority of HALSIII becomes even more pronounced in ﬁlms of thickness below 200 mm. In linear low density polyethylene (LLDPE) also, the polymeric HALS-II and HALS-III show much better performance than other commercial light stabilizers. Blooming is observed with low molecular weight HALS-I, similar to that found with LDPE. In more polar substances such as ethylene-vinyl acetate copolymers (EVA), the low molecular weight HALS-I can be used. However, in this substrate too, the polymeric HALS-II and HALS-III are signiﬁcantly superior to the low molecular weight HALS. 1.20.3 Styrenic Polymers Double bonds are not regarded as the chromophores responsible for initiation of photooxidation in polystyrenes because they absorb below 300 nm. However, peroxide groups in the polymer chain, resulting from co-polymerization of oxygen with styrene, are deﬁnitely photolabile. Moreover, oxidation products such as aromatic ketones of acetophenone type, which have been detected by emission spectroscopy, are formed during processing of styrene polymers at high temperatures. Aromatic ketones (AK) in the triplet state are able to abstract hydrogen from polystyrene (PSH) (Reaction 1.143): AK ! AK* AK* + PSH ! AKH• + PS• (1.143) This reaction is considered the most important initiation mechanism for styrene photooxidation in the presence of aromatic ketones. Styrenic plastics such as acrylonitrile/butadiene/styrene graft copolymers (ABS) and impact-resistant polystyrenes are very sensitive towards oxidation, mainly because of their butadiene content. Degradation on weathering starts at the surface and results in rapid loss of mechanical properties such as impact strength. Because of the lack of efﬁcient light stabilizers, ABS has not been used outdoors on a large scale. However, by combining two light stabilizers with different protection mechanisms, e.g., a UV absorber of the benzotriazole class and the sterically hindered amine HALS-I, it is possible to achieve good stabilization even in ABS. This is a case of synergism in which the UV absorber protects the deeper layers, while HALS-I assures surface protection. At the same time discoloration of the ABS polymer is also reduced signiﬁcantly. The same holds for polystyrene and styrene–acrylonitrile copolymers (SAN), and the best 130 Plastics Technology Handbook protection is obtained with HALS/UV absorber combinations. Light stabilization is necessary for articles of these polymers for which UV exposure can be expected (e.g., covers for ﬂuorescent lights). 1.20.4 Poly(Vinyl Chloride) “Pure” poly(vinyl chloride) (PVC) does not absorb any light above 220 nm. Different functional groups and structural irregularities that may arise during polymerization and processing have thus been considered as possible initiating chromophores. They include irregularities in the polymer chain as well as hydroperoxides, carbonyl groups, and double bonds. Thermal stabilizers used in PVC also confer some degree of light stability. Ba/Cd salts and organic tin carboxylates, for example, confer already some UV stability to PVC on outdoor exposure. However, for transparent and translucent PVC articles requiring high UV stability, the light stability conferred by thermal stabilizers is not sufﬁcient. The addition of light stabilizers in such cases is therefore mandatory. So far, UV absorbers yield the best results in practical use. HALS-I has almost no effect. 1.20.5 Polycarbonate Bisphenol-A polycarbonate (PC) absorbs UV light below 360 nm but its absorption is intense only below 300 nm. Insufﬁcient light stability of PC on outdoor use is manifested by yellowing, which increases rapidly. Studies indicate that on absorption of UV light, PC undergoes photo-Fries rearrangement, which gives ﬁrst a phenyl salicylate, and after absorption of a second photon and subsequent rearrangement, it gives 2,2′-dihydroxybenzophenone groups (Figure 1.76). The absorption of these groups reaches into the visible region, and PC yellowing has been essentially attributed to them. In addition to the Fries reaction, the formation of O2-charge transfer complexes in PC, similar to those found in polyoleﬁns and leading to the formation of hydroperoxides, is considered to contribute signiﬁcantly to photooxidation in the early stages. Among the stabilizer classes, only UV absorbers are in use for stabilization of PC. In choosing a UV absorber, intrinsic performance, volatility, adequate thermal stability at the elevated processing temperatures (about 320°C), and effect on initial color of the PC articles should be considered. Benzotriazole, oxanilide, and cinnamate-type UV absorbers are effective photostabilizers for PC with benzotriazoles giving the best performance among the three types. O C O O OH hν C O O HO OH hν C O FIGURE 1.76 Photo-Fries rearrangement on polycarbonate. Characteristics of Polymers and Polymerization Processes 131 1.20.6 Polyacrylates Poly(methyl methacrylate) (PMMA) is highly transparent in the UV region, and thus much more light stable than other thermoplastics. UV absorbers may therefore be used to confer a UV ﬁlter effect to PMMA articles. PMMA window panes for solar protection containing 0.05%–0.2% 2-(2′-hydroxy-5′methylphenyl)-benzotriazole are well-known examples of this application. The rear lights of motor cars, electric signs, and covers for ﬂuorescent lights are some applications for which PMMA is UV-stabilized. The excellent light-stabilizing performance of HALS is also found with PMMA. 1.20.7 Polyacetal Polyacetal is markedly unstable towards lights because even UV radiation of wavelengths as high as 365 nm may initiate its degradation. Polyacetal cannot therefore be used outdoors if it does not contain any light stabilizers. Even after a short weathering, surface crazes and pronounced chalking are observed. Carbon black (0.5%–3%) is a good stabilizer for polyacetal when sample color is not important. Other possibilities for stabilization are the use of 2-hydroxybenzophenone and, especially, hydroxyphenylbenzotriazole-type UV absorbers. Stabilization with HALS/UV absorber is superior to that with UV absorber alone. 1.20.8 Polyurethanes Lights stability of polyurethanes depends to a large extent on their chemical structure, and both components (i.e., isocyanate and polyol) have an inﬂuence. Polyurethanes based on aliphatic isocyanates and polyester diols show the best light stability if yellowing is considered, whereas polyurethanes based on aromatic isocyanates and polyether diols are worst in this respect. Light stabilizers are used mainly in the coatings industry (textile coatings, synthetic leather). In addition to some UV absorbers of the 2-(2′-hydroxyphenyl)-benzotriazole type, HALS used alone or in combination with benzotriazoles are especially effective stabilizers. 1.20.9 Polyamides The absorption of aliphatic polyamides in the short wavelength region of sunlight is attributed largely to the presence of impurities. Direct chain scission at wavelengths below 300 nm and photosensitized oxidation above 300 nm have been considered a long ago as responsible for photooxidation. Antioxidants used in polyamides often confer good light stability as well. However, enhanced performance is obtained with the combination of an antioxidant with a light stabilizer. The sterically hindered amines are signiﬁcantly superior. For example, molded polyamide samples stabilized with HALS (0.5%) are found to exhibit approximately twice the light stability of the samples stabilized with a phenolic antioxidant (0.5%) or combination of the latter with a UV absorber (0.5%). 1.21 Diffusion and Permeability There are many instances where diffusion and permeation of a gas, vapor, or liquid through a plastics material is of considerable importance in the processing and usage of the material. For example, dissolution of a polymer in a solvent occurs through diffusion of the solvent molecules into the polymer, which thus swells and eventually disintegrates. In a plastisol process the gelation of PVC paste, which is a suspension of PVC particles in a plasticizer such as tritolyl phosphate, involves diffusion of the plasticizer into the polymer mass, resulting in a rise of the paste viscosity. Diffusion processes are involved in the production of cellulose acetate ﬁlm by casting from solution, as casting requires removal of the solvent. Diffusion also plays a part in plastic molding. For 132 Plastics Technology Handbook example, lubricants in plastics compositions are required to diffuse out of the compound during processing to provide lubrication at the interface of the compound and the mold. Incompatibility with the base polymer is therefore an important criterion in the choice of lubricants in such cases. Permeability of gases and vapors through a ﬁlm is an important consideration in many applications of polymers. A high permeability is sometimes desirable. For example, in fruit-packaging applications of plastics ﬁlm it is desirable to have high permeability of carbon dioxide. On the other hand, for making inner tube and tubeless tires, or in a child’s balloon, the polymer used must have low air permeability. 1.21.1 Diffusion Diffusion occurs as a result of natural processes that tend to equalize the concentration gradient of a given species in a given environment. A quantitative relation between the concentration gradient and the amount of one material transported through another is given by Fick’s ﬁrst low: dm = −D dc A dt dx (1.144) where dm is the number of grams of the diffusing material crossing area A (cm2)of the other material in time dt (s). D is the diffusion coefﬁcient (cm2/s) whose value depends on the diffusing species and the material in which diffusion occurs, and dc/dx is the concentration gradient, where the units of x of and c are centimeters and grams per cubic centimeter. Diffusion in polymers occurs by the molecules of the diffusing species passing through voids and other gaps between the polymer molecules. The diffusion rate will therefore heavily depends on the molecular size of the diffusing species and on the size of the gaps. Thus, if two solvents have similar solubility parameters, the one with smaller molecules with diffuse faster in a polymer. On the other hand, the size of the gaps in the polymer depend to a large extent on the physical state of the polymer—that is, whether it is crystalline, rubbery, or glassy. Crystalline structures have an ordered arrangement of molecules and a high degree of molecular packing. The crystalline regions in a polymer can thus be considered as almost impermeable, and diffusion can occur only in amorphous regions or through region of imperfection: hence, the more crystalline the polymer, the greater will be its tendency to resist diffusion. Amorphous polymers, as noted earlier, exist in the rubbery state above the glass transition temperature and in the glassy state below this temperature. In the rubbery state there is an appreciable “free volume” in the polymer mass, and molecular segments also have considerable mobility, which makes it highly probable that a molecular segment will at some stage move out of the way of the diffusing molecule, the contributing to a faster diffusion rate. In the glassy state, however, the molecular segments cease to have mobility, and there is also a reduction in free volume or voids in the polymer mass, both of which lower the rate of diffusion. Thus, diffusion rates will be highest in rubbery polmers and lowest in crystalline polymers. 1.21.2 Permeability Permeation of gas, vapor, or liquid through a polymer ﬁlm consists of three steps: (1) a solution of permeating molecules in the polymer, (2) diffusion through the polymer due to concentration gradient, and (3) emergence of permeating molecules at the outer surface. Permeability is therefore the product of solubility and diffusion; so where the solubility obeys Henry’s law one may write P = DS where P is the permeability, D is the diffusion coefﬁcient, and S is the solubility coefﬁcient [64]. (1.145) Characteristics of Polymers and Polymerization Processes 133 Hence, factors which contribute to greater solubility and higher diffusivity will also contribute to greater permeability. Thus, a hydrocarbon polymer like polyethylene should be more permeable to hydrocarbons than to liquids of similar molecular size but of different solubility parameter, and smaller hydrocarbons should have higher permeability. The permeabilities of a number of polymers to the common atmospheric gases [65,66], including water vapor, are given in Table 1.17. It appears that regardless of the ﬁlm material involved, oxygen permeates about four times a fast as nitrogen, and carbon dioxide about 25 times as fast. The fact that the ratios of the permeabilities for all gases, apart from water vapor, are remarkably constant, provided there is no interaction between the ﬁlm material and the diffusing gas, leads one to express the permeability as the product of three factors [65]: one determined by the nature of the polymer ﬁlm, one determined by the nature of the gas, and one accounting for the interaction between the gas and the ﬁlm; i.e., Pi,k = Fi Gk Hi,k (1.146) where Pi,k is the permeability for the system polymer i and gas k; Fi, Gk and Hi,k are the factors associated with the ﬁlm, gas, and interaction respectively. When Hi,k≈1, there is little or no interaction, and as the degree of interaction increases, Hi,k becomes larger. With little or no interaction, Equation 1.146 becomes Pi,k = Fi Gk (1.147) It then appears that the ratio of the permeability of two gases (k,l) in the same polymer (i) will be the same as the ratio between the two G factors Pi,k Gk = Pi,l Gl (1.148) If the G value of one of the gases, usually nitrogen, is taken as unity, G values of other gases and F values of different polymers can be calculated from Equation 1.148 and Equation 1.147. These values are reliable for gases but not for water vapor. Some F and G values for polymers are given in Table 1.18. Evidently, the F values correspond to the ﬁrst column of Table 1.17, since G for N2 is 1. The G values for O2 and CO2 represent the averages of PO2/PN2 and PCO2/PN2 in columns 5 and 6, respectively. 1.22 Polymer Compounding In many commercial plastics, the polymer is only one of several constituents, and it is not necessarily the most important one. Such systems are made by polymer compounding—a term used for the mixing of polymer with other ingredients. These other ingredients or supplementary agents are collectively referred to as additives. They may include chemicals to act as plasticizing agents, various types of ﬁlling agents, stabilizing agents, antistatic agents, colorants, ﬂame retardants, and other ingredients added to impart certain speciﬁc properties to the ﬁnal product [61,67,68]. The properties of compounded plastics may often be vastly different from those of the base polymers used in them. A typical example is SBR, which is the largest-volume synthetic rubber today. The styrenebutadiene copolymer is a material that does not extrude smoothly, degrades rapidly on exposure to warm air, and has a tensile strength of only about 500 psi (3.4 × 105 N/m2). However, proper compounding changes this polymer to a smooth-processing, heat-stable rubber with a tensile strength of over 3,000 psi 20 × 10 N/m2). Since all properties cannot be optimized at once, compounders have developed thousands of specialized recipes to optimize one or more of the desirable properties for particular applications (tires, fan belts, girdles, tarpaulins, electrical insulation, etc.). In SBR the compounding ingredients can be (1) reinforcing ﬁllers, such as carbon black and silica, which improve tensile strength or tear strength; (2) inert ﬁllers and pigments, such as clay, talc, and 11 23 13.0 2.9 – 3.12 64.5 80.8 Polystyrene Polypropylene (d = 0.910) Butyl rubber Polybutadiene Natural rubber 7.8 1,310 1,380 92 51.8 88 35 352 68 1.6 10 1.7 0.72 1.53 0.29 CO2 (30°C) – 4.1 3.0 2.9 – – 3.8 3.9 2.9 2.8 3.8 3.0 3.8 3.3 4.4 5.6 PO2 =PN2 680 – 12,000 130 800 75,000 7,000 1,560 240 2.9 1,300 14 H2O (25°C, 90% RH) Source: From Stannett, V. T. and Szwarc, M. 1955. J. Polym. Sci., 16, 89 and Paine, F. A. 1962. J. Roy. Inst. Chem., 86, 263. 233 191 10.6 55 2.8 2.7 19 Polyethylene (d = 0.954, 0.960) Polyethylene (d = 0.922) 0.38 1.20 0.30 0.10 0.22 0.053 O2 (30°C) Cellulose acetate 0.08 0.10 0.40 Polyamide (Nylon 6) Poly(vinyl chloride) (unplasticized) 0.03 0.05 Rubber hydrochloride (Plioﬁlm ND) 0.0094 Polychlorotriﬂuoro-ethylene Poly(ethylene terephthalate) (Mylar A) N2 (30°C) Poly(vinylidene chloride) (Saran) Polymer Permeability (P × 1010 cm3/cm2/mm/s/cm Hg) TABLE 1.17 Permeability Data for Various Polymers 16.2 21.4 – 16.2 30 13 19 24 16 25 21 24 31 31 PCO2 =PN2 – – – – 4,100 48 42 2,680 70,000 3,900 3,000 97 26,000 1,400 PH2 O =PN2 Rubbery Rubbery Crystalline Rubbery Glassy Crystalline Semicrystalline Glassy Crystalline Semicrystalline Crystalline Crystalline Crystalline Crystalline Nature of Polymer Ratios (to N2 Permeability as 1.0) 134 Plastics Technology Handbook 135 Characteristics of Polymers and Polymerization Processes TABLE 1.18 F and G Constants for Polymers and Gases Polymer F Poly(vinylidene chloride) (Saran) 0.0094 Poly(chlorotriﬂuoroethylene) 0.03 Poly(ethylene terephthalate) Rubber hydrochloride (Plioﬁlm) Gas G N2 1.0 0.05 0.08 O2 H2S 3.8 21.9 Nylon 6 0.1 CO2 24.2 Cellulose acetate (+15% plasticizer) Polyethylene (d = 0.922) 5 19 Ethyl cellulose (plasticized) 84 Natural rubber Butyl rubber 80.8 3.12 Nitrile rubber 2.35 Polychloroprene Polybutadiene 11.8 64.5 Source: From Stannett, V. T. and Szwarc, M. 1955. J. Polym. Sci., 16, 89. calcium carbonate, which make the polymer easier to mold or extrude and also lower the cost; (3) plasticizers and extenders, such as mineral oils, fatty acids, and esters; (4) antioxidants, basically amines or phenols, which stop the chain propagation in oxidation; and (5) curatives, such as sulfur for unsaturated polymers and peroxides for saturated polymers, which are essential to form the network of cross-links that ensure elasticity rather than ﬂow. Polymer applications which generally involve extensive compounding are rubbers, thermosets, adhesives, and coatings. Fibers and thermoplastic polymers (with the exception of PVC) are generally not compounded to any signiﬁcant extend. Fibers, however, involve complex after-treatment processes leading to the ﬁnal product. PVC, which by itself is a rigid solid, owes much of its versatility in applications to compounding with plasticizers. The plasticize content varies widely with the end use of the product but is typically about 30% by weight. Of the compounding ingredients, ﬁllers and plasticizers are more important in terms of quantities used. Other additives used in smaller quantities are antioxidants, stabilizers, colorants, ﬂame retardants, etc. The ingredients used as antioxidants and light stabilizers, and their effect have been discussed previously. Fillers, plasticizers and ﬂame retardants are described next. 1.22.1 Fillers Fillers play a crucial role in the manufacture of plastics. Alone many plastics are virtually useless, but they are converted into highly useful products by combining them with ﬁllers. For example, phenolic and amine resins are almost always used in combination with substances like wood ﬂour, pure cellulose, powdered mica, and asbestos. Glass ﬁber is used as a ﬁller for ﬁber-reinforced composites with epoxy or polyester resins. Another extremely important example is the use of carbon ﬁller for rubber. Rubber would be of little value in modern industry were it not for the fact that the ﬁller carbon greatly enhances its mechanical properties lie tensile strength, stiffness, tear resistance, and abrasion resistance. Enhancement of these properties is called reinforcement, and the ﬁllers which produce the strengthening effect are known as reinforcing ﬁllers. Other ﬁllers may not appreciably increase strength, but they may improve other properties of the polymer, thus, making it easier to mold, which reduces cost. Fillers used in plastics can be divided into two types: particulate and ﬁbrous. Typical ﬁllers in these two categories and the improvements they bring about are summarized in Table 1.19. In some instances they are added to perform one or more prime functions or to provide special properties. Asbestos, for example, 136 Plastics Technology Handbook + Calcium carbonate + + + + Calcium silicate Carbon black + + + + + + + + S S/P + + + + + + Graphite + Jute Kaolin + + + + + + + + + + + + + + + + + + + + S S/P + Carbon ﬁber Cellulose Cotton (macerated/ chopped ﬁbers) Fibrous glass Recommend forUseina + Processability + Moisture Resistance + Thermal Conductivity Aluminum powder Asbestos Electrical Conductivity + Electrical Insulation Tensile Strength + Hardness Dimensional Stability + + Aipha cellulose Impact Strength Heat Resistance + Alumina Fillers Stiffness Chemical Resistance Properties Improved Lubricity TABLE 1.19 Some Fillers and Their Effects on Plastics + + + + + + + S/P + S S/P S S/P S + + + + + + + + + S/P + S/P + + + S S/P Kaolin (calcined) + + + + + + + + S/P Mica Molybdenum disulﬁde Nylon (macerated/ chopped ﬁbers) Acrylic ﬁber (Orion) + + + + + + + + + + + + + S/P P + + + S/P + + S/P + + + S/P S/P + + + S/P + + + + + + + + + + + + + + + + + + + + + Rayon Silica, amorphous TFE-ﬂuorocarbon Talc Wood ﬂour + + + + + + + + + + + + S + S Note: P = in thermoplastics only; S = in thermosets only; S/P = in both thermoplastics and thermosets. provides high-temperature resistance and improves dimensional stability. Mica improves the electrical and thermal properties of all compounds. Glass ﬁbers produce high strength. Carbon black is the only important reinforcing ﬁller for most elastomers. It also imparts UV resistance and coloring. Beryllium oxide-ﬁlled resins gain high conductivity without loss of electrical properties. Metal particles have been used as ﬁllers for plastics to improve or impart certain properties. Thus, aluminum has been used for applications ranging from making a decorative ﬁnish to improving thermal conductivity. Copper particles are used in plastics to provide electrical conductivity. Lead is used because it dampens vibrations, acts as barrier to gamma-radiation, and has high density. Of the new space-age products used as reinforcing ﬁllers, carbon ﬁbers and boron ﬁbers have the greatest potential for use in high-strength advanced composites. Carbon ﬁbers made by pyrolizing organic ﬁbers such as rayon and polyacrylonitrile have tensile strengths approaching 3.3 × 105 psi (2.3 (109 N/m2). Boron ﬁbers made by depositing boron from a BCl3–H2 mixture onto tungsten wire have tensile strengths approaching 5 × 105 psi (3.5 × 109 N/m2). The speciﬁc strengths and speciﬁc moduli of polymer composites made with these ﬁbers are far above those attainable in monolithic structural materials such s highstrength aluminum, steel, and titanium. These composites can thus lead to signiﬁcant weight savings in Characteristics of Polymers and Polymerization Processes 137 actual applications. A relatively recent addition to the high-performance ﬁber ﬁeld is the organic polymeric ﬁber Kevlar-49, developed by DuPont. It has a higher speciﬁc strength than glass, boron, or carbon. Furthermore, Kevlar, a polyamide, is cheap, having one-sixth the price of acrylic-based graphite ﬁbers. The extensive range of ﬁllers and reinforcing agents used nowadays indicates the importance that these materials have attained. The main difference between inert and reinforcing ﬁllers lies in the fact that modulus of elasticity and stiffness are increased to a greater or les extent by all ﬁllers, including the spherical types such as chalk or glass spheres, whereas tensile strength can be appreciably improved only by a ﬁber reinforcement. The heat deﬂection temperature, i.e. stiffness at elevated temperatures, cannot be increased by spherical additives to the same extent as by ﬁber reinforcement. On the other hand, ﬁllers in ﬂake form, such as talc or mica, likewise produce a marked improvement in the heat deﬂection temperature but lead to a decrease in tensile strength and to elongation at break. The inﬂuences of common ﬁllers on the properties of polyoleﬁns are compared in Table 1.20. 1.23 Plasticizers Plasticizers are organic substances of low volatility that are added to plastics compounds to improve their ﬂexibility, extensibility, and processability. They increase ﬂow and thermoplasticity of plastic materials by decreasing the viscosity of polymer melts, the glass transition temperature (Tg) the melting temperature (Tm), and the elasticity modulus of ﬁnished products [69]. Plasticizers are particularly used for polymers that are in a glassy state at room temperature. These rigid polymers become ﬂexible by strong interactions between plasticizer molecules and chain units, which lower their brittle-tough transition or brittleness temperature (Tb) (the temperature at which a sample breaks when struck) and their Tg value, and extend the temperature range for their rubbery or viscoelastic state behavior (see Figure 1.52). Mutual miscibility between plasticizers and polymers is an important criterion from a practical point of view. If a polymer is soluble in a plasticizer at a high concentration of the polymer, the plasticizer is said to be a primary plasticizer. Primary plasticizers should gel the polymer rapidly in the normal processing temperature range and should not exude from the plasticized material. Secondary plasticizers, on the other hand, have lower gelation capacity and limited compatibility with the polymer. In this case, two phases are present after plasticization process—one phase where the polymer is only slightly plasticized, and one phase where it is completely plasticized. Polymers plasticized with secondary plasticizers do not, therefore, deform homogeneously when stressed as compared to primary plasticizers. The deformation appears only in the plasticizer-rich phase and the mechanical properties of the system are poor. Unlike primary plasticizers, secondary plasticizers cannot be used alone and are usually employed in combination with a primary plasticizer. Plasticizer properties are determined by their chemical structure because they are affected by the polarity and ﬂexibility of molecules. The polarity and ﬂexibility of plasticizer molecules determine their interaction with polymer segments. Plasticizers used in practice contain polar and nonpolar groups, and their ratio determines the miscibility of a plasticizer with a given polymer. Plasticizers for PVC can be divided into two main groups [67] according to their nonpolar part. The ﬁrst group consists of plasticizers having polar groups attached to aromatic rings and is termed the polar aromatic group. Plasticizers such as phthalic acid esters and tricresyl phosphate belong to this group. An important characteristic of these substances is the presence of the polarizable aromatic ring. It has been suggested that they behave like dipolar molecules and form a link between chlorine atoms belonging to two polymer chains or to two segments of the same chain, as shown in Figure 1.77a. Plasticizers belonging to this group are introduced easily into the polymer matrix. They are characterized by ability to produce gelation rapidly and have a temperature of polymerplasticizer miscibility that is low enough for practical use. These plasticizers are therefore called solvent-type plasticizers, and their kerosene extraction (bleeding) index is very low. They are, however, not recommended for cold-resistant materials. 30 30 Talc Mica Glass spheres (<50 mm) Quartz powder 6 11 10 – 0.90 1.20 40 – 30 40 40 30 40 40 Asbestos Polypropylene Short glass ﬁber Asbestos Talc Mica Chalk Wood ﬂour 180 8 – 7 11 25 20 30 45 30 31 49 30 28 26 20 27 60 12 10 16 13 20 16 24 10 Tensile Strength (N/mm2) 3,400 2,200 6,900 5,000 3,100 1,600 7,500 2,100 2,100 2,000 1,900 1,400 7,800 400 290 600 440 670 900 1,200 210 Modulus of Elasticity (N/mm2) – 63 – – 84 83 75 100 – – – 48 46 68 23 73 – – 96 95 62 120 77 – – 2.5 – – 4.0 3.0 1.0 6.0 0.7 – 2.5 18 2.5 – 19 – 2.6 – – 22 6 1.1 1.2 – 30 27 50 108 2.3 – – 22 0.2 1.6 35 1.7 Melt Index 190/5 (g/10 min) – 33 Heat Deﬂection Temperature (1.85 N/mm2) (°C) 16 Ball Indentation (N/mm2) Source: From Gächter, R. and Müller, H., eds. 1987. Plastics Additives Handbook. Hanser Publishers, Munich/New York. 1.23 – – 1.28 1.21 620 5 9 – 1.24 30 40 Wood ﬂour Kaolin 1.17 30 Chalk >550 2 77 73 40 46 20 220 65 500 Elongation (%) – 30 0.95 1.16 1.14 1.11 1.14 1.16 1.13 1.26 1.11 0.92 Density (g/cm3) Polyethylene Glass ﬁber High density 30 30 30 Asbestos – 40 30 Chalk Glass ﬁber Filler Content (% by wt) Polyethylene Low density Material TABLE 1.20 Inﬂuence of Fillers on the Properties of Polyoleﬁns 138 Plastics Technology Handbook 139 Characteristics of Polymers and Polymerization Processes PVC chains O Cl RO OR Cl Cl C Cl O C C O OR O RO O C C Cl OR Cl Cl Cl PVC chains Screened Cl atom Cl O (a) FIGURE 1.77 RO C Cl Cl (b) Action of (a) a polar aromatic plasticizer and (b) a polar aliphatic plasticizer on PVC chains. The second group consists of plasticizers having polar groups attached to aliphatic chains and is called the polar aliphatic group. Examples are aliphatic alcohols and acid or alkyl esters of phosphoric acid (such as trioctyl phosphate). Their polar groups interact with polar sites on polymer molecules, but since their aliphatic part is rather bulky and ﬂexible other polar sites on the polymer chain may be screened by plasticizer molecules. This reduces the extent of intermolecular interactions between neighboring polymer chains, as shown in Figure 1.77b. Polar aliphatic plasticizers mix less well with polymers than do polar aromatics and, consequently, may exude (bloom) from the plasticized polymer more easily. Their polymer miscibility temperature is higher than that for the ﬁrst group. These plasticizers are called oil-type plasticizers, and their kerosene extraction index is high. Their plasticization action is, however, more pronounced than that of polar aromatic plasticizers at the same molar concentration. Moreover, since the aliphatic portions of the molecules retain their ﬂexibility over a large temperature range, these plasticizers give a better elasticity to ﬁnished products at low temperature, as compared to polar aromatic plasticizers, and allow the production of better cold-resistant materials. In PVC they also cause less coloration under heat exposure. In practice, plasticizers usually belong to an intermediate group. Mixtures of solvents belonging to the two groups discussed above are used as plasticizers to meet the requirements for applications of the plasticized material. Plasticizers can also be divided into groups according to their chemical structure to highlight their special characteristics. Several important plasticizers in each group (with their standard abbreviations) are cited below. 1.23.1 Phthalic Acid Esters Di(2-ethyl hexyl) phthalate (DOP) and diisooctyl phthalate (DIOP) are largely used for PVC and copolymers of vinyl chloride and vinyl acetate as they have an afﬁnity to these polymers, produce good solvation, and maintain good ﬂexibility of ﬁnished products at low temperature. The use of n-octyl-ndecyl phthalate in the production of plastics materials also allows good ﬂexibility and ductility at low temperature. Diisodecyl phthalate (DDP), octyl decyl phthalate (ODP), and dicapryl phthalate (DCP) have a lower solvency and are therefore used in stable PVC pastes. Butyl octyl phthalate (BOP), butyl decyl phthalate (BDP, and butyl benzyl phthalate (BBP) have a good solvency and are used to adjust melt 140 Plastics Technology Handbook viscosity and fusion time in the production of high-quality foams. They are highly valued for use in expandable plasticized PVC. Dibutyl phthalate (DBP) is not convenient for PVC plasticization because of its relatively high volatility. It is a good gelling agent for PVC and vinyl chloride-vinyl acetate copolymer (PVCA) and so is sometimes used as a secondary plasticizer in plasticizer mixers to improve solvation. DBP is mainly used for cellulose-based varnishes and for adhesives. It has a high dissolving capacity for cellulose nitrate (CN). Dimethyl phthalate (DMP) also has high dissolving capacity for CN. It has good compatibility with cellulose esters and are used in celluloid made from CN and plastic compounds or ﬁlms made from other cellulosic polymers, cellulose acetate (CA), cellulose acetate-butyrate (CAB), cellulose acetate-propionate (CAP), and cellulose propionate (CP). It is light stable but highly volatile. Diethyl phthalate (DEP) possesses properties similar to DMP and is slightly less volatile. 1.23.2 Phosphoric Acid Esters Tricresyl phosphate (TCP), trioctyl phosphate, diphenyl 2-ethylhexyl phosphate, and tri(2-ethylhexyl) phosphate (TOP) are used as plasticizers as well as ﬂame retardants. They have a low volatility, resist oil extraction well, and are usually combined with other plasticizers. TCP is a good ﬂame retardant plasticizer for technical PVC, PVAC, NC, and CAB. It is used for PVC articles especially in electrical insulation, but it is not recommended for elastic materials to be used at low temperature. Trioctyl phosphate is a better choice in low temperature applications but it offers a lower resistance to kerosene and oil extraction. Diphenyl 2-ethylhexyl phosphate is a good gelling agent for PVC. It also can be used for materials designed for low temperature application. TOP gels NC, PVCA, and PVC. It has markedly higher volatility than DOP, and it gives plastisols of low viscosity. 1.23.3 Fatty Acid Esters The esters of aliphatic dicarboxylic acids, mainly adipic, azelaic, and sebacic acid, are used as plasticizers for PVC and PVCA. Di-2-ethylhexyl adipate (DOA), benzyl butyl adipate, di-2-ethylhexyl azelate (DOZ), and di-2-ethylhexyl sebacate (DOS) are good examples. They give the polymer outstanding lowtemperature resistance and are distinguished by their high light stability. Another characteristic is their low viscosity, which is valuable in the manufacture of PVC plastisols. The solvating action of these esters on PVC at room temperature is weak. This also has a favorable effect on the initial viscosity and storage of plastisols, which is observed even when, for example, plasticizer mixtures of DOP and an aliphatic dicarboxylic ester with DOP contents of up to 80% are used. Such combinations of plasticizers help to improve gelation. DOS exceeds the low-temperature resistance of all other products in the group. It has the least sensitivity to water and has a relatively low volatility. DOS is used most often because of these properties and because of its high plasticization efﬁciency. Monoisopropyl citrate, stearyl citrate, triethyl citrate (TEC), butyl and octyl maleates, and fumarates are other important plasticizers for the preparation of stable PC pastes and of low-temperature-resistant PVC products. Triethyl citrate and triethyl acetylcitrate are among the few plasticizers to have good solvency for cellulose acetate. In comparison with diethyl phthalate, with which they are in competition in this application, there are slight but not serious differences in volatility and water sensitivity. The interest in citrate esters is due to a favorable assessment of their physiological properties. They are intended for plastic components used for packaging of food products. Butyl and octyl maleates, being unsaturated, are used for copolymerization with vinyl chloride, vinyl acetae, and styrene to provide internal plasticization. In the production of poly(vinyl butyral), which is used as an adhesive interlayer ﬁlm between glass plates for safety glass, triethyleneglycol di(2-ethyl) butyrate is an extremely valuable plasticizer that has a 141 Characteristics of Polymers and Polymerization Processes proven record of success over many years. It has the required light stability and a plasticizing effect tailored to the requirements of poly(vinyl butyral) that gives the ﬁlms suitable adhesion to glass ensuring good splinter adhesion over the temperature range from −40 to +70°C. The ethylhexanoic esters of triethylene glycol and polyethylene glycol are good plasticizers for cellulose acetate butyrate (CAB). They, however, have only limited compatibility with poly(vinyl butyral). Through the action of peracids on esters of unsaturated fatty acids, oxygen adds to the double bound forming an epoxide group. For example, epoxidized oleates have the following structure: H3C (CH2)7 CH CH (CH2)7 CO OR O Since the epoxide group reacts with acids, epoxidized fatty acid esters have become popular as plasticizer for PVC with a good stabilizing effect. Such compounds are not only able to bind hydrogen chloride according to the reaction R CH CH R + HCI R O CH CH OH Cl R But they are also able to substitute labile chlorine atoms in PVC under the catalytic inﬂuence of metal; R CH CH R + PVC Zn2+ or Cd2+ PVC O Cl Cl O R CH CH R′ 1.23.4 Polymeric Plasticizers Polymeric plasticizers can be divided into two main types. Oligomers or polymers of molecular weight ranging from 600 to 8,000 belong to the ﬁrst group. They include poly(ethylene glycol) and polyesters. High-molecular plasticizers comprise the second group. Poly(ethylene glycol) is used to plasticize proteins, casein, gelatin, and poly(vinyl alcohol). Polyester plasticizers are condensation products of dicarboxylic acid with simple alcohols corresponding to the following two general formulas: Ac – G – AcD – G – Ac Al – AcD – G – AcD – Al where Ac, monocarboxylic acid; G, glycol; AcD, dicarboxylic acid; Al, monofunctional alcohol In practice, esters from adipic, sebacic, and phthalic acid are frequently used as polyester plasticizers. The value of n may vary from 3 to 40 for adipates and from 3 to 35 for sebacates. Polyester plasticizers are seldom used alone. They are used in combination with monomeric plasticizers to reduce the volatility of the mixed solvents. They offer a higher resistance to plasticizer migration and to extract by kerosene, oils, water, and surfactants. Polyester plasticizers are used specially in PVC-based blends and in nitrocellulose varnishes. 142 Plastics Technology Handbook Ethylene-vinyl acetate (EVA) polymers (containing 65%–70% by weight of vinyl acetate) are of industrial interest as high-molecular weight plasticizers for PVC, mainly because of their low cost. A polymeric plasticizer PB-3041 available from DuPont allows the preparation of a highly permanent plasticized PVC formulation. It is believed to be a terpolymer of ethylene, vinyl, acetate, and carbon monoxide. Also, butylene terephthalate-tetrahydrofuran block copolymers, with the trade name of Hytrel (DuPont), are used as excellent permanent plasticizers of PVC. 1.23.5 Miscellaneous Plasticizers Hydrocarbons and chlorinated hydrocarbons (chloroparafﬁns) belong to the secondary plasticizer type. Aromatic and aliphatic hydrocarbons are used as extenders, particularly in the manufacture of PVC plastisols that must maintain as stable a viscosity as possible for relatively long periods of time (dip and rotational molding). The petroleum industry offers imprecisely deﬁned products with an aliphaticaromatic structure for use as extenders. They are added to plastisols in small amounts as viscosity regulators. Dibenzyl toluene also serves the same purpose. A disadvantage of this type of extenders is the risk of exudation if excessive quantities are used. The normal liquid chlorinated parafﬁns used as plasticizers for PVC have viscosities ranging from 100 to 40,000 MPa.s at 20°C. Products with chlorine contents ranging from 30 to 70% are on the market. Compatibility with PVC increases with increasing chlorine content but the plasticizing effect is reduced. The low viscosity products (chlorine content 30%–40%) are used as secondary plasticizers for PVC. They have a stabilizing effect on viscosity in plastisols. Chlorinated parafﬁns can be used up to a maximum of 25% of the total plasticizer content of the PVC plastisol without the risk of exudation. As chlorinecontaining substances, these plasticizers also have a ﬂame-retarding effect. At a chlorine content of 50%, chlorinated parafﬁns are sufﬁciently compatible with PVC to be used alone, i.e., as primary plasticizers, in semirigid compounds under certain circumstances. Products with extremely good compatibility, however, have low plasticizing action, requiring larger amounts to be used. Among other plasticizers, o- and p-toluene-sulfonamides are used to improve the processibility of urea and melamine resins, and cellulose-based adhesives. For the same reason, N-ethyl-o-toluenesulfonamide and N-ethyl-p-toluenesulfonamide are used for polyamides, casein, and zein. Diglycol benzoates are liquid plasticizers rather like benzyl butyl phthalate (BBP) in their plasticizing action in PVC. In the manufacture of PVC ﬂoor coverings, the benzoates offer the advantage of a highly soil resistant surface. In this respect, they are even superior to BBP, which itself performs well here. Pentaerythritol esters can be used instead of phthalic acid esters when low volatility is one of the major factors in the choice of plasticizer for PVC. Pentaerythriol triacrylate can be used to plasticize PVC and to produce-cross-links under UV radiation. Such plasticized cross-linked PVC has some application in the cable industry. Commonly used plasticizers for some polymers beside PVC are listed below: Cellulosics: Dimethyl phthalate (DMP), diethyl phthalate (DEP), dibutyl phthalate (DBP), dimethylcyclohexyl phthalate, dimethyl glycol phthalate, trichloroethyl phosphate, cresyldiphenyl phosphate, some glycolates, some sulfonamides. Poly(vinyl butyrals): Triethyleneglycol di(2-ethyl)butyrate, triethyleneglycol propionate, some adipates, some phosphates. Polyurethanes: Dibutyl phthalate (DBP), diethyleneglycol dibenzoate (DEGB), and most of the plasticizers used for PVC are preferred for their good plasticization action. 1.24 Antistatic Agents By virtue of their chemical constitution, most plastics are powerful insulators, a property that makes them useful for many electrical applications. A disadvantage of the insulation property, however, is the accumulation of static electricity, which is not discharged fast enough due to the low surface conductivity of Characteristics of Polymers and Polymerization Processes 143 most plastics—a difference between plastics and metals. On plastics and other nonmetallic materials, frictional contact is necessary to generate static charges. On immobile objects, for example, static electricity may build up simply by friction with the ambient air. Electrostatic charges can also be produced on the surfaces of polymeric materials in the course of processing operations such as extrusion, calendaring, and rolling up of plastic sheets or ﬁlms. The superﬁcial electrical potential generated by friction may reach values up to a few tens of kilovolts, and this presents serious difﬁculties for practical applications and to users. Spark formation through heavy charging with subsequent ignition of dust or solvent/air mixtures have been a cause of many destructive explosions. In the service life of plastics products, static charging can give rise to many other troublesome effects, as for example, interference with the sound reproduction of records by dust particles picked up by electrostatic charges, production delays due to clinging of adjacent ﬁlms or sheets, lump formation during pneumatic transportation, and static build-up on people passing over synthetic ﬁber carpeting or plastic ﬂoor-coverings, with a subsequent “shock” as the charge ﬂows off, usually when the person touches a door handle. There are many ways to eliminate surface electrostatic, for example, by increasing the humidity or the conductivity of the surrounding atmosphere, or by increasing the electric conductance of materials with the use of electroconducting carbon blacks, powdered metals, or antistatic agents. Electroconducting carbon blacks are largely utilized to increase the electric conductivity of organic polymers. The electric conductivity of carbon blacks depends, inter alia, on the capacity to form branched structures in the polymer matrix, and on the size and size distribution of carbon black particles. The branched and tentacular structures of carbon in the polymer matrix are responsible for the electric conductivity, as is the case for lamp, acetylene, and furnace carbon blacks. The speciﬁc resistance of the carbon particles decreases with their size and then increases with further diminution of the size. A wide size distribution is believed to favor the formation of branched structures contributing to greater conductivity. In spite of the effectiveness of some carbon blacks in reducing surface charges on plastics materials, the use of antistatic agents have increased steadily. The simplest antistatic agent is water. It is adsorbed on the surface of objects exposed to a humid atmosphere, and it forms a thin electroconducting layer with impurities adsorbed from the air. Such a layer is even formed on the surface of hydrophobic plastics, probably because of the existence of a thin layer of dirt. Antistatic agents commonly used are substances that are added to plastics molding formulations or to the surface of molded articles in order to minimize the accumulation of static electricity. In general terms, antistatic agents can be divided according to the method of application into external and internal agents. 1.24.1 External Antistatic Agents External antistatic agents are applied to the surface of the molded article from aqueous or alcoholic solution by wetting, spraying, or soaking the plastic object in solution, followed by drying at room temperature or under hot blown air. Their concentration varies between 0.1 and 2% by weight. Almost all surface active compounds are effective, as well as numerous hygroscopic substances such as glycerin, polyols, and polyglycols, which lack the surface activity feature. The most important external antistatic agents from the practical point of view are quaternary ammonium salts and phosphoric acid derivatives. The advantage of external agents lies in the fact that, in their performance, the properties of compatibility with the polymer and controlled migration in the polymer, which play an important part in the performance of internal antistatic agents described later, are not of any consideration. It is generally assumed that surface-active molecules accumulate on the surface and are oriented such that the hydrophobic part containing the hydrocarbon chain extends into the plastic, and the hydrophilic part points outwards there it is able to absorb water on the surface. The phase boundary angle between water and plastic is reduced by the surface active antistatic agents thus allowing the absorbed water to be uniformly distributed on the plastic surface. A water ﬁlm forms on the surface thus increasing the 144 Plastics Technology Handbook conductivity by means of an ion conduction process. This also explains why surface conductivity, and hence the antistatic action, are found to increase with increasing atmospheric humidity. For the ion conduction in the surface ﬁlm, a conductivity mechanism that is similar to protonic conductivity in water has been suggested: H H+ O H O H H O H H H H O H O H O H+ H H This mechanism is based on a comparison of the conductivities of substances of different chemical structure. For instance, primary amines are efﬁcient as antistatic agents but secondary amines are not. The conductivity of tertiary amines, on the other hand, depends on the nature of N-hydroxyalkyl substituents. Among amides, only N,N-disubstituted derivatives derivatives and mainly those having two hydroxyalkyl substituents are effective. The presence of many OH groups in the molecule makes the efﬁciency of the antistatic agent less dependent on the humidity. Antistatic agents bearing OH or NH2 groups are able to associate in chain form via hydrogen bonding and display antistatic activity even at low atmospheric humidity, unlike compounds that are able to form only intramolecular hydrogen bonds. Many antistatic agents also show hygroscopic properties, thereby intensifying the attraction of water to the surface. At constant atmospheric humidity, a hygroscopic compound attracts more water to the surface and so increases antistatic effectiveness. 1.24.2 Internal Antistatic Agents Internal antistatic agents are incorporated in the polymer mass as additives, either before or during the molding process. However, to be functional, they must be only partially miscible with the polymer so that they migrate slowly to the surface of the plastics material. Their action therefore appears after a few hours and even a few days after compounding, depending on the mutual miscibility of the agent and the polymer. The concentration of internal antistatic agents in plastics varies from 0.1 to 10% by weight. Interfacial antistatic agents are all of interfacially active character, the molecule being composed of a hydrophilic and hydrophobic component. The hydrophobic part confers a certain compatibility with the particular polymer and is responsible for anchoring the molecule on the surface, while the hydrophilic part takes care of the binding and exchange of water on the surface. 1.24.3 Chemical Composition of Antistatic Agents The selection of an antistatic agent for a given polymer is based on its chemical composition. Accordingly, antistatic agents can be divided in many groups, as presented below. 1.24.3.1 Antistatic Agents Containing Nitrogen Antistatic agents containing nitrogen are mainly amines, amides, and their derivatives, such as amine salts, and addition compounds between oxiranes and aminoalcohols. Pyrrolidone, triazol, and polyamine derivatives also belong to this group. A few representative examples belonging to this group are (CH2CH2O)x H Ethoxylated amines of the type R N (CH2CH2O)x H 145 Characteristics of Polymers and Polymerization Processes R1 Amine oxides of the type R2 N O R3 O Fatty acid polyglycolamides of the type R C NH(CH2CH2O)3H The following two commercial products are typical examples: (1) Alacstat C-2 (Alcolac Chemical Corp., U.S.A.). This is an N,N-bis(2-hydroxyethyl)-alkylamine used for polyoleﬁns at 0.1% by weight. (2) Catanac 477 (American Cyanamid Co., U.S.A.). It is N-(3-dodecyloxy-2-hydroxypropyl)ethanolamine (C12H25OCH2CHOHCH2NHCH2CH2OH) used for linear polyethylene (0.15% by weight), for polystyrene (1.5% by weight), and for polypropylene (1% by weight). These compounds are not recommended for PVC. Other compounds containing nitrogen and used as antistatic agents are quaternary ammonium salts, quaternized amines, quaternized heterocycles obtained from imidazoline and pyridine, and amides of quaternized fatty acids. Some of these compounds can be utilized for PVC, for example, + R2 Ammonium salts R3 X– R1 N R4 + R2 Quaternized ethoxylated amine R3 N (CH2CH2O)xH X– R4 O Amide of quaternized fatty acid R1 C NH R Some typical examples of commercial products are Catanac SN, Catanac 609, Catanac LS, and Catanac SP, all of American Cyanamid Co. Catanac SN is stearamidopropyl-dimethyl-b-hydroxyethylammonium nitrate, while Catanac 609 is N,N-bis(2-hydroxyethyl)-N-(3′-dodecyloxy-2′-hydroxypropyl) methylammonium methylsulfate: CH2CH2OH [C12H25OCH2CHOHCH2 N CH3]+ CH3SO4– CH2CH2OH It is supplied mainly as a 50% by weight solution in a water-propanol mixture. It is applied as an external antistatic agent at 2% by weight concentration (pH 4–6) and is recommended for phonograph records and other products made of PVC and its copolymers. 146 Plastics Technology Handbook Catanac LS is 3-lauramidopropyl-trimethylammonium methylsulfate: CH3 CH3]+ CH3SO4– N [C11H23CONHCH2CH2CH2 CH3 It exists in a crystalline form with a melting point of 99°C–103°C. Catanac SP is stearamidopropyl-dimethyl-b-hydroxyethyl-ammonium dihydrogenphosphate: CH3 [C17H35CONHCH2CH2CH2 CH2CH2OH]+ H2PO4– N CH3 supplied as a 35% by weight solution in a water-isopropanol mixture (pH 6–8). It beings to compose slightly at 200°C, and at 250°C its decomposition is rapid. It is soluble in water, acetone and alcohols, and is not corrosive to metals even during prolonged contact. Catanac SP can be used as an internal and external antistatic agent at 1%–3% by weight. 1.24.3.2 Antistatic Agents Containing Phosphorus Antistatic agents that contain phosphorus can be used for all polymers, although they are recommended mainly for PVC with which they also function as plasticizers. They are phosphoric acid derivatives, phosphine oxide, triphosphoric acid derivatives, and substituted phosphoric amides. Typical examples are the following: Phosphoric acid esters O = P(OR)3 Ethoxylated alcohols and phosphoric acid esters O = P½O(CH2 CH2 O)x − R3 + – Ammonium salts of phosphoric acid esters O P O O O R1 R2 R3 H N H R4 1.24.3.3 Antistatic Agents Containing Sulfur Antistatic agents that contain sulfur include compounds such as sulfates, sulfonates, derivatives of aminosulfonic acids, condensation products of ethyleneoxide and sulfonamides, and sulfonates of alkylbenzimidazoles, dithiocarbamides, and sulﬁdes. Sulfur-containing antistatic agents are recommended mainly for PVC and PS because they do not interfere with heat stabilizers. They are not suitable for PMMA, polyoleﬁns, and polyamides. As examples, there are alkylpolyglycolether sulfates, RO(CH2 CH2 O)x SO−3 Me+ , which are marketed under the trade name Statexan HA (Bayer). Characteristics of Polymers and Polymerization Processes 147 1.24.3.4 Betaine-Type Antistatic Agents Betaines are trialkyl derivatives of glycine, which exist as dipolar ions of the formula R+3 NCH2 CO−2 . Typical examples of betaine-type antistatic agents are stearylbetaine and dodecyldimethyl–ethanesulfobetaine. They are used mainly for polyoleﬁns. 1.24.3.5 Nonionic Antistatic Agents Nonionic antistatic agents are nonionizable, interfacially active compounds of low polarity. The hydrophilic portion of the molecule is usually represented by hydroxyl groups and the hydrophobic portion by organic groups. This group of antistatic agents includes the following: polyhydroxy derivatives of glycine, sugars, and fatty acids, sometimes modiﬁed by addition of oxiranes, most of the products being nontoxic and hygienically acceptable but having low antistatic effects; heavy alcohol derivatives such as alkylpolyglycol ethers, which are recommended for polyoleﬁns and PVC; alkylphenolpolyglycol ethers having good heat stability, which are used mainly for polyoleﬁns; polyglycol ethers ether obtained from the reaction of glycols with oxiranes, which are used for polyoleﬁns; and fatty acid polyglycol esters RCO2 (CH2CH2O)xH, which are used for many type of plastics. Nonionic antistatic agents are supplied for the most parting liquid form or as waxes with a low softening region. The low polarity of this class makes its members ideal internal antistatic agents for polyethylene and polypropylene. A number of substances that cannot be included in any of the groups above are also good antistatic agents. They are silicone copolymers, organotin derivatives, oxazoline derivatives, organoboron derivatives, and perﬂuorated surfactants. Some of them show excellent heat stability. 1.25 Flame Retardants Efforts to develop ﬂame-retarding plastics materials have been focusing on the increasing use of thermoplastics. As a result, ﬂame-retarding formulations are available today for all thermoplastics, which strongly reduce the probability of their burning in the initiating phase of a ﬁre. The possibility of making plastics ﬂame-retardant increases their scope and range of application. (It must be noted, however, that in the case of ﬁre, the effectiveness of ﬂame retardants depends on the period of time and intensity of the ﬁre. Even a product containing the most effective ﬂame-retardants cannot resist a strong and long-lasting ﬁre) [70]. Flame retardants are deﬁned as chemical compounds that modify pyrolysis reactions of polymers or oxidation reactions implied in combustion by slowing them down or by inhibiting them. The combustion of thermo-plastics (see “Fire Resistant Polymers” in Chapter 5) is initiated by heating the material up to its decomposition point through heat supplied by radiation, ﬂame, or convection. During combustion, free radials are formed by pyrolysis and, besides noncombustible gases such as carbon dioxide and water, combustible gases are formed–mainly hydrocarbons, hydrogen, and carbon monoxide. The radicals combine with oxygen and combustible gases in a radical chain reaction, thereby causing heat release and further decomposition of the plastics material. For continued combustion it is necessary to have sufﬁcient oxygen as well as combustible gaseous compounds. The combustion will be slowed down or inhibited if free radicals, which are evolved by pyrolysis, are blocked. For example, it is presumed that the following reactions take place when thermoplastics containing organobromine compounds as ﬂame retardants are used: Exothermic propagation Chain branching Chain transfer HO• + CO = CO2 + H• (1.149) H• + O2 = HO• + O• (1.150) O• + HBr = HO• + Br• (1.151) 148 Plastics Technology Handbook Inhibition HO• + HBr = H2 O + Br• (1.152) RH + Br• = R• + HBr (1.153) Regeneration The radical chain reaction is interrupted when the highly reactive HO• radical, which plays a key role in the combustion process, is replaced by the less reactive Br• radical (Reaction 1.152). While Equation 1.151 and Equation 1.152 show reactions with HBr (which is released during combustion), experiments have shown that halogen-containing organic compounds may also react as undecomposed molecules. Flame retardants, in view of the above discussion, can be deﬁned as chemical compounds that modify pyrolysis reactions of polymers or oxidation reactions implied in the combustion by slowing these reactions down or by inhibiting them. In practice, mixtures of ﬂame retardants are often used to combine different types of retardancy effects. In spite of a few thousand references on ﬂame retardants, only a small number of compounds are produced commercially. They are mainly chlorine, bromine, phosphorus, antimony, aluminum, and boron-containing compounds. Most ﬂame retarding agents may be divided into three groups: (1) halogen compounds, (2) phosphorous compounds, and (3) halogen–antimony synergetic mixtures. 1.25.1 Halogen Compounds As ﬂame retardants, bromine compounds show superiority over chlorine compounds. For example, in case of combustion of n-heptane/air mixture, the minimum concentrations (in vol. %) found necessary for the extinction of ﬂame are 5.2% for dibromomethane and 17.5% for trichloromethane. The efﬁciency of halogen-containing agents depends on the strength of the carbon-halogen bond. Thus the relative efﬁciency of aliphatic halides decreases in the following order: Aliphatic bromides > aliphatic chlorides = aromatic bromides > aromatic chlorides It is also known that aromatic bromides have a higher decomposition temperature (250°C–300°C) than aliphatic bromides (200°C–250°C). Bromine-containing compounds on a weights basis are at least twice as effective as chlorine-containing ones. Because of the smaller quantities of bromine compounds needed, their used hardly inﬂuences the mechanical properties of the base resins and reduces markedly the hydrogen halide content of the combustion gases. Aliphatic, cycloaliphtic, as well as aromatic and aromatic-aliphatic bromine compounds are used as ﬂame retardants. Bromine compounds with exclusively aromatic-bound bromine are produced in large quantities. Occupying the ﬁrst place is tetrabromobios-phenol-A, which is employed as a reactive ﬂame retardant in polycarbonates and epoxy resins. Similarly, tetrabromophthalic anhydride has been used in the production of ﬂame-retarded unsaturated polyester resins. Brominated diphenyls and brominated diphenylethers—for instance, octabromodiphenyl ether (melting point 200°C–290°C) and decabromodiphenyl ether (melting point 290°C–310°C)-have gained great signiﬁcance. Possessing excellent heat stability, they are excellent ﬂame retardants for those thermoplastics that have to be processed at high temperatures, such as linear polyesters and ABS. Aromatic-aliphatic bromine compounds, like the bis(dibromopropyl)-ether of tetrabromobisphenol A, or bromoethylated and aromatic ring-brominated compounds, such as 1,4-bis-(bromoethyl)tetrabromobenzene, combine the high heat stability of aromatic-bound bromine with the outstanding ﬂame retardancy of aliphatic-bound bromine. They are used mainly as ﬂame retardants for polyoleﬁns, including polyoleﬁn ﬁbers. Polymeric aromatic bromine compounds, such as poly(tribromostyrene), poly(pentabromobenzyl acrylate) or poly(dibromophenylene oxide) are suitable for application in polyamides, ABS, and polyesters. Characteristics of Polymers and Polymerization Processes 149 Oligomeric bromine-containing ethers on the basis of tetrabromoxylene dihalides and tetrabromobisphenol-A (molecular weight between 3,000 and 7,000) are useful ﬂame retardants for polypropylene, polystyrene, and ABS. The following cycloaliphatic compounds are suitable as ﬂame retardants in polyoleﬁns: hexabromocyclododecane and bromine-containing Diels-Alder reaction products based on hexachloro cyclopentadiene. 1.25.2 Phosphorus Compounds Whereas halogen acids, which are released during the combustion of thermoplastics containing organic halogen compounds, interrupt the chain reaction (e.g., Reaction 1.133), the ﬂame-retarding action of phosphorus compounds is not yet fully understood. It is supposed, however, that in case of ﬁre, phosphorus compounds facilitate polymer decomposition, whereby the very stable poly(meta-phosphoric acid) formed creates an insulating and protecting surface layer between the polymer and the ﬂame. The phosphoric acid also reacts with the polymer and produces a char layer protecting the surface. Phosphorus-containing ﬂame retardants may be divided according to the oxidation state of the element into phosphates [(RO)3PO], phsophites [(RO)3P], phosphonites [(RO)2PR′], phosphinates [(RO) R2′PO], phosphines [R3P], phosphine oxides [R3PO], and phosphonium salts [R4PX]. The halogenated phosphorus-containing compounds represent a very important group of ﬂame retardants with high efﬁciency A few examples are tris-(tribromoneopentyl)-phosphite, tris-(dibrompropyl)-phosphate, tris(di-bromophenyl)-phosphate, and tris-(tribromophenyl)-phosphate. A similar ﬂame retardancy may also be obtained by the combinations of phosphorus-containing compounds with various halogen derivatives. These combinations are relatively universal in their action. Phosphorus-containing ﬂame retardants are suitable for polar polymers such as PVC, but for polyoleﬁns their action is not sufﬁcient. In this case they are used together with Sb2O3 and with halogenated compounds. Alkyl-substituted aryl phosphates are incorporated into plasticized PVC and modiﬁed PPO (Noryl) to a great extent; for other plastics they are less important. Quaternary phosphonium compounds are recommended as ﬂame retardants for ABS and polyoleﬁns. 1.25.3 Halogen–Antimony Synergetic Mixtures Antimony oxide is an important component of ﬂame retarded thermoplastics containing halogen compounds. Used alone, the ﬂame-retardant effect of antimony oxide is not sufﬁcient, but the effect increases signiﬁcantly as halogen is added to the system. The basis of this antimony-halogen synergism is probably the formation of volatile ﬂame-retardant SbX3 (see below), where X is Cl or Br. Antimony–halogen combinations have be widely used in polymer for ﬂame retardation, and the interaction of antimony (mostly as antimony oxide) with halogenated polymers or polymers containing halogenated additives represents the classic case of ﬂame-retardant synergism. Experimental results indicate that the best Sb:X ratio is 1 to 3. More effective are ternary synergetic mixtures such as Sb2O3− pentabromotoluenechloroparafﬁn. The antimony oxide-alkyl (aryl) halide system functions on heating according to the following mechanism (shown for chloride): RCl ! HCl + R0 CH = CH2 (1.154) Sb2 O3 + 6HCl ! 2SbCl3 + 3H2 O (1.155) Sb2 O3 + 2HCl ! 2SbOCl + H2 O (1.156) 150 Plastics Technology Handbook Volatile SbCl3 reduces the formation of radicals in the ﬂame and also affects the oxidation process strongly, thus terminating the combustion. As with phosphorus compounds, it promotes carbonization of the polymer. It is believed that SbOCl and SbCl3 already function as dehydrogenation agents in the solid phase of the polymer in ﬂame. At 240°C–280°C, SbOCl is transformed to higher oxychlorides and SbCl3, and at temperatures around 500°C the higher antimony oxychlorides are converted back into antimony oxide by releasing again SbCl3: 240°C–280°C ⟶ 5SbOCl 4Sb4 O5 Cl2 Sb4 O5 Cl2 + SbCl3 410°C–475°C 3Sb3 O4 Cl ⟶ 5Sb3 O4 Cl + SbCl3 475°C–565°C ⟶ 4Sb2 O3 + SbCl3 (1.157) (1.158) (1.159) The ﬂame retardation is also attributed to the consumption of thermal energy by these endothermic reactions, to the inhibition reactions in the ﬂame, and to the release of hydrochloric acid directly in the ﬂame. The pigmentation effect of mixtures containing Sb2O3 is a disadvantage that limits their use for light-colored and transparent products. The demands on a ﬂame retardant and thermoplastics formulated with such agents are manifold. The ﬂame retardant should provide a durable ﬂame-retarding effect by the addition of only small quantities of the additive; it should be as cheap as possible and the manner of incorporation should be easy; it should be nontoxic and should not produce ﬁre efﬂuents with increased toxicity; it should not decompose at the processing temperatures; it should not volatilize and smell; and the mechanical, optical, and physical properties of the thermoplastics should be affected as little as possible. It is understandable that these farranging demands cannot be satisﬁed by only one ﬂame retardant and for all thermoplastics with their manifold applications. One is therefore forced to seek the optimum ﬂame retarding formulation for each thermoplastic and the speciﬁc application. Examples of typical formulations of several ﬂame-retarded thermoplastics are given in Appendix A3. 1.25.4 Intumescent Flame Retardants While halogenated compounds are good ﬁre-retardant additives for polyoleﬁns, especially in synergistic combination with antimony trioxide, there are serious disadvantages, such as evolution of toxic gases and corrosive smoke. In the search for halogen-free ﬂame retardants, intumescent ﬂame-retardants (IFRs) have received considerable attention recently, and they have shown particularly good efﬁciency in the ﬂame retardation of polypropylene (PP) [71]. IFRs that contain ammonium polyphosphate (APP) and pentaerythritol (petol) work mainly through a condensed phase mechanism [72]. APP is most often used as the acid source and petol as a carbon source. These APP-based products also have a low smoke density and do not emit corrosive gases during combustion. Recently, the synergistic activity of IFRs with heavy metal ions has received wide attention as this can render an IFR more efﬁcient at lower concentrations [73]. It has been found that divalent metal compounds and their derivatives have an excellent synergistic effect when used together with the polyphosphate and petol. Thus, using nickel formate, Ni(HCOO)2 the limiting oxygen index (LOI) of PP/ APP/petol (90: 6.75: 3.25 by wt.) has been found [74] to have the highest value of 30% at 2–3 wt.% Ni (HCOO)2. It is proposed that during burning, the petol is ﬁrst phosphorylated by APP, with the release of water and ammonia, while the phosphate ester is pyrolyzed with the formation of double bonds that subsequently cross-link, resulting in a three-dimensional structure. 151 Characteristics of Polymers and Polymerization Processes 1.26 Smoke Suppressants In recent years, the hazards of smoke and toxic gas production by burning have received much greater recognition. Since burning of a large amount of polymer represents a fuel rich combustion, it produces large amounts of toxic carbon monoxide and carbon-rich black smoke, causing death and inhibiting escape from the burning environment. A comparison of smoke production from burning different polymers is given in Table 1.21. The problem of smoke suppression has not yet been solved satisfactorily, because the demands on ﬂame retardancy and reduced smoke development are principally of antagonistic nature. Most of the developmental work on smoke suppressants has been carried out in the ﬁeld of PVC. For this polymer, highly dispersed aluminum trihydrate is widely used as a ﬂame retardant and as a smoke suppressant. Antimony trioxide-metal borates (barium borate, calcium borate, and zinc borate) are good smoke reducers. Zinc borate of the formula 2ZnO:3B2O3:3.5H2O has shown the best effect with regard to reduced ﬂammability and reduced smoke development of PVC. It is possible to reduce smoke development of rigid PVC signiﬁcantly by adding rather small quantities of molybdic oxide. For example, 2% of molybdic oxide reduces the smoke density to about 35%, but does not decrease the limiting oxygen index value of rigid PVC. Several methods have been proposed for the determination of the smoke density. Reproducible values are obtained with the laboratory method worked out by the National Bureau of Standards, whereby a specimen of 7.6 × 7.6 cm is heated by radiation. The smoke density is then measured optically. 1.27 Colorants A wide variety of inorganic acid organic materials are added to polymers to impart color. For transparent colored plastics materials, oil-soluble dyes or organic pigments (such as phthalocyanines) having small particle size and refractive index near that of the plastic are used. Others, including inorganic pigments, impart opaque color to the plastic. Some of the common colorants for plastics, among many others, are barium sulfate and titanium dioxide (white), ultramarine blues, phthalocyanine blues and greens, chrome green, molybdate organs, chrome yellow, cadmium reds and yellows, quinacridone reds and magentas, and carbon black. Flake aluminum is added for a silver metallic appearance, and lead carbonate or mica for pearlescence. The principal requirements of a colorant are that it have a high covering power-cost ratio and that it withstand processing and service conditions of the plastic. The colorant must not catalyze oxidation of the polymer nor adversely affect its desirable properties. Colorants are normally added to the powered plastic and mixed by tumbling and compounding on a hot roll or in an extruder. TABLE 1.21 Smoke Emission on Burning (NBS Smoke Chamber, Flaming Condition) Polymer ABS Poly(vinyl chloride) Maximum Smoke Density (Dmax) 800 520 Polystyrene 475 Polycarbonate Polyamide-imide 215 169 Polyarylate 109 Polytetraﬂuoroethylene Polyethersulfone 95 50 Polyether ether ketone 35 152 Plastics Technology Handbook 1.28 Antimicrobials Antimicrobials impart protection against mold, mildew, fungi, and bacterial growth to materials. Without antimicrobials, polymers can experience surface growths, producing allergic reactions, unpleasant odor, staining, embrittlement, and even permanent product failure. The effectiveness of an antimicrobial depends on its ability to migrate to the surface of the product where microbial attack ﬁrst occurs. Antimicrobials are thus usually carried in plasticizers which are mobile. High mobility, however, can result in leaching of the additives due to weather exposure. The durability of protection is, therefore, determined by a proper balance between migration and leaching. Of the hundreds of chemicals that are effective as antimicrobials, only a few are used in plastics applications. The most common group of active ingredients consists of 10,16-oxybisphenoxy arsine (OBPA, tradenames: Intercide, Akzo Chemical; Vinyzene, Morton International), 2-n-octyl-4isothiazolin-3-one (tradenames: Micro-Check 11, Ferro Corporation; Vinyzene IT, Morton International), N-trichloro-methylthio-4-cyclohexene-1,2-dicarboximide (Captan, tradenames: Vancide 89, R.T. Vanderbilt; Fungitrol C, Hüls America), and N-(trichloromethylthio) phthalimide (Flopet, trade name; Fungitrol 11, Huls America). Other antimicrobial active ingredients include amine-neutralized phosphate and zinc 2-pyridinethiol-1-oxide. OBPA is the most toxic and requires approximately 0.04% active ingredient in the ﬁnal product. Less toxic active ingredients such as Flopet require a loading of 1.0% to achieve a similar level of protection. Antimicrobials are generally formulated with a carrier into concentrations of 2–5% active ingredient and are available to plastic processors in powder, liquid, or solid-pellet form. The carrier is usually a plasticizer, such as epoxidized soybean oil or diisodecyl phthalate. Among the antimicrobials, OBPA has broad-spectrum effectiveness against gram-positive and gramnegative bacteria and fungi. Another additive, 2-n-octyl-4-isothiazolin-3-one, is a fungicide only and is commonly used to prevent molds in paints. Captan is an agricultural fungicide that has important plastics applications. The majority of all antimicrobial additives are used in ﬂexible PVC. The remaining portion is used in polyurethane foams and other resins. PVC applications using antimicrobials include ﬂooring, garden hose, pool liners, and wall coverings. 1.29 Toxicity of Plastics With the expanding use of plastics in all walks of life, including clothing, food packaging, and medical and paramedical applications, attention has been focused on the potential toxic liability of these man-made materials. Toxic chemicals can enter the body in various ways, particularly by skin absorption, inhalation, and swallowing. Although some chemicals may have an almost universal effect on humans, others may attack few persons. The monomers used in the synthesis of many of the polymers are unsaturated compounds with reactive groups such as vinyl, styrene, acrylic, epoxy, and ethylene imine groups. Such compounds can irritate the mucous membranes of the eyes and respiratory tract and sensitize the skin. They are suspected of including chemical lesions and carcinogenic and mutagenic effects. It was long ago that carcinogenic properties of the monomers vinylchloride and chloroprene were reported, and there may be more trouble ahead in this respect. However, the monomers used in the production of polyester or polyamide resins are usually less reactive and may be expected to be less harmful. Although many monomers are harmful chemicals, the polymers synthesized from them are usually harmless macromolecules. But then one has to take into account that the polymers may still contain small amounts of residual monomer and catalyst used in the polymerization process. Moreover, polymers are rarely used as such but are compounded, as we saw earlier, with various additives such as plasticizers, stabilizers, curing agents, etc., for processing into plastic good. Being relatively smaller in molecular size, Characteristics of Polymers and Polymerization Processes 153 the residual monomer, catalyst residues, and the additives can migrate from the plastic body into the environment. They may eventually migrate into food products packed in plastic containers, or they may interact with the biological substrate when plastics materials are used as parts of tissue and organs implanted into humans. The toxic potential of thermodegradation and combustion products of plastics when these materials are burned, either deliberately or by accident, is also an important consideration in view of the widening use of plastics as structural and decorative lining material in buildings, vehicles, and aircraft. 1.29.1 Plastic Devices in Pharmacy and Medicine Within the past two decades there has been an increase in the use of a variety of plastic materials in the pharmaceutical, medical, dental, and biochemical engineering ﬁelds. Such plastic devices can be classiﬁed according to use into six basic groups: (1) collection and administrative devices—e.g., catheters, blood transfusion sets, dialyzing units, injection devices; (2) storage devices—e.g., bags for blood and blood products, containers for drug products, nutritional products, diagnostic agents; (3) protective and supportive devices—e.g., protective clothes, braces, ﬁlms; (4) implants having contact with mucosal tissue— e.g., dentures, contact lenses, intrauterine devices; (5) permanent implants—e.g., orthopedic implants, heart valves, vascular grafts, artiﬁcial organs, and (6) nanomedicines and drug delivery. The toxic potential of plastic devices becomes more relevant in those applications which involve long periods of contact with the substrate. For short periods of contact, however, a great number of plastic materials manufactured today as medical and paramedical devices will produce little or no irritant response. With polyethylene, polypropylen, Teﬂon, Dacron, polycarbonate, and certain types of silicone rubbers, the migration of additives from the material is so small that a biological response cannot be detected. Besides, many of these materials contain additives only in extremely low concentrations, some perhaps having only a stabilizing agent in concentrations of less than 0.1 wt.%. Incidence of tissue response increases when the plastic materials require greater concentrations of various types of additives. It should be stressed, however, that the toxic effects of a material will depend on the intrinsic toxicity of the additives and their ability to migrate from the material to the tissue in contact with it. It is important to know precisely the toxic potential of each of the ingredients in the ﬁnal polymerized and formulated material to be used in an implantable or storage device. 1.29.1.1 Packing Many pharmaceutical manufacturers have now taken to packaging their drug product in plastic containers. The plastics used most in these applications are those manufactured from polyethylene and polypropylene, though other materials have also been used. Since both polyethylene and polypropylene contain extremely small amounts of additives (mostly as antioxidants and antistatic agents), the possibility of their release in sufﬁcient concentrations to endanger to patient is negligible. 1.29.1.2 Tubings and Blood Bag Assemblies Plastic tubings are used for many medical applications, such as catheters, parts of dialysis and administration devices, and other items requiring clear, ﬂexible tubings. The most successful tubing is PVC, which has been plasticized to give it the desired ﬂexibility. Plasticized PVC is also the chief material used in America for making bags for blood and products. In Europe polyethylene and polypropylene have been mainly used. The plasticizers used most for ﬂexible PVC are the esters of phthalic acid; one that is most employed from this group is di-2-ethylhexyl phthalate. Long-term feeding studies have demonstrated the nontoxic nature of the plasticizer under the experimental conditions used. Organotin compounds, which are one of the best groups of stabilizers for vinyl polymers, have been used to stabilize PVC, but, in general, their toxicity has decreased their use in medical applications. 154 Plastics Technology Handbook 1.29.1.3 Implants Man-made materials have been used for making implants to save human life. It is possible that these materials, depending on their speciﬁc nature and the site of implant, will degrade with long periods of contact and release polymer fragments in the body. These in turn may elicit one or more biological responses, including carcinogenesis. However, no well-substantiated evidence has been reported showing that man-made materials have caused cancer in humans, although the real answer will be available only when these materials have been implanted for periods of time exceeding 20 to 30 years. Perhaps the most commonly used implantable material is silicone rubber. This material, if properly prepared by the manufacturer, does not cause local toxic response. Various types of epoxy polymers and polyurethane materials also have found one or more medical applications. 1.29.1.4 Adhesives In most medical and paramedical applications of plastics, the materials used are those already produced by the manufacturers in a polymerized or formulated form. Certain surgical and dental applications, however, require that the material be polymerized or formulated just prior to use. Surgical cements and adhesives, a host of dental ﬁlling materials, materials for dentures, cavity liners, and protective coatings for tooth surfaces are in this category. Cyanoacrylates have become very useful as tissue adhesives in surgical applications, because they polymerize rapidly in contact with moisture and create an extremely tenacious ﬁlm. Methyl cyanoacrylate, used initially, has now fallen out of favor because of its toxic properties. The butyl and heptyl analogs are, however, quite satisfactory and do not produce objectionable tissue response. They also degrade at a much slower rate than the methyl compound in a biological environment. 1.29.1.5 Dental Materials Sometimes known as white ﬁlling or synthetic porcelain, polymeric dental composites are commonly used as a tooth-colored restorative material, for example, in the fabrication of ﬁllings and veneers, and the cementation of crowns. Dental composites are complex mixtures generally containing an organic resin matrix, reinforcing inorganic ﬁller (such as Sr-glass, 0.7 mm), a silane coupling agent, which connects the ﬁller and the resin matrix, and stabilizers to maximize storage [75]. Composites without the ﬁller and coupling agent are commonly used as sealants, which effectively isolate pits and ﬁssures to help prevent caries. The dental resins are commonly based on the highly viscous bisphenol A glycidyl methacrylate (abbreviated as bis-GMA, also known as Bowen’s monomer after the inventor) have been used for over four decades. Because of its high viscosity (1,369 Pa.s), bis-GMA is blended with diluent, lower molecular weight monomers, to provide a workable matrix resin for composites, for example, bisphenolA dimethacrylate (bis-DMA), ethylene glycol dimethacrylate (EGDMA), and triethylene glycol dimethacrylate (TEGDMA). Camphorquinone (CQ) is traditionally used as photosensitizer for dental composites. It undergoes a redox reaction with a tertiary amine to produce radicals for free-radical polymerization of the acrylate resin. In a typical procedure, a mixture of bis-GMA and TEGDMA with 1 wt% of CQ and tertiary amine is photopolymerized using visible light device with an intensity of 900 mW/cm2. The photopolymerization of the dental resin is fast with most of the reaction taking place within 40 s, causing gelation and vitriﬁcation accompanied by shrinkage of about 8%. Shrinkage continues, as also stiffening, during post-polymerization at a much slower rate (see Table 1.22). Based on data reported in several studies involving application of sealant to teeth, it appears that low levels of bisphenol A (BPA) may be released from certain sealants, although only during a short time period immediately after application of the sealant. Further, no detectable levels of BPA have been found in blood after application of a sealant that releases low levels of BPA into saliva and the sealant possesses no risk to human health [76]. 155 Characteristics of Polymers and Polymerization Processes TABLE 1.22 Young’s Modulus, Density, and Shrinkage of Dental Resin Matrix after Photopolymerization Young Modulus (GPa) Density (g/cm3) Shrinkage (%) 0 2.22 1.203 8.02 24 3.27 1.220 9.44 48 3.31 1.222 9.67 Time after Photopolymerization (s) Source: From Trufﬁer-Boutry, D., Demonstier-Champagne, S., Devaux, J., Brebuyck, J.-J., Larbanois, P., Mestdagh, M., and Leloup, G. 2005. Eur. Cell Mater., 9, Suppl. 1, 60. 1.29.1.6 Nanomedicines and Drug Delivery The term “nano” refers to all molecules and devices/technologies in the size range 1–1,000 nm (a nanometer is 10−9 m). Nanomedicine has been deﬁned as “the science and technology of diagnosing, treating, and preventing disease, relieving pain, and preserving and improving human health by using molecular tools and molecular knowledge of the human body.” New drugs and new delivery systems both thus come under the umbrella of nanomedicines. Innovative devices and drug delivery systems are needed to exploit many of the drugs developed from advances in molecular biology, to guide the therapeutic to its correct location of action and ensure that pharmacological activity is maintained for an adequate duration once it is there. In many cases, water soluble polymers are used to carry the active drug into the body without affecting the sites other than the target sites, where the local environment breaks down the polymer releasing the active drug. “Nanoparticles” are solid colloidal polymeric carriers (less than 1 mm in size) that have received much attention over the recent years due to their ability to control drug release and distribution and due to their biodegradability [77]. Furthermore, these systems have proven their potential to administer peptides or other drugs either by intravenous or oral routes, increasing their bioavailability and protection of the drug against degradation, and reducing the associated adverse effects [78,79]. Most commonly used methods for preparing polymer-based nanoparticles include emulsion evaporation [80], in situ monomer polymerization [81], a method based on the salting out effect [82] and nanoprecipitation [83]. The last named method represents an easy and reproducible technique, which has been widely explored for producing both vesicle and matrix type nanoparticles, also termed nanocapsule and nanosphere, respectively [84]. For the last decade, surfactant-free nanoparticles formation has been investigated by many researchers. Fessi et al. [83] developed surfactant-free nanocapsules of poly(D,Llactide) (PLA) by the nanoprecipitation technique, using a novel and simple procedure which involved interfacial deposition of a preformed, well-deﬁned and biodegradable polymer following displacement of a semi-polar solvent miscible with water from a lipophilic solution. Nanocapsules of PLA containing indomethacin as a drug model were thus prepared [83]. More recently, poly(lactide-co-glycolide) nanoparticles were prepared by using poly(ethylene glycol)-based block copolymers as substitutes for conventional surfactants [85]. Progress in the development of nano-sized hybrid therapeutics and nano-sized drug delivery systems over the last decade has been remarkable. A growing number of products have already secured regulatory authority approval and, in turn, are supported by a healthy clinical development pipeline. They include products used to treat cancer, hepatitis, muscular sclerosis, and growth hormone deﬁciency. (See “Therapeutics Packaging and Nanomedicines” in Chapter 7.) 1.29.2 Biodegradable Plastics and Bioplastics The ability to undergo biodegradation producing nontoxic by-products is a useful property for some medical applications. Biodegradable polymers [86] have been formulated for uses such as sutures, vascular grafts, drug delivery devices, and scaffolds for tissue regeneration, artiﬁcial skin, orthopedic implants, and others. The polymers commonly known in the medical ﬁeld for such applications include poly 156 Plastics Technology Handbook (a-hydroxy esters), poly(ϵ-caprolactone), poly(ortho esters), polyanhydrides, poly(3-hydroxybutyrate), polyphosphazenes, polydioxanones, and polyoxalates (see Chapter 5). The homopolymers poly(L-lactic acid) (PLLA) and poly(glycolic acid) (PGA) as well as poly (DLlactic-co-glycolic acid) (PLGA) copolymers are poly(a-hydroxy esters). They are biocompatible, biodegradable and are among the few biodegradable polymers with Food and Drug Administration (FDA) approval for human clinical use. PLLA, PLGA copolymers, and PGA are produced by a catalyzed ring— opening polymerization of lactide and/or glycolide. Their degradation in vivo is brought about by simple, non-enzymatic hydrolysis of the ester bond leading, eventually, to lactic acid and/or glycolic acid. These acids are processed through normal metabolic pathways and eliminated from the body ultimately through the respiratory system as carbon dioxide. PGA is a highly crystalline, hydrophilic poly(a-hydroxy ester) and was one of the ﬁrst synthetic degradable polymers to ﬁnd application as a biomaterial. One of the ﬁrst applications of PGA was as a biodegradable suture material. PLLA is one of the strongest known biodegradable polymers and has therefore found many applications in areas, such as orthopedics. Although organ transplantation has saved many lives, the harsh reality remains that the need for donor organs far outweighs the supply. It is recognized that tissue engineering may provide an alternative to organ transplantation. Tissue engineering involves the creation of natural tissue with the ability to restore missing organ function. This may be achieved either by transplanting cells seeded into a porous material or, in some cases, by relying on ingrowth of tissue and cells into such a material. PGA, PLLA and PLGA copolymers have been used as an artiﬁcial scaffold [87] in cell transplantation and organ regeneration (see Chapter 5). Two different groups of products fall under the term “bioplastics,” namely, biobased plastics and compostable plastics. Biobased materials or plastics are partly or entirely made of renewable raw materials, a few examples being polylactic acid (PLA), polyhydroxyalkanoate, starch, cellulose, chitin, and gelatin. Biobased plastics can be biodegradable, but they are not so always, examples of the latter being composites of natural ﬁber, wood, and plastic. Compostable plastics, on the other hand, can be completely biodegraded by microorganisms. Special bacteria release enzymes capable of breaking down the polymer chains of compostable plastics into small parts, which are then digested by the bacteria together with other organic matter present in the waste, leaving behind water, CO2, and biomass. The biodegradability of a compostable polymer depends entirely on its chemical structure and the polymer can be produced from renewable raw materials or from fossil sources. A commercial example of compostable bioplastics is Ecovio marketed by BASF. It consists of the biodegradable BASF polymer Ecoﬂex and PLA, derived from corn, and also consists partially of renewable raw materials. Compared to simple starch-based bioplastics, Ecovio is more resistant to mechanical stress and moisture. 1.29.3 Oxo-Biodegradable Plastics Discarded conventional plastics remain in the environment for decades. They block sewers and drains, disﬁgure the streets, beaches and countryside, and kill wildlife on land, in rivers and oceans. To overcomethese problems increasing attention has been paid to the development of degradable plastics: (1) starch-based, biodegradable, (2) aliphatic polyesters, biodegradable, (3) photodegradable, and (4) oxobiodegradable. The starch-based plastics do not degrade totally, since only the starch constituent is consumed by microbial activity, and the plastic residues can be harmful to the soil and to birds and insects. Aliphatic polyesters, described above, are relatively expensive. In the same way as starch, they rely on microbial activity in compost or soil to degrade. Both these products degrade by a process of hydrodegradation. Photodegradable plastics degrade after prolonged exposure to sunlight, so will not degrade if buried in a landﬁll, a compost heap, or other dark environment, or if heavily overprinted. Oxo-biodegradable Characteristics of Polymers and Polymerization Processes 157 plastics (Symphony Plastic Technologies, Borehamwood, Herts, U.K.) are a new development. The plastics degrade by a process of oxo-degradation. The technology is based on a small amount (typically 3%) of prodegradent additive (proprietary) being introduced into the conventional manufacturing process, thereby making the plastic degradable. The degradation, which does not rely on microbes, starts immediately after manufacture and accelerate when exposed to heat, light or stress. The process is irreversible and continues in air until the whole material reduces to CO2 and water, leaving no fragments in the soil. The oxobiodegradable plastics (OBPs) are also consumed by bacteria and fungi after the additive has broken down the molecular structure sufﬁciently, allowing access to living micro-organisms. So organic wastes in homes, restaurants, hospitals etc. can be put into oxo-biodegradable plastic sacks for disposing straight into the composting plant without emptying the sacks. Even if OBPs were eaten by cow, deer, turtle or other animal whilst still intact, they would degrade even faster due to the temperature and bacteria present in the gut without causing blockage, unlike conventional plastic bags which could kill the animal. OBPs have an advantage over plastics produced from starch or other agricultural products in that they biodegrade and can be composed but do not need to be buried in a compost heap or landﬁll in order to degrade. The fact that they can degrade in a normal environment is a signiﬁcant factor in relation to litter, because a large amount of plastic waste on land and at sea cannot be collected and buried. OBPs can be made transparent and can be used for direct food contact. The length of time OBP takes to degrade totally can be “programmed” at the time of manufacture by varying the amount of additive and can be as little as a few months or as much as a few years. 1.29.4 Toxicity of Plastic Combustion Products Fires involving plastics produce not only smoke but also other pyrolysis and combustion products. Since most of the polymeric materials contain carbon, carbon monoxide is one of the products generated from the heating and burning of these materials. Depending on the material, the temperature, and the presence or absence of oxygen, other harmful gases may also evolve. These include HCl, HCN, NO2, SO2, and ﬂuorinated gases. Presently, various ﬂame-retarding agents are added to plastics to reduce their combustion properties. When heated or subjected to ﬂame, these agents can change the composition of the degradation products and may produce toxic responses not originally anticipated [73]. 1.29.5 Toxicity Testing Ideally, materials for medical and paramedical applications should be tested or evaluated at three level: (1) on the ingredients used to make the basic resin, (2) on the ﬁnal plastic or elastomeric material, and (3) on the ﬁnal device. Organizations such as the Association for the Advancement of Medical Instrumentation, the U.S.A. Standard Institute, and the American Society for Testing and Materials (F4 Committee) have developed toxicity testing programs for materials used in medical applications. The American Dental Association has recommended standard procedures for biological evaluation of dental materials [89]. References 1. Seymour, R. B. and Carraher, C. E. 1992. Polymer Chemistry, An Introduction. Marcel Dekker, New York. 2. Billmeyer, F. W. 1969. Appl. Polymer Symposia, 10, 1. 3. Debye, P. 1947. J. Phys. Coll. Chem., 51, 18. 4. Price, B. A., Halwer, M., and Speiser, R. 1950. J. Opt. Soc. Am., 40, 768. 5. Chanda, M. 2013. Introduction to Polymer Science and Chemistry, A Problem Solving Approach. CRC Press, Boca Raton, FL. 6. Onyon, P. F. 1951. Viscometry, Chap. 6. In Techniques of Polymer Characterization, P. W. Allen, ed. Butterworth, London, UK. 158 Plastics Technology Handbook 7. Patat, F. and Sinn, H. 1958. Angew. Chem., 70, 496. 8. Cossee, P. 1967. The mechanism of Ziegler-Natta polymerization II. Quantum-chemical and crystalchemical aspects, Chap. 3. In The Stereochemistry of Macromolecules, Vol. 6, A. D. Ketley, ed., Marcel Dekker, New York. 9. Brunsveld, L., Folmer, B. J. B., Meijer, E. W., and Sijbesma, R. P. 2001. Chem. Rev., 101, 4071. 10. Sijbesma, R. P., Beijer, F. H., Brunsveld, L., Folmer, B. J. B., Hirschberg, J. H. K., Lange, R. F. M., Lowe, J. K. L., and Meijer, E. W. 1997. Science, 278, 1601. 11. Folmer, B. J. B., Sijbesma, R. P., Versteegen, R. M., van der Rijt, J. A. J., and Meijer, E.W. 2000. Adv. Mater., 12, 12, 874. 12. Lenz, R. W. 1970. Ind. Eng. Chem., 62, 2, 54. 13. Wall, F. T. 1944. J. Am. Chem. Soc., 66, 2050 14. Odian, G. 2004. Principles of Polymerization, Wiley, New York. 15. Harkins, W. D. 1947. J. Am. Chem. Soc., 69, 1428. 16. Smith, W. V. and Ewart, R. W. 1948. J. Chem. Phys., 16, 592. 17. Rudin, A. 1982. The Elements of Polymer Science and Engineering, Academic Press, Orlando, FL. 18. Langevin, D. 1992. Annu. Rev. Phys. Chem., 43, 341. 19. Winsor, P. A. 1954. Solvent Properties of Amphiphilic Compounds, Butterworth, London, UK. 20. Chanda, M. 2016. Direct synthesis of nano-size polymers by microemulsion polymerization, Chap. 3. In Nano-Size Polymers, Preparation, Properties, Applications, S. Fakirov, ed. Springer Nature, New York. 21. Stoffer, J. O. and Bone, T. 1980. J. Polym. Sci. Polym. Chem. Ed., 18, 2641. 22. Atik, S. S. and Thomas, K. 1981. J. Am. Chem. Soc., 103, 4279. 23. Guo, G. S., El-Asser, M. S., and Vanderhoff, J. W. 1989. ACS Symp. Ser., 384, 86. 24. P’erez-Luna, V. H., Puig, J. E., Castano, V. M., Rodriguez, B. E., Murthy, A. K., and Kaler, E. W. 1990. Langmuir, 6, 1040. 25. Leong, Y. S. and Candau, F. 1982. J. Phys. Chem., 86, 2269. 26. Candau, F., Leong, Y. S., and Fitch, R. M. 1985. J. Polym. Sci. Polym. Chem. Ed., 23, 193. 27. Gan, L. M., Chew, C. H., Lim, J. H., Lee, K. S., and Gan, L. H. 1994. Colloid. Polym. Sci., 272, 1082. 28. Pinto Reis, C., Neufeld, R. J., Rebeiro, A. J., and Veiga, F. 2006. Nanomedicine, 2, 1, 6. 29. Levenspiel, O. 1979. The Chemical Reactor Omnibus, OSU Book Stores, Corvalis, OR. 30. Flory, P. J. 1944. Chem. Rev., 35, 51. 31. Nielsen, L. C. 1969. J. Macromol. Sci. Rev. Macromol. Chem. C369. 32. Boyer, R. F. 1963. Rubber Chem. Tech., 36, 1303. 33. Boyer, R. F. 1954. J. Appl. Phys., 25, 825. 34. Beamen, R. G. 1952. J. Polym. Sci., 9, 470. 35. Mark, H. F. 1967. Sci. Am., 217(3), Sept. 19, 148, see also p. 156. 36. Overberger, C. G. and Moore, J. A. 1970. Fortschr. Hochpolym. Forsch., 7, 113. 37. Hildebrand, J. H., Prausnitz, J. M., and Scott, R. L. 1970. Regular and Related Solutions, Van Nostrand Reinhold, New York. 38. Burrell, H. 1955. Ofﬁcial Dig., 27, 726. 39. Nelson, R. C., Hemwall, R. W., and Edwards, G. D. 1970. J. Paint Technol., 42, 636. 40. Small, P. A. 1953. J. Appl. Chem., 3, 71. 41. Hoy, K. L. 1970. J. Paint Technol., 42, 76. 42. Brandrup, J. and Immergut, E., eds. 1975. Polymer Handbook, p. 251. Wiley Interscience, New York. 43. Burrell, H. 1975. Solubility parameter values. In Polymer Handbook, 2nd Ed., J. Brandrup and E. Immergut, eds., p. 251. Wiley Interscience, New York, pp. IV-337–359. 44. Krause, S. 1972. J. Macromol. Sci. Macromol. Rev., C7, 251. 45. Hansen, C. M. 1967. J. Paint Technol., 39, 104, see also p. 511. 46. Barton, A. F. M. 1983. Handbook of Solubility Parameters and Other Cohesion Parameters. CRC Press, Boca Raton, FL. Characteristics of Polymers and Polymerization Processes 159 47. Hansen, C. M. and Beerbower, A. 1971. Kirk Othmer Encycl. Chem. Technol., 2nd Ed., A. Standen, ed., Suppl. 889, 910, Wiley Interscience, New York. 48. Teas, J. P. 1968. J. Paint Technol., 40, 519. 49. Bagley, E. B., Nielsen, T. P., and Scigliano, J. M. 1971. J. Paint Technol., 43, 35. 50. Anon. 1968. Plastics World, 26, 3, 30. 51. Lee, H., Stoffey, D., and Neville, K. 1967. New Linear Polymers, McGraw-Hill, New York. 52. Pielichowski, K. and Njuguna, J. 2005. Thermal Degradation of Polymeric Materials. Rapra Technology, Shrewsbury, Shropshire, UK. 53. Bhuiyan, A. L. 1982. Adv. Polym. Sci., 47, 43. 54. Schmidt, D. L. Mod. Plastics, 37, 131 (Nov.1960), 147 (Dec. 1960). 55. Kuryla, W. C. and Papa, A. J., eds. 1973. Flame Retardancy of Polymeric Materials, 3, p. 129. Marcel Dekker, New York. 56. Halim Hamid, S., Amin, M. B., and Maadhah, A. G. eds. 1992. Handbook of Polymer Degradation, p. 129. Marcel Dekker, New York. 57. Rabek, J. F. 1996. Photodegradation of Polymers, Springer, Berlin-Heidelberg, New York. 58. Rivaton, A., Gardette, J.-L., Mailhot, B., and Morlat-Therlas, S. 2005. Macromol. Symp., 225, 129. 59. Grassie, N. and Scott, G. 1985. Polymer Degradation and Stabilization. Cambridge Univ. Press, Cambridge. 60. Hawkins, W. L. 1972. Polymer Stabilization. Wiley, New York. 61. Gächter, R. and Müller, H. eds. 1987. Plastics Additives Handbook, p. 89. Hanser Publishers, Munich/ New York. 62. Henman, T. J. 1982. World Index of Polyoleﬁn Stabilizers, Kogan Page Ltd., London, UK. 63. Ranby, B. and Rabek, J. F. 1975. Photodegradation, Photooxidation, and Photostabilization of Polymers, Interscience, New York. 64. Massey, L. 2002. Permeability Properties of Plastics and Elastomers, 2nd Ed. ChemTec Publishing, Toronto-Scarborough. 65. Stannett, V. T. and Szwarc, M. 1955. J. Polym. Sci., 16, 89. 66. Paine, F. A. 1962. J. Roy. Inst. Chem., 86, 263. 67. Seymour, R. B. 1978. Additives for Plastics., Vol. 2., Academic Press, New York. 68. Stepak, J. and Daoust, H. 1982. Additives for Plastics, Springer, New York. 69. Wypych, G. 2004. Handbook of Plasticizers, ChemTec Publishing, Toronto-Scarborough. 70. Dufton, P. W. 2003. Flame Retardants for Plastics, ChemTec Publishing, Toronto-Scarborough. 71. Camino, G. and Delobel, R. 2000. Fire Retardancy of Polymeric Materials, A. F. Grand and C. A. Wilke, eds., p. 217. Marcel Dekker, New York. 72. Baljinder, K. K. and Horrocks, A. R. 1996. Polym. Degrad. Stab., 54, 289. 73. Lewin, M. and Endo, M. 2003. Polym. Adv. Technol., 14, 3. 74. Chen, X., Ding, Y., and Tang, T. 2005. Polym. Int., 54, 904. 75. Soderholm, K. and Mariotti, A. 1999. J. Am. Dent. Assoc., 130, 201. 76. Fung, E. Y. K., Ewoldsen, N. O., St. Germain, H. S., Marx, D. B., Miaw, C., Siew, C., Chou, H., Grunniger, S. E., and Meyer, D. M. 2000. J. Am. Dent. Assoc., 131, 51. 77. Kreuter, J. 1991. J. Controlled Release, 16, 169. 78. Maincent, P., Le Verge, R., Sado, P., Couvreur, P., and Devissaguet, J. P. 1986. J. Pharma. Sci., 75, 955. 79. Couvreur, P., Grislain, L., Lenaerts, V., Brasseur, P., Guiot, P., and Biernacki, A. 1986. Polymeric Nanoparticles and Microspheres, P. Guiot and P. Couvreir, eds., pp. 27–93. CRC Press Inc., Boca Raton, FL. 80. Zambaux, M. F., Bonneaux, F., Gref, R., Maincent, P., Dellacherie, E., Alonso, M. J., Labrude, P., and Vigneron, C. 1998. J. Controlled Release, 50, 31. 81. Hármia, H., Speiser, P., and Kreuter, J. 1986. J. Microencapsul., 3, 3. 82. Alle´man, E. Gumy, E., and Doelkar, E. 1992. Int. J. Pharm., 87, 247. 83. Fessi, H., Puisieux, F., Devissaguet, J.-P., Ammoury, N., and Benita, S. 1989. Int. J. Pharm., 55, R1. 160 Plastics Technology Handbook 84. Chacon, M., Berges, L., Molpaceres, J., Aberturass, M. R., and Guzman, M. 1996. Int. J. Pharm., 141, 81. 85. Chacon, M., Berges, L., Molpaceres, J., Aberturass, M. R., and Guzman, M. 1996. Int. J. Pharm., 141, 81. 86. Siggs, L. J. and Mikos, A. G. 1996. Synthetic biodegradable polymers for medical applications. In Physical Properties of Polymers Handbook, J. E. Mark, ed., Chap. 44, Am. Inst. of Physics, Woodbury, New York. 87. Thomson, R. C., Wake, M. C., Yaszemski, M. J., and Mikos, A. G. 1995. Adv. Polym. Sci., 122, 245. 88. Autian, J. 1970. J. Fire Flammability, 1, 239. 89. American Dental Association. 1972. J. Am. Dent. Assoc., 84, 382. 2 Fabrication Processes 2.1 Types of Processes As indicated in Chapter 1, the family of polymers is extraordinarily large and varied. There are, however, some fairly broad and basic approaches that can be followed when designing or fabricating a product out of polymers or, more commonly, polymers compounded with other ingredients. The type of fabrication process to be adopted depends on the properties and characteristics of the polymer and on the shape and form of the ﬁnal product. In the broad classiﬁcation of plastics there are two generally accepted categories: thermoplastic resins and thermosetting resins. Thermoplastic resins consist of long polymer molecules, each of which may or may not have side chains or groups. The side chains or groups, if present, are not linked to other polymer molecules (i.e., are not cross-linked). Thermoplastic resins, usually obtained as a granular polymer, can therefore be repeatedly melted or solidiﬁed by heating or cooling. Heat softens or melts the material so that it can be formed; subsequent cooling then hardens or solidiﬁes the material in the given shape. No chemical change usually takes place during this shaping process. In thermosetting resins the reactive groups of the molecules from cross-links between the molecules during the fabrication process. The cross-linked or “cured” material cannot be softened by heating. Thermoset materials are usually supplied as a partially polymerized molding compound or as a liquid monomer–polymer mixture. In this uncured condition they can be shaped with or without pressure and polymerized to the cured state with chemicals or heat. With the progress of technology the demarcation between thermoplastic and thermoset processing has become less distinct. For thermosets processes have been developed which make use of the economic processing characteristics or thermoplastics. For example, cross-linked polyethylene wire coating is made by extruding the thermoplastic polyethylene, which is then cross-linked (either chemically or by irradiation) to form what is actually a thermoset material that cannot be melted again by heating. More recently, modiﬁed machinery and molding compositions have become available to provide the economics of thermoplastic processing to thermosetting materials. Injection molding of phenolics and other thermosetting materials are such examples. Nevertheless, it is still a widespread practice in industry to distinguish between thermoplastic and thermosetting resins. Compression and transfer molding are the most common methods of processing thermosetting plastics. For thermoplastics, the more important processing techniques are extrusion, injection, blow molding, and calendaring; other processes are thermoforming, slush molding, and spinning. 2.2 Tooling for Plastics Processing Tooling for plastics processing deﬁnes the shape of the part. It falls into two major categories, molds and dies. A mold is used to form a complete three-dimensional plastic part. The plastics processes that use molds are compression molding, injection molding, blow molding, thermoforming, and reaction injection 161 162 Plastics Technology Handbook molding (RIM). A die, on the other hand, is used to form two of the three dimensions of a plastic part. The third dimension, usually thickness or length, is controlled by other process variables. The plastics processes that use dies are extrusion, pultrusion and thermoforming. Many plastics processes do not differentiate between the terms mold and die. Molds, however, are the most predominant form of plastics tooling. 2.2.1 Types of Molds The basic types of molds, regardless of whether they are compression, injection, transfer, or even blow molds, are usually classiﬁed by the type and number of cavities they have. For example, Figure 2.1 illustrates three mold types: (a) single-cavity, (b) dedicated multiple-cavity, and (c) family multiple-cavity. Single-cavity mold (Figure 2.1a) represents one of the simplest mold concepts. This design lends itself to low-volume production and to large plastic part designs. The multiple-cavity molds may be of two types. A dedicated multiple-cavity mold (Figure 2.1b) has cavities that produce the same part. This type of mold is very popular because it is easy to balance the plastic ﬂow and establish a controlled process. In a family multiple-cavity mold (Figure 2.1c), each cavity may produce a different part. Historically, family mold designs were avoided because of difﬁculty in ﬁlling uniformly; however, recent advances in mold making and gating technology make family molds appealing. This is the case especially when a processor has a multiple-part assembly and would like to keep inventories balanced. 2.2.2 Types of Dies Within the plastics industry, the term die is most often applied to the processes of extrusion (see “Extrusion”). Extrusion dies may be categorized by the type of product being produced (e.g., ﬁlm, sheet, proﬁle, or coextrusion), but they all have some common features as described below. 1. Steel. The extrusion process being continuous, both erosion and corrosion are signiﬁcant factors. Hence the dies must be made of a high-quality tool steel, hardened so that the areas that contact the (a) (b) (c) FIGURE 2.1 Three basic types of molds. (a) single-cavity; (b) dedicated multiple-cavity; (c) family multiple-cavity. Fabrication Processes 2. 3. 4. 5. 163 plastic material do not erode. Additionally, many dies have a dense, hard chrome plating in the area where plastic melt contacts the die. Heaters. Extrusion dies are to be heated in order to maintain a melt ﬂow condition for the plastic material. Most of the heaters are cartridge-type elements that slip ﬁt into the die at particular locations. In addition to the heaters, the dies have to accommodate temperature sensors, such as thermocouples. Melt Pressure. Many sophisticated dies are equipped with sensors that monitor melt pressure. This allows the processor to better monitor ad control the process. Parting Line. Large extrusion dies must be able to separate at the melt ﬂow line for easier fabrication and maintenance. Smaller extrusion dies may not have a parting area, because they can be constructed in one piece. Die Swell Compensation. The polymer melt swells when it exits the die, as explained previously. This die swell is a function of the type of plastic material, the melt temperature, the melt pressure, and the die conﬁguration. The die must be compensated for die swell so that the extruded part has the corrected shape and dimensions. Molds and dies for different fabrication processes will be described later in more detail when the processes are discussed. 2.2.3 Tool Design The design of the tooling to produce a speciﬁc plastics part must be considered during the design of the part itself. The tool designer must consider several factors that may affect the fabricated part, such as the plastics material, shrinkage, and process equipment. Additionally, competitive pressures within the plastics industry require the tool designer to consider how to facilitate tool changeovers, optimize tool maintenance, and simplify (or eliminate) secondary operations. Historically, plastics molds and dies were built by toolmakers who spent their lives learning and perfecting their craft. Today the void created by the waning numbers of these classically trained toolmakers is being ﬁlled by the development of numerically controlled (NC) machinery centers, computer-based numerically controlled (CNC) machinery centers, and computer-aided design (CAD) systems. Molds and dies can now be machined on computer-controlled mills, lathes, and electric discharge machines that require understanding of computers and design, rather than years of experience and machining skills. The quality of tool components is now more a function of the equipment than of the toolmaker’s skill. The high costs of molds and the fact that many production molds are built under extreme time constraints leave no room for trial and error. Though prototyping has been widely used to evaluate smaller part designs when circumstances and time allow, prototyping is not always feasible for larger part designs. There are, however, several alternatives to prototyping, e.g., CAD, ﬁnite-element analysis (FEA), and rapid prototyping. While CAD allows a tool designer to work with a three-dimensional computer model of the tool being designed and to analyze the design, FEA allows the tool to be evaluated (on a computer) for production worthiness. The mold is then fabricated from the computer model, a process called computeraided manufacturing (CAM). Rapid prototyping is a relatively new method of producing a plastics part by using a three-dimensional computer drawing. A sophisticated prototyping apparatus interprets the drawing and guides an articulating laser beam across a speciﬁc medium such as a photopolymer plastic or laminated paper, the result being a physical representation of the computer-based drawing. Prototyped parts can be produced in less than 24 h, and part designs can be scaled to ﬁt the size of the prototyping equipment. Another trend is the introduction of molds that accept interchangeable modules. Modules take less time to manufacturing, and in turn, cut down on the delivery time and costs. In addition, it usually takes less time to change the module than the entire mold frame. 164 Plastics Technology Handbook 2.3 Compression Molding Compression molding is the most common method by which thermosetting plastics are molded [1–3]. In this method the plastic, in the form of powder, pellet, or disc, is dried by heating and then further heated to near the curing temperature; this heated charge is loaded directly into the mold cavity. The temperature of the mold cavity is held at 150°C–200°C, depending on the material. The mold is then partially closed, and the plastic, which is liqueﬁed by the heat and the exerted pressure, ﬂows into the recess of the mold. At this stage the mold is fully closed, and the ﬂow and cure of the plastic are complete. Finally, the mold is opened, and the completely cured molded part is ejected. Compression-molding equipment consists of a matched mold, a means of heating the plastic and the mold, and some method of exerting force on the mold halves. For severe molding conditions molds are usually made of various grades of tool steel. Most are polished to improve material ﬂow and overall part quality. Brass, mild steel, or plastics are used as mold materials for less severe molding conditions or shortrun products. In compression molding a pressure of 2,250 psi (158 kg/cm2)–3,000 psi (211 kg/cm2) is suitable for phenolic materials. The lower pressure is adequate only for an easy-ﬂow materials and a simple uncomplicated shallow molded shape. For a medium-ﬂow material and where there are average-sized recesses, cores, shapes, and pins in the molding cavity, a pressure of 3,000 psi (211 kg/cm2) or above is required. For molding urea and melamine materials, pressures of approximately one and one-half times that needed for phenolic material are necessary. The time required to harden thermosetting materials is commonly referred to as the cure time. Depending on the type of molding material, preheating temperature, and the thickness of the molded article, the cure time may range from seconds to several minutes. In compression molding of thermosets the mold remains hot throughout the entire cycle; as soon as a molded part is ejected, a new charge of molding powder can be introduced. On the other hand, unlike thermosets, thermoplastics must be cooled to harden. So before a molded part is ejected, the entire mold must be cooled, and as a result, the process of compression molding is quite slow with thermoplastics. Compression molding is thus commonly used for thermosetting plastics such as phenolics, urea, melamine, an alkyds; it is not ordinarily used for thermoplastics. However, in special cases, such as when extreme accuracy is needed, thermoplastics are also compression molded. One example is the phonograph records of vinyl and styrene thermoplastics; extreme accuracy is needed for proper sound reproduction. Compression molding is ideal for such products as electrical switch gear and other electrical parts, plastic dinnerware, radio and television cabinets, furniture drawers, buttons, knobs, handles, etc. Like the molding process itself, compression molding machinery is relatively simple. Most compression presses consist of two platens that close together, applying heat and pressure to the material inside a mold. The majority of the presses are hydraulically operated with plateau ranging in size from 6 in. square to 8 ft square or more. The platens exert pressures ranging from 6 up to 10,000 tons. Virtually all compression molding presses are of vertical design. Most presses having tonnages under 1000 are upward-acting, while most over 1,000 tons act downward. Some presses are built with a shuttle-clamp arrangement that moves the mold out of the clamp section to facilitate setup and part removal. Compression molds can be divided into hand molds, semiautomatic molds, and automatic molds. The design of any of these molds must allow venting to provide for escape of steam, gas, or air produced during the operation. After the initial application of pressure the usual practice is to open the mold slightly to release the gases. This procedure is known as breathing. Hand molds are used primarily for experimental runs, for small production items, or for molding articles which, because of complexity of shape, require dismantling of mold sections to release them. Semiautomatic molds consist of units mounted ﬁrmly on the top and bottom platens of the press. The operation of the press closes and opens the mold and actuates the ejector system for removal of the molded article. However, an operator must load the molding material, actuate press controls for the molding sequence, and remove the ejected piece from the mold. This method is widely used. Fabrication Processes 165 Fully automatic molds are specially designed for adaptation to a completely automatic press. The entire operation cycle, including loading and unloading of the mold, is performed automatically, and all molding operations are accurately controlled. Thermosetting polymers can be molded at rates up to 450 cycles/h. Tooling must be of the highest standard to meet the exacting demands of high-speed production. Automatic molds offer the most economical method for long production runs because labor costs are kept to a minimum. The three common types of mold designs are open ﬂash, fully positive, and semipositive. 2.3.1 Open Flash In an open ﬂash mold a slight excess of molding powder is loaded into the mold cavity (Figure 2.1a) [4]. On closing the top and bottom platens, the excess material is forced out and ﬂash is formed. The ﬂash blocks the plastic remaining in the cavity and causes the mold plunger to exert pressure on it. Gas or air can be trapped by closing the mold too quickly, and ﬁnely powdered material can be splashed out of the mold. However, if closing is done carefully, the open ﬂash mold is a simple one, giving very good results. Since the only pressure on the material remaining in the ﬂash mold when it is closed results from the high viscosity of the melt which did not allow it to escape, only resins having high melt viscosities can be molded by this process. Since most rubbers have high melt viscosities, the ﬂash mold is widely used for producing gaskets and grommets, tub and ﬂash stoppers, shoe heels, door mats, and many other items. Because of lower pressure exerted on the plastic in the ﬂash molds, the molded products are usually less dense than when made using other molds. Moreover, because of the excess material loading needed, the process is somewhat wasteful as far as raw materials are concerned. However, the process has the advantage that the molds are cheap, and very slight labor costs are necessary in weighing out the powder. 2.3.2 Fully Positive In the fully positive molds (Figure 2.2b) no allowance is made for placing excess powder in the cavity [4]. If excess powder is loaded, the mold will not close; an insufﬁcient charge will result in reduced thickness of the molded article. A correctly measured charge must therefore be used with this mold—it is a disadvantage of the positive mold. Another disadvantage is that the gases liberated during the chemical curing reaction are trapped inside and may show as blisters on the molded surface. Excessive wear on the sliding ﬁt surface on the top and bottom forces and the difﬁculty of ejecting the molding are other reasons for discarding this type of mold. The mold is used on a small scale for molding thermosets, laminated plastics, and certain rubber components. 2.3.3 Semipositive The semipositive mold (Figure 2.2c and d) combines certain features of the open ﬂash and fully positive molds and makes allowance for excess powder and ﬂash [4]. It is also possible to get both horizontal and vertical ﬂash. Semipositive molds are more expensive to manufacture and maintain than the other types, but they are much better from an applications point of view. Satisfactory operation of semipositive molds is obtained by having clearance (0.025/25 mm of diameter) between the plunger (top force) and the cavity. Moreover, the mold is given a 2–3° taper on each side. This allows the ﬂash to ﬂow on and the entrapped gases to escape along with it, thereby producing a clean, blemish-free mold component. 2.3.4 Process Applicability Compression molding is most cost-effective when used for short-run parts requiring close tolerances, high-impact strength, and low mold shrinkage. Old as the process may be, new applications continue to 166 Plastics Technology Handbook Plunger Plunger Land Parting line Cavity depth Molded piece Clearance Loading space Depth of cavity Cavity Knockout pins (b) (a) Knockout pins Cavity Plunger a b a a Land a Knockout pin (c) Positive portion Cavity well Cavity (d) FIGURE 2.2 Compression molds. (a) A simple ﬂash mold. (b) A positive mold. Knockout pins could extend through plunger instead of through cavity. (c) Semi-positive mold as it appears in partly closed position before it becomes positive. Material trapped in area b escapes upward. (d) Semipositive mold in closed position. evolve compression molding. For example, in the dental and medical ﬁelds, orthodontic retainers, and pacemaker casings are now mostly compression molded because of low tool costs. Injection molding tools to produce the same part would cost as much as eight times more. Manufacturers of gaskets and seals who started out with injection-molded products to take advantage of the faster cycle times, are now switching back to compression molding to maintain quality level required for these parts. The use of compression molding has expanded signiﬁcantly in recent years due to the development of new materials, reinforced materials in particular. Molding reinforced plastics (RPs) requires two matched dies usually made of inexpensive aluminum, plastics, or steel and used on short runs. Adding vacuum chambers to compression molding equipment in recent years has reduced the number of defects caused by trapped air or water in the molding compound, resulting in higher-quality ﬁnished parts. Another relatively new improvement has been the addition of various forms of automation to the process. For example, robots are used both to install inserts and remove ﬁnished parts. 2.4 Transfer Molding In transfer molding, the thermosetting molding powder is placed in a chamber or pot outside the molding cavity and subjected to heat and pressure to liquefy it [1–6]. When liquid enough to start ﬂowing, the material is forced by the pressure into the molding cavity, either by a direct sprue or though a system of runners and gates. The material sets hard to the cavity shape after a certain time (cure time) has elapsed. When the mold is disassembled, the molded part is pushed out of the mold by ejector pins, which operate automatically. Figure 2.3 shows the molding cycle of pot-type transfer molding, and Figure 2.4 shows plunger-type transfer molding (sometime called auxiliary raw transfer molding). The taper of the sprue is pot-type transfer is such that, when the mold is opened, the sprue remains attached to the disc of material left in the pot, known as cull, and is thus pulled away from the molded part, whereas the latter is lifted out of the cavity by the ejector pins (Figure 2.3c). In plunger-type transfer molding, on the other hand, the cull and the sprue remains with the molded piece when the mold is opened (Figure 2.4c). 167 Fabrication Processes Cull Sprue Molding compound Sprue bush Molded part (a) (b) (c) FIGURE 2.3 Molding cycle of a pot-type transfer mold. (a) Molding compound is placed in the transfer pot and then (b) forced under pressure when hot through an oriﬁce and into a closed mold. (c) When the mold opens, the sprue remains with the cull in the pot, and the molded part is lifted out of the cavity by ejector pins. (After Frados, J. ed. 1976. Plastics Engineering Handbook, 4th Ed., Van Nostrand Reinhold, New York.) Another variation of transfer molding in screw transfer molding (Figure 2.5). In this process the molding material is preheated and plasticized in a screw chamber and dropped into the pot of an inverted plunger mold. The preheated molding material is then transferred into the mold cavity by the same method as shown in Figure 2.4. The screw-transfer-molding technique is well suited to fully automatic operation. The optimum temperature of a phenolic mold charge is 240°F ± 20°F (155°C ± 11°C), the same as that for pot-transfer and plunger molding techniques. For transfer molding, generally pressures of three times the magnitude of those required for compression molding are required. For example, usually a pressure of 9,000 psi (632 kg/cm2) and upward is required for phenolic molding material (the pressure referred to here is that applied to the powder material in the transfer chamber). The principle of transferring the liqueﬁed thermosetting material from the transfer chamber into the molding cavity is similar to that of the injection molding of thermoplastics (described later). Therefore the same principle must be employed for working out the maximum area which can be molded—that is, the projected area of the molding multiplied by the pressure generated by the material inside the cavity must be less than the force holding the two halves together. Otherwise, the molding cavity plates will open as the closing force is overcome. Transfer molding has an advantage over compression molding in that the molding powder is ﬂuid when it enters the mold cavity. The process therefore enables production of intricate parts and molding around thin pins and metal inserts (such as an electrical lug). Thus, by transfer molding, metal inserts can be molded into the component in predetermined positions held by thin pins, which would, however, bend or break under compression-molding conditions. Typical articles made by the transfer molding process are terminal-bloc insulators with many metal inserts and intricate shapes, such as cups and caps for cosmetic bottles. 168 Plastics Technology Handbook Transfer ram Molding compound Molded part Cull (a) (b) (c) FIGURE 2.4 Molding cycle of a plunger-type transfer mold. (a) An auxiliary ram exerts pressure on the heatsoftened material in the pot and (b) forces it into the mold. (c) When the mold is opened, the cull and sprue remain with the molded piece. (After Frados, J. ed. 1976. Plastics Engineering Handbook, 4th Ed., Van Nostrand Reinhold, New York.) 2.4.1 Ejection of Molding Ejection of a molded plastic article from a mold can be achieved by using ejector pins, sleeves, or stripper plates. Ejector pins are the most commonly used method because they can be easily ﬁtted and replaced. The ejector pins must be located in position where they will eject the article efﬁciently without causing distortion of the part. They are worked by a common ejector plate or a bar located under the mold, and operated by a central hydraulic ejector ram. The ejector pins are ﬁtted either to the bottom force or to the top force depending on whether it is necessary for the molding to remain in the bottom half of the female part or on the top half of the male part of the tool. The pins are usually constructed of a hardened steel to avoid wear. 2.4.2 Heating System Heating is extremely important in plastics molding operations because the tool and auxiliary parts must be heated to the required temperature, depending on the powder being molded, and the temperature must be maintained throughout the molding cycle. The molds are heated by steam, hot waters, and induction heaters. Steam heating is preferred for compression and transfer molding, although electricity is also used because it is cleaner and has low installation costs. The main disadvantage of the latter method is that the heating is not fully even, and there is tendency to form hot spots. 169 Fabrication Processes Clamping ram Molded piece Hopper Mold halves Drive motor Plasticizer screw Heater Transfer ram FIGURE 2.5 Drawing of a screw-transfer molding machine. (After Frados, J. ed. 1976. Plastics Engineering Handbook, 4th Ed., Van Nostrand Reinhold, New York.) 2.4.3 Types of Presses Presses used for compression and transfer molding of thermosets can be of many shapes and designs, but they can be broadly classiﬁed as hand, mechanical, or hydraulic types. Hand presses have relatively lower capacity, ranging from 10 to 100 tons, whereas hydraulic presses have considerably higher capacity (500 tons). Hydraulic presses may be of the upstroke or downstroke varieties. In the simple upstroke press, pressure can be applied fairly quickly, but the return is slow. In the downstroke press ﬁtted with a preﬁlling tank, this disadvantage of the upstroke press is removed, and a higher pressure is maintained by preﬁlling with liquid from a tank. The basic principles of hydraulics are used in the presses. Water or oil is used as the main ﬂuid. Water is cheap but rusts moving parts. Oil is more expensive but it does not corrode and it does lubricate moving parts. The main disadvantage of oil is that it tends to form sludge due to oxidation with air. The drive for the presses is provided by single pumps or by central pumping stations, and accumulators are used for storing energy to meet instantaneous pressure demand in excess of the pump delivery. The usual accumulator consists of a single-acting plunger working in a cylinder. The two main types of accumulators used are the weight-loaded type and the air-loaded type. The weight-loaded type is heavy and therefore not very portable. There is also an initial pressure surge on opening the valve. The pressuresurge problem is overcome in the air- or gas (nitrogen)-loaded accumulator. This type is more portable but suffers a small pressure loss during the molding cycle. 2.4.4 Preheating To cut down cycle times and to improve the ﬁnished product of compression molding and transfer molding, the processes of preheating and performing are commonly used. With preheating, relatively thick sections can be molded without porosity. Other advantages of the technique include improved ﬂow of resin, lower molding pressures, reduced mold shrinkage, and reduced ﬂash. 170 Plastics Technology Handbook Preheating methods are convection, infrared, radio frequency, and steam. Thermostatically controlled gas or electrically heated ovens are inexpensive methods of heating. The quickest, and possibly the most efﬁcient, method is radio-frequency heating, but it is also the most expensive. Preheaters are located adjacent to the molding press and are manually operated for each cycle. 2.4.5 Preforming Preforming refers to the process of compressing the molding powder into the shape of the mold before placing it in the mold or to pelleting, which consists of compacting the molding powder into pellets of uniform size and approximately known weight. Preforming has many advantages, which include avoiding waste, reduction in bulk factor, rapid loading of charge, and less pressure than uncompacted material. Preformers are basically compacting presses. These presses may be mechanical, hydraulic, pneumatic, or rotary cam machines. 2.4.6 Flash Removal Although mold design takes into consideration the fact that ﬂash must be reduced to a minimum, it still occurs to some extent on the molded parts. It is thus necessary to remove the ﬂash subsequent to molding. This removal is most often accomplished with tumbling machines. These machines tumble molded parts against each other to break off the ﬂash. The simplest tumbling machines are merely wire baskets driven by an electric motor with a pulley belt. In more elaborate machines blasting of molded parts is also performed during the tumbling operation. 2.5 Injection Molding of Thermoplastics Injection molding is the most important molding method for thermoplastics [7–9]. It is based on the ability of thermoplastic materials to be softened by heat and to harden when cooled. The process thus consists essentially of softening the material in a heated cylinder and injecting it under pressure into the mold cavity, where it hardens by cooling. Each step is carried out in a separate zone of the same apparatus in the cyclic operation. A diagram of a typical injection-molding machine is shown in Figure 2.6. Granular material (the plastic resin) falls from the hopper into the barrel when the plunger is withdrawn. The plunger then pushes the material into the heating zone, where it is heated and softened (plasticized or plasticated). Rapid heating Plunger Back pressure plate Clamp pressure Hopper Gating Torpedo Barrel Clamp cylinder Mold Nozzle Runner Mold cavity Ram pressure Resin Heater Injection chamber Cooling zone Hydraulic cylinder FIGURE 2.6 Cross section of a typical plunger injection-molding machine. (After Lukov, L. J. 1963. SPE J., 13(10), 1057.) 171 Fabrication Processes takes place due to spreading of the polymer into a thin ﬁlm around a torpedo. The already molten polymer displaced by this new material is pushed forward through the nozzle, which is in intimate contact with the mold. The molten polymer ﬂows through the sprue opening in the die, down the runner, past the gate, and into the mold cavity. The mold is held tightly closed by the clamping action of the press platen. The molten polymer is thus forced into all parts of the mold cavities, giving a perfect reproduction of the mold. The material in the mold must be cooled under pressure below Tm or Tg before the mold is opened and the molded part is ejected. The plunger is then withdrawn, a fresh charge of material drops down, the mold is closed under a locking force, and the entire cycle is repeated. Mold pressures of 8,000–40,000 psi (562–2,812 kg/cm2) and cycle times as low as 15 sec are achieved on some machines. Note that the feed mechanism of the injection molding machine is activated by the plunger stroke. The function of the torpedo in the heating zone is to spread the polymer melt into thin ﬁlm in close contact with the heated cylinder walls. The ﬁns, which keep the torpedo centered, also conduct heat from the cylinder walls to the torpedo, although in some machines the torpedo is heated separately. Injection-molding machines are rated by their capacity to mold polystyrene in a single shot. Thus a 2-oz machine can melt and push 2 oz of general-purpose polystyrene into a mold in one shot. This capacity is determined by a number of factors such as plunger diameter, plunger travel, and heating capacity. The main component of an injection-molding machine are (1) the injection unit which melts the molding material and forces it into the mold; (2) the clamping unit which opens the mold and closes it under pressure; (3) the mold used; and (4) the machine controls. 2.5.1 Types of Injection Units Injection-molding machines are known by the type of injection unit used in them. The oldest type is the single-stage plunger unit (Figure 2.6) described above. As the plastic industry developed, another type of plunger machine appeared, known as a two-stage plunger (Figure 2.7a). It has two plunger units set one on top of the other. The upper one, also known as a preplasticizer, plasticizes the molding material and feeds it to the cylinder containing the second plunger, which operates mainly as a shooting plunger, and pushes the plasticized material through the nozzle into the mold. Later, another variation of the two-stage plunger unit appeared, in which the ﬁrst plunger stage was replaced by a rotation screw in a cylinder (Figure 2.7b). The screw increases the heat transfer at the walls Preplasticizing cylinder Torpedo Heaters Valve Screw Preplasticizing plunger Heaters Shooting plunger 3-Way (a) valve Injection chamber (b) Injection chamber Shooting plunger FIGURE 2.7 Schematic drawings of (a) a plunger-type preplasticizer and (b) a screw-type preplasticizer atop a plunger-type injection molding machine. (After Lukov, L. J. 1963. SPE J., 13(10), 1057.) 172 Plastics Technology Handbook Hopper Injection chamber Heating (a) Sprue Cylinder Heater (b) FIGURE 2.8 Cross section of a typical screw-injection molding machine, showing the screw (a) in the retracted position and (b) in the forward position. (After Lukov, L. J. 1963. SPE J., 13(10), 1057.) and also does considerable heating by converting mechanical energy into heat. Another advantage of the screw is its mixing and homogenizing action. The screw feeds the melt into the second plunger unit, where the injection ram pushes it forward into the mold. Although the single-stage plunger units (Figure 2.6) are inherently simple the limited heating rate has caused a decline in popularity: they have been mostly supplanted by the reciprocating screw-type machines [10]. In these machines (Figure 2.8) the plunger and torpedo (or spreader) that are the key components of plunger-type machines are replaced by a rotating screw that moves back and forth like a plunger within the heating cylinder. As the screw rotates, the ﬂights pick up the feed of granular material dropping from the hopper and force it along the heated wall of the barrel, thereby increasing the rate of heat transfer and also generating considerable heat by its mechanical work. The screw, moreover, promotes mixing and homogenization of the plastic material. As the molten plastic comes off the end of the screw, the screw moves back to permit the melt to accumulate. At the proper time the screw is pushed forward without rotation, acting just like a plunger and forcing the melt through the nozzle into the mold. The size of the charge per shot is regulated by the back travel of the screw. The heating and homogenization of the plastics material are controlled by the screw rotation speed and wall temperatures. 2.5.2 Clamping Units The clamping unit keeps the mold closed while plasticized material is injected into it and opens the mold when the molded article is ejected. The pressure produced by the injection plunger in the cylinder is transmitted through the column of plasticized material and then through the nozzle into the mold. The Fabrication Processes 173 unlocking force, that is, the force which tends to open the mold, is given by the product of the injection pressure and the projected area of the molding. Obviously, the clamping force must be greater than the unlocking force to hold the mold halves closed during injection. Several techniques can be used for the clamping unit: (1) hydraulic clamps, in which the hydraulic cylinder operates on the movable parts of the mold to open and close it; (2) toggle or mechanical clamps, in which the hydraulic cylinder operates through a toggle linkage to open and close the mold; and (3) various types of hydraulic mechanical clamps that combine features of (1) and (2). Clamps are usually built as horizontal units, with injection taking place through the center of the stationary platen, although vertical clamp presses are also available for special jobs. 2.5.3 Molds The mold is probably the most important element of a molding machine. Although the primary purpose of the mold is to determine the shape of the molded part, it performs several other jobs. It conducts the hot plasticized material from the heating cylinder to the cavity, vents the entrapped air or gas, cools the part until it is rigid, and ejects the part without leaving marks or causing damage. The mold design, construction, the craftsmanship largely determine the quality of the part and its manufacturing cost. The injection mold is normally described by a variety of criteria, including (1) number of cavities in the mold; (2) material of construction, e.g., steel, stainless steel, hardened steel, beryllium copper, chromeplated aluminum, and epoxy steel; (3) parting line, e.g., regular, irregular, two-plate mold, and three-plate mold; (4) method of manufacture, e.g., machining, hobbing, casting, pressure casting, electroplating, and spark erosion; (5) runner system, e.g., hot runner and insulated runner; (6) gating type, e.g., edge, restricted (pinpoint), submarine, sprue, ring, diaphragm, tab, ﬂash, fan, and multiple; and (7) method of ejection, e.g., knockout (KO) pins, stripper ring, stripper plate, unscrewing cam, removable insert, hydraulic core pull, and pneumatic core pull. 2.5.3.1 Mold Designs Molds used for injection molding of thermoplastic resins are usually ﬂash molds, because in injection molding, as in transfer molding, no extra loading space is needed. However, there are many variations of this basic type of mold design. The design most commonly used for all types of materials is the two plate design (Figure 2.9). The cavities are set in one plate, the plungers in the second plate. The sprue bushing is incorporated in that plate mounted to the stationary half of the mold. With this arrangement it is possible to use a direct center gate that leads either into a single-cavity mold or into a runner system for a multi-cavity mold. The plungers are ejector assembly and, in most cases, the runner system belongs to the moving half of the mold. This is the basic design of an injection mold, though many variations have been developed to meet speciﬁc requirements. A three-plate mold design (Figure 2.10) features a third, movable, plate which contains the cavities, thereby permitting center or offset gating into each cavity for multicavity operation. When the mold is opened, it provides two openings, one for ejection of the molded part and the other for removal of the runner and sprue. Moldings with inserts or threads or coring that cannot be formed by the normal functioning of the press require installation of separate or loose details or cores in the mold. These loose members are ejected with the molding. They must be separated from the molding and reinstalled in the mold after every cycle. Duplicate sets are therefore used for efﬁcient operation. Hydraulic or pneumatic cylinders may be mounted on the mold to actuate horizontal coring members. It is possible to mold angular coring, without the need for costly loose details, by adding angular core pins engaged in sliding mold members. Several methods may be used for unscrewing internal or external threads on molded parts: For high production rates automatic unscrewing may be done at relatively low cost by the use of rack-and-gear mechanism actuated by a double-acting hydraulic long-stroke cylinder. 174 Plastics Technology Handbook 1 3 2 5 4 6 7 8 10 11 9 19 12 18 16 13 14 17 15 FIGURE 2.9 A two-plate injection-mold design: (1) locating ring; (2) clamping plate; (3) water channels; (4) cavity; (5) sprue bushing; (6) cavity retainer; (7) gate; (8) full round runner; (9) sprue puller pin; (10) plunger; (11) parting line; (12) ejector pin; (13) stop pin; (14) ejector housing; (15) press ejector clearance; (16) pin plate; (17) ejector bar; (18) support plate; (19) plunger retainer. Runner and sprue Cavity plate Molded part Force FIGURE 2.10 A diagram of a three-plate injection mold. Fabrication Processes 175 Other methods of unscrewing involve the use of an electric gear-motor drive or friction-mold wipers actuated by double-acting cylinders. Parts with interior undercuts can be made in a mold which has provision for angular movement of the core, the movement being actuated by the ejector bar that frees the metal core from the molding. 2.5.3.2 Number of Mold Cavities Use of multiple mold cavities permits greater increase in output speeds. However, the greater complexity of the mold also increases signiﬁcantly the manufacturing cost. Note that in a single-cavity mold the limiting factor is the cooling time of the molding, but with more cavities in the mold the plasticizing capacity of the machine tends to be the limiting factor. Cycle times therefore do not increase prorate with the number of cavities. There can be no clear-cut answer to the question of optimum number of mold cavities, since it depends on factors such as the complexity of the molding, the size and type of the machine, cycle time, and the number of moldings required. If a fairly accurate estimate can be made of the costs and cycle time for molds with each possible number of cavities and a cost of running the machine (with material) is assumed, a break-even quantity of the number of moldings per hour can be calculated and compared with the total production required. 2.5.3.3 Runners A thermoplastic resin is melted in the barrel of the injection molding machine and is injected into the mold. The channels, or the pathways, through which the melted resin enters the gate areas of the mold cavities are called runners. There are two general types of plastic injection molds, deﬁned by the types of runners used, namely, hot or cold. In a hot runner mold, the runner is internal in the mold and, during operation, is kept constantly at a temperature above the melting point of the plastic so that the resin always remains in a liquid state within the tool except when it passes through the gate into the mold cavity. This means that the material in the channel, also referred to as runner, is not ejected with the ﬁnished part. Instead, it stays in the mold ready to ﬁll the cavity to make the next part. Hot runner molds thus eliminate runners entirely and the mold runs automatically, eliminating variations caused by operators. In a long-running job, hot runner molding (also known as “runnerless” molding) is the most economical way of molding as there is no regrinding (of runners) with its attendant cost of handling and loss of material. Hot runner molds consist of two plates and include a heated manifold and a number of heated nozzles. The manifold distributes the melted plastic to the various nozzles, which then meter it precisely to the injection points of the cavities. There are several types of hot runner systems. In general, they fall into two main categories, namely, externally heated and internally heated. The externally heated systems are well suited to polymers that are sensitive to thermal variations. The internally heated systems, however, allow better ﬂow control. The advantages of hot runner systems are as follows: (1) they eliminate runners and avoid potential waste; (2) they allow potentially shorter cycle time (i.e., time required to mold a part); (3) they are better for high-volume production; and (4) they can accommodate larger parts. The disadvantages are as follows: (1) they have more expensive molds; (2) they have higher maintenance costs; (3) the material and color cannot be changed easily; and (4) they may not be suited to certain types of engineering plastic resins and thermally sensitive thermoplastic resins. Cold runner molds usually consist of either two or three plates that are held within the mold base. The plastic is injected into the mold via the sprue and runners that lead to the parts in the cavity. In two-plate molds, the material in the runner system and the parts remain attached, and an ejection system is used to separate the pair from the mold. Two-plate molds have one parting line along which the mold is split into two halves. In three-plate molds, in contrast, the runner is contained on a separate plate, leaving the parts to be ejected alone. These molds have two parting lines. When a molded part is ejected, the mold splits into three 176 Plastics Technology Handbook sections. It is important that the runner dimension is thicker than the component because this ensures that the molten plastic can be packed into the component as it cools without any restriction. While in both two-plate and three-plate systems the runner is reground and recycled to reduce plastic waste, a three-plate cold runner system offers greater design ﬂexibility and allows gates to be installed according to application requirements. Tunnel gates (submarine gates, see below) are the most frequently used in combination with cold runner systems because trimming of the gate from the part takes place automatically. Typically, the action of either the mold opening or ejecting the part also removes the gate from the part. The main advantages of cold runner systems are as follows: (1) they are comparatively cheaper to produce and maintain; (2) they can accommodate a wide variety of polymers; (3) material and color changes can be made easily; and (4) short cycle time and fast production rate can be achieved by using robotic assist in removing runners. The disadvantages are as follows: (1) cycle times are longer than hot runner systems; and (2) plastic waste is generated from runners (if they cannot be reground and recycled). 2.5.3.4 Gating The gate provides the connection between the runner and the mold cavity. It must permit enough material to ﬂow into the mold to ﬁll out the cavity. The type of the gate and its size and location in the mold strongly affect the molding process and the quality of the molded part. There are two types of gates: large and restricted. A restricted (pin-pointed) thermal gate is a very small oriﬁce between runner and cavity. It is usually circular in cross section and, for most thermoplastics, does not exceed 0.060 in. in diameter. The apparent viscosity of a thermoplastic is a function of shear rate—the viscosity decreases as the shear rate and, hence, the velocity increases. The use of the restricted gate is therefore advantageous, because the velocity of the plastic melt increases as it is forced through the small opening; in addition, some of the kinetic energy is transformed into heat, raising the local temperature of the plastic and thus further reducing its viscosity. The passage through a restricted area also results in higher mixing. Hot-tip gates are identiﬁed as restricted thermal gates, which are typically located at the tip of the part rather than on the parting line and are ideal for round or conical shapes where uniform ﬂow is necessary. Hot tip gates are only used with hot runner molding systems. The gate leaves a small raised nub on the surface of the part. The most common type of gate in injection molding is the edge gate (Figure 2.11a), where the part is gated either as a restricted or larger gate at some point on the edge of the part. The edge gate is easy to construct and is best suited for ﬂat parts. Edge gates are ideal for medium and thick sections and can be used on multicavity two-plate molds. This gate will leave a scar on the parting line. The edge gate can be fanned out for large parts or when there is a special reason. Then, it is called a fan gate (Figure 2.11b). A fan gate reduces stress concentration in gate area by spreading the opening over a wider area. With this type of gate, less warping of parts can usually be expected. When it is required to orient the ﬂow pattern in one direction, a ﬂash gate (also known as ﬁlm gate) may be used (Figure 2.11c). It is very thin, involves extending the fan gate over the full length of the part, and has parallel runners before the gate. It is generally used in thin and ﬂat requirement, like ﬂat mobile phone cap, ipod cap, and others. The most common gate for single-cavity molds is the direct gate or sprue gate (Figure 2.11d). It is located in the center of the plastic part, has a circular cross section, is slightly tapered, and merges with its largest cross section into the part. It is suitable for a big part with a deep cavity. No runner is required. As the gate feeds directly from the nozzle of the machine into the molded part, the pressure loss is minimum. But the sprue gate has the following disadvantages: a bigger cross-sectional gate area and a high stress concentration around the area, the need for gate removal, lack of a cold slug, more difﬁcult degating, and degating relics affecting product appearance since the gate mark is visible in molded components, for example, bucket molding (backside cylindrical gate mark visible). [It may be mentioned that pin gate, especially used for three-part molds, is a reverse taper sprue gate. The runner channel is 177 Fabrication Processes Top view Side view (a) (b) (d) (c) (e) (f ) Runner Plastic Tab Gate (g) Runner (h) Knockout pins FIGURE 2.11 Gating design: (a) edge; (b) fan; (c) ﬂash; (d) sprue; (e) diaphragm; (f) ring; (g) tab; (h) submarine. located in a separate runner plate. The melt ﬂow is divided into several directions and led into the cavity by separate gate locations. The gate point is designed to be very small. It is trimmed off by the action of injection mold opening (see Figure 2.10).] A diaphragm gate (Figure 2.11e) has, in addition to the sprue, a circular area leading from the sprue to the piece. The molten material ﬂows from the center toward the outer circumference. A diaphragm gate is thus used in symmetrical cavity ﬁlling to reduce weld line formations and improve ﬁlling rate. This type of gate is suitable for gating concentric molded products, such as annular and tubular articles, hollow tubes, and so on. The diaphragm eliminates stress concentration around the gate because the whole area is removed, but the cleaning of this gate is more difﬁcult than a sprue gate. Ring gates (Figure 2.11f) are annular gates, which accomplish the same purpose as gating internally in a hollow tube, but from the outside. The gate is between the runner and the cavity. It encircles the core to permit the melt to ﬁrst move around the core before ﬁlling the cavity. Its easy ﬁlling and venting characteristics could avoid weld lines and reduce stress. Ring gates are thus most suitable for cylindrical components to eliminate weld line defect. When the gate leads directly into the part, there may be surface imperfection attributed to jetting. This may be overcome by extending a tab from the part into which the gate is cut. This procedure is called tab gating (Figure 2.11g). The tab has to be removed as a secondary operation. A submarine gate (Figure 2.11h) is one that goes through the steel of the cavity. It is very often used in automatic molds as a type of gate having a structure that automatically cuts the molded item and gate at the time of opening and closing the parting surface. The submarine gate allows one to gate away from the parting line, allowing more ﬂexibility to place the gate at an optimum location on the part. The gate leaves a pin-sized scar on the part. 178 Plastics Technology Handbook The tunnel gate is a variation of the submarine gate. Classifying tunnel gates broadly into those provided on the ﬁxed half and those on the moving half of the parting surface, there are four patterns of gate and runner combinations. When a tunnel gate is provided on the ﬁxed half, the molded product is cut off from the gate at the time of opening the parting surface. On the other hand, when a tunnel gate is provided on the moving half, the molded product is cut off from the gate at the time that the runner is ejected by the runner ejector pin. On the basis of trimming modes, the various gates fall into two main types, namely, manually trimmed gates and automatically trimmed gates. Manually trimmed gates are those that require an operator to separate parts from runners. These gates are used if the required gate is too large to be sheared from the part, if the part is too thin, or if the material is shear sensitive such that the high shear rate of automatic trimming would damage the part. Gate types trimmed from the part manually include sprue gate, diaphragm gate, ﬂash or ﬁlm gate, edge gate, fan gate, ring gate, and tab gate. Automatically trimmed gates incorporate features in the tool to break or shear the gate as the mold is opened to eject the part. This type of gate is used to avoid the need for a press operator, decrease cycle time, maintain consistent cycle times for all shots, and minimize gate scrap. Automatically trimmed gates include pin gates, submarine (tunnel) gates, sub gates, hot runner gates, and valve gates (see below). 2.5.3.5 Valve Gates In injection molding with thermal gating or hot tip gating, the melt in the gate area (after cavity ﬁll) cools and solidiﬁes, forming a small slug inside the gate, which remains in the gate during the phases of mold open, part ejection, and mold close motion. During the next injection, the pressurized melt ﬂushes the cold slug into the melt stream that ﬁlls the empty cavity, while the slug liqueﬁes from shear heating and mixes with the melt. A part molded with thermal gating retains a standing vestige at the gate interface. Thermal gate vestige is highly dependent on the gate diameter, a larger gate producing a larger vestige. Gate cooling optimization is critical in thermal gate design since mold open cannot occur until the gate is sealed enough to break clearly from the part as well as “hold back” the melt in the hot runner. Mold open before complete gate solidiﬁcation results in drool (extrudation or leakage of molten resin) or stringing (insufﬁcient cooling of melt in area between ﬁnished part and sprue), whereas excessive gate cooling can produce a frozen gate to prevent or delay gate opening and result in short shots or unﬁlled cavities. All molding parameters (pressure, temperature, and time) that determine the quality of a molded part are also responsible for the formation of the thermal gate at every cycle to avoid drooling and stringing of the gate. If gate quality, as discussed above, is a critical factor and gate vestiges are unacceptable, valve-gated hot runner systems are generally recommended, as they offer more processing control. In valve-gated hot runner systems, the ﬂow of plastic into the mold cavity is controlled with a valve stem or mechanical shutoff pins. Through mechanical action, the valve stem moves forward and seals the gate oriﬁce. The valve stem remains in the closed position during mold open and part ejection, preventing drool and stringing. A typical example is polyamide (PA). It may drool when processed in a thermally gated hot runner, but any possibility of drooling is eliminated if it is processed in a valve-gated hot runner. Another drawback of thermal gated systems is that melt decompression, which can lead to splay (“splash-like” appearance) and other imperfections on the surface of the molded part, is often required on these systems to relieve pressure in the manifold. With valve-gated nozzles, however, melt decompression is not needed since the seal is robust even if the hot runner manifold remains pressurized. The molded part separates from the valve gate without breaking or shearing plastic. Any discoloration or deformation as a result of gate break is thus unlikely. However, valve gate nozzles leave a small witness mark (of the same size as the gate diameter) on the part. Nowadays, the commonly used valve gates are pneumatic valve gates, hydraulic valve gates, and electrical valve gates (eGate). Pneumatic valve gates are the most widely used and are common for small parts in packaging, electronic, Fabrication Processes 179 and medical applications. Though the pneumatic valve gates are proven and are cleaner than the hydraulic ones, they have certain limitations like slow response (owing to the high compressibility of air), lower closing force, and proneness to inconsistent opening/closing times, no monitoring capability, single speed, and single stroke. Hydraulic valve gates are the second most common actuation tool often used in large part molding in automotive applications. Their main advantages are that they are quicker and more powerful compared to pneumatic valve gates and have a smaller piston size, while their limitations are that they are service intensive with need for oil changes, bleeding air, and replacement of seals, besides being single speed and single stroke, and lacking the ability to monitor pin position and repeatability. They are also energy intensive, consuming about nine times more energy compared to electric valve gates (eGate), which are a new alternative, having features that were available never before with conventional valve gates. Featuring adjustable pin position, speed, acceleration, stroke, immediate response (<1 ms) with precise repeatability, ability to record and monitor real-time pin, besides being clean, quiet, energy efﬁcient, and having lowest maintenance and easiest serviceability, the new e-gate technology overcomes the limitations of all the other valve gates. The key beneﬁts of the e-gate technology are superior part quality owing to improved gate quality and reduction of blush in critical applications, elimination of service defects by providing a controlled ﬂow front velocity, and ability to control knit lines and improve cosmetic ﬁnish. Other beneﬁts are better tolerance, greater part-to-part consistency, and individual control of valve pin speed and stroke that provides independent ﬂow rate adjustability at each nozzle. This technology allows independent control of up to 64 valve pins and ﬂexibility to synchronize all pins, if desired. With its advanced ﬂow control capability, e-gate is a powerful new tool for the processor to optimize balance in multicavity and family molding. With very low operating cost and energy consumption, e-gate is also rated to be ∼93% more efﬁcient than traditional valve gates. 2.5.3.6 Venting When the melted plastic ﬁlls the mold, it displaces the air. The displaced air must be removed quickly, or it may ignite the plastic and cause a characteristic burn, or it may restrict the ﬂow of the melt into the mold cavity, resulting in incomplete ﬁlling. For venting the air from the cavity, slots can be milled, usually opposite the gate. The slots usually range from 0.001 to 0.002 in. deep and from 3/8 to 1 in. wide. Additional venting is provided by the clearance between KO pins and their holes. Note that the gate location is directly related to the consideration of proper venting. 2.5.3.7 Parting Line If one were inside a closed mold and looking outside, the mating junction of the mold cavities would appear as a line. It also appears as a line on the molded piece and is called the parting line. A piece may have several parting lines. The selection of the parting line in mold design is inﬂuenced by the type of mold, number of cavities, shape of the piece, tapers, method of ejection, method of fabrication, venting, wall thickness, location and type of gating, inserts, postmolding operations, and aesthetic consideration. 2.5.3.8 Cooling The mold for thermoplastics receives the molten plastic in its cavity and cools it to solidity to the point of ejection. The mold is provided with cooling channels. The mold temperature is controlled by regulating the temperature of the cooling ﬂuid and its rate of ﬂow through the channels. Proper cooling or coolant circulation is essential for uniform repetitive mold cycling. The functioning of the mold and the quality of the molded part depend largely on the location of the cooling channel. Since the rate of heat transfer is reduced drastically by the interface of two metal pieces, no matter how well they ﬁt, cooling channels should be located in cavities and cores themselves rather than only in the supporting plates. The cooling channels should be spaced evenly to prevent uneven 180 Plastics Technology Handbook temperatures on the mold surface. They should be as close to the plastic surface as possible, taking into account the strength of the mold material. The channels are connected to permit a uniform ﬂow of the cooling or heating medium, and they are thermostatically controlled to maintain a given temperature. Another important factor in mold temperature control is the material the mold is made from. Beryllium copper has a high thermal conductivity, about twice that of steel and four times that of stainless steel. A beryllium copper cavity should thus cool about four times as fast as a stainless steel one. A mold made of beryllium copper would therefore run signiﬁcantly faster than one of stainless steel. 2.5.3.9 Ejection Once the molded part has cooled sufﬁciently in the cavity, it has to be ejected. This is done mechanically by KO pins, KO sleeves, stripper plates, stripper rings or compressed air, used either singly or in combination. The most frequent problem in new molds is with ejection. Because there is no mathematical way of predicting the amount of ejection force needed, it is entirely a matter of experience. Since ejection involves overcoming the forces of adhesion between the mold and the plastic, the area provided for the knockout (KO) is an important factor. If the area is too small, the KO force will be concentrated, resulting in severe stresses on the part. As a result, the part may fail immediately or in later service. In materials such as ABS and high-impact polystyrene, the severe stresses can also discolor the plastic. Sticking in a mold makes ejection difﬁcult. Sticking is often related to the elasticity of steel and is called packing. When injection pressure is applied to the molten plastic and force it into the mold, the steel deforms; when the pressure is relieved, the steel retracts, acting as a clamp on the plastic. Packing is often eliminated by reducing the injection pressure and/or the injection forward time. Packing is a common problem is multicavity molds and is caused by unequal ﬁlling. Thus, if a cavity seals off without ﬁlling, the material intended for the cavity is forced into other cavities, causing overﬁlling. 2.5.3.10 Standard Mold Bases Standardization of mold bases for injection molding, which was unknown prior to 1940, was an important factor in the development of efﬁcient mold making. Standard mold bases were pioneered by the D-M-E Co., Michigan, to provide the mold maker with a mold base at lower cost and with much higher quality than if the base were manufactured by the mold maker. Replacement parts, such as locating ring and sprue bushings, loader pins and bushings, KO pins and push-back pins of high quality are also available to the molder. Since these parts are common for many molds, they can be stocked by the molder in the plant and thus down time is minimized. An exploded view of the components of a standard injection-mold base assembly is shown in Figure 2.12. 2.5.4 Structural Foam Injection Molding Structural foam injection molding produces parts consisting of solid external skin surfaces surrounding an inner cellular (or foam) core, as shown in Figure 2.13. Large, thick structural foam parts can be produced by this process with both low and high pressure and using either nitrogen gas or chemical blowing agents (see “Foaming Process”). 2.5.5 Co-Injection (Sandwich) Molding Co-injection molding is used to produce parts that have a laminated structure with the core material embedded between the layers of the skin material. As shown in Figure 2.14, the process involves sequential injection of two different but compatible polymer melts into a cavity where the materials laminate and solidify. A short shot of skin polymer melt is ﬁrst injected into the mold (Figure 2.14a), followed by core polymer melt which is injected until the mold cavity is nearly ﬁlled (Figure 2.14b); the skin polymer is then injected again to purge the core polymer away from the spruce (Figure 2.14c). The process offers the 181 Fabrication Processes Locating ring Sprue bushing Top clamp plate Front cavity plate (‘A’ plate) Rear cavity plate (‘B’ plate) Support plate Ejector retainer plate Spruepuller pin Return pin Ejector plate Ejector housing FIGURE 2.12 Exploded view of a standard mold base showing component parts. (a) (b) FIGURE 2.13 Structural foam injection molding. (a) During injection under high pressure there is very little foaming. (b) After injection, pressure drops and foaming occurs at hot core. inherent ﬂexibility of using the optimal properties of each material or modifying the properties of each material or those of the molded part. 2.5.6 Gas-Assisted Injection Molding The gas-assisted injection molding process begins with a partial or full injection of polymer melt into the mold cavity. Compressed gas is then injected into the core of the polymer melt to help ﬁll and pack the mold, as shown in Figure 2.15 for the Asahi Gas Injection (AGI) molding process. This process is thus capable of producing hollow rigid parts, free of sink marks. The hollowing out of thick sections of moldings results in reduction in part weight and saving of resin material. Other advantages include shorter cooling cycles, reduced clamp force tonnage and part consolidation. The process allows high precision molding with greater dimensional stability by eliminating uneven mold shrinkage and makes it possible to mold complicated shapes in single form, thus reducing product assembly work and simplifying mold design. 182 Plastics Technology Handbook (b) (a) (c) FIGURE 2.14 Three stages of co-injection (sandwich) molding. (a) Short shot of skin polymer melt (shown in black) is injected into the mold; (b) injection of core polymer melt until cavity is nearly ﬁlled; and (c) skin polymer melt is injected again, pushing the core polymer away from the sprue. The formation of thick walled sections of a molding can be easily achieved by introducing gas in the desired locations. The gas channels thus formed also effectively support the ﬂow of resin, allowing the molding pressure to be greatly reduced, which in turn reduces internal stresses, allows uniform mold shrinkage, and reduces sink marks and warpage. 2.6 Injection Molding of Thermosetting Resins 2.6.1 Screw-Injection Molding of Thermosetting Resins The machines used earlier were basically plunger-type machines [6,10–12]. But in the late 1960s’ shortly after the development of screw-transfer machines, the concept of screw-injection molding of thermosets, also known as direct screw transfer (or DST), was introduced. The potential of this techniques for lowcost, high-volume production of molded thermoset parts was quickly recognized, and today screwinjection machines are available in all clamp tonnages up to 1,200 tons and shot sizes up to 10 lb. Coupled with this, there has been a new series of thermosetting molding materials developed speciﬁcally for injection molding. These materials have long life at moderate temperature (approximately 200°F), which permits plastication in screw barrel, and react (cure) very rapidly when the temperature is raised to 350°F– 400°F (177°–204°C), resulting in reduced cycle time. A typical arrangement for a direct screw-transfer injection-molding machine for thermosets is shown in Figure 2.16. The machine has two sections mounted on a common base. One section constitutes the plasticizing and injection unit, which includes the feed hopper, the heated barrel that encloses the screw, the hydraulic cylinder which pushes the screw forward to inject the plasticized material into the mold, and a motor to rotate the screw. The other section clamps and holds the mold halves together under pressure during the injection of the hot plastic melt into the mold. 183 Fabrication Processes Gas Resin injection Injection of gas, pressure holding, cooling Gas Opening of mold and removal of molding FIGURE 2.15 Gas release Schematic of the Asahi Gas Injection (AGI) molding process. The thermosetting material (in granular or pellet form) suitable for injection molding is fed from the hopper into the barrel and is then moved forward by the rotation of the screw. During its passage, the material receives conductive heat from the wall of the heated barrel and frictional heat from the rotation of the screw. For thermosetting materials, the screw used is a zero-compression-ratio screw—i.e., the depths of ﬂights of the screw at the feed-zone end and at the nozzle end are the same. By comparison, the screws used in thermoplastic molding machines have compression ratios such that the depth of ﬂight at the feed end is one and one-half to ﬁve times that at the nozzle end. This difference in screw conﬁguration is a major difference between thermoplastic- and thermosetting-molding machines. As the material moves forward in the barrel due to rotation of the screw, it changes in consistency from a solid to semiﬂuid, and as it starts accumulating at the nozzle end, it exerts a backward pressure on the screw. This back pressure is thus used as a processing variable. The screw stops turning when the required amount of material—the charge—has accumulated at the nozzle end of the barrel, as sensed by a limit switch. (The charge is the exact volume of material required to ﬁll the sprue, runners, and cavities of the mold.) the screw is then moved forward like a plunger by hydraulic pressure (up to 20,000 psi) to force the hot plastic melt through the sprue of the mold and into the runner system, gates, and mold cavities. The nozzle temperature is controlled to maintain a proper balance between a hot mold (350°F–400°F), and a relatively cool barrel (150°F–200°F). Molded-in inserts are commonly used with thermosetting materials. However, since the screwinjection process is automatic, it is desirable to use postassembled inserts rather than molded-in inserts 184 Plastics Technology Handbook Hopper Drive Heating or cooling Fixed platen Tie bar Barrel Mold clamping cylinder Reciprocating screw Runner Gate Mold cavity Clamp cylinder piston Cylinder to retreat barrel from mold FIGURE 2.16 Schematic of a direct screw-transfer molding machine for thermosets. (After Frados, J. ed. 1976. Plastics Engineering Handbook, 4th Ed., Van Nostrand Reinhold, New York.) because molded-in inserts require that the mold be held open each cycle to place the inserts. A delay in the manual placement disrupts an automatic cyclic operation, affecting both the production rate and the product quality. Tolerances of parts made by injection molding of thermosetting materials are comparable to those produced by the compression and transfer methods described, earlier. Tolerances achieved are as low as ±0.001 in./in., although ordinarily tolerance of ±0.003–0.005 in./in. are economically practical. Thermosetting materials used in screw-injection molding are modiﬁed from conventional thermosetting compounds. These modiﬁcations are necessary to provide the working time-temperature relationship required for screw plasticating. The most commonly used injection-molding thermosetting materials are the phenolics. Other thermosetting materials often molded by the screw-injection process include melamine, urea, polyester, alkyd, and dially phthalate (DAP). Since the mid-1970s the injection molding of glass-ﬁber-reinforced thermosetting polyesters gained increasing importance as better materials (e.g., low shrinkage resins, palletized forms of polyester/glass, etc.), equipment, and tooling became available. Injection-molded reinforced thermoset plastics have thus made inroads in such markets as switch housings, fuse blocks, distributor caps, power-tool housings, ofﬁce machines, etc. Bulk molding compounds (BMC), which are puttylike FRP (ﬁbrous glass-reinforced plastic) materials, are injection molded to make substitutes of various metal die castings. For injection molding, FRP should have some speciﬁc characteristics. For example, it must ﬂow easily at lower-than-mold temperatures without curing and without separating into resin, glass, and ﬁller components, and it should cure rapidly when in place at mold temperature. A traditional FRP material shrinks about 0.003 in./in. during molding, but low-shrink FRP materials used for injection molding shrink as little as 0.000–0.0005 in./in. Combined with proper tooling, these materials thus permit production of pieces with dimensional tolerances of ±0.0005 in./in. Proper design of parts for injection molding requires an understanding of the ﬂow characteristics of material within the mold. In this respect, injection-molded parts of thermosets are more like transfermolded parts than to compression-molded parts. Wall-section uniformity is an important consideration in part design. Cross sections should be as uniform as possible, within the dictates of part requirements, since molding cycles, and therefore costs, depend on the cure time of the thickest section. 185 Fabrication Processes (For thermoplastics, however, it is the cooling time that is critical). A rule of thumb for estimating cycle times for a 1/4-in. wall section is 30 sec for injection-molded thermosets (compared to 45 sec for thermoplastics). As a guideline for part design, a good working average for wall thickness is 1/8–3/16 in., with a minimum of 1/16 in. 2.7 Extrusion The extrusion process is basically designed to continuously convert a soft material into a particular form [13–15]. An oversimpliﬁed analogy may be a house-hold meat grinder. However, unlike the extrudate from a meat grinder, plastic extrudates generally approach truly continuous formation. Like the usual meat grinder, the extruder (Figure 2.17) is essentially a screw conveyor. It carries the cold plastic material (in granular or powdered form) forward by the action of the screw, squeezes it, and, with heat from external heaters and the friction of viscous ﬂow, changes it to a molten stream. As it does this, it develops pressure on the material, which is highest right before the molten plastic enters the die. The screen pack, consisting of a number of ﬁne or coarse mesh gauzes supported on a breaker plate and placed between the screw and the die, ﬁlters out dirt and unfused polymer lumps. The pressure on the molten plastic forces it through an adapter and into the die, which dictates the shape of the ﬁne extrudate. A die with a round opening as shown in Figure 2.17, produces pipe; a square die opening produces a square proﬁle, etc. Other continuous shapes, such as the ﬁlm, sheet, rods, tubing, and ﬁlaments, can be produced with appropriate dies. Extruders are also used to apply insulation and jacketing to wire and cable and to coat substrates such as paper, cloth, and foil. When thermoplastic polymers are extruded, it is necessary to cool the extrudate below Tm or Tg to impart dimensional stability. This cooling can often be done simply by running the product through a tank of water, by spraying cold water, or, even more simply, by air cooling. When rubber is extruded, dimensional stability results from cross-linking (vulcanization). Interestingly, rubber extrusion for wire coating was the ﬁrst application of the screw extruder in polymer processing. Extruders have several other applications in polymer processing: in the blowmolding process they are used to make hollow objects such as bottles; in the blow-ﬁlm process they are used for making wide ﬁlms; they are also used for compounding plastics (i.e., adding various ingredients to a resin mix) and for converting plastics into the pellet shape commonly used in processing. In this last operation specialized equipment, such as the die plate-cutter assembly, is installed in place of the die, and an extrusion-type screw is used to provide plasticated melt for various injection-molding processes. Resin Hopper Thermocouples Hardened liner Screw Screen pack Breaker plate Barrel Gear reducer and motor drive Feed section Hopper cooling jacket FIGURE 2.17 Compression or transition section Metering section Indicates heaters Scheme for a typical single-screw extruder showing extruding pipe. Die 186 Plastics Technology Handbook 2.7.1 Extruder Capacity Standard sizes of single-screw extruders are 1½, 2, 2½, 3¼, 3½, 4½, 6, and 8 in., which denote the inside diameter (ID) of the barrel. As a rough guide, extruder capacity Qe, in pounds per hour, can be calculated from the barrel diameter Db, in inches, by the empirical relation [15] Qe = 16Db2:2 (2.1) Another estimate of extruder capacity can be made by realizing that most of the energy needed to melt the thermoplastic stems from the mechanical work, whereas the barrel heaters serve mainly to insulate the material. If we allow an efﬁciency from drive to screw of about 80%, the capacity Qe (lb/h) can be approximately related to the power supplied Hp (horsepower), the heat capacity of the material Cp [Btu/lb °F], and the temperature rise from feed to extrudate DT (°F) by Qe = 1:9 103 Hp =Cp DT (2.2) Equation 2.2 is obviously not exact since the heat of melting and other thermal effects have been ignored. Equation 2.2 coupled with Equation 2.1 enables one to obtain an estimate of DT. Thus, for processing of poly(methyl methacrylate), for which Cp is about 0.6 Btu/lb °F, in a 2-in. extruder run by a 10-hp motor, Equation 2.1 gives Qe = 74 lb/h, and Equation 2.2 indicates that DT ≅ 430°F. In practice, a DT of 350°F is usually adequate for this polymer. 2.7.2 Extruder Design and Operation The most important component of any extruder is the screw. It is often impossible to extruder satisfactorily one material by using a screw designed for another material. Therefore screw designs vary with each material. 2.7.2.1 Typical Screw Construction The screw consists of a steel cylinder with a helical channel cut into it (Figure 2.18). The helical ridge formed by the machining of the channel is called the ﬂight, and the distance between the ﬂights is called the lead. The lead is usually constant for the length of the screw in single-screw machines. The helix angle is called pitch. Helix angles of the screw are usually chosen to optimize the feeding characteristics. An angle of 17.5° is typical, though it can be varied between 12 and 20°. The screw outside diameter is generally just a few thousandths of an inch less than the ID of the barrel. The minimal clearance between screw and barrel ID prevents excessive buildup of resin on the inside barrel wall and thus maximizes heat transfer. The screw may be solid or cored. Coring is used for steam heating or, more often, for water cooling. Coring can be for the entire length of the screw or for a portion of it, depending on the particular Flight Lead Screw outside dia. Root Pitch FIGURE 2.18 Detail of screw. Flight depth Fabrication Processes 187 application. Full length coring of the screw is used where large amounts of heat are to be removed. The screw is cored only in the initial portions at the hopper end when the objective is to keep the feed zone cooler for resins which tend to soften easily. Screws are often fabricated from 4140 alloy steel, but other materials are also used. The screw ﬂights are usually hardened by ﬂame-hardening techniques or inset with a wear resistant alloy (e.g., Stellite 6). 2.7.2.2 Screw Zones Screws are characterized by their length-diameter ratio (commonly written as L/D ratios). L/D ratios most commonly used for single-screw extruders range from 15:1 to 30:1. Ratios of 20:1 and 24:1 are common for thermoplastics, whereas lower values are used for rubbers. A long barrel gives a more homogeneous extrudate, which is particularly desirable when pigmented materials are handled. Screws are also characterized by their compression ratios—the ratio of the ﬂight depth of the screw at the hopper end to the ﬂight depth at the die end. Compression ratios of single-screw extruders are usually 2:1–5:1. The screw is usually divided into three sections, namely, feed, compression, and metering (Figure 2.17). One of the basic parameters in screw design involves the ratio of lengths between the feed, compression (or transition), and metering sections of the screw. Each section has its own special rate. The feed section picks up the powder, pellets, or beads from under the hopper mouth and conveys them forward in the solid state to the compression section. The feed section is deep ﬂighted so that it functions in supplying enough material to prevent starving the forward sections. The gradually diminishing ﬂight depth in the compression section causes volume compression of the melting granules. The volume compression results in the trapped air being forced back through the feed section instead of being carried forward with the resin, thus ensuring an extrudate free from porosity. Another consequence of volume compression is an increase in the shearing action on the melt, which is caused by the relative motion of the screw surfaces with respect to the barrel wall. The increased shearing action produces good mixing and generates frictional heat, which increases ﬂuidity of the melt and leads to a more uniform temperature distribution in the molten extrudate. The resin should be fully melted into a reasonably uniform melt by the time it enters the ﬁnal section of the screw, known as the metering section. The function of the metering section is to force the molten polymer through the die at a steady rate and iron out pulsations. For many screw designs the compression ratio is 3–5; i.e., the ﬂight depth in the metering section is one-third to one-ﬁfth that in the feed section. 2.7.2.3 Motor Drive The motor employed for driving the screw of an extruder should be of more than adequate power required for its normal needs. Variable screw speeds are considered essential. Either variable-speed motors or constant-speed motors with variable-speed equipment, such as hydraulic systems, step-change gear boxes, and inﬁnitely variable-speed gear boxes may be used. Thrust bearings of robust construction are essential because of the very high back pressure generated in an extruder and the trend towards higher screw speeds. Overload protection in the form of an automatic cut-out should be ﬁtted. 2.7.2.4 Heating Heat to melt the polymer granules is supplied by external heaters or by frictional heat generated by the compression and shearing action of the screw on the polymer. Frictional heat is considerable, and in modern high-speed screw extruders it supplies most of the heat required for steady running. External heaters serve only to insulate the material and to prevent the machine from stalling at the start of the run when the material is cold. The external heater may be an oil, steam, or electrical type. Electrical heating is most popular because it is compact, clean, and accurately controlled. Induction heating is also used because it gives quicker heating with less variation and facilitates efﬁcient cooling. The barrel is usually divided into three or four heating zones; the temperature is lowest at the feed end and highest at the die end. The temperature of each zone is controlled by carefully balancing heating and cooling. Cooling is done automatically by either air or water. (The screw is also cored for heating and 188 Plastics Technology Handbook cooling.) The screw is cooled where the maximum amount of compounding is required, because this improves the quality of the extrudate. 2.7.2.5 Screw Design Metering section Transition section Feed section Mixing section Metering section Transition section First stage Second stage Elective screw length Overall screw length FIGURE 2.19 Single-ﬂight, two-stage extrusion screw with mixing section. Screw root diameter Flight depth Lead φ Pitch Screw outside diameter Flight depth Flight depth Flight width Flight depth (Channel depth) The screw we have described is a simple continuous-ﬂight screw with constant pitch. The more sophisticated screw designs include ﬂow disrupters or mixing sections (Figure 2.19). These mixer screws have mixing sections which are designed as mechanical means to break up and rearrange the laminar ﬂow of the melt within the ﬂight channel, which results in more thorough melt mixing and more uniform heat distribution in the metering section of the screw. Mixer screws have also been used to mix dissimilar materials (e.g., resin and additives or simply dissimilar resins) and to improve extrudate uniformity at higher screw speeds (>100 rpm). A few typical mixing section designs are shown in Figure 2.20. The ﬂuted-mixing-section-barrier-type design (Figure 2.20a) has proved to be especially applicable for extrusion of polyoleﬁns. For some mixing problems, such as pigment mixing during extrusion, it is convenient to use rings (Figure 2.20b) or mixing pins (Figure 2.20c) and sometimes parallel interrupted mixing ﬂights having wide pitch angles (Figure 2.20d). A later development in extruder design has been the use of venting or degassing zones to remove any volatile constituents from the melt before it is extruded through the die. This can be achieved by placing an obstruction in the barrel (the reverse ﬂights in Figure 2.21) and by using a valved bypass section to step down the pressure developed in the ﬁrst stage to atmospheric pressure for venting. In effect, two screws are used in series and separated by the degassing or venting zone. Degassing may also be achieved by having a deeper thread in the screw in the degassing section than in the ﬁnal section of the ﬁrst screw, so the polymer melt suddenly ﬁnds itself in an increased volume and hence is at a lower pressure. The volatile vapors released from the melt are vented through a hold in the top of the extruder barrel or through a hollow core of the screw by way of a hole drilled in the trailing edge of one of the ﬂights in the degassing zone. A vacuum is sometimes applied to assist in the extraction of the vapor. Design and operation must be suitably controlled to minimize plugging of the vent (which, as noted above, is basically an open area) or the possibility of the melt escaping from this area. Many variations are possible in screw design to accommodate a wide range of polymers and applications. So many parameters are involved, including such variables as screw geometry, materials characteristics, operating conditions, etc., that the industry now uses computerized screw design, which permits analysis of the variables by using mathematical models to derive optimum design of a screw for a given application. Various screw designs have been recommended by the industry for extrusion of different plastics. For polyethylene, for example, the screw should be long with an L/D of at least 16:1 or 30:1 to provide a large area for heat transfer and plastication. A constant-pitch, decreasing-channel-depth, metering-type polyethylene screw or constant-pitch, constant-channel-depth, metering-type nylon screw with a compression Feed section Screw hub Screw shank 189 Fabrication Processes Out In (a) (b) (c) (d) FIGURE 2.20 Mixing section designs: (a) ﬂuted-mixing-section-barrier type; (b) ring-barrier type; (c) mixing pins; (d) parallel interrupted mixing ﬂights. (After Frados, J. ed. 1976. Plastics Engineering Handbook, 4th Ed., Van Nostrand Reinhold, New York.) Reverse ﬂights Valve Vent Stage 2 Stage 2 Bypass Pressure gauge Valve FIGURE 2.21 A two-stage vented extruder with a valved bypass. (After Fisher, E. G. 1971. Blow Molding of Plastics. Iliffe, London.) ratio between 3–1 and 4–1 (Figure 2.22) is recommended for polyethylene extrusion, the former being preferable for ﬁlm extension and extrusion coating. Nylon-6, 6 melts at approximately 260°C (500°F). Therefore, an extruder with an L/D of at least 16:1 is necessary. A screw with a compression ratio of 4:1 is recommended. 2.7.3 Multiple-Screw Extruders Multiple-screw extruders (that is, extruders with more than a single screw) were developed largely as a compounding device for uniformly blending plasticizers, ﬁllers, pigments, stabilizers, etc., into the polymer. Subsequently, the multiple-screw extruders also found use in the processing of plastics. Multiple-screw extruders differ signiﬁcantly from single-screw extruders in mode of operation. In a single-screw machine, friction between the resin and the rotating screw, makes the resin rotate with the screw, and the friction between the rotating resin and the barrel pushes the material forward, and this also generates heat. Increasing the screw speed and/or screw diameter to achieve a higher output rate in a single-screw extruder will therefore result in a higher buildup of frictional heat and higher temperatures. In contrast, in twin-screw extruders with intermeshing screws the relative motion of the ﬂight of one screw inside the channel of the other pushes the material forward almost as if the machine were a positivedisplacement gear pump which conveys the material with very low friction. 190 Plastics Technology Handbook Metering section Compression section Feed section Screw diameter Land width of the ﬂights (a) Lead Channel or ﬂight depth Compression section Metering section Feed section (b) FIGURE 2.22 (a) Constant pitch, decreasing channel depth, metering-type polyethylene screw. (b) Constant pitch, constant-channel-depth, metering-type nylon screw (not to scale). (After Frados, J. ed. 1976. Plastics Engineering Handbook, 4th Ed., Van Nostrand Reinhold, New York.) In two-screw extruders, heat is therefore controlled independently from an outside source and is not inﬂuenced by screw speed. This fact become especially important when processing a heat-sensitive plastic like poly(vinyl chloride) (PVC). Multiple-screw extruders are therefore gaining wide acceptance for processing vinyls, although they are more expensive than single-screw machines. For the same reason, multiple-screw extruders have found a major use in the production of high-quality rigid PVC pipe of large diameter. Several types of multiple-screw machines are available, including intermeshing corotating screws (in which the screws rotate in the same direction, and the ﬂight of one screw moves inside the channel of the other), intermeshing counterrotating screws (in which the screws rotate in opposite directions), and nonintermeshing counterrotating screws. Multiple-screw extruders can involve either two screws (twin-screw design) or four screws. A typical four-screw extruder is a two-stage machine, in which a twin-screw plasticating section feeds into a twinscrew discharge section located directly below it. The multiple screws are generally sized on output rates (lb/h) rather than on L/D ratios or barrel diameters. 2.7.4 Blown-Film Extrusion The blown-ﬁlm technique is widely used in the manufacture of polyethylene and other plastic ﬁlms [14,15]. A typical setup is shown in Figure 2.23. In this case the molten polymer from the extruder head enters the die, where it ﬂows round a mandrel and emerges through a ring-shaped opening in the form of a tube. The tube is expanded into a bubble of the required diameter by the pressure of internal air admitted through the center of the mandrel. The air contained in the bubble cannot escape because it is sealed by the die at one end and by the nip (or pinch) rolls at the other, so it acts like a permanent shaping mandrel once it has been injected. An even pressure of air is maintained to ensure uniform thickness of the ﬁlm bubble. The ﬁlm bubble is cooled below the softening point of the polymer by blowing air on it from a cooling ring placed round the die. When the polymer, such as polyethylene, cools below the softening point, the 191 Fabrication Processes Pinch rolls Collapsing plate Wind up Gusset bars Blown tube Guide rollers Mandrel Frost line Cooling ring Air inlet Extruder Adjustable section of die Die Valve Air supply FIGURE 2.23 Typical blow-ﬁlm extrusion setup. crystalline material is cloudy compared with the clear amorphous melt. The transition line which coincides with this transformation is therefore called the frost line. The ratio of bubble diameter to die diameter is called the blowup ratio. It may range as high as 4 or 5, but 2.5 is a more typical ﬁgure. Molecular orientation occurs in the ﬁlm in the hoop direction during blowup, and orientation in the machine direction, that is, in the direction of the extrudate ﬂow from the die, can be induced by tension from the pinch rolls. The ﬁlm bubble after solidiﬁcation (at frost line) moves upward through guiding devices into a set of pinch rolls which ﬂatten it. It can then be slit, gusseted, and surface-treated in line. (Vertical extrusion, shown in Figure 2.23, is most common, although horizontal techniques have been successfully used.) Blown-ﬁlm extrusion is an extremely complex subject, and a number of problems are associated with the production of good-quality ﬁlm. Among the likely defects are variation in ﬁlm thickness, surface imperfections (such as “ﬁsh eyes,” “orange peel,” haze), wrinkling, and low tensile strength. The factors affecting them are also numerous. “Fish eyes” occur due to imperfect mixing in the extruder or due to contamination of the molten polymer. Both factors are controlled by the screen pack. The blown-ﬁlm technique has several advantages: the relative ease of changing ﬁlm width and caliber by controlling the volume of air in the bubble and the speed of the screw; the elimination of the end effects (e.g., edge bead trim and nonuniform temperature that result from ﬂat ﬁlm extrusion); and the capability of biaxial orientation (i.e., orientation both in the hoop direction and in the machine direction), which results in nearly equal physical properties in both directions, thereby giving a ﬁlm of maximum toughness. After extrusion, blown-ﬁlm is often slit and wound up as ﬂat ﬁlm, which is often much wider than anything produced by slot-die extrusion. Thus, blown-ﬁlms of diameters 7 ft. or more have been produced, giving ﬂat ﬁlm of widths up to 24 ft. One example is reported [16] of a 10-in. extruder with 5-ft diameter and a blowup ratio of 2.5, producing 1,100 lb/h of polyethylene ﬁlm, which when collapsed and slit in 40 ft wide. Films in thicknesses of 0.004–0.008 in. are readily produced by the blown-ﬁlm process. Polyethylene ﬁlms of such large widths and small thicknesses ﬁnd extensive uses in agriculture, horticulture, and building. 192 Plastics Technology Handbook 2.7.5 Flat Film or Sheet Extrusion In the ﬂat-ﬁlm process the polymer melt is extruded through a slot die (T-shaped or “coat hanger” die), which may be as wide as 10 ft. The die has relatively thick wall sections on the ﬁnal lands (as compared to the extrusion coating die) to minimize deﬂection of the lips from internal melt pressure. The die opening (for polyethylene) may be 0.015–0.030 in.—even for ﬁlms that are less than 0.003 in. thick. The reason is that the speed of various driven rolls used for taking up the ﬁlm is high enough to draw down the ﬁlm with a concurrent thinning. (By deﬁnition, the term ﬁlm is used for material less than 0.010 in. thick, and sheet for that which is thicker.) Figure 2.24 illustrates the basic components of a sheet extrusion die. The die is built in two halves the part for easier construction and maintenance. The use of a jack bolt facilitates separation of the die halves when the die is full of plastic. The die lip can be adjusted (across the entire lip length) to enable the processor to keep the thickness of the extruded sheet within speciﬁcation. Figure 2.25 illustrates a T-type die and a coat-hanger-type die, which are used for both ﬁlm and sheet extrusion. The die must produce a smooth and uniform laminar ﬂow of the plastic melt which has already been mixed thoroughly in the extruder. The internal shape of the die and the smoothness of the die surface are critical to this ﬂow transition. The deckle rods illustrated in Figure 2.25 are used by the processor to adjust the width of the extruded sheet or ﬁlm. Body bolt Jack bolt Adjusting bolt Die lip Heater holes Adapter face Plastic melt FIGURE 2.24 Sheet extrusion die. V-shaped cross section Adjustable jaw (a) (b) FIGURE 2.25 Adjustable jaw Deckle channel Die land Die land Internal deckle External deckle Schematic cross sections (a) T-type and (b) coat-hanger-type extrusion dies. 193 Fabrication Processes Following extrusion, the ﬁlm may be chilled below Tm or Tg by passing in through a water bath or over two or more chrome-plated chill rollers which have been cored for water cooling. A schematic drawing of a chill-roll (also called cast-ﬁlm) operation is shown in Figure 2.26. The polymer melt extruded as a web from the die is made dimensionally stable by contacting several chill rolls before being pulled by the powered carrier rolls and wound up. The chrome-plated surface of the ﬁrst roll is highly polished so that the product obtained is of extremely high gloss and clarity. In ﬂat-ﬁlm extrusion (particularly at high takeoff rates), there is a relatively high orientation of the ﬁlm in the machine direction (i.e., the direction of the extrudate ﬂow) and a very low one in the traverse direction. Biaxially oriented ﬁlm can be produced by a ﬂat-ﬁlm extrusion by using a tenter (Figure 2.27). Polystyrene, for example, is ﬁrst extruded through a slit die at about 190°C and cooled to about 120°C by Nip roll (rubber) Extruder Nip roll (stainless steel) Treater bar Slitter Die Powered carrier Idler rolls rolls Nip roll (rubber) Polished chill rolls (water-cooled) Nip roll (stainless steel) Wind up FIGURE 2.26 Sketch of chill-roll ﬁlm extrusion. (After Lukov, L. J. 1963. SPE J., 13, 10, 1057.) Across stretch Extruder Extruded ﬁlm Along stretch Heater Stretched ﬁlm cooling To winder Heater Slit die Water cooled rolls Edge grips mounted on endless chain belt Rollers FIGURE 2.27 Plax process for manufacture of biaxially stretched polystyrene ﬁlm. (After Brydson, J. A. 1982. Plastics Materials, Butterworth Scientiﬁc, London, UK.) 194 Plastics Technology Handbook passing between rolls. Inside a temperature-controlled box and moving sheet, rewarmed to 130°C, is grasped on either side by tenterhooks which exert a drawing tension (longitudinal stretching) as well as widening tension (lateral stretching). Stretch ratios of 3:1–4:1 in both directions are commonly employed for biaxially oriented polystyrene ﬁlm. Biaxial stretching leads to polymers of improved tensile strength. Commercially available oriented polystyrene ﬁlm has a tensile strength of 10,000–12,000 psi (703–843 kg/cm2), compared to 6,000–8,000 psi (422–562 kg/cm2) for unstretched material. Biaxial orientation effects are important in the manufacture of ﬁlms and sheet. Biaxially stretched polypropylene, poly(ethyleneterephthalate) (e.g., Melinex) and poly(vinylidene chloride) (Saran) produced by ﬂat-ﬁlm extrusion and tentering are strong ﬁlms of high clarity. In biaxial orientation, molecules are randomly oriented in two dimensions just as ﬁbers would be in a random mat; the orientationinduced crystallization produces structures which do not interfere with the light waves. With polyethylene, biaxial orientation often can be achieved in blown-ﬁlm extrusion. 2.7.6 Pipe or Tube Extrusion The die used for the extrusion of pipe or tubing consists of a die body with a tapered mandrel and an outer die ring which control the dimensions of the inner and outer diameters, respectively. Since this process involves thicker walls than are involved in blown-ﬁlm extrusion, it is advantageous to cool the extrudate by circulating water through the mandrel (Figure 2.28) as well as by running the extrudate through a water bath. The extrusion of rubber tubing, however, differs from thermoplastic tubing. For thermoplastic tubing, dimensional stability results from cooling below Tg or Tm, but rubber tubing gains dimensional stability due to a cross-linking reaction at a temperature above that in the extruder. The high melt viscosity of the rubber being extruded ensures a constant shape during the cross-linking. A complication encountered in the extrusion of continuous shapes in die swell. Die swell is the swelling of the polymer when the elastic energy stored in capillary ﬂow is relaxed on leading the die. The extrusion of ﬂat sheet or pipe is not sensitive to die swell, since the shape remains symmetrical even through the dimensions of the extrudate differ from those of the die. Unsymmetrical cross sections may, however, be distorted. Adaptor for connecting to extruder Insulator Water in Water out Spiral baﬄes Tapered mandrel (cooling and sizing) FIGURE 2.28 An extrusion die ﬁtted with a tapered cooling and sizing mandrel for use in producing either pipe or tubing. (After Fisher, E. G. 1971. Blow Molding of Plastics, Iliffe, London.) 195 Fabrication Processes 2.7.7 Wire and Cable Coverings The covering or coating of wire and cable in continuous lengths with insulating plastics is an important application of extrusion, and large quantities of resin are used annually for this purpose. This application represented one of the ﬁrst uses of extruders for rubber about 100 years ago. The wire and cable coating process resembles the process used for pipe extrusion (Figure 2.28) with the difference that the conductor (which may be a single metal strand, a multiple strand, or even a bundle of previously individually insulated wires) to be covered is drawn through the mandrel on a continuous basis (Figure 2.29). For thermoplastics such as polyethylene, nylon, and plasticized PVC, the coating is hardened by cooling below Tm or Tg by passing through a water trough. Rubber coatings, on the other hand, are to be cross-linked by heating subsequent to extrusion. 2.7.8 Extrusion Coating Many substrates, including paper, paperboard, cellulose ﬁlm, ﬁreboard, metal foils, and transparent ﬁlms, are coated with resins by direct extrusion. The resins most commonly used are the polyoleﬁns, such as polyethylene, polypropylene, ionomer, and ethylene–vinyl acetate copolymers. Nylon, PVC, and polyester are used for a lesser extent. Often combinations of these resins and substrates are used to provide a multiplayer structure. [A related technique, called extrusion laminating, involves two or more substrates, such as paper and aluminum foil, combined by using a plastic ﬁlm, (e.g., polyethylene) as the adhesive and as a moisture barrier.] Coatings are applied in thicknesses of about 0.2–15 mils, the common average being 0.5–2 mils, and the substrates range in thickness from 0.5 to more than 24 mils. The equipment used for extrusion coating is similar to that used for the extrusion of ﬂat ﬁlm. Figure 2.30 shows a typical extrusion coating setup. The thin molten ﬁlm from the extruder is pulled down into the nip between a chill roll and a pressure roll situated directly below the die. The pressure between these two rolls forces the ﬁlm on to the substrate while the substrate, moving at a speed faster than the extruded ﬁlm, draws the ﬁlm to the required thickness. The molten ﬁlm is cooled by the watercooled, chromium-plated chill roll. The pressure roll is also metallic but is covered with a rubber sleeve, usually neoprene or silicone rubber. After trimming, the coated material is wound up on conventional windup equipment. Polymer melt Bare wire Coated wire Die Crosshead FIGURE 2.29 Crosshead used for wire coating. 196 Plastics Technology Handbook Extruder Water-cooled chill roller (driven) Wind-up Slitter (driven) Reel of uncoated paper Die Coated paper Pressure roll (idler) FIGURE 2.30 Sketch of paper coating for extrusion process. 2.7.9 Proﬁle Extrusion Proﬁle extrusion is similar to pipe extrusion (Figure 2.28) except that the sizing mandrel is obviously not necessary. A die plate, in which an oriﬁce of appropriate geometry has been cut, is placed on the face of the normal die assembly. The molten polymer is subjected to surface drag as it passes through the die, resulting in reduced ﬂow through the thinner sections of the oriﬁce. This effect is countered by altering the shape of the oriﬁce, but often this results in a wide difference in the oriﬁce shape from the desired extrusion proﬁle. Some examples are shown in Figure 2.31. 2.8 Blow Molding Basically, blow molding in intended for use in manufacturing hollow plastic products, such as bottles and other containers [17]. However, the process is also used for the production of toys, automobile parts, accessories, and many engineering components. The principles used in blow molding are essentially similar to those used in the production of glass bottles. Although there are considerable differences in the process available for blow molding, the basic steps are the same: (1) melt the plastic; (2) form the molten plastic into a parison (a tubelike shape of molten plastic); (3) seal the ends of the parison except for one area through which the blowing air can enter; (4) inﬂate the parison to assume the shape of the mold in which it is placed; (5) cool the blowmolded part; (6) eject the blow-molded part; Extruded Oriﬁce (7) trim ﬂash if necessary. section Two basic processes of blow molding are extrusion blow molding and injection blow molding. These processes differ in the way in which the parison is made. The extrusion process utilizes an unsupported parison, whereas the injection process utilizes a parison supported on a metal core. Oriﬁce Extruded section The extrusion blow-molding process by far accounts for the largest percentage of blowmolded objects produced today. The injecFIGURE 2.31 Relationships between extruder die oriﬁce and tion process is, however, gaining acceptance. extruded section. 197 Fabrication Processes Although any thermoplastic can be blow-molded, polyethylene products made by this technique are predominant. Polyethylene squeeze bottles form a large percentage of all blow-molded products. 2.8.1 Extrusion Blow Molding Extrusion blow molding consists basically of the extrusion of a predetermined length of parison (hollow tube of molten plastic) into a split die, which is then closed, sealing both ends of the parison. Compressed air is introduced (through a blowing tube) into the parison, which blows up to ﬁt the internal contours of the mold. As the polymer surface meets the cold metal wall of the mold, it is cooled rapidly below Tg or Tm. When the product is dimensionally stable, the mold is opened, the product is ejected, a new parison is introduced, and the cycle is repeated. The process affords high production rates. In continuous extrusion blow molding, a molten parison is produced continuously from a screw extruder. The molds are mounted and moved. In one instance the mold sets are carried on a rotating horizontal table (Figure 2.32a), in another on the periphery of a rotating vertical wheel (Figure 2.32b). Such rotary machines are best suited for long runs and large-volume applications. In the ram extrusion method the parison is formed in a cyclic manner by forcing a charge out from an accumulated molten mass, as in the preplasticizer injection-molding machine. The transport arm cuts and holds the parison and lowers it into the waiting mold, where shaping under air pressure takes place (Figure 2.33). A variation of the blow-extrusion process which is particularly suitable for heat-sensitive resins such as PVC is the cold perform molding. The parison is produced by normal extrusion and cooled and stored until needed. The required length of tubing is then reheated and blown to shape in a cold mold, as in conventional blow molding. Since, unlike in the conventional process, the extruder is not coupled directly to the blow-molding machine, there is less chance of a stoppage occurring, with consequent risk of holdup and degradation of the resin remaining in the extruder barrel. There is also less chance of the occurrence of “dead” pockets and consequent degradation of resin in the straight-through die used in this process than in the usual crosshead used with a conventional machine. Extruder Knife Die head Parison Eject Extruder Cooling Closed Knife Die head Eject Parison Blowing (a) Rotation Blowing Cooling (b) FIGURE 2.32 Continuous extrusion blow molding. (a) Rotating horizontal table carrying mold sets. (b) Continuous vertical rotation of a wheel carrying mold sets on the periphery. (After Morgan, B. T., Peters, D. L., and Wilson, N. R. 1967. Mod. Plastics, 45, 1A, Encyl. Issue, 797.) 198 Plastics Technology Handbook Die head Parison transport Parison of semimolten plastic Air pressure Mold closed around parison Bottle Mold FIGURE 2.33 Continuous extrusion blow molding with parison transfer. The transport arm cuts the extruded parison from the die head and lowers it into the waiting mold. (After Morgan, B. T., Peters, D. L., and Wilson, N. R. 1967. Mod. Plastics, 45, 1A, Encyl. Issue, 797.) 2.8.2 Injection Blow Molding In this process the parison is injection molded rather than extruded. In one system, for example, the parison is formed as a thick-walled tube around a blowing stick in a conventional injection-molding machine. The parison is then transferred to a second, or blowing, mold in which the parison is inﬂated to the shape of the mold by passing compressed air down the blowing stick. The sequence is shown in Figure 2.34. Injection blow molding is relatively slow and is more restricted in choice of molding materials as compared to extrusion blow molding. The injection process, however, affords good control of neck and wall thicknesses of the molded object. With this process it is also easier to produce unsymmetrical molding. 2.8.3 Blow Molds Generally, the blow mold is a cavity representing the outside of a blow-molded part. The basic structure of a blow mold consists of a cast or machined block with a cavity, cooling system, venting system, pinchoffs, ﬂash pockets, and mounting plate. The selection of material for the construction of a blow mold is based on the consideration of such factors as thermal conductivity, durability, cost of the material, the resin being processed, and the desired quality of the ﬁnished parts. Commonly used mold materials are beryllium, copper, aluminum, ampcoloy, A-2 steel, and 17-4 and 420 stainless steels. Beryllium–copper (BeCu) alloys 165 and 25 are normally used for blow molds. These materials display medium to good thermal conductivity with good durability. Stainless steels such as 17-4 and 420 are also frequently employed in blow molds where durability and resistance to hydrochloric acid are required. Heat-treated A-2 steel is often used as an insert in pinchoffs where thermal conductivity is not a concern and high quality parts are required. For blow molding HDPE parts, aluminum is commonly employed for the base material, with BeCu or stainless steel inserts in the pinchoff areas. For PVC parts BeCu, ampcoloy, or 17-4 stainless steel are used as the base material, with A-2 or stainless inserts in the pinchoff area. For PET parts, the base mold is typically made of aluminum or BeCu, with A-2 or stainless steel pinchoffs. The production speed of blow-molded parts is generally limited by one of two factors: extruder capacity or cooling time in the mold. Cooling of mold is accomplished by a water circuit into the mold. Flood cooling and cast-in tubes are most common in cast molds; drilled holes and milled slots are the norm in machined blow molds. 199 Fabrication Processes (a) (b) (c) Injection cycle (d) FIGURE 2.34 (e) (f ) Sketch of injection-blow-molding process. In multiple-cavity molds, series and parallel cooling circuits are used. Series cooling enters and cools one cavity, then moves to the next until all the cavities are cooled. The temperature of the water increases as it moves through the mold, and this results in non-uniform cooling. Parallel cooling, on the other hand, enters and exists all cavities simultaneously, thereby cooling all cavities at a uniform rate. Parallel cooling is thus the preferred method but it is not always possible due to limitations. 2.9 Calendering Calendering is the leading method for producing vinyl ﬁlm, sheets, and coatings [18]. In this process continuous sheet is made by passing a heat-softened material between two or more rolls. Calendering was originally developed for processing rubber, but is now widely used for producing thermoplastic ﬁlms, sheets, and coatings. A major portion of thermoplastics calendered is accounted for by ﬂexible (plasticized) PVC. Most plasticized PVC ﬁlm and sheet, ranging from the 3-mil ﬁlm for baby pants to the 0.10 in “vinyl” tile for ﬂoor coverings, is calendered. The calendaring process consists of feeding a softened mass into the nip between two rolls where it is squeezed into a sheet, which then passes round the remaining rolls. The processed material thus emerges as a continuous sheet, the thickness of which is governed by the gap between the last pair of rolls. The surface quality of the sheet develops on the last roll and may be glossy, matt, or embossed. After leaving the calender, the sheet is passed over a number of cooling rolls and then through a beta-ray thickness gage before being wound up. The plastics mass fed to the calender may be simply a heat-softened material, as in the case of, say, polyethylene, or a rough sheet, as in the case of PVC. The polymer PVC is blended with stabilizers, 200 Plastics Technology Handbook Feed Hot roll Fixed roll Hot roll Feed Cold roll (a) (b) Engraved (replaceable) Uneven speed Even or uneven speed Even speed Uneven speed Bank Calendered sheet Bank Fabric Feed Conveyor Friction (c) (d) FIGURE 2.35 Typical arrangements of calender rolls: (a) single-ply sheeting; (b) double-ply sheeting; (c) applying rubber to fabrics; (d) proﬁling with four-roll engraving cylinder. plasticizers, etc., in ribbon blenders, gelated at 120°C–160°C for about 5–10 min in a Banbury mixer, and the gelated lumps are made into a rough sheet on a two-roll mill before being fed to the calender. Calenders may consist of two, three, four, or ﬁve hollow rolls arranged for steam heating or water cooling and are characterized by the number of rolls and their arrangement. Some arrangements are shown in Figure 2.35. Thick sections of rubber can be made by applying one layer of polymer upon a previous layer (double plying) (Figure 2.35b). Calenders can be used for applying rubber or plastics to fabrics (Figure 2.35c). Fabric or paper is fed through the last two rolls of the calender so that the resin ﬁlm is pressed into the surface of the web. For proﬁling, the plastic material is fed to the nip of the calender, where the material assumes the form of a sheet, which is then progressively pulled through two subsequent banks to resurface each of the two sides (Figure 2.35d). For thermoplastics the cooling of the sheet can be accomplished on the rolls with good control over dimensions. For rubber, cross-linking can be carried out with good control over dimensions, with the support of the rolls. Despite the simple appearance of the calender compared to the extruder, the close tolerances involved and other mechanical problems make for the high cost of a calendaring unit. 2.10 Spinning of Fibers The term spinning, as used with natural ﬁbers, refers to the twisting of short ﬁbers into continuous lengths [19–21]. In the modern synthetic ﬁber industry, however, the term is used for any process of producing continuous lengths by any means. (A few other terms used in the ﬁber industry should also be deﬁned. A ﬁber may be deﬁned as a unit of matter having a length at least 100 times its width or diameter. An individual strand of continuous length is called a ﬁlament. Twisting together ﬁlaments into a strand gives continuous ﬁlament yarn. If the ﬁlaments are assembled in a loose bundle, we have tow or roving. These can be chopped into small lengths (an inch to several inches long), referred to as staple. Spun yarn is made 201 Fabrication Processes by twisting lengths of staple into a single continuous strand, and cord is formed by twisting together two or more yarns.) The dimensions of a ﬁlament or yarn are expressed in terms of a unit called the “tex” which is a measure of the ﬁneness or linear density. One tex is 1 gram per 1,000 meters or 10−6 kg/m. The tex has replaced “denier” as a measure of the density of the ﬁber. One denier is 1 gram per 9,000 meters, so 1 denier = 0.1111 tex. The primary fabrication process in the production of synthetic ﬁbers is the spinning—i.e., the formation—of ﬁlaments. In every case the polymer is either melted or dissolved in a solvent and is put in ﬁlament form by forcing through a die, called spinneret, having a multiplicity of holes. Spinnerets for rayon spinning, for example, have as many as 10,000 holes in a 15-cm-diameter platinum disc, and those for textile yarns may have 10–120 holes; industrial yarns such as tire core might be spun from spinnerets with up to 720 holes. Three major categories of spinning processes are melt, dry, and wet spinning [19]. The features of the three processes are shown in Figure 2.36, and the typical cross sections of the ﬁbers produced by them are shown in Figure 2.37. The products of all the above processes are micro-size ﬁbers. In contrast, ultraﬁne ﬁbers or nanoﬁbers are produced by electrospinning, which is described in Section 2.11. Polymer solution Polymer chips Heating grid Pump Polymer melt Pump Filter and spinnerette Filter and spinnerette Fiber cooling and solidiﬁcation Air diﬀuser Hot chamber Steam chamber Roll and guide Yarn drawing (a) Bobbin Polymer solution (b) Packaging Pump Fiber solidiﬁcation by precipitation Filter and spinnerette Coagulation bath (c) FIGURE 2.36 Schematic of the three principal types of ﬁber spinning: (a) melt spinning; (b) dry spinning; (c) wet spinning. (After Carraher, C. E., Jr. 2002. Polymer News, 27, 3, 91.) 202 Plastics Technology Handbook 2.10.1 Melt Spinning In melt spinning, which is the same as melt extrusion, the polymer is heated and the viscous melt is pumped through a spinneret. An inert atmosphere is provided in the melting chamber before the pump. Special pumps are used to operate in the temperature range necessary to (a) (b) (c) produce a manageable melt (230–315°C). For nylon, for example, a gear pump is used to feed FIGURE 2.37 Typical cross section of ﬁbers produced by the melt to the spinneret (Figure 2.36a). For a different spinning processes: (a) melt-spun nylon from polymer with high melt viscosity such as polyvarious shaped oriﬁces; (b) dry-spun cellulose acetate from propylene, a screw extruder is used to feed a round oriﬁce; (c) wet-spun viscose rayon from round heated spinneret. Dimensional stability of the oriﬁce. ﬁber is obtained by cooling under tension. Typical melt spinning temperatures are given in Table 2.1. 2.10.2 Dry Spinning In dry spinning, a polymer is dissolved in a solvent and the polymer solution (concentration on the order of 20–40%) is ﬁltered and then forced through a spinneret into a chamber through which heated air is passed to achieve dimensional stability of the ﬁber by evaporation of the solvent (Figure 2.36b). For economical reasons, the gas is usually air, but inert gases such as nitrogen and superheated steam are sometimes used. The skin which forms ﬁrst on the ﬁber by evaporation from the surface gradually collapses and wrinkles as more solvent diffuses out and the diameter decreases. The cross section of a dryspun ﬁber thus has an irregularly lobed appearance (Figure 2.37). Recovery of the solvent used for dissolving the polymer is important to the economics of the process. Cellulose acetate dissolved in acetone and polyacrylonitrile (PAN) dissolved in dimethylformamide are two typical examples. The hot solution (dope) of PAN in DMF is extruded directly into a hot stream of nitrogen at 300°C. The residual DMF is recovered in subsequent water washing steps. Dry spun ﬁbers have lower void concentrations than wet spun ﬁbers. This is reﬂected in greater densities and lower dyeability for the dry spun ﬁbers. 2.10.3 Wet Spinning Wet spinning also involves pumping a solution of the polymer to the spinneret. However, unlike dry spinning, dimensional stability is achieved by precipitating the polymer in a nonsolvent (Figure 2.36c). For example, PAN in dimethylformamide can be precipitated by passing a jet of the solution through a bath of water, which is miscible with the solvent but coagulates the polymer. For wet-spinning cellulose TABLE 2.1 Typical Spinning Temperatures for Selected Polymers Polymer Melting Point (°C) Typical Spinning Temperature (°C) Nylon-6 220 280 Nylon-6,6 Poly(ethylene terephthalate) 260 260 290 290 ∼130 220–230 170 120–140 250–300 180 Polyethylene Polypropylene Poly(vinylidene chloride) copolymers Source: Carraher, C. E. Jr. 2002. Polymer News, 27(3), 91. 203 Fabrication Processes triacetate a mixture methylene chloride and alcohol can be used to dissolve the polymer, and a toluene bath can be used for precipitation of the polymer. In some cases the precipitation can also involve a chemical reaction. An important example is viscose rayon, which, is made by regenerating cellulose from a solution of cellulose xanthate in dilute alkali. S R OH CS2, NaOH R O C S– Na+ + H2O Cellulose xanthate H2O H2SO4 NaHSO4 R OH + CS2 Cellulose + Na+ (HSO4–) If a slot die rather than a spinneret is used, the foregoing process would yield cellulose ﬁlm (cellophane) instead of ﬁber. Wet spinning is the most complex of the three spinning processes, typically including washing, stretching, drying, crimping, ﬁnish application, and controlled relaxation to form tow material [22]. A simpliﬁed sketch of the Asahi wet spinning process for making PAN (acrylic) ﬁber tows is shown in Figure 2.38. The polymer solution (dope) is made in concentrated HNO3 (67%) at low temperatures using pulverizer and mixer, ﬁltered, and deaerated. The dope, containing 14–15% polymer and maintained at −7°C, is extruded at a pumping pressure of 10–15 atm through the spinnerets immersed in the coagulation bath. Each spinneret has about 46,000–73,000 holes (different sizes for different grades). The acid concentration in the bath is 37% and the temperature is −5°C. The ﬁlaments from 5 spinnerets are collected into a tow. The spinning speed in this step is about 7 meters per minute. The dilute nitric acid from the bath goes to the concentration section where 67% HNO3 is obtained for re-use in dope preparation. In the next pre-ﬁnishing step, the ﬁbers (tows) are repeatedly washed with water by spraying and immersion to remove the acid. The ﬁber is stretched in three stages, a 1:10 stretch being obtained in the last hot water bath at 100°C. In the ﬁnishing section, the tows pass through a water bath containing 1% oil to impart antistaticity and reduce friction. The tows are then dried in a hot air (135°C) dryer over rollers. Subsequent treatments include a second ﬁnishing oil spray, crimping, second hot air drying, plaiting, thermosetting, and cutting into staple ﬁbers. Wash and stretching Polymer Acid Finishing Dope preparation Water + oil 3–4 stages Slurry tank Acid out In Out In N2 Heat treatment FIGURE 2.38 Crimping N2 Drying A simpliﬁed sketch of the Asahi wet spinning process for polyacrylonitrile ﬁber. 204 Plastics Technology Handbook Fibers made from wet spinning generally have high void contents in comparison to all of the other processes giving them increased dyeability and the surface is rougher with longitudinal serrations. Hollow ﬁbers for gas and liquid separation are prepared by passing air through the material just prior to entrance into the non-solvent bath. 2.10.4 Cold Drawing of Fibers Almost all synthetic ﬁbers are subjected to a drawing (stretching) operation to orient the crystalline structure in the direction of the ﬁber axis. Drawing orients crystallites in the direction of the stretch so that the modulus in that direction is increased and elongation at break is decreased. Usually the drawing is carried out at a temperature between Tg and Tm of the ﬁber. Thus, polyethylene (Tg = −115°C) can be drawn at room temperature, whereas nylon−6,6 (Tg = 53°C) should be heated or humidiﬁed to be drawn. Tg is depressed by the presence of moisture, which acts as a plasticizer. The drawing is accomplished by winding the yarn around a wheel or drum driven at a faster surface velocity than a preceding one. 2.11 Electrospinning of Polymer Nanoﬁbers Nanoﬁbers are deﬁned as ﬁbers with diameters of less than 100 nm. In the textile industry, however, this deﬁnition is often extended to include ﬁbers with diameters as large as 1000 nm. Polymer nanoﬁbers are generally produced by electrospinning, which is a simple yet versatile method for producing ultrathin ﬁbers from a variety of materials and is currently the most important technique for converting polymers into nanosized ﬁber materials. When the dimensions of polymer ﬁber materials are reduced from micrometers (e.g., 10–100 µm) to submicrons or nanometers (e.g., 10 × 10−3 to 100 × 10−3 µm), there appear several outstanding properties, such as very large surface area-to-volume ratio (for a nanoﬁber, this can be as large as 103 times that of a microﬁber) and superior mechanical properties (e.g., stiffness and tensile strength) compared with any other form of the material. These features make polymer nanoﬁbers ideally suited for many important applications. In recent years, a number of processing techniques have been developed to produce polymer nanoﬁbers such as drawing [23], templating [24], self-assembly [25], and electrospinning [26]. Though the drawing process can produce one-by-one very long single nanoﬁbers, only a viscoelastic material that can undergo strong deformations during pulling can be made into nanoﬁbers through drawing. The template method, besides depending on the use of a nanoporous membrane as template to make nanoﬁbers, cannot produce one-by-one continuous nanoﬁbers. The self-assembly, on the other hand, is time-consuming in processing continuous polymer nanoﬁbers. It thus appears that the electrospinning process is the only method that can be developed for mass production of one-by-one continuous nanoﬁbers from solutions or melts of various polymers. Although the term “electrospinning” (derived from “electrostatic spinning”) was used relatively recently (around 1994), its fundamental idea, namely, production of polymer ﬁlaments using an electrostatic force, dates back more than 70 years earlier. This is evident from a series of patents [27] that appeared from 1934 to 1944 describing an experimental setup for the production of polymer ﬁlaments (from polymer solution) between two electrodes bearing electrical charges of opposite polarity. One of the electrodes was placed into the solution and the other was placed onto a collector. On being ejected through a small hole of a metal spinneret, the charged solution jets evaporated to become solid ﬁbers and deposited on the collector. The potential difference to be applied between the electrodes depended on the properties of the spinning solution, such as polymer molecular weight and solution viscosity. Subsequently, over the years, numerous variations of electrospinning setup were devised and used, and nearly 100 different polymers, mostly dissolved in solvent (and some heated to melts), were spun into ultraﬁne ﬁbers using this technique. A schematic diagram of the basic setup for electrospinning of polymer nanoﬁbers is shown in Figure 2.39. It has three basic components: a high-voltage (usually direct current, DC) power supply, a 205 Fabrication Processes Syringe Polymer solution Needle Taylor cone Liquid jet V High voltage power supply Smaller diameter ﬁber Fiber jet Nonwoven nanoﬁber mat Pore Larger diameter ﬁber Collector FIGURE 2.39 Schematic diagram of the basic setup for electrospinning. The inset shows a diagram of the electriﬁed Taylor cone and SEM image (schematic) of the randomly oriented nonwoven mat of polymer nanoﬁbers deposited on the collector. spinneret (a metallic needle), and a metal collector (grounded conductor). The spinneret is connected to a syringe containing the polymer solution (or melt), which is fed through the spinneret at a constant and controllable rate with the use of a syringe pump. When a high voltage (usually in the range of 1 to 30 kV) is applied, the pendant drop of liquid that is held by surface tension at the tip of the spinneret becomes highly electriﬁed with the induced charges evenly distributed over the surface. The drop then experiences two major types of electrostatic forces, namely, repulsion between the surface charges and the coulombic force exerted by the external ﬁeld. Under the inﬂuence of these two forces, the hemispherical surface of the drop elongates to form a conical shape, known as the Taylor cone. Further increasing the electric ﬁeld, a critical value is attained at which the repulsive electrostatic force can overcome the surface tension of the polymer solution (or melt) and force the ejection of a charged jet of the ﬂuid from the tip of the Taylor cone. The electriﬁed jet then undergoes a stretching and whipping (rapid bending) process leading to the formation of a very long and thin thread. During this process, the solvent (of polymer solution) also evaporates, leaving behind a charged polymer ﬁber, while its diameter can be greatly reduced from hundreds of micrometers to as small as tens of nanometers. This charged ﬁber is often deposited as a randomly oriented, nonwoven mat (see Figure 2.39) on the grounded metal collector placed under the spinneret. This relatively simple and straightforward technique has been used to process more then 50 different types of organic polymers. Polymers melted at a high temperature can also be processed into nanoﬁbers by electrospinning. Instead of a solution, the polymer melt is introduced into the capillary tube and the electrospinning process is performed in a vacuum condition, which means that the whole assembly of capillary tube, the charged jet ejection, and the collector must be fully encapsulated within a vacuum. A number of variations [28] of conventional electrospinning of polymer solution or melt have been developed and used for different objectives, a few of which are brieﬂy highlighted below. Electroblown spinning (EBS), or electroblowing, is an electrospinning process in the presence of a controlled airﬂow. In this process, two forces are simultaneously applied for producing nanoﬁbers, 206 Plastics Technology Handbook namely, electrical force and air-blowing shear force. The method is especially useful for highly viscous solutions in which applying high voltage alone is not sufﬁcient to overcome surface tension. Centrifugal electrospinning (CFS) makes combined use of an electrical ﬁeld and a centrifugal ﬁeld. Compared with pure centrifugal spinning that is required to rotate at thousands of revolutions per minute (rpm), the rotational speed in CFS can decrease to 300–600 rpm with more aligned ﬁbers. Combining centrifugal and electrical forces into the spinning process, CFS leads to further orientation of polymer chains in nanoﬁbers and a higher production rate can be achieved at a lower working voltage or lower rpm. Near-ﬁeld electrospinning (NFES) is a potential approach for easier and more predictable location control for the deposition of nanoﬁbers, which is impossible to be achieved by conventional electrospinning owing to the whipping action of ﬁbers. By reducing the distance between the nozzle and the substrate to less than a few millimeters, NFES allows the ﬁbers to land on the substrate before the onset of whipping so that nonwoven nanoﬁbers can be deposited precisely along a predetermined pattern. Coaxial electrospinning (CES) is an innovatively extended form of electrospinning that uses two concentrically aligned capillaries to enforce ﬁber formation with a core–shell structure. CES is of particular interest for those core materials that cannot form ﬁbers via electrospinning by themselves, such as conductive polymers, metals, or some natural polymers. While the process is conceptually similar to that of conventional electrospinning, two dissimilar materials can be delivered independently through the coaxial capillary and drawn to generate nanoﬁbers in a core–shell conﬁguration. The shell and the core may not be miscible owing to the short process duration at which the jet becomes solidiﬁed into ﬁbers. Emulsion electrospinning (EES) is similar to normal solution electrospinning, except that the solution is replaced by an emulsion (water-in-oil or oil-in-water type). Using water-in-oil type of emulsions comprising a water phase that is a drug/protein dissolved in water, and an oil phase, which is a polymer dissolved in an organic solvent, the process allows for the encapsulation of a wide range of bioactive molecules (with different solubilities) into polymeric nanoﬁbers. For a polymer that can be electrospun into nanoﬁbers, the ideal targets of manufacturing would be the following: (1) consistent and controllable ﬁber diameters, (2) defect-free or defect-controllable ﬁber surface, and (3) continuous single nanoﬁbers. However, researches so far have shown that these three targets are not easily achievable. This is because many parameters can inﬂuence the transformation of a polymer solution into nanoﬁbers by electrospinning. These include (a) solution properties such as viscosity, elasticity, surface tension, and electrical conductivity; (b) operating variables such as hydrostatic pressure in the spinneret, electric potential at the tip of the spinneret, and the gap between the tip and the collector; and (c) ambient parameters such as solution temperature, humidity, and air velocity in the electrospinning chamber. Since nanoﬁbers result from evaporation or solidiﬁcation of polymer ﬂuid jets, the ﬁber diameters should depend primarily on the jet sizes and the polymer contents in the jets. Since the polymer solution viscosity is proportional to the polymer concentration, one of the most signiﬁcant parameters inﬂuencing the ﬁber diameter is the solution viscosity. A higher viscosity results in larger ﬁber diameter. The parameters other than the solution viscosity that affect the ﬁber diameter are strength of applied potential, solution conductivity, polymer feeding rate, capillary size, and the distance between the capillary and the collector. Polymer solutions with high conductivity have high surface charge density that leads to ﬁner ﬁbers. A challenge with electrospinning lies in the fact that the ﬁber diameters obtained are seldom uniform. Another problem is that defects such as beads and pores (voids) may occur in polymer nanoﬁbers. Bead formation can be reduced by using higher polymer concentrations for electrospinning [29]. Fibers without beads may also result from reduction of surface tension [30] that is more likely to be a function of solvent compositions and not polymer concentration [31]. Adding some ﬁller material into the polymer solution for electrospinning can also result in ﬁbers free of voids [32]. This apart, the idea of incorporating ﬁllers was used to prepare composite nanoﬁbers by dispersing carbon SWNTs (single-wall nanotubes) in polyacrylonitrile solution that was electrospun into ultraﬁne ﬁbers. Fabrication Processes 207 The basic setup for electrospinning being very simple, it has found widespread use in many laboratories for making ultrathin ﬁbers. Because electrospinning is a continuous process, the ﬁbers could be as long as several kilometers and comparable to ﬁbers manufactured by conventional drawing or spinning techniques. However, because of the bending instability associated with a spinning jet, electrospun ﬁbers are mostly deposited on the surface of a collector in randomly oriented, nonwoven form that can be useful, however, for a relatively small number of applications, such as ﬁltration [33], tissue scaffolds [34], implant coating ﬁlm [35], and wound dressing [36]. Since applications can be expanded only when continuous single nanoﬁbers or uniaxial ﬁber bundles are available, considerable research has been focused on devising possible means to align electrospun nanoﬁbers. Over the last decade, a number of approaches have been demonstrated to directly produce electrospun nanoﬁbers as uniaxially aligned arrays. Two of the earliest such approaches involved the use of a rotating drum (or frame) or a pair of split electrodes as the collector. Thus, aligned poly(glycolic acid) nanoﬁbers [37] and collagen nanoﬁbers [38] were obtained by using a rotating cylindrical collector at 1000 and 4500 rpm, respectively. It was demonstrated that by using a collector consisting of two conductive strips separated by a void gap of variable widths (up to several centimeters), electrospun ﬁbers could be axially aligned over long length scales during the spinning process [39]. Bundles of uniaxially aligned PAN nanoﬁbers were prepared by this method using a solution of 15 wt% PAN/DMF for electrospinning. [Note: Electrospun PAN nanoﬁbers can be used to make carbon nanoﬁbers by a series of heat treatments. See Section 2.14.2.3.] The lengths of the bundles were 3 cm, which was ∼85% of the width of the gap between the collectors [40]. This method allows the aligned ﬁbers to be transferred onto other solid substrates for further processing steps and applications. Single ﬁbers can also be collected across the gap and transferred onto a substrate for the fabrication of single-ﬁber-based devices [39]. Though electrospinning is a well-established process capable of producing nonwoven webs as well as single or well-aligned arrays of continuous nanoﬁbers with controlled morphology and size, the major challenge associated with the process is its production rate, compared with that of conventional ﬁber spinning. Of the two main parameters related to production efﬁciency, namely, ﬂow rate and ﬁber diameter, the ﬂow rate in electrospinning is largely determined by the strength of the electric ﬁeld, which, in turn, is limited by the electric breakdown strength of the spinning atmosphere (usually air) [41], while ﬁber diameters are two to three orders of magnitude smaller than conventional polymeric ﬁbers. Thus, the throughput of a single electrospinning needle is typically 0.1–1.0 g h−1 by ﬁber weight or 1.0–5.0 mL h−1 by ﬂow rate, depending on the polymer solution, while, in comparison, millions of tons of ﬁber are produced per annum by conventional spinning methods. Since the low production rate of conventional needle electrospinning setup, as mentioned above, hinders commercialization and limits the application scope of electrospun nanoﬁbers, productivity enhancement on a comparable industrial scale to that of conventional polymeric ﬁbers has been under active investigation over the last 15 years. The emphasis in this drive has been on multi-jet electrospinning, which is the straightforward way to increase the throughput (as the ﬁber productivity can be simply increased by increasing the jet number). In order to obtain multi-jets, an array of multiple needles can be used as the spinneret [39]. However, in such multi-jet electrospinning, a strong repulsion occurs among the jets, and this may lead to reduced ﬁber production as the jets have to be set at an appropriate distance from one another to reduce the jet repulsion, thus requiring a large space to accommodate the needles for the mass nanoﬁber production. Kim et al. [42] have shown that by using an extra-cylindrical electrode as an auxiliary electrode to cover the multi-jet spinneret, the ﬁber deposition area can be dramatically reduced, thus improving the ﬁber production rate. It, however, leads to the formation of coarser ﬁbers. Needle conﬁguration, needle number, and needle spacing are three key parameters for designing the needle array in a multi-needle electrospinning setup, while the needle conﬁguration is of two types, namely, linear arrays and twodimensional (i.e., square, circular and elliptic, hexagonal and triangular) arrays with signiﬁcant effects on ﬂow rate and electric ﬁeld distribution [41]. The adoption of an array of needles as the spinneret for 208 Plastics Technology Handbook electrospinning can also facilitate the production of multicomponent blend nanoﬁbrous mats from different polymer solutions. Besides using multi-needle spinneret, many other devices have been used as electrospinning spinneret. Methods have also been explored to fabricate arrayed capillaries, such as porous membranes with patterned arrays of micrometer-sized channels, or an assembly of tens to hundreds of polyimide-coated glass capillaries of the type used in electrophoresis. Many research efforts have also been directed at improving the electrospinning productivity and/or quality of the electrospun ﬁber by replacing the needle spinneret with other spinnerets, such as conical wire coil, plate, splashing spinneret, rotary cone, and bowl edge. More recently, a number of methods have been developed to enhance the electrospinning throughput that can be roughly classiﬁed as needleless electrospinning methods. Among these, the most successful design for practical applications is the upward needleless electrospinning, which has been shown to have the ability to mass produce nanoﬁbers. In a typical upward needleless electrospinning setup, a two-layerﬂuid system is used [43], in which the lower ﬂuid layer is a ferromagnetic suspension and the upper layer is a polymer solution to be spun. During electrospinning, when a normal magnetic ﬁeld is applied to the system from a permanent magnet or coil, steady vertical spikes are formed perturbing the interlayer interface as well as the free surface of the uppermost polymer layer. Then, as a result of applying high voltage to the ﬂuid at the same time, the perturbations of the free surface become sites of jetting directed upward. As thousands of jetting eject upward, they undergo strong stretching (by the electric ﬁeld) and bending instability, solvent evaporates, and solidiﬁed nanoﬁbers deposit on the upper counterelectrode. A complicated setup is, however, required for the process and the nanoﬁbers formed have a large diameter and a wide diameter distribution. Though nanoﬁbers prepared by electrospinning usually exhibit a solid interior and smooth surface, nanoﬁbers with some speciﬁc secondary (e.g., core-sheath, hollow, and porous) structures can also be prepared if appropriate processing parameters or new designs of spinnerets are employed. For example, CES (described earlier) has been used to produce nanoﬁbers with core-sheath structures [44]. 2.12 Thermoforming When heated, thermoplastic sheet becomes as soft as a sheet of rubber, and it can then be stretched to any given shape [45]. This principle is utilized in thermoforming processes which may be divided into three main types: (a) vacuum forming, (2) pressure forming (blow forming), and (3) mechanical forming (e.g., matched metal forming), depending on the means used to stretch the heat softened sheet. Since fully cured thermoset sheets cannot be resoftened, forming is not applicable to them. Common materials subjected to thermoforming are thermoplastics such as polystyrene, cellulose acetate, cellulose acetate butyrate, PVC, ABS, poly(methyl methacrylate), low- and high-density polyethylene, and polypropylene. The bulk of the forming is done with extruded sheets, although cast, calendered, or laminated sheets can also be formed. In general, thermoforming techniques are best suited for producing moldings of large area and very thin-walled moldings, or where only short runs are required. Thermoformed articles include refrigerator and freezer door liners complete with formed-in compartments for eggs, butter, and bottles of various types, television masks, dishwasher housings, washing machine covers, various automobile parts (instrument panels, arm rests, ceilings, and door panels), large patterned diffusers in the lighting industry, displays in advertising, various parts in aircraft industry (windshields, interior panels, arm rests, serving trays, etc.), various housing (typewriters, Dictaphones, and duplicating machines), toys, transparent packages, and much more. 2.12.1 Vacuum Forming In vacuum forming, the thermoplastic sheet can be clamped or simply held against the RIM of a mold and then heated until it becomes soft. The soft sheet is then sealed at the RIM, and the air from the mold cavity 209 Fabrication Processes (a) Heaters active, stock heating (b) Heaters active, stock heating FIGURE 2.40 Stock on mold, healer’s idle Stock on mold, heaters idle Vacuum applied, stock cooling Plug assist loweredvacuum applied (a) Vacuum forming. (b) Plug-assist forming using vacuum. is removed by a suction pump so that the sheet is forced to take the contours of the mold by the atmospheric pressure above the sheet (Figure 2.40a). The vacuum in the mold cavity is maintained until the part cools and becomes rigid. Straight cavity forming is not well adapted to forming a cup or box shape because as the sheet, drawn by vacuum, continues to ﬁll out the mold and solidify, most of the stock is used up before it reaches the periphery of the base, with the result that this part becomes relatively thin and weak. This difﬁculty is alleviated and uniformity of distribution in such shapes is promoted if the plug assist is used (Figure 2.40b). The plug assist is any type of mechanical helper which carries extra stock toward an area where the part would otherwise be too thin. Plug-assist techniques are adaptable both to vacuum-forming and pressure forming techniques. The system shown in Figure 2.40b is thus known as plug assist vacuum forming. 2.12.2 Pressure Forming Pressure forming is the reverse of vacuum forming. The plastic sheet is clamped, heated until it becomes soft, and sealed between a pressure head and the RIM of a mold. By applying air pressure (Figure 2.41), one forces the sheet to take the contours of the mold. Exhaust holes in the mold allow the trapped air to escape. After the part cools and becomes rigid, the pressure is released and the part is removed. As compared to vacuum forming, pressure forming affords a faster production cycle, greater part deﬁnition, and greater dimensional control. A variation of vacuum forming or pressure forming, called free forming or free blowing, is used with acrylic sheeting to produce parts that require superior optical quality (e.g., aircraft canopies). In this process the periphery is deﬁned mechanically by clamping, but no bolt is used, and the depth of draw or height is governed only by the vacuum or compressed air applied. 210 Plastics Technology Handbook Compressed air open Clamps (a) (b) Vent holes FIGURE 2.41 Pressure forming: (a) heated sheet is clamped over mold cavity; (b) compressed air pressure forces the sheet into the mold. 2.12.3 Mechanical Forming Various mechanical techniques have been developed for thermoforming that use neither air pressure nor vacuum. Typical of these is matched mold forming (Figure 2.41). A male mold is mounted on the top or bottom platen, and a matched female mold is mounted on the other. The plastic sheet, held by a clamping frame, is heated to the proper forming temperature, and the mold is then closed, forcing the plastic to the contours of both the male and the female molds. The molds are held in place until the plastic cools and attains dimensional stability, the latter facilitated by internal cooling of the mold. The matched mold technique affords excellent reproduction of mold detail and dimensional accuracy. 2.13 Casting Processes There are two basic types of casting used in plastics industry: simple casting and plastisol casting. 2.13.1 Simple Casting In simple casting, the liquid is simply poured into the mold without applying any force and allowed to solidify. Catalysts that cause the liquid to set are often added. The resin can be a natural liquid or a granular solid liqueﬁed by heat. After the liquid resin is poured into the closed mold, the air bubbles are removed and the resin is allowed to cure either at room temperature or in an oven at low heat. When completely cured, the mold is split apart and the ﬁnished casting is removed. In the production of simple shapes such as rods, tubes, etc., usually two-piece metal mold with an entry hole for pouring in the liquid resin is used. For making ﬂat-cast acrylic plexiglass or lucite sheets, two pieces of polished plate glass separated by a gasket with the edge sealed and one corner open are usually used as a mold. Both thermosets and thermoplastics may be cast. Acrylics, polystyrene, polyesters, phenolics, and epoxies are commonly used for casting. 2.13.2 Plastisol Casting Plastisol casting, commonly used to manufacture hollow articles, is based on the fact that plastisol in ﬂuid form is solidiﬁed as it comes in contact with a heated surface [46]. A plastisol is a suspension of PVC in a liquid plasticizer to produce a ﬂuid mixture that may range in viscosity from a pourable liquid to a heavy paste. This ﬂuid may be sprayed onto a surface, poured into a mold, spread onto a substrate, etc. The plastisol is converted to a homogeneous solid (“vinyl”) product through exposure to heat [e.g., 350°F (176°C)], depending on the resin type and plasticizer type and level. The heat causes the suspended 211 Fabrication Processes resin to undergo fusion—that is, dissolution in the plasticizer (Figure 2.42)—so that on cooling, a ﬂexible vinyl product is formed with little or no shrinkage. The product possesses all the excellent qualities of vinyl plastics. Dispersion-grade PVC resins are used in plastisols. These resins are of ﬁne particle size (0.1–2 mm in diameter), as compared to suspension type resins (commonly 75–200 mm in diameter) used in calender and extrusion processing. A plastisol is formed by simply mixing the dispersion-grade resin into the plasticizer with sufﬁcient shearing action to ensure a reasonable dispersion of the resin particles. (PVC plasticizers are usually monomeric phthalate esters, the most important of them being the octyl esters based on 2-ethylhexyl alcohol and isooctyl alcohol, namely dioctyl phthalate and diisooctyl phthalate, respectively.) The ease with which virtually all plastisol resins mix with plasticizer to form a smooth stable dispersion/paste is due to the ﬁne particle size and the emulsiﬁer coating on the resin particles. (The emulsiﬁer coating aids the wetting of each particle by the plasticizer phase.) The liquid nature of the plastisol system is the key to its ready application. The plastisol may be spread onto a cloth, paper, or metal substrate, or otherwise cast or slushed into a mold. After coating or molding, heat is applied, which causes the PVC resin particles to dissolve in the plasticizer and form a cohesive mass, which is, in effect, a solid solution of polymer in plasticizer. The various changes a plastisol system goes through in the transformation from a liquid dispersion to a homogeneous solid are schematically shown in Figure 2.43. At 280°F (138°C) the molecules of plasticizer begin to enter between the polymer units, and fusion beings. If the plastisol were cooled after being brought to this temperature, it would give a cohesive mass with a minimum of physical strength. Full fusion occurs and full strength is accomplished when the plastisol is brought to approximately 325°F (163°C) before cooling. The optimum fusion temperature, however, depends on resin type and plasticizer type. For coating applications it is common practice to add solvent (diluent) to a plastisol to bring down viscosity. This mixture is referred to as organosol. It may be applied by various coating methods to form a ﬁlm on a substrate and then is heated to bring about fusion, as in the case of plastisol. Female die Press Heater Plastic sheet Sheet clamp (a) Male die Press Forming plastic sheet Sheet clamp Female die (b) FIGURE 2.42 Male die Matched mold forming: (a) heating; (b) forming. 212 Plastics Technology Handbook 27°C 54°C 80°C Resin particles Gelation: Swollen particles touch; all plasticizer taken by resin Plasticizer Pre-gelation: Resin particle swollen with absorbed plasticizer Liquid plastisol 160°C Fusion: Plasticizer uniformly distributed along polymer chain FIGURE 2.43 solid. 140°C Partial fusion: Plasticizer begins to dissolve polymer Various changes in a plastisol system in the transformation from a liquid dispersion to a homogenous Unlike coating applications, there are some applications where it is desirable to have an inﬁnite viscosity at low shear stress. For such applications, a plastisol can be gelled by adding a metallic soap (such as aluminum stearate) or ﬁnely divided ﬁller as a gelling agent to produce a plastigel. A plastigel can be cold molded, placed on a pan, and heated to fusion without ﬂow. The whole operation is like baking cookies. A rigidsol is a plastisol of such formulation that it becomes a rigid, rather than a ﬂexible, solid when fused. A very rigid product can be obtained when the plasticizer is polymerized during or right after fusion. For example, a rigidsol can be made from 100 parts of PVC resin, 100 parts of triethylene glycol dimethacrylate (network forming plasticizer) and 1 part of di-tert-butyl peroxide (initiator). This mixture has a viscosity of only 3 poises compared with 25 poises for phthalate-based plastisol. However, after being heated for 10 min at 350°F (176°C), the resin solvates and the plasticizer polymerizes to a network structure, forming a hard, rigid glassy solid with a ﬂexural modulus of over 2.5 × 105 psi (1.76 × 104 kg/cm2) at room temperature. Three important variations of the plastisol casting, are dip casting, slush casting, and rotational casting. 2.13.2.1 Dip Casting A heated mold is dipped into liquid plastisol (Figure 2.44a) and then drawn at a given rate. The solidiﬁed plastisol (with mold) is then cured in an oven at 350°F–400°F (176°C–204°C). After a cools, the plastic is stripped from the mold. Items with intricate shapes such as transparent ladies’ overshoes, ﬂexible gloves, etc., can be made by this process. The dipping process is also used for coating metal objects with vinyl plastic. For example, wire dish drainers, coat hangers, and other industrial and household metal items can be coated with a thick layer of ﬂexible vinyl plastic by simply dipping in plastisol and applying fusion. 2.13.2.2 Slush Casting Slush casting is similar to slip casting (drain) of ceramics. The liquid is poured into a preheated hollow metal mold, which has the shape of the outside of the object to be made (Figure 2.44b). The plastisol in immediate contact with the walls of the hot mold solidiﬁes. The thickness of the cast is governed by the time of stay in the mold. After the desired time of casting is ﬁnished, the excess liquid is poured out and the solidiﬁed plastisol with the mold is kept in an oven at 350°F–400°F (176°C–204°C). The mold is then opened to remove the plastic part, which now bears on its outer side the pattern of the inner side of the 213 Fabrication Processes Plastisol (solidiﬁed) Plastisol (solidiﬁed) Plastisol (solidiﬁed) Heated mold Liquid plastisol Liquid plastisol Split mold (a) (b) FIGURE 2.44 Split mold (c) Plastisol casting processes: (a) dip casting; (b) slush casting; (c) rotational casting. metal mold. Slush molding is used for hollow, open articles. Squeezable dolls or parts of dolls and boot socks are molded this way. 2.13.2.3 Rotational Casting In rotational casting a predetermined amount of liquid plastisol is placed in a heated, closed, twopiece mold. The liquid is uniformly distributed against the walls of the mold in a thin uniform layer (Figure 2.43c) by rotating the mold in two planes. The solidiﬁed plastisol in the mold is cured in an oven; the mold is then opened, and the part is removed. The method is used to make completely enclosed hollow objects. Doll parts, plastic fruits, squeeze bulbs, toilet ﬂoats, etc. can be made by rotational casting of plastisols. 2.14 Reinforcing Processes An RP consists of a polymeric resin strengthened by the properties of a reinforcing material [47]. Reinforced plastics occupy a special place in the industry. They are at one and the same time both unique materials into themselves and part and parcel of virtually every other segment of the plastics industry. Reinforced plastics are composites in which a resin is combined with a reinforcing agent to improve one or more properties of the resin matrix. The resin may be either thermosetting or thermoplastic. Typical thermosetting resins used in RPs include unsaturated polyester, epoxy, phenolic, melamine, silicone, alkyd, and diallyl phthalate. In the ﬁeld of reinforced thermoplastics (RTPs), virtually every type of thermoplastic material can be, and has been, reinforced and commercially molded. The more popular grades include nylon, polystyrene, polycarbonate, polyporpylene, polyethylene, acetal, PVC, ABS, styrene-acrylonitrile, polysulfone, polyphenylene sulﬁde, and thermoplastic polyesters. The reinforcement used in RP is a strong inert material bound into the plastic to improve its strength, stiffness, or impact resistance. The reinforcing agent can be ﬁbrous, powdered, spherical, crystalline, or whisker, and made of organic, metallic, or ceramic material. Fibrous reinforcements are usually glass, although asbestos, sisal, cotton and high-performance ﬁbers (discussed later) are also used. To be structurally effective, there must be a strong adhesive bond between the resin and the reinforcement. Most reinforcements are thus treated with sizes or ﬁnishes to provide maximum adhesion by the resins. Although by deﬁnition, all RPs are composites (i.e., combinations of two materials—resin and reinforcement—that act synergistically to form a new third material RP with different properties than the original components) the term advanced composites, or high-strength composites, has taken on a special meaning. The term is applied to stiffer, higher modulus combinations involving exotic reinforcements such as graphite, boron, or other high-modulus ﬁbers like aromatic polyamide ﬁbers (Nomex and Kevlar) and extended-chain polyethylene ﬁbers (Spectra ECPE). And resins like epoxy or some of the newer high 214 Plastics Technology Handbook heat-resistant plastics—polyamides, polyamideimide, polyquinoxalines, and polyphenylquinoxalines. Prime outlets for these materials are in the aerospace and aircraft industries. 2.14.1 Molding Methods As mentioned above, either thermosetting and thermoplastic resin can be used as the matrix component in RPs. Today, the RTP have become an accepted part of the RPs business, although smaller than the reinforced thermosets. RTP are generally made available to the processor in the form of injection molding pellets (into which glass has been compounded) or concentrates (also a pellet but containing a much higher percentage of reinforcement, and designed to be mixed with nonreinforced pellets). It is also possible for the processor to do his own compounding of chopped glass and thermoplastic powder in injection molding. In rotational molding and casting techniques, the processor usually adds his own reinforcement. For thermoforming, glass-RTP laminates are sold commercially. Structural foam molding of glass-RTPs and reinforcing molded urethane foams appeared as later developments in the ﬁeld of RTPs. Among thermosetting resins, unsaturated polyesters are by far the most widely used in RPs, largely because of their generally good properties, relatively easy handling, and relatively low cost. For special uses, however, other types are signiﬁcant: epoxies for higher strength, phenolics for greater heat resistance, and silicones for their electrical properties and heat resistance. All these resins must be used in conjunction with a system of catalysts or curing agents in molding thermoset composites. The type and amount strongly affect the properties, working life, and molding characteristics of the resin. The polyester and epoxies are most often mixed with a catalyst just prior to molding. The most widely used catalyst for polyesters is benzoyl peroxide. Where heat is not available for curing, special catalyst-promoter systems can be used. With epoxies, an amine curing agent that reacts with the resin is most often used. However, there are many other types to choose from (see Chapter 4). The polyester, epoxy, and thermosetting acrylic resins are usually thick liquids that become hard when cured. For this reason, they are most often combined with the reinforcement, by the molder, by dipping or pouring. There are available, however, preimpregnated reinforcements (prepregs) for the molder who wants to keep the operations as simple as possible. Several methods are employed to make RPs. Although each method has the characteristics of either molding or casting, the process may be described as (1) hand lay-up. (2) spray-up, (3) matched molding, (4) vacuum-bag molding, (5) pressure-bag molding, (6) continuous pultrusions, (7) ﬁlament winding, and (8) prepreg molding. 2.14.1.1 Hand Lay-Up or Contact Molding A mold is ﬁrst treated with a release agent (such as wax or silicone-mold release), and a coating of the liquid resin (usually polyester or epoxy) is brushed, rolled, or sprayed on the surface of the mold. Fiberglass cloth or mat is impregnated with resin and placed over the mold. Air bubbles are removed, and the mat or cloth is worked into intimate contact with the mold surface by squeegees, by rollers, or by hand (Figure 2.45a). Additional layers of glass cloth or mat are added, if necessary, to build up the desired thickness. The resin hardens due to curing, as a result of the catalyst or hardener that was added to the resin just prior to its use. Curing occurs at room temperature, though it may be speeded up by heat. Ideally, any trimming should be carried out before the curing is complete, because the material will still be sufﬁciently soft for knives or shears. After curing, special cutting wheels may be needed for trimming. Lowest-cost molds such as simple plaster, concrete, or wood are used in this process, since pressures are low and little strength is required. However, dimensional accuracy of the molded part is relatively low and, moreover, maximum strength is not developed in the process because the ratio of resin to ﬁller is relatively high. 215 Fabrication Processes Resin brushed on Chopped ﬁber Spray gun Roller Roller Molding cures in air Mold (a) FIGURE 2.45 Layers of resin and ﬁbers (b) Molding (a) Basic hand lay-up method. (b) Spray-up technique. The hand lay-up process can be used for fabricating boat hulls, automobile bodies, swimming pools, chemical tanks, ducts, tubes, sheets, and housings, and for building, machinery, and autobody repairs. 2.14.1.2 Spray-Up A release agent is ﬁrst applied on the mold surface, and measured amounts of resin, catalyst, promoter, and reinforcing material are sprayed with a multiheaded spray gun (Figure 2.44b). The spray guns used for this work are different from those used for spraying glazes, enamels, or paints. They usually consist of two or three nozzles, and each nozzle is used to spray a different material. One type, for example, sprays resin and promoter from one nozzle, resin and catalyst from another, and chopped glass ﬁbers from a third. The spray is directed on the mold to build up a uniform layer of desired thickness on the mold surface. The resin sets rapidly only when both catalyst and promoter are present. This method is particularly suitable for large bodies, tank linings, pools, roofs, etc. 2.14.1.3 Matched Metal Molding Matched metal molding is used when the manufacture of articles of close tolerances and a high rate of production are required. Possible methods are perform molding, sheet molding, and dough molding. In preform molding the reinforcing material in mat or ﬁber form is preformed to the approximate shape and placed on one-half of the mold, which was coated previously with a release agent. The resin is then added to the perform, the second half of the mold (also coated previously with a release agent) is placed on the ﬁrst half, and the two halves of the mold are then pressed together and heated (Figure 2.46). The resin ﬂows, impregnates the perform, and becomes hard. The cured part is removed by opening the mold. Because pressures of up to 200 psi (14 kg/cm2) can be exerted upon the material to be molded, a higher ratio of glass to resin may be used, resulting in a stronger product. The cure time in the mold depends on the temperature, varying typically from 10 min at 175°F (80°C) to only 1 min at 300°F (150°C). The cure cycle can thus be very short, and a high production rate is possible. The molding of sheet-molding compounds (SMC) and dough-molding compounds (DMC) is done “dry”—i.e., it is not necessary to pour on resins. SMC, also called prepreg, is basically a polyester resin mixture (containing catalyst and pigment) reinforced with chopped strand mat or chopped roving and formed into a pliable sheet that can be handled easily, cut to shape, and placed between the halves of the heated mold. The application of pressure then forces the sheet to take up the contours of the mold. DMC is a doughlike mixture of chopped strands with resin, catalyst, and pigment. The charge of dough, also called premix, may be placed in the lower half of the heated mold, although it is generally wise to perform it to the approximate shape of the cavity. When the mold is closed and pressure is applied, DMC ﬂows readily to all sections of the cavity. Curing generally takes a couple of minutes for mold temperatures 216 Plastics Technology Handbook Press ram Press platen Steam cores Female mold Molding Resin Preform Guide pins Stops Male mold Press platen (a) FIGURE 2.46 (b) Matched metal molding: (a) before closing of die; (b) after closing of die. from 250 to 320°F (120°C–160°C). This method is used for the production of switch gear, trays, housings, and structural and functional components. 2.14.1.4 Vacuum-Bag Molding In vacuum-bag molding the reinforcement and the resin mixed with catalyst are placed in a mold, as in the hand layup method, and an airtight ﬂexible bag (frequently rubber) is place over it. As air is exhausted from the bag, atmospheric air forces the bag against the mold (Figure 2.47). The resin and reinforcement mix now takes the contours of the mold. If the bag is placed in an autoclave or pressure chamber, higher pressure can be obtained on the surface. After the resin hardens, the vacuum is destroyed, the bag opened and removed, and the molded part obtained. The technique has been used to make automobile body, aircraft component, and prototype molds. 2.14.1.5 Pressure-Bag Molding In pressure-bag molding the reinforcement and the resin mixed with catalyst are placed in a mold, and a ﬂexible bag is placed over the wet lay-up after a separating sheet (such as cellophane) is laid down. The bag is then inﬂated with an air pressure of 20–50 psi (1.4–3.5 kg/cm2). The resin and reinforcement follow the contours of the mold (Figure 2.48). After the part is hardened, the bag is deﬂated and the part is removed. The technique has been used to make radomes, small cases, and helmets. Clamp To vacuum To vacuum Gasket Molded part Glass resin lay-up Flexible bag Flexible bag Mold (a) FIGURE 2.47 (b) Vacuum-bag molding: (a) before vacuum applied; (b) after vacuum applied. 217 Fabrication Processes Rubber bag (not inﬂated) Rubber bag (inﬂated) Air pressure line Pressure back-up plate Clamps Glass resin lay-up Air Molded part Cellophane Mold (a) FIGURE 2.48 (b) Pressure-bag molding: (a) during lay-up; (b) during curing. 2.14.1.6 Filament Winding In the ﬁlament-winding method, continuous strands of glass ﬁber are used in such a way as to achieve maximum utilization of the ﬁber strength. In a typical process, rovings or single strands are fed from a reel through a bath of resin and wound on a suitably designed rotating mandrel. Arranging for the resin impregnated ﬁbers to tranverse the mandrel at a controlled and predetermined (programmed) manner (Figure 2.49) makes it possible to lay down the ﬁbers in any desired fashion to give maximum strengths in the direction required. When the right number of layers have been applied, curing is done at room temperature or in an oven. For open-ended structures, such as cylinders or conical shapes, mandrel design is comparatively simple, either cored or solid steel or aluminum being ordinarily used for the purpose. For structures with integrally wound end closures, such as pressure vessels, careful consideration must be given to mandrel design and selection of mandrel material. A sand-poly(vinyl alcohol) combination, which disintegrates readily in hot water, is an excellent choice for diameters up to 5 ft (1.5 m). Thus, a mandrel made of sand with water-soluble poly(vinyl alcohol) as a binder can be decomposed with water to recover the ﬁlament-wound part. Other mandrel materials include low-melting alloys, eutectic salts, soluble plasters, frangible or breakout plasters, and inﬂatables. Because of high glass content, ﬁlament-wound parts have the highest strength-to-weight ratio of any reinforced thermoset. The process is thus highly suited to pressure vessels where reinforcement in the Rotating mandrel Traversing resin bath Supply of roving FIGURE 2.49 Sketch of ﬁlament winding. 218 Plastics Technology Handbook highly stressed hoop direction is important. Pipe installation, storage tanks, large rocket motor cases, interstage shrouds, high-pressure gas bottles, etc., are some of the products made of ﬁlament winding. The main limitation on the process is that it can only be used for fabricating objects which have some degree of symmetry about a central axis. 2.14.1.7 Pultrusion FIGURE 2.50 Pultruded FRP Pull mechanism Die and heat source Resin soaked ﬁber Resin impregnator Tension roller Reinforcement material (ﬁbers, or woven or braided strands) Pultrusion is the process of “pulling” raw composites (i.e., resin-impregnated ﬁber or cloth) through a heated die, creating a continuous composite proﬁle of high ﬁber content. The term pultrusion combines the words pull and extrusion to signify that material is forced through the die by pulling, instead of pushing. In the process (Figure 2.50), continuous reinforcement materials like ﬁbers or oven mat or braided strands are impregnated with resin and then pulled through a long, heated stationary die, where the resin undergoes polymerization, while the die controls the resin content. The impregnation is done either by pulling the reinforcement through a resin bath or by injecting the resin into an injection chamber that is connected to the die. Resin can also be injected directly into the die in some pultrusion systems. Most often, the reinforcement is glass ﬁber, but it can also be carbon, aramid, or a mixture. The impregnating resin is almost always a thermosetting resin with unsaturated polyesters accounting for nearly 90% and epoxies for the balance. Other thermosetting resin types, such as polyurethane and vinyl esters, are also used. Though economic and environmental factors favor the use of a thermoplastic resin as matrix, the high viscosity of thermoplastic melts makes high productivity and high degree of resin impregnation of the ﬁber difﬁcult to achieve. In the standard pultrusion process, the raw ﬁber is pulled off the racks/rolls and guided through a resin bath or resin impregnation system whereby the ﬁber reinforcement becomes fully impregnated (wetted out) with the resin such that all the ﬁber ﬁlaments are thoroughly saturated with the resin mixture. The uncured composite material is then guided through a series of tooling (known as “pre-former”) that helps arrange and organize the ﬁber into the correct shape, while excess resin is squeezed out (known as “debulking”). The raw composite then passes through a precisely machined steel die, which is heated to a constant temperature, and may have several zones of temperature throughout the length to cure the thermosetting resin. The proﬁle that exits from the die is a cured, pultruded FRP. The process yields straight constant cross-section parts of virtually any shippable length with high unidirectional strength (e.g., I-beams, rods, and shafts). Schematic of the pultrusion process. (After https://en.wikipedia.org/wiki/Pultrusion.) 219 Fabrication Processes The pultrusion technology is not limited to thermosetting polymers. It has been used successfully with thermoplastic polymers, such as poly(ethylene terephthalate) (PET) and poly(butylene terephthalate) (PBT), by impregnating the glass ﬁber with resin powder or surrounding it with sheet material of the thermoplastic resin, followed by heating to fuse the resin [48]. Further, in a novel adaptation of the pultrusion process for thermoplastic polymers, known as reactive thermoplastic pultrusion (RTP), very low viscosity reactive monomers of thermoplastic polymers have been used to enable impregnation of a great quantity (up to 85%) of ﬁbers in the composite. One of the ﬁrst companies to pioneer the RTP process is the French company “CQFD Composites,” which commercially developed this technology to produce linear or curved structural proﬁles. The process (Figure 2.51) has two steps. In Step 1, a proprietary formulation of low-viscosity monomer (caprolactam), containing catalyst, activator, additives, and suitable ﬁbers, is introduced under pressure into a pultrusion die. In Step 2, a thermoplastic polymer (nylon-6) is synthesized (see Section 4.3.2.1) in situ among the ﬁbers in the pultrusion die, while shaping of the proﬁle takes place under heat and pressure. (The polymerization reaction being sensitive to humidity and oxygen, it requires a well-controlled process to proceed efﬁciently.) Structural proﬁles of composites can thus be produced with extremely high content (up to 85 wt%) of reinforcing ﬁbers, which can generate strategic advantages in construction, transport, or electrical applications. With modulus ranging from 50 to 60 GPa, such RTP proﬁles, unlike conventional thermoset proﬁles, are also heat-deformable and hence can be post-shaped under heat. Structural RTP proﬁles have been used as insert to be the structural “backbone” of injection-molded parts, especially in the automobile industry. Since the standard pultrusion process involves pulling the materials through a stationary die, it is suited to manufacture only straight proﬁles (see Figure 2.52b). In a later modiﬁcation of the process, developed by Thomas GmbH + Co.Technik + Innovation KG, the die is not stationary but moves back and forth along the proﬁle to be manufactured. The modiﬁed process, known as Radius-Pultrusion, enables production of two- and three-dimensional curved proﬁles in endless circles and arches of any radius (see Figure 2.52c). The proﬁles can be reinforced with endless ﬁbers (glass, carbon, or natural) in a unidirectional or with the help of netting or webbing [strong fabric commonly made of synthetic ﬁbers, such as nylon, polypropylene, or polyester, and woven as a ﬂat strip (Figure 2.52a) or tube of varying width, often used in place of rope] in a bidirectional way. The Radius-Pultrusion technology allows nearly unlimited application of ﬁber-reinforced materials for engineers and architects in various ﬁelds, for example, automotive and transportation industry (e.g., ﬁxtures for bumpers; dashboards; structural proﬁles; springs; frame and body for lorries, buses, and trains; container walls; aircraft bodies; and naval architecture), building industry/architecture (e.g., window proﬁles, arched proﬁles, bridges, scaffolding Fibers Catalyst + activator Monomer Additives Composite Pultrusion die (heated) FIGURE 2.51 Schematic of the reactive thermoplastic pultrusion process of CQFD composites. 220 Plastics Technology Handbook (a) (c) (b) FIGURE 2.52 (a) Webbing, (b) pultruded straight proﬁles from the standard pultrusion process, (c) pultruded curved proﬁles from the Radius-Pultrusion process. and ladders, stairs and rails, and grids), and sports (e.g., tennis rackets, squash courts, golf clubs, battens for surﬁng and sailing, surfboards, and ﬁshing rods). 2.14.1.8 Prepreg Molding Optimal strength and stiffness of continuous ﬁber–reinforced polymeric composites is obtained through controlled orientation of the continuous ﬁbers. One means to achieve this is by prepreg molding [49]. In this process, unidirectionally oriented layers of ﬁbers are pre-impregnated with the matrix resin and cured to an intermediate stage of polymerization (B-stage). When desired, this pre-impregnated composite precursor, called a prepreg, can be laid up in the required directions in a mold for quick conversion into end components through the use of hot curing techniques. Prepregs can thus be described as pre-engineered laminating materials for the manufacture of ﬁber–reinforced composites with controlled orientation of ﬁbers. For the designer, a precisely controlled ply of prepreg represents a building block, with well-deﬁned mechanical properties from which a structure can be developed with conﬁdence. For the fabricator, on the other hand, prepregs provide a single, easy-to-handle component that can be applied immediately to the lay-up of the part to be manufactured, be it aircraft wing skin or ﬁshing rod tube. The prepreg has the desired handleability already built in to suit the lay-up and curing process being utilized, thus improving efﬁciency and consistency. Prepregs have been used since the late 1940s, but they have only achieved wide prominence and recognition since the development of the higher performance reinforcing ﬁbers, carbon, and kevlar. The quantum leap in properties provided by these new ﬁbers generated a strong development effort by prepreg manufacturers and there is no doubt that new developments made in this area have been as signiﬁcant, if not as evident, as that of the introduction of the new ﬁbers. The use of prepreg in the manufacture of a composite components offers several advantages over the conventional wet lay-up formulations: 1. Being a readily formulated material, prepreg minimizes the material’s knowledge required by a component manufacturer. The cumbersome process of stocking various resins, hardeners, and reinforcements is avoided. Fabrication Processes 221 2. With prepreg a good degree of alignment in the required directions with the correct amount of resin is easily achieved. 3. Prepreg offers a greater design freedom due to simplicity of cutting irregular shapes. 4. Material wastage is virtually eliminated as offcuts of prepregs can be used as random molding compounds. 5. Automated mass-production techniques can be used for prepreg molding, and the quality of molded product is reproducible. 6. Toxic chemical effects on personnel using prepregs are minimized or eliminated. A ﬂowchart showing the key stages in the fabrication of composite structures from raw materials with an intermediate prepregging step is given in Figure 2.53. It can be seen that there are two basic constituents to prepreg—the reinforcing ﬁber and the resin system. All of the advanced reinforcing ﬁbers are available in continuous form, generally with a ﬁxed ﬁlament diameter. The number of ﬁlaments that the supplier arranges into a “bundle” (or yarn) varies widely, and is an important determinant of the ability to weave fabric and make prepregs to a given thickness. In the majority of cases, the yarn is treated with a size to protect it from abrasion during the weaving or prepregging process. Often, the size is chosen such that it is compatible with the intended resin system. Resin systems have developed into extremely complex multi-ingredient formulations in an effort to ensure the maximum property beneﬁt from the ﬁber. Normally there are four methods of impregnation: (1) solution dip; (2) solution spray; (3) direct hot-melt coat; and (4) ﬁlm calendaring. The solution dip and solution spray impregnation techniques work with a matrix resin dissolved in a volatile carrier. The low viscosity of the resin solution allows good penetration of the reinforcing ﬁber bundles with resin. In solution dipping, the ﬁber, in yarn or fabric form, is passed through the resin solution and picks up an amount of solids dependent upon the speed of through-put and the solids level. With solution spraying, on the other hand, the required amount of resin formulation is metered directly onto the ﬁber. In both cases, the impregnated ﬁber is then put through a heat cycle to remove the solvent and “advance” the chemical reaction in the resin to give the correct degree of tack. Direct hot melt can be performed in a variety of ways. In one method, the reinforcing ﬁber web is dipped into a melt resin bath. A doctor blade, scraper bar, or metering roller controls the resin content. Alternately, the melt resin is ﬁrst applied to a release paper, the thickness of the resin being determined by a doctor blade. The melt resin on the release paper is then brought into contact with a collimated ﬁber bundle and pressed into it in a heated impregnation zone. In ﬁlm calendaring, which is a variation of the above method, the resin formulation is cast into a ﬁlm from either hot-melt or solution and then stored. Thereafter in a separate process, the reinforcing ﬁber is sandwiched between two ﬁlms and calendered so that the ﬁlm is worked into the ﬁber. The decision of which method to use is dependent upon several factors. Holt melt ﬁlm processes are faster and cheaper, but certain resin formulations cannot be handled in this way, and hence solution methods have to be used. The solution dip method is often preferred for fabrics as the need to squeeze hot-melt and ﬁlm into the interstices of the fabric can cause distortion of the weave pattern. In a process based on a biconstituent two impregnation concept [49], the polymeric matrix is introduced in ﬁbrous form and a comingled two of polymer and reinforcing ﬁbers are fed into a heated impregnation zone (Figure 2.54). In this method, it may be possible to effect better wetout of the reinforcing ﬁbers with the matrix polymer, especially with high viscosity thermoplastic matrix polymers, through intimate comingling. The machines necessary to accomplish the above prepregging procedures are many and varied. There are three distinct aspect to quality control: raw material screening, on-line control, and batch testing. All three are obviously important, but the ﬁrst two are more critical. Fibers and base resins are supplied against certiﬁcates of conformance and often property test certiﬁcates. On-line control during the manufacture of prepreg revolves around the correct ratio of ﬁber to resin. This is done by a traversing Beta-gauge, which scans the dry ﬁber (either unidirectional of Elastomers Epoxy Phenolic Polyester Polyimide Solvents Hardeners Thermoplastic Tape Thickness 50–300 μm Width 25–1200 mm Single yam Crossplied packs Resin contents 34% – 45% Woven Thickness 100–500 μm Width 500–1500 mm Resin contents 30% – 60% STD weaves Bias weaves Triaxial Hybrids Prepregs Woven fabrics Intermediate Tape winding Automatic cutting Hand lay-up Tape laying Lay-up Oven/bag Autoclave Press Curing Fastening Bonding Drilling Trimming Finishing Structure FIGURE 2.53 Flow chart showing key stages in the fabrication of composite structures from raw materials by prepreg molding. (After Lee, W. J., Seferis, J. C., and Bonner, D. C. 1986. SAMPE Q., 17, 2, 58.) Resin systems Fibers Glass Kevlar Carbon Boron Others Raw material 222 Plastics Technology Handbook 223 Fabrication Processes Heat Polymer ﬁber Reinforcing ﬁber Polymer matrix Reinforcing ﬁber FIGURE 2.54 Biconstituent tow impregnation using a comingled tow of polymer and reinforcing ﬁbers. (After Lee, W. J., Seferis, J. C., and Bonner, D. C. 1986. SAMPE Q., 17(2), 58.) fabric) and then the impregnated ﬁber and provides a continuous real-time plot of the ratio across the width of the material. This can be linked back to the resin system application point for continuous adjustment. Batch testing is carried out to verify prepreg properties, such as resin content, volatile level, and ﬂow. The resin “advancement” (chemical reaction) is monitored via a Differential Scanning Calorimeter (DSC) and the formulation consistency by testing the Tg via DSC or Dynamic Mechanical Analyzer (DMA). The laminate properties are also determined. All are documented and quoted on a Release Certiﬁcate. Commercial prepregs are available with different trade names. Fibredux 914 of Ciba–Geigy is a modiﬁed epoxy resin preimpregnated into unidirectional ﬁbers of carbon (HM or HT), glass (E type and R type), or aramid (Kevlar 49) producing prepregs that, when cured to form ﬁber reinforced composite components, exhibit very high strength retention between −60°C and +180°C operational temperatures. “Scotchply” brand RPs of Industrial Specialties Division of 3M Company are structural-grade thermosetting molding materials, consisting of unidirectional nonwoven glass ﬁbers embedded in an epoxy resin matrix. The product is available both in prepreg form in widths up to 48 in. (1.22 m) and in ﬂat sheet stock in sizes up to 48 in. (1.22 m) × 72 in. (1.83 m). The cured product is claimed to possess extraordinary fatigue life, no notch sensitivity, high ultimate strength, and superior corrosion resistance. The range of application of the RP includes vibratory springs, sonar housings, landing gear, picker blades, snowmobile track reinforcement, helicopter blades, seaplane pontoons, missile casing, and archery bow laminate. 2.14.2 Fibrous Reinforcements Although many types of reinforcements are used with plastics, glass ﬁbers predominate. Fibrous glass reinforcements are available in many forms (described below). Asbestos is used in the form of loose ﬁber, paper, yarn, felt, and cloth. The two largest uses of asbestos in plastics are with PVC in vinyl asbestos tile and with polyesters and polypropylene. Most natural and synthetic ﬁbers do not have the strength required for a RPs part. However, when intermediate strengths are satisfactory, they can be used. In this category are nylon, rayon, cotton fabrics, and paper. Sisal ﬁbers have also found use as a low cost reinforcing material in premix molding compounds. High-modulus graphite and carbon ﬁbers, aramid ﬁbers and ECPE ﬁbers are playing a more and more important role in RPs. Boron ﬁlaments, with outstanding tensile strengths, are usually used in the form of prepreg tapes and have been primarily evaluated for the aerospace and aircraft industry. 2.14.2.1 Glass Fibers A high-alkali “A-glass” and a low-alkali “E-glass” are used as reinforcements for polymer composites, the latter being used most often. Since the modulus of E-glass is 10.5 × 106 psi and the tensile strength 224 Plastics Technology Handbook upwards of 250,000 psi, it is not surprising that the stiffness and strength of most plastics can be increased by compounding with glass. A more chemical resistant glass, sodium borosilicate (C-glass), and a highertensile-strength glass, S-glass, are also available. E-glass is a calcium alumino-silicate, and S-glass is a magnesium aluminosilicate. Fiberglass is available as a collection of parallel ﬁlaments (roving), chopped strands, mat, and woven fabric. Glass ﬁlaments are produced by melting a mixture of silica, limestone, and other reactants, depending on the type of glass and forcing the molten product through small holes (bushings). The hot ﬁlaments are gathered together and cooled by a water spray. These multiple glass ﬁlaments are gathered together into a bundle, called a strand, which is wound up on a coil. Short ﬁbers (staple) are produced by passing a stream of air across the ﬁlaments as they emerge from the bushings. Rovings are rope-like bundles of continuous untwisted strands for use in such processes as perform press molding, ﬁlament winding, spray-up, pultrusion, and centrifugal casting. They can also be converted into chopped strand mats or cut into short ﬁbers for molding compounds. Chopped strands of glass 1/32–1/2 in. in length can be incorporated in thermoset or thermoplastic materials about as easily as the particulate ﬁllers. Each strand may be made up of 204 individual ﬁlament whose diameter is 2–7.5 × 10−4 in. Chopped strands several inches long can be loosely bound as a mat that is porous and in which the strands are randomly oriented in two dimensions. This form is suitable for impregnation by a liquid polymer. After polymerization or cross-linking (curing) under pressure, the composite will comprise a polymer-network matrix in which the individual strands are embedded. Chopped strand mats provide nondirectional reinforcement (i.e., strengths in many directions, as contrasted to unidirectional forms which are continuous ﬁbers, like roving, that provide strength in one direction). These mats are available in a variety of thicknesses, usually expressed in weight per square foot. In order to hold the ﬁbers together, a resin binder is generally used, the type depending on the resin and molding process. In some cases, the mats are stitched or needled, instead of using the resin binder. A woven glass fabric (cloth) might be used in place of the mat or in combination with it. In this case there will be a variation in strength with the angle between the axis of the ﬁbers and the direction of stress. Twisted yarns are generally woven into fabrics of varying thicknesses and with tight or loose weaves, depending upon the application. Most are balanced weaves (i.e., equal amounts of yarn in each direction), although some are unidirectional (more ﬁbers running in one direction). Although costlier, they offer a high degree of strength. Rovings can also be woven into a fabric that is less costly than the woven yearn fabrics, coarser, heavier, and easier to drape. For optimum adhesion at the interface between the ﬁber surface (stationary phase) and the resin matrix (continuous phase), the glass ﬁbers must be treated with coupling agents to improve the interfacial adhesion. The pioneer coupling agent (linking agent) was methacrylatochromic chloride (Volan). This has been supplanted by organosilanes, organotitanates, and organozirconates. These coupling agents contain functional groups, one of which is attracted to the ﬁber surface and the other to the resin (Figure 2.55). 2.14.2.2 Graphite/Carbon Fibers, the Beginning Graphite carbon ﬁbers are the predominant high-strength, high-modulus reinforcing agent used in the fabrication of high-performance polymer composites. In general, the term graphite ﬁber refers to ﬁbers that have been treated above 1,700°C (3,092°F) and have tensile moduli of elasticity of 5 × 105 psi (3,450 MPa) or greater. Carbon ﬁbers are those products that have been processed below 1,700°C (3092°F) and consequently exhibit elastic moduli up to 5 × 105 psi (3,450 MPa) [47]. A further distinction is that the carbon content of carbon ﬁbers is 80%–95%; and that of graphite, above 99%. However, the industry has universally adopted the term “graphite.” It will therefore be used to describe both product forms in this section. Graphite ﬁbers were ﬁrst utilized by Thomas Edition in 1880 for his incandescent lamps. The ﬁlaments were generated by the carbonization of bamboo in the absence of air. When tungsten ﬁlaments replaced the graphite in lamps, interest in graphite materials waned until the mid 1950s when rayon-based graphite 225 Fabrication Processes HO CH = CH2 Si OH OH Si Si O O + Cl CH = CH2 O O Si Si Si O O Cl Cl Si Slightly hydrolyzed glass surface FIGURE 2.55 Si + Vinyl trichlorosilane Attachment to glass surface Mechanism of functioning of a glass surface ﬁnish. ﬁbers were created. These products exhibited relatively high tensile strengths of about 4 × 105 psi (2,760 MPa) and were designed for rocket/missile ablative component applications. A signiﬁcant event that led to the development of today’s graphite industry was the utilization of PAN as a graphite precursor material by Tsunoda in 1960 [50]. Subsequent work led to continued improvement of PAN-based graphite ﬁber properties by numerous researchers. These developments focused on stretching the PAN precursor to obtain a high degree of molecular orientation of the polymer molecules followed by stabilizing it under tensile load, carbonization, and graphitization. PAN-based graphite ﬁbers are now available with tensile moduli of up to 1.2 × 106 psi (8,280 MPa) and tensile strengths above 8 × 105 psi (5,516 MPa). Pitch was ﬁrst identiﬁed as a graphite precursor by Otani in 1965 [51]. These ﬁbers are made by melt spinning a low-cost isotropic molten (petroleum) pitch and then oxidizing the ﬁlaments as they are spun. This step is followed by carbonization at 1,000°C (1832°F) in an inert atmosphere. Process modiﬁcations to improve the ﬁber properties evolved through the 1970s until pitch-based (including mesophase liquid crystal pitch) graphite ﬁbers with tensile strengths up to 3.75 × 105 psi (2,590 MPa) and tensile moduli to 1.2 × 106 psi (8,300 MPa) were achievable. 2.14.2.3 Manufacture of Graphite (Carbon) Fibers The pyrolysis of organic ﬁbers used as graphite precursors is a multistage process. The three principal graphite precursors are PAN, pitch, and rayon, with PAN as the predominant product. The commercial production of PAN precursor ﬁber is based on either dry or wet spinning technology. In both instances, the polymer is dissolved in either an organic or inorganic solvent at a concentration of 5–10% by weight. The ﬁber is formed by extruding the polymer solution through spinneret holes into hot gas environment (dry spinning) or into a coagulating solvent (wet spinning). The wet spinning is more popular as it produces ﬁbers with round cross section, whereas the dry spinning results in ﬁber with a dog bone cross section. The wet-spun precursor making process [52] includes three basic steps: polymerization, spinning, and after treatments (Figure 2.56). Acrylonitrile monomer and other comonomers (methyl acrylate or vinyl acetate) are polymerized to form a PAN copolymer. The reactor efﬂuent solution, called “dope,” is puriﬁed, the unreacted monomers removed, and the solid contaminants ﬁltered off. The spinning process next extrudes the puriﬁed dope through holes in spinnerettes into a coagulating solution. The spun gel ﬁber then passes through a series of after-treatments such as stretching, oiling, and drying. The product is the PAN precursor. In order to produce high-strength, high-modulus graphite ﬁbers from the PAN precursor, it is essential to produce preferred molecular orientation parallel to the ﬁber axis and then “stabilize” the ﬁber against relaxation phenomena and chain scission reactions that may occur in subsequent carbonization steps. A typical step-by-step PAN-based graphite manufacturing process begins with the aforementioned 226 Plastics Technology Handbook Comonomers + catalysts Solvent Polymerization Filtration Precursor ﬁber FIGURE 2.56 Winding or piddling Extrusion coagulation Washing After treatments (stretching, oiling, drying) Typical PAN precursor manufacturing steps based on solution polymerization and wet spinning. precursor stabilization which is followed by carbonization, graphitization, surface treatment, and sizing, as shown schematically in Figure 2.57. The stabilization of the PAN precursor involves “preoxidation” by heating the ﬁber in an air oven at 200°C–300°C (392°F–572°F) for approximately one hour while controlling the shrinkage/tension of the ﬁber so that the PAN polymer is converted into a thermally infusible aromatic ladder-like structure. The next step is the process of carbonization, which pyrolyzes the stabilized PAN-based ﬁbers until they are transformed into graphite (carbon) ﬁbers. The carbonization treatment is done in an inert atmosphere (generally nitrogen) at temperatures greater than 1,200°C (2,192°F). This step removes hydrogen, oxygen, and nitrogen atoms from the ladderlike polymers whose aromatic rings then collapse into a graphitelike polycrystalline structure. It is during this stage that high-mechanical-property characteristics of graphite ﬁbers are developed. The development of these properties is directly related to the formation and orientation of graphitelike ﬁbers or ribbons within each individual ﬁber. Graphitization performed at temperatures above 1,800°C (3,272°F) is an optional treatment. Its purpose is to improve the tensile modulus of elasticity of the ﬁber by improving the crystalline structure and orientation of graphitelike crystallites within each individual ﬁber. The higher heat-treatment temperature used in graphitization also results in a higher carbon content of that ﬁber. The ﬁnal step in the process of producing graphite or carbon ﬁber is surface treatment and sizing prior to bobbin winding the continuous ﬁlaments. The surface treatment is essentially an oxidation of the ﬁber surface to promote wettability and adhesion with the matrix resin in the composite. Sizing improves handleability and wettability of the ﬁber with the matrix resin. Typical sizing agents are poly(vinyl alcohol), epoxy, and polyimide. Pitch-based graphite ﬁbers are produced by two processes. The precursor of one of these processes is a low-softening-point isotropic pitch and the process scheme includes the following steps: (1) melt-spin isotropic pitch; (2) thermoset at relatively low temperatures for long periods of time; (3) carbonize in an inert atmosphere at 1,000°C (1,832°F); (4) stress graphitize at high temperatures 3,000°C (5,432°F). The high-performance ﬁbers produced in this manner are relatively expensive because of the very long thermosetting time required and the need for high-temperature stress graphitization. The commercially more signiﬁcant process for making pitch-based ﬁbers is the mesophase process, which involves the following steps: (1) heat treat in an inert atmosphere at 400–450°C (752–842°F) for an extended period of time in order to transform pitch into a liquid-crystalline (mesophase) state; (2) spin the mesophase pitch into ﬁbers; (3) thermoset the ﬁbers at 300°C (572°F) for 2½ h; (4) carbonize the ﬁbers at 1,000°C (1,832°F); and (5) graphitize the ﬁbers at 3,000°C (5,432°F). Since long thermosetting times and 227 Fabrication Processes Pan precursor Carbonization Stabilization Surface treatment Sizing Carbon fiber Graphitization O Stabilization N N N N N (200°C – 300°C in air) Pan precursor O C H2N C N N N N OH Carbonization 1,000 – 3,000°C in inert gas Graphite (carbon) fiber FIGURE 2.57 Schematic of a typical step-by-step PAN-based graphite manufacturing process. stress graphitization treatment are not required, the high-temperature graphite ﬁbers produced by this process are lower in cost. The process by which rayon precursor is converted to graphite ﬁbers includes four steps: (1) ﬁber spinning; (2) stabilization at 400°C (752°F) for long periods of time; (3) carbonization at 1,300°C (2,372°F); and (4) stress graphitization at high temperatures 3,000°C (5,432°F). The rayon-based graphite ﬁbers produced in this manner tend to be relatively expensive because of the very long stabilization times required and the need for stress graphitization at high temperatures. 2.14.2.4 Graphite/Carbon Fibers and Fabrics The excellent properties of graphite are directly attributable to the highly anisotropic nature of the graphite crystal. The standard-grade PAN-based graphite ﬁbers, which make up the largest part of both the commercial and aerospace markets, have tensile strengths ranging from 4.5 × 105 to 5.5 × 105 psi (3100– 3800 MPa) and moduli of approximately 340,000 psi (2345 MPa). Further, a family of intermediatemodulus/high-strain ﬁbers with tensile strengths up to 7 × 105 psi (4800 MPa) and modulus above 4 × 105 psi (2760 MPa) have been developed to meet high-performance aerospace requirements. The highmodulus ﬁbers (both PAN- and pitch-based), used in high-stiffness/low-strength applications, such as space hardware, have tension moduli ranging from 5 × 105 to 1.2 × 106 psi (3450–8280 MPa) and strain to failures generally greater than or equal to 1%. Graphite ﬁbers are available to the user in a variety of forms—continuous ﬁlament for ﬁlament winding, braiding or pultrusion, tow for creation of ﬁber cloth and fabrics, chopped ﬁber for injection or compression molding, impregnated woven fabrics for lay-ups, and unidirectional tapes for lamination. [It may be mentioned that man-made ﬁbers are usually extruded into ﬁlaments, which are then converted into ﬁlament yarn (composed of continuous ﬁlaments assembled with or without twist), staple (cut lengths from ﬁlaments), or tow (a large strand of continuous ﬁber ﬁlaments collected in loose, rope-like form).] 228 Plastics Technology Handbook A carbon ﬁber ﬁlament is much thinner than a human hair. The strands of carbon ﬁber are bundled together to create what is known as tow and the tow is used to create the carbon ﬁber cloth. Carbon ﬁber ﬁlaments created in different ways by different processes have their own set of structural values and properties like strength and stiffness as noted above. There are several variations of tow sizes, the most common sizes being 3k, 6k, and 12k, where “k” stands for thousands (thus, there are 3000 individual ﬁlaments in a 3k tow). While there are several types of carbon ﬁber, differing in strength and modulus, there are several tow size differences and many different cloth patterns [53], such as twill, satin, plain, unidirectional (Uni), triaxial, and so on to choose from. A common way carbon ﬁber cloth is listed is similar to the following two examples: (a) T700S-12k, 7 oz plain weave and (b) AS4-6k, 11 oz harnesssatin 4 weave. While the tags “T700S” and “AS4” refer to the brand or type of carbon ﬁber, the numbers “12k” and “6k” relate to the tow size; the “7 oz” and “11 oz” are usually the weights of the cloth per yard; the “plain weave” and “4 harness-satin weave” indicate how the tow is woven into cloth. The more underlapping and overlapping (or “bumpy”) the ﬁbers within the weave, the weaker it will be because as the cloth begins to come under tension, the straightening tow in it is subjected to a shear force and it snaps. Thus, a plain weave will be more prone to breaking than a 4 harness-satin weave using the same type of tow. A harness-satin 4 weave also has fewer overlapping bumps than a 2 × 2 twill weave, which has less overlapping than a plain weave. The strongest clothes are the nonwoven types like unidirectional cloth. [A unidirectional fabric is one in which the primary ﬁbers (i.e., tow bundles) run in one direction only, usually in 0° direction (i.e., the warp direction), while a small amount of ﬁber or other material may run in other directions mainly to hold the primary ﬁbers in position.] Among the weaves, the plain weave is the most difﬁcult to drape (the ability of a fabric to conform to a complex surface). Woven fabrics are produced by the interlacing of warp (0°) ﬁbers and weft (90°) ﬁbers in a regular pattern or weave style. Weaves are generally referred and deﬁned by notation such as 1 × 1, 2 × 2, 4 × 4, and 3 × 1. The ﬁrst number in a set refers to how many strands are crossed “over” before going “under” the perpendicular strands (in a 90° weave). The second number refers to how many strands are crossed “under” before going back “over” the perpendicular strands. A plain weave, which is deﬁned as a 1 × 1 weave, would thus run as over, under, over, under, over, under, and so on. Similarly, a 3 × 1 weave would run as over, over, over, under, over, over, over, under, over, over, over, and so on. A twill weave is deﬁned as a set of identical number (>1) of weave both under and over, as for example, 2 × 2, 4 × 4 (see Figure 2.58). This produces the visual effect of a straight or broken diagonal “rib” to the fabric. A twill weave thus has a 3D “look” to it, which is so often desired. It is also much easier to bend around complex curves than a plain weave because its weave is more loose. Superior wet out and drape is seen in the twill weave over the plain weave with only a small reduction in stability. A satin weave is characterized by three or more weft yarns passing (“ﬂoating”) over a warp yarn or vice versa. A satin fabric tends to have a high luster because of the high number of “ﬂoats” (i.e., missed interlacings) on the fabric. The “harness” number used in the designation of satin (typically 4, 5, and 8) is the total number of ﬁbers crossed and passed under, before the ﬁber repeats the pattern. A 3 × 1 harnesssatin (see Figure 2.58) is referred to as a harness-satin 4, H4, or 4HS and a 4 × 1 harness-satin is referred to as a harness-satin 5 (5HS or H5). Satin weaves are very ﬂat and have good wet out and a high degree of drape. A harness-satin bends over complex curves better than either a plain or a twill weave, and it almost always has more weaves per inch than the other two. For applications of carbon ﬁbers, the most important properties to consider are the tensile strength, modulus, and elongation/strain. Tensile strength is how hard it is to break by pulling it apart from end to end. The higher the tensile strength, the harder it is to break. Modulus is a measure of the stiffness. The higher the modulus, the stiffer the carbon ﬁber is. A high-modulus carbon ﬁber will be stiffer but also weaker in tensile strength than a low or standard modulus ﬁber. Thus, a high-modulus carbon ﬁber will give stiffer but weaker and more brittle parts, whereas a high-strength/low-modulus carbon ﬁber will give stronger but ﬂexible parts. To consider a typical application such as bike frames, one may choose highmodulus carbon ﬁber as this will give a stiffer bike frame while using less material (hence making the frame lighter) and because the loss of strength is not large enough to compromise the safety of the frame. 229 Fabrication Processes (a) (b) (c) FIGURE 2.58 Different types of woven fabrics: (a) a plain (1 × 1) weave, (b) a twill (2 × 2) weave, and (c) a satin (3 × 1) weave. With low-modulus carbon ﬁber, on the other hand, to get a frame that is as stiff as a high-modulus frame, one will need more carbon ﬁber and the frame will be heavier. In general, if there are no complex curves to be covered and aesthetics are not important, a plain weave is the best option. If, however, aesthetics are very important, a twill weave is generally selected, while for a sophisticated look, a harness-satin H7 or H8 is often used, the latter being the best choice for very complex curves. 2.14.2.5 Graphite/Carbon Fiber-Reinforced Plastics Graphite (or carbon) ﬁber-reinforced plastics (CFRP is preferred to the term GFRP, which also stands for glass ﬁber-reinforced plastics) having graphite/carbon, or simply carbon, ﬁbers as reinforcement are strong and light, relatively more expensive to produce, but are commonly used wherever high strength-to-weight ratio and rigidity are required, such as aerospace, automotive, sports goods, and many other consumer and technical applications. The matrix or binding polymer is often a thermoset resin, such as epoxy, but other thermoset or thermoplastic polymers, such as polyester, vinyl ester, or nylon are sometimes used. Besides carbon ﬁber as reinforcement, CFRPs may also contain other ﬁbers such as aramid (e.g., Kevlar, Twaron), aluminum, ultrahigh-molecular-weight polyethylene (UHMWPE), and additives, such as silica, rubber, and carbon nanotubes. Because CFRP consists of two distinct components, namely, carbon ﬁber reinforcement, which provides the strength, and a polymer resin (matrix), which binds the reinforcements together, the material properties of the composite will depend on the respective properties of these two components. However, unlike isotropic materials like metals and alloys, CFRP will have directional strength properties depending on the layouts of the carbon ﬁber and the properties of the carbon ﬁbers relative to the polymer matrix. The two different equations governing the net elastic modulus of a composite (using the properties of the ﬁber reinforcement and the matrix resin) can also be applied to CFRP. Thus, the following equation (see Equation 3.129 for derivation) Ec = Em fm + Ef ff (2.3) is valid for CFRP with the ﬁbers oriented in the direction of the applied load; Ec is the modulus of the composite; fm and ff are the volume fractions of the matrix and ﬁber, respectively, in the composite; and Em and Ef are the elastic moduli of the matrix and ﬁbers, respectively. The other extreme case of the elastic modulus of the composite with the ﬁbers oriented transverse to the applied load can be found from the following equation: Ec = Em Ef Em ff + Ef fm (2.4) Although CFRPs with epoxy have high strength and elastic modulus, they exhibit virtually no plasticity, with less than 0.5% strain to failure. Efforts to toughen CFRPs include replacing epoxy with alternative 230 Plastics Technology Handbook matrix resins. One such resin with high promise is PEEK, which exhibits an order of magnitude greater toughness with similar elastic modulus and strength. However, PEEK is considerably more difﬁcult to process and is also more expensive. While having the advantage of high initial strength-to-weight ratio, compared to common structural materials, CFRP has a design limitation that arises from the fact that because of its complex failure modes, the fatigue failure properties are difﬁcult to predict and design for. As a consequence, when using CFRP for critical cyclic-loading applications, considerable strength safety margins should be provided in design to ensure adequate component reliability during service. Like other polymer-based composites, CFRPs can also be profoundly affected by humidity and temperature, their combined action leading to degradation of the composites’ mechanical properties, particularly at the matrix–ﬁber interface, where the diffusing moisture plasticizes the polymer matrix. The carbon ﬁbers, moreover, can cause galvanic corrosion when CFRP parts are attached to aluminum. 2.14.2.6 Manufacture of CFRP Parts The process by which most CFRP products are made varies depending on the pieces being created, the ﬁnish (outside gloss) required, and the number of the particular piece to be produced. The molding methods described in Section 2.14.1 for FRPs, in general, can also be used for producing CFRP parts. Thus, using the hand lay-up method, highly corrosion-resistant, stiff, and strong CFRP parts can be made by layering sheets of carbon ﬁber cloth into a mold (see Figure 2.45) in the shape of the ﬁnal product, choosing the alignment and weave of the cloth ﬁbers to optimize the strength and stiffness of the product. The carbon cloth or mat is impregnated with the matrix resin and is air- or heat-cured. Parts used in less critical areas are manufactured by draping cloth over a mold, with the resin either pre-impregnated into it (known as prepreg) or sprayed over it. High-performance parts using single molds are often vacuum-bagged (see Figure 2.47) or autoclavecured in order to avoid the presence of small air bubbles in the material as such bubbles reduce the strength. Using vacuum-bagged (atmospheric) oven-cured material systems for secondary aircraft structures (such as ﬂaps, fairings, etc.) is well established. However, to produce composite materials with less than 1% void content and superior mechanical properties required for aerospace primary structures, such as wings, fuselage, and empennage components with integrated stiffeners, autoclave curing is employed. An alternative to the autoclave method is to use internal pressure via inﬂatable air bladders or expandable polystyrene foam inside the noncured laid-up carbon fabric. Vacuum-bagging is generally used for CFRP molding when simple objects and relatively few pieces (1–2 per day) are to be made. Typically, an aluminum mold is polished and waxed, and a release agent is applied, followed by carbon fabric and resin. Vacuum is then applied and the assembly is set aside to allow the piece to harden (cure). The resin can be applied in three ways to the fabric in a vacuum mold. One is wet lay-up, where the two-part epoxy resin is mixed, poured, and spread on the carbon fabric, and the fabric, then referred to as “wet prepreg cloth,” is laid in the mold and placed in the bag. The other is infusion, where the fabric and the mold are placed inside the bag while the resin is pulled in by vacuum through a tube and then through a device that causes even spreading of the resin throughout the fabric. The third method of applying resin for vacuum-bagging is a dry lay-up, where a carbon ﬁber prepreg (“dry prepreg cloth”) is laid in the mold in a way similar to adhesive ﬁlm. The assembly is then placed in a vacuum and allowed to cure (∼120°C). [Note that a carbon ﬁber prepreg, usually referred to as “dry carbon cloth,” is carbon cloth that has been pre-impregnated with epoxy, which is not wet like a regular lay-up epoxy, but it is usually a bit sticky at room temperature. The dry carbon cloth is usually stored in a freezer.] As there is no excess resin bleed-out and resin waste is minimal in the process, the resulting prepreg parts are generally lighter and have fewer pinholes than parts made by wet lay-up. However, to achieve still better pinhole elimination by purging gases along with minimal resin use, autoclaving is generally required. Fabrication Processes 231 Compression molding, also known as matched metal molding (Section 2.14.1.3), is a high-volume, high-pressure method that can be used for molding complex, high-strength CFRP parts. The process uses a two-piece (male and female) mold (see Figure 2.46), commonly made of aluminum or steel, that is pressed together with the fabric and resin between the two, while curing takes place under heat and pressure. The cured part is removed by opening the mold. Close tolerance of the molded parts and very high rate of productions can be achieved. However, initial cost may be high since the molds require machining of very high precision, such as computer numerical control (CNC) prototype machining. [Note: CNC is the automation of machine tools that are operated by precisely programmed commands encoded on a storage medium, as opposed to manual controlling. Machine movements that are controlled by cams, gears, levers, or screws in conventional machines are directed by computer and digital circuitry in CNC machines ensuring high precision [54].] For difﬁcult and convoluted shapes, ﬁlament winding (Section 2.14.1.6) can be used to create strong and durable CFRP structures. Filament winding (Figure 2.49) is the process of winding resin-impregnated ﬁber on a mandrel surface in a precise geometric pattern. This is accomplished by rotating the mandrel while a delivery head precisely positions the carbon ﬁbers or ﬁlaments on the mandrel surface. Thus, CFRP structures can be made with properties stronger than steel at much lighter weights. Filamentwound CFRP pressure vessel has been one of the most effective solutions for high-pressure storage. From pressure vessels to tubing to intricate components for medical devices to aerospace and military parts, ﬁlament winding can be used to meet high-performance demands of many critical applications. The pultrusion process (see Section 2.14.1.7) can be used to produce CFRP proﬁles, such as rods, angles, tubes, and sheets with maximum rigidity and minimum mass. The proﬁles are produced by pulling carbon ﬁbers and resins through a heated die in a continuous process (Figure 2.50) that aligns the ﬁbers lengthwise (in the direction of pulling). Carbon being 70% lighter than steel, 40% lighter than aluminum, and having three times the stiffness of either for the same weight, the aligned carbon ﬁbers contribute greatly to the rigidity, while minimizing weight. For example, 0.125-in-diameter pultruded carbon rods with 67% ﬁber volume and bisphenol epoxy vinyl ester as the matrix resin has the following typical properties: density, 1.5 g/cm3; tensile strength, 2.34 GPa; tensile modulus, 134 GPa; compressive strength, 1.90 GPa; compressive modulus, 131 GPa; ultimate tensile strain, 1.30%; and glass transition temperature, 100°C [55]. 2.14.2.7 Applications of CFRP Products Graphite composites have exceptional mechanical properties that are unmatched by other materials. The principal advantage of graphite composites are high speciﬁc stiffness (stiffness divided by density), high speciﬁc strength (strength divided by density), and extremely low coefﬁcient of thermal expansion (CTE). Graphite composites are also nonpoisonous, biologically inert, and transparent to x-rays. Historically, graphite composites have been very expensive, which limited its use to only special applications. However, over the past 15 years, the manufacturing processes have improved and the prices of graphite composites have steadily declined. Consequently, graphite composites are now economically viable in many applications, such as sporting goods and high-performance vehicles, boats, and industrial machinery. Table 2.2 gives a comparison of costs and mechanical properties of graphite composites and several other materials. The properties are listed in ranges as there are a wide variety of graphite ﬁbers and resins, thus allowing numerous combinations and variation of properties [56]. For example, PAN-based carbon ﬁber has higher strength than pitch-based carbon ﬁber, while the latter has higher stiffness and lower (negative) CTE than the former. Any material that is strong and light is characterized by a high strength-to-weight ratio (also known as speciﬁc strength). While strength is resistance to breaking, rigidity or stiffness (measured by Young’s modulus) is resistance to bending or stretching. It is seen from Table 2.2 that both speciﬁc strength and speciﬁc stiffness of graphite composites are several times higher than those of common structural materials like ﬁber glass composites, aluminum, and steel. CFRP is thus the material of choice for applications where lightweight structures need to carry extremely high loads, such as components of spacecraft, ﬁghter aircraft, and race cars. 0.050 1 × 106 to 1.8 × 106 160 × 106 to 200 × 106 1 × 10−6 to 2 × 10−6 0.050 1.8 × 106 to 4 × 106 200 × 106 to 1,000 × 106 −1 × 10−6 to 1 × 10−6 Density (lb/in3) Speciﬁc strength(strength/density) Speciﬁc stiffness (Stiffness/Density) CTE (in/in-°F) Source: www.performancecomposites.com/about-composites-technical-info/124-designing-with-carbon-ﬁber.html. Note: Conversion factors: 1000 psi = 6.895 MPa; 1 lb/in3 = 27,675 kg/m3; 1 in/in-°F = 1.8/K. 50,000–90,000 8 × 106 to 10 × 106 90,000–200,000 10 × 106 to 50 × 106 $5–$20 $20–$250+ Strength (psi) Stiffness (psi) Graphite Composite (Commercial Grade) Cost $/lb Graphite Composite (Aerospace Grade) TABLE 2.2 Comparison of Properties of Graphite Composites with Those of Other Structural Materials 6 × 10−6 to 8 × 10−6 363,640–636,360 18 × 106 to 27 × 106 0.055 20,000–35,000 1 × 106 to 1.5 × 106 $1.50–$3.00 Fiber Glass Composite 13 × 10−6 350,000 100 × 106 0.10 35,000 10 × 106 $3 Aluminum 6061 T-6 7 × 10−6 200,000 100 × 106 0.30 60,000 30 × 106 $0.30 Steel, Mild 232 Plastics Technology Handbook Fabrication Processes 233 Carbon ﬁber has been described as a “game-changing material.” This could not be truer as it is in sporting goods. The high-strength and lightweight properties of carbon ﬁber have taken sporting goods to the next level of performance. Thus, golf shafts, racquets, skis, snowboards, hockey sticks, ﬁshing rods, and bicycles have all been advanced through CFRP. As carbon ﬁber reduces weight without sacriﬁcing strength, its use in marine applications means that yachts, cruisers, and racing vessels will be lighter and stronger when made with CFRP. Tough and durable CFRP material stands up to the extremes of marine environments and its high speciﬁc stiffness lends itself well to use in such applications as masts, hulls, and propellers. CTE is a measure of how much a material expands and contracts when the temperature goes up or down. As the data in Table 2.3 show, carbon ﬁber can have a broad range of CTEs, −1 to +8. The variation occurs because of the direction of measurement, the fabric weave, and the precursor material (e.g., higher CTE of PAN-based ﬁber and lower CTE of pitch-based ﬁber). A negative CTE for graphite ﬁber means that when the ﬁber is heated, it will shrink. So when such graphite ﬁber is put into a resin matrix (positive CTE), the composite can be tailored to have zero or very small CTE (see Table 2.2). Such graphite composites are therefore used for applications where small movements owing to temperature change can be critical, for example, telescope and other optical machinery, high-precision antennas, and scanning and imaging machines. The application of carbon ﬁber in the wind energy industry has led to the development of a lighter, longer, stiffer, and stronger wind turbine blade, pushing the industry to higher levels of performance, as longer blades mean more energy output per revolution. Being nonpoisonous, biologically inert, and x-ray permeable, carbon ﬁber is useful in medical applications, for example, prosthesis, implants, tendon repair, x-ray accessories, surgical instruments, and so on. However, the matrix, either epoxy or polyester, can be toxic and proper care needs to be exercised. Carbon ﬁber is used for 3D printing applications in both milled and chopped grades. In combination with a wide range of resins, parts that have exceptional mechanical properties can be created. While the major markets for advanced graphite ﬁber composites, as mentioned above, are aerospace, marine, automotive, industrial equipment, and recreation, military aerospace applications dominate the market and military consumptions are slated to increase rapidly as programs, which utilize a very high percentage of composites, move from development to large-scale production. Graphite ﬁber usage in space applications is in a large measure linked to space station programs and production activities. Examples of non-aerospace military applications include portable, rapid deployment bridges for the army, and propeller shafts for submarines. Fiber usage in the commercial aerospace sector is also growing. Commercial planes such as Boeing 767 and the Airbus A320 utilize two to three times the graphite ﬁber per plane that is used in older commercial models. The biggest industrial market potential of graphite ﬁbers is in the automotive sector. The graphite composite usage in this area should increase as lower-cost ﬁbers become available. A major and growing use of chopped graphite ﬁbers in the industrial market is as a reinforcement for thermoplastic injection molding compounds. The advantages of such use include greater strength and stiffness, higher creep and fatigue resistance, increased resistance to wear, higher electrical conductivity, and improved thermal stability and conductivity. Continuous tow, chopped ﬁbers, and milled ﬁbers are produced for both general plastics (e.g., nylons, polycarbonates) and high-temperature engineering thermoplastics (such as PEI, PEEK, PPS, etc). Carbon ﬁber-reinforced carbon (also known as carbon–carbon or C/C) is a composite material consisting of carbon ﬁber reinforcement and graphite matrix. It was developed for the nose cones of intercontinental ballistic missiles and wing leading edges of the Space Shuttle Orbiter. It has also been used in the brake systems (brake discs and pads) of Formula One racing cars since 1976. The carbon–carbon composite is well suited to structural applications at high temperatures, or where thermal shock resistance and/or a low coefﬁcient of thermal expansion is needed. It, however, lacks impact resistance and this must be taken into consideration in designing so as to avoid the likelihood of violent impacts. 234 Plastics Technology Handbook TABLE 2.3 Comparison of Thermal Expansion Coefﬁcients Material CTE (in/in-°F) Steel 7 Aluminum 13 Kevlar Carbon ﬁber woven 3 or lower 3 or lower Carbon ﬁber unidirectional −1 to +8 Fiber glass Brass 7–8 11 Source: www.christinedemerchant.com/carboncharacteristics.html. A carbon–carbon composite material is made in three stages [57]. In the ﬁrst stage, carbon ﬁlament or cloth, surrounded by an organic binder such as plastic or pitch, is ﬁrst laid-up in its intended ﬁnal shape. In the second stage, the lay-up is heated so that the binder undergoes pyrolysis to relatively pure carbon. This is, however, accompanied by the formation of voids. The void formation is reduced by the addition of coke or some other carbon aggregate in the initial lay-up but is not eliminated. In the third stage, the voids are gradually ﬁlled by forcing a carbon-forming gas such as acetylene through the material at a high temperature over several days. This long heat treatment process, which also helps carbon to form into larger graphite crystals, is the main reason for the high cost of carbon–carbon composite. The way the initial carbon ﬁber scaffold is laid up and the quality of the matrix ﬁller strongly inﬂuence the properties of the ﬁnal material such as hardness and thermal properties like resistance to thermal expansion, temperature gradients, and thermal cycling. 2.14.2.8 Aramid Fibers Aramid ﬁber is the generic name for aromatic polyamide ﬁbers. As deﬁned by the U.S. Federal Trade Commission, an aramid ﬁber is a “manufactured ﬁber in which the ﬁber forming substance is a long chain synthetic polyamide in which at least 85% of the amide linkages are attached directly to two aromatic rings”: O O H H O O H H C C N C C N N N n (I) n (II) Among the commercially available aramid ﬁbers are DuPont’s Nomex (I) and Kelvar (II); in fact these trade names are commonly used in lieu of the generic name. Kelvar 49 is a high-modulus aramid ﬁber and is the most widely used reinforcing aramid ﬁber. Kevlar 29 has a lower modulus and Kevlar 149 has a higher modulus than Kevlar 49. Aromatic polyamides are described in greater detail in Chapter 4. Aramid ﬁbers can be used to advantage to obtain composites having lighter weight, greater stiffness, higher tensile strength, higher impact resistance, and lower notch sensitivity than composites incorporating E-glass or S-glass reinforcement. Weight savings over glass result from the lower speciﬁc gravity of aramid ﬁbers, 90.4 lb/in.3 (1.45 g/cm3) versus E-glass, 159.0 lb/in.3 (2.55 g/cm3). Higher stiffnesses are reﬂected in a Young’s modulus up to 19 × 106 psi (1.31 × 105 MPa) for Kevlar 49 and 27 × 106 psi (1.86 × 105 MPa) for Kevlar 149, compared to 107 psi (6.9 × 104 MPa) for E-glass and 12 × 106 psi (8.6 × 104 MPa) for S-glass. Aramid composites are more insulating than their glass counterparts, both electrically and thermally, more damped to mechanical and sonic vibrations, and are transparent to radar and sonar. Fabrication Processes 235 Despite their outstanding mechanical properties, these high-modulus organic ﬁbers have the processability normally associated with conventional textiles. This leads to wide versatility in the form of the reinforcement, e.g., yarns, rovings, woven and knit goods, felts, and papers. 2.14.2.9 Applications Fabrics woven from Kevlar 49 aramid ﬁber are often used as composite reinforcement, since fabrics offer biaxial strength and stiffness in a single ply. The mechanical properties of Kevlar 49 aramid are dependent on the fabric construction. The composite properties are functions of the fabric weave and the ﬁber volume fraction (typically 50%–55% with ply thickness 5–10 mils, depending on fabric construction). In 1987, DuPont introduced high-modulus Kevlar 149. Compared to Kevlar 49 it has higher performance (47% modulus increase) and lower dielectric properties (65% decrease in moisture regain). Over the past three decades, Kevlar has gained wide acceptance as a ﬁber reinforcement for composites in many end uses, such as tennis rackets, golf clubs, shafts, skis, ship masts, and ﬁshing rods. The boating and aircraft industries make extensive use of advanced composites. The advanced composites have allowed innovative designers to move ahead in designing aircraft with unprecedented performance. The cost of high-modulus aramid ﬁbers is higher than E-glass and equivalent to some grades of S-glass on a unit-weight basis. Price differences versus glass are, however, reduced by about half on a unit volume basis when lower density of the aramids is taken into account. For many applications, fabrics containing more than one ﬁber type offer signiﬁcant advantages. Hybrids of carbon and Kevlar 49 aramid yield greater impact resistance over all-carbon construction and higher compressive strength over all-Kevlar construction. Hybrids of Kevlar 49 and glass offer enhanced properties and lower weight than constructions containing glass as the sole reinforcement and are less expensive than constructions using only Kevlar 49 reinforcement. 2.14.2.10 Extended-Chain Polyethylene Fibers Extended-chain polyethylene (ECPE) ﬁbers are relatively recent entrants into the high-performance ﬁbers ﬁeld. Spectra ECPE, the ﬁrst commercially available ECPE ﬁber and the ﬁrst in family of extended chain polymers manufactured by Allied-Signal, Inc. was introduced in 1985. ECPE ﬁbers are arguably the highest modulus and highest strength ﬁbers made. These are being utilized as a reinforcement in such applications as ballistic armor, impact shields, and radomes to take advantage of the ﬁber’s unique properties. Polyethylene is a ﬂexible molecule that normally crystallizes by folding back on itself (see Chapter 1). Thus ﬁbers made by conventional melt spinning do not possess outstanding physical properties. ECPE ﬁbers, on the other hand, are made by a process that results in most of the molecules being fully extended and oriented in the ﬁber direction, producing a dramatic increase in physical properties. Using a simple analogy, the structure of ECPE ﬁbers can be described as that of a bundle of rods, with occasional entangled points that tie the structure together. Conventional PE, by comparison, is comprised of a number of short-length chain folds that do not contribute to material strength (see Figure 2.59). ECPE ﬁbers are, moreover, made from ultrahigh molecular weight polyethylene (UHMWPE) with molecular weight generally 1–5 million that also contributes to superior mechanical properties. Conventional PE ﬁbers, in comparison, have molecular weights in the range 50,000 to several hundred thousand. ECPE ﬁbers exhibit a very high degree of crystalline orientation (95%–99%) and crystalline content (60%–85%). High-modulus PE ﬁbers can be produced by melt extrusion and solution spinning. The melt extrusion process leads to a ﬁber with high modulus but relatively low strength and high creep whereas solution spinning in which very high-molecular-weight PE is utilized yields a ﬁber with both high modulus and high strength. The solution spinning process for a generalized ECPE ﬁber starts with the dissolution of polyethylene of approximately 1–5 million molecular weight in a suitable solvent. This serves to disentangle the polymer chains, a key step in achieving an extended chain polymer structure. The solution must be fairly dilute to facilitate this process, but viscous enough to be spun using conventional melt spinning equipment. The cooling of the extrudate lends to the formation of a ﬁber that can be continuously dried to 236 Plastics Technology Handbook Conventional ﬁber Extended-chain ﬁber (a) (b) FIGURE 2.59 Fiber morphology of polyethylene. (a) Conventional PE ﬁber characterized by relatively low molecular weight, moderate orientation, and crystalline regions chain folded. (b) Extended-chain PE ﬁber characterized by very high molecular weight, very high degree of orientation, and minimum chain folding. remove solvent or later extracted by an appropriate solvent. The ﬁbers are generally postdrawn prior to ﬁnal packaging. The solution spinning process is highly ﬂexible and can provide an almost inﬁnite number of process and product variations of ECPE ﬁbers. Fiber strengths of (3.75–5.60) × 105 psi (2,890–3,860 MPa) and tensile moduli of (15−30) × 106 psi [(103−207) × 103 MPa] have been achieved. The properties are similar to other high-performance ﬁbers; however, because the density of PE is approximately two-thirds that of high modulus aramid and half that of high-modulus carbon ﬁber, ECPE ﬁbers possess extraordinarily high speciﬁc strengths and speciﬁc moduli. Figure 2.60 compares the speciﬁc strength versus speciﬁc modulus for currently available ﬁbers. Traditional binders and wetting agents are ineffective in improving resin adhesion to polyethylene. For ECPE ﬁbers, this characteristic is actually advantageous in speciﬁc areas. For instance, ballistic performance is inversely related to the degree of adhesion between the ﬁber and the resin matrix. However, for applications requiring higher levels of adhesion and wetout, it has been shown that by submitting ECPE ﬁber to speciﬁc surface treatments, such as corona discharge or plasma treatments, the adhesion of the ﬁber to various resins can be dramatically increased. The chief application areas being explored and commercialized for ECPE ﬁbers are divided between traditional ﬁber applications and high-tech composite applications. The former include sailcloth, marine ropes, cables, sewing thread, nettings, and protective clothing. The latter includes impact shields, ballistics, radomes, medical implants, sports equipment, pressure vessels, and boat hulls. ECPE ﬁbers (such as Spectra 1,000) are well suited for high-performance yachting sails, offering, in addition, resistance to sea water and to typical cleaning solutions used in the sailing industry, such as 15 Speciﬁc strength (106 in.) Spectra 1000 Spectra 900 10 Kevlar 29 ‘S’ glass 5 Kevlar 149 Kevlar 49 ‘E’ glass HT graphite Boron HM graphite Steel 0 FIGURE 2.60 100 300 500 700 Speciﬁc modulus (106 in.) Comparative properties of various reinforcing ﬁbers. 900 Fabrication Processes 237 bleach. The major sport equipment applications to date have been canoes, kayaks, and snow and water skis. Numerous other sport applications are under development. The high-strength, lightweight, low-moisture absorption and excellent abrasion resistance of ECPE make it a natural candidate for marine rope. In marine rope applications, load, cycling, and abrasion resistance are critical. Thus a 12-strand ECPE braid, for example, is reported to withstand about eight times the number of cycles that cause failure in 12-strand aramid braid. Specially toughened and dimensionally stabilized ECPE yarn has been used in a revolutionary new line of cut-resistant products. ECPE ﬁbers are being used to produce cut-resistant gloves, arm guards, and chaps in such industries as meat packing, commercial ﬁshing, and poultry processing and in sheet metal work, glass cutting, and power tool use. ECPE’s high strength and modulus and low speciﬁc gravity offer higher ballistic protection at lower density per area than is possible with currently used materials. The signiﬁcant applications include ﬂexible and rigid armor. Flexible armor is manufactured by joining multiple layers of fabric into the desired shape, the ballistic resistance being determined by the style of the fabric and the number of layers. Traditional rigid armor can be made by utilizing woven ECPE ﬁber in either thermoset or thermoplastic materials. Ballistics are currently the dominant market segment. Products include helmets, helicopter seats, automotive and aircraft armor, armor radomes, and other industrial structures. The radome (radome protective domes) market is also important for ECPE ﬁbers. ECPE composite systems act as a shield that is virtually transparent to microwave signals, even in high-frequency regimes. 2.15 Reaction Injection Molding A new type of injection molding called Reaction Injection Molding (RIM) has become important for fabricating thermosetting polymers [58,59]. RIM differs from the conventional injection molding in that the ﬁnished product is made directly from monomers or low-molecular-weight polymeric precursors (liquid reactants), which are rapidly mixed and injected into the mold even as the polymerization reaction is taking place. Thus, synthesis of polymers prior to molding is eliminated, and the energy requirements for handling of monomers are much less than those for viscous polymers. For RIM to be successful, the monomers or liquid reactants must be fast reacting, and the reaction rates must be carefully synchronized with the molding process. Thus the polymers most commonly processed by RIM are polyurethanes and nylons though epoxies, and certain other polymers such as polycyclopentadiene have been processed by RIM. The process uses equipment that meters reactants to an accuracy of 1%, mixes them by high-pressure impingement, and dispenses the mixture into a closed mold. The mold is, in fact, a chemical reactor. The reaction in RIM takes place in a completely ﬁlled mold cavity. Reinforcing ﬁllers are sometimes injected into the mold along with the reactants, a process called reinforced reaction injection molding (RRIM). In the mold the functional groups of the liquid reactants react to form chemical linkages, producing solid polymers, which comprise polymeric chains or networks depending on the starting materials. The temperature of the mold plays a vital role in the polymerization of reactants. To produce a molded part by the RIM process requires precise but realistic process control. Figure 2.61 shows a simpliﬁed schematic of the RIM process. The important elements of the process are conditioning, metering, mixing, and molding. All the liquid reactants require precise temperature control. The ﬂow property (viscosity) of the liquid reactants usually varies with temperature as does the density. For accurate metering the temperature must be controlled within very narrow limits. This is usually accomplished by recirculating reactants from conditioning or storage tanks designed to maintain raw material temperatures speciﬁed by the system supplier. These conditions are normally quite moderate (30°C–38°C) for polyurethanes. For some polymerization systems, such as nylon which is processed at a high temperature, the machines are designed with heated lines and temperature control devices for pumps, mixers, and other components. The molds are also designed to control temperatures as the reaction characteristics of the RIM process are exothermic. A higher temperature increases the reaction rate, which results in an decrease of cycle 238 Plastics Technology Handbook Liquid component B Recirculation Liquid component A Conditioning Metering Mixing Reaction/molding FIGURE 2.61 Schematic of RIM process. time. For polyurethane, however, speeding the reaction rate by operating the mold at elevated temperatures is to be avoided, as this changes the types of linkages produced (see “Polyurethanes” in Chapter 4). Polyurethane RIM systems have been commercial in the United States for about 60 years and a bit longer in Europe. It is still a rapidly growing ﬁeld of technology. The automotive industries in the United States account for most of the commercial RIM production. A later development for RIM polyurethane, and to a lesser extent RIM nylons, is the application for housings of various instruments and appliances: computer housings, business machine housings, TV and radio cabinets, instrument cases, and similar electronic product enclosures. While elastomeric RIM is most commonly used in these applications, some housings are also molded from RIM structural foam. Though systems suppliers do not always clearly differentiate between elastomeric and structural RIM, elastomeric RIM is molded in thin cross section (usually 0.125 in.) at high density while structural foam has an interior foam structure, a density about one-third that of elastomeric RIM, and is molded in thicker cross section (usually 0.375 in.). 2.15.1 Machinery Conditioning and temperature control are accomplished by recirculating reactants from storage tanks which are jacketed and/or contain tempering coils to maintain the process temperature required by the chemical system. RIM parts (especially polyurethanes) are usually removed from the mold before the chemical reactions that develop the physical properties are complete. The part is placed on a support jig that holds it in its ﬁnal shape until it is fully cured. In some cases this is done by simply setting the supported part aside for 12–24 h. More often the supported part is postcured in an oven for several hours at temperatures of about 180°F (82°C). Nylons are completely reacted in the mold and postcuring is not necessary. RIM polyurethanes made with aromatic isocyanates (such as pure or polymeric MDI) have a tendency to darken as a result of the effect of UV light on the chemical ring structure of the MDI component. Soft white limestone or ﬁne carbon black is often used as ﬁller to mask the effect of this color change. Polyurethanes manufactured with aliphatic isocyanates are light stable, and products are molded in a wide range of bright colors. Especially interesting is the development of equipment to add color concentrate, usually dispersed in a polyol, directly into the mixing head attached to a given mold. The basic urethane formula is adjusted to compensate for the additional reactive polyol. Using this technique with a 239 Fabrication Processes multiclamp RIM line, it is possible to mold different colors using a single RIM machine. In some cases, aromatic RIM systems are molded in color then painted the same color. This technique eliminates the need to touch up every dent or scratch which would otherwise show up tan or white. There are several systems for painting RIM parts. Usually, RIM parts are simply primed and painted. Before painting, however, the part is cleaned to remove mold-release agents. The most common mold releases are metal stearates (or soaps) that can be removed from the part by a water wash and parafﬁn waxes that are usually removed by solvent vapor degreasing. Silicone mold releases are to be avoided as they are very difﬁcult to remove from the part, and paint will not stick to the silicone surface ﬁlm. 2.15.2 Polyurethanes The most common chemicals used in the RIM process for poyurethanes are isocyanates containing two or more isocyanate (–N═C═O) groups and polyols, which contain two or more hydroxyl (–OH) groups. These reactive end-groups, so named because they occur at the ends of the chemical structure, react chemically to form a urethane linkage: ( H O N C O ) The chemical system, must be adjusted so that the number of isocyanates and hydroxyls balance and that all reactive end-groups are used in the formation of urethane linkages. The number of polymer structures that can be formed using the urethane reaction is quite large. There are ways to produce polyurethanes having different physical properties (see “Polyurethanes” in Chapter 4). If linear polyols are reacted with diisocyanates, a ﬂexible polyurethane will be formed. If a low boiling liquid, such as Refrigerant-11 (R-11), is incorporated into the system, the heat of reaction will produce a cellular structure. The resulting product will be ﬂexible polyurethane foam. The physical properties of these materials can be varied by selecting polyols with shorter or longer polyol chains. The most common polyol or macroglycol chains are polyethers and polyesters (Chapter 4). The composition of these thermoplastic chains also plays a role in the physical properties of the end product. These chain segments in the block copolymer are often referred to as “soft” blocks, or segments, while the polyurethane segments formed by the reaction of diisocyante with glycol are referred to as “hard” blocks or segments. In addition to changing the chain composition and length, the physical properties can be varied by blending up to approximately 10% of a long-chain triol (such as a triol adduct of ethylene oxide and propylene oxide with glycerol) into the basic resin system formulation. This produces branching in the “soft” segment of the block copolymer. Excessive triol modiﬁcation may, however, diminish physical properties. Use of short-chain triols such as glycerol will produce cross-linking in the “hard” segments of the polymer chain. The formation of hard blocks and cross-linking in the hard block tend produce a stiffer, more rigid product. The hard blocks tend to be crystalline and reinforce the amorphous polymer, improving its strength. 2.15.3 Nylons RIM nylons, like polyurethanes, form polymers very rapidly by the reaction of chemical end-groups. The linkages produced are as follows: Polyesteramide prepolymer + (CH2)5 CONH Caprolactam Nylon block copolymer 240 Plastics Technology Handbook Equipment used to manufacture RIM products must be extensively rebuilt to process RIM nylons. Because the viscosity of nylon RIM systems is low and the ingredients are quite reactive, leakage at the seals and the volumetric efﬁciency of the metering pumps (that is, the amount of material actually pumped divided by the volume displaced by the metering pumps) may cause problems, which require special attention. Hence, most manufacturers recommend having the machine designed speciﬁcally for nylon RIM systems. The ﬁrst commercial product made from nylon RIM was a front quarter panel (fender) for the Oldsmobile Omega Sport. Because of the excellent impact strength of nylon RIM, it has been used for bumper covers and automobile fascia. It also ﬁnds application in housings for business machines and electronics. 2.16 Structural Reaction Injection Molding Structural reaction injection molding (SRIM) may be considered as a natural evolution of RIM. It is a very attractive composite manufacturing process for producing large, complex structural parts economically. The basic concepts of the SRIM process are shown in Figure 2.62. A preformed reinforcement is placed in a closed mold, and a reactive resin mixture that is mixed by impingement under high pressure in a specially designed mix head (like that in RIM) ﬂows at low pressure through a runner system to ﬁll the mold cavity, impregnating the reinforcement material in the process. Once the mold cavity is ﬁlled, the resin quickly completes its reaction. A completed component can often be removed from the mold in as little as one minute. SRIM is similar to RIM in its intensive resin mixing procedures and its reliance on fast resin reaction rates. It is also similar to resin transfer molding (RTM) (discussed later in this chapter) in employing performs that are preplaced in the cavity of a compression mold to obtain optimum composite mechanical properties. The term structural is added to the term RIM to indicate the more highly reinforced nature of the composite components manufactured by SRIM. The key to SRIM is the perform. It is a preshaped, three-dimensional precursor of the part to be molded and does not contain the resin matrix. It can consist of ﬁbrous reinforcements, core materials, metallic inserts, or plastic inserts. The reinforcements, cores, or inserts can be anything available that meets the economic, structural, and durability requirements of the parts. This tremendous manufacturing freedom allows a variety of alternative perform constructions. Most commercial SRIM applications have been in general industry or in the automotive industry. The reinforcement material most commonly used has been ﬁberglass, due to the low cost. Fiberglass has been used in the form of woven cloth, continuous strand mat, or chopped glass. Space-shaping cores can be used in the SRIM process to fabricate thick, three-dimensional parts with low densities. Speciﬁc grades of urethane-based foams, having densities of 6–8 lb/ft3 and dimensional stability at SRIM molding temperatures, are commonly used as molded core materials. Fiberglass reinforcements and inserts can be placed around these cores, resulting in SRIM parts, molded in one piece, that are very light-weight and structurally strong and stiff. Metallic inserts can be used in SRIM parts as local stiffeners, stressed attachment points, or weldable studs. The metallic material of choice is usually steel. SRIM is a very labor-intensive process, and the consistency from preform to preform is usually poor. However, for very low manufacturing volumes this process can be cost-effective. Most SRIM resins have several characteristics in common: their liquid reactants have roomtemperature viscosity below 200 cps; their viscosity-cure curves are sigmoidal in shape, the typical moldﬁll time being 10–90 sec; and their demold time is from 60 to 180 sec, varying with catalyst concentration. The low viscosity of SRIM resins and their relatively long ﬁll times are crucial in allowing them to penetrate and ﬂow through their reinforcing performs. The design of the gating and runner conﬁguration (if any) is usually kept proprietary by the molder. However, it appears that most SRIM parts are center-gated, with vents located along the periphery of the part. This conﬁguration allows the displaced air in the mold cavity to be expelled uniformly. 241 Fabrication Processes Tank (component A) Metering cylinder Recirculation Air pressure Tank (component B) Metering cylinder Static mixer Mixing head Preform FIGURE 2.62 Schematic of structural reaction injection molding (SRIM). 2.16.1 Applications The ability of SRIM to fabricate large, lightweight composite parts, consisting of all types of precisely located inserts and judiciously selected reinforcements, is an advantage that other manufacturing processes ﬁnd difﬁcult to match. Moreover, large SRIM parts can often be molded in 2–3 min, using clamping pressures as low as 100 psi. The capital requirements of SRIM are thus relatively low. The ﬁrst commercially produced SRIM part was the cover of the sparetire well in several automobiles produced by General Motors. Since then, SRIM automotive structural parts have included foamed door panels, instrument panel inserts, sunshades, and rear window decks. Nonautomotive applications include satellite dishes and seat shells for the furniture market. 2.17 Resin Transfer Molding RTM is similar to SRIM. In its common form, RTM is a closed-mold, low-pressure process in which dry, preshaped reinforcement material is placed in a closed mold and a polymer solution or resin is injected at a low pressure, ﬁlling the mold and thoroughly impregnating the reinforcement to form a composite part. The mold pressure in the RTM process is lower than in both SRIM and RIM/RRIM and the molding cycle time is much longer. The reinforcement and resin may take many forms, and the low pressure combined with the preoriented reinforcement package, affords a large range of component sizes, geometries, and performance options. RTM is an excellent process choice for making prototype components. It allows representative prototypes to be molded at low cost, unlike processes such as compression molding and injection molding, which require tools and equipment approaching actual production level. When prototyping with RTM, less reactive resins are generally used, allowing long ﬁll times and easier control of the vents. Sizes can range from small components to very large, complex, three-dimensional 242 Plastics Technology Handbook structures. RTM provides two ﬁnished surfaces and controlled thickness, while other processes used for prototyping, such as hand lay-up and wet molding, give only a single ﬁnished surface. 2.18 Foaming Processes Plastics can be foamed in a variety of ways. The foamed plastics, also referred to as cellular or expanded plastics, have several inherent features which combine to make them economically important. Thus, a foamed plastic is a good heat insulator by virtue of the low conductivity of the gas (usually air) contained in the system, has a higher ratio of ﬂexural modulus to density than when unfoamed, has greater loadbearing capacity per unit weight, and has considerably greater energy-storing or energy-dissipating capacity than the unfoamed material. Foamed plastics are therefore used in the making of insulation, as core materials for load-bearing structures, as packaging materials used in product protection during shipping, and as cushioning materials for furniture, bedding, and upholstery. Among those plastics which are commercially produced in cellular form are polyurethane, PVC, polystyrene, polyethylene, polypropylene, epoxy, phenol-formaldehyde, urea-formaldehyde, ABS, cellulose acetate, styrene-acrylonitrile, silicone, and ionomers. However, note that it is possible today to produce virtually every thermoplastic and thermoset material in cellular form. In general, the basic properties of the respective polymers are present in the cellular products except, of course, those changed by conversion to the cellular form. Foamed plastics can be classiﬁed according to the nature of cells in them into closed-cell type and opencell type. In a closed-cell foam each individual cell, more or less spherical in shape, is completely closed in by a wall of plastic, whereas in an open-cell foam individual cells are inter-connecting, as in a sponge. Closed-cell foams are usually produced in processes where some pressure is maintained during the cell formation stage. Free expansion during cell formation typically produces open-cell foams. Most foaming processes, however, produce both kinds. A closed-cell foam makes a better buoy or life jacket because the cells do not ﬁll with liquid. In cushioning applications, however, it is desirable to have compression to cause air to ﬂow from cell to cell and thereby dissipate energy, so the open-cell type is more suitable. Foamed plastics can be produced in a wide range of densities—from 0.1 lb/ft.3 (0.0016 g/cm3) to 60 lb/ft.3 (0.96 g/cm3)—and can be made ﬂexible, semirigid, or rigid. A rigid foam is deﬁned as one in which the polymer matrix exists in the crystalline state or, if amorphous, is below its Tg. Following from this, a ﬂexible cellular polymer is a system in which the matrix polymer is above its Tg. According to this classiﬁcation, most polyoleﬁns, polystyrene, phenolic, polyycarbonate, polyphenylene oxide, and some polyurethane foams are rigid, whereas rubber foams, elastomeric polyurethanes, certain polyoleﬁns, and plasticized PVC are ﬂexible. Intermediate between these two extremes is a class of polymer foams known as semirigid. Their stress–strain behavior is, however, closer to that of ﬂexible systems than to that exhibited by rigid cellular polymers. The group of rigid cellular polymers can be further subdivided according to whether they are used (1) for non-load-bearing applications, such as thermal insulation; or as (2) load-bearing structural materials, which require high stiffness, strength and impact resistance. The description of cellular foams as low, medium or high density is very common in practice. This is, however, not exact as the different density ranges which correspond to each of these items are not strictly deﬁned. The following ﬁgures can, however, serve as a rough general guide: lb/ft.3 Kg/m3 Low density 0.1–3 2–50 Medium density 3–21 50–350 High density 21–60 350–960 243 Fabrication Processes (Note that the density of a polymer foam refers to its bulk density, deﬁned by the ratio of total weight/total volume of the polymer and gaseous component. Obviously the gas phase contributes considerably to the volume of the end product, while the solid component contributes almost to the entire weight.) Obtained forms of foamed plastics are blocks, sheets, slabs, boards, molded products, and extruded shapes. These plastics can also be sprayed onto substrates to form coatings, foamed in place between walls (i.e., poured into the empty space in liquid form and allowed to foam), or used as a core in mechanical structures. It has also become possible to process foamed plastics by conventional processing machines like extruders and injection-molding machines. Polymer foams may be homogeneous with a uniform cellular morphology throughout or they may be structurally anisotropic. They may have an integral solid polymer skin or they may be multicomponent in which the polymer skin is of different composition to the polymeric cellular core. Schematic representations of the different physical forms of cellular polymers are given in Figure 2.63. Some special types of foams, namely, structural foams, reinforced foams, and syntactic foams are represented by Figure 2.63c to Figure 2.63f. These are described in a later section. Foaming of plastics can be done in a variety of ways. Most of them typically involve creating gases to make foam during the foaming cycle. Once the polymer has been expanded or “blown,” the cellular structure must be stabilized rapidly; otherwise it would collapse. Two stabilization methods are used. First, if the polymer is a thermoplastic, expansion is carried out above the softening or melting point, and the form is then immediately cooled to below this temperature. This is called physical stabilization. The second method—chemical stabilization—requires the polymer to be cross-linked immediately following the expansion step. Common foaming processes are the following: 1. Air is whipped into a dispersion or solution of the plastic, which is then hardened by heat or catalytic action or both. 2. A low-boiling liquid is incorporated in the plastic mix and volatilized by heat. (a) (b) (c) (d) (e) (f ) FIGURE 2.63 Schematic representations of section through different types of cellular polymer. (a) Low-density open-cell foam. (b) High-density closed-cell foam. (c) Single-component structural foam with cellular core and integral solid skin. (d) Multicomponent structural foam. (e) Fiber-reinforced closed-cell foam. (f) Syntactic foam. 244 Plastics Technology Handbook 3. Carbon dioxide gas is produced within the plastic mass by chemical reaction. 4. A gas, such as nitrogen, is dissolved in the plastic melt under pressure and allowed to expand by reducing the pressure as the melt is extruded. 5. A gas, such as nitrogen, is generated within the plastic mass by thermal decomposition of a chemical blowing agent. 6. Microscopically small hollow beads of resin or even glass (e.g., microballoons) are embedded in a resin matrix. Foams can be made with both thermoplastic and thermosetting plastics. The well known commercial thermoplastic foams are polystyrene, PVC, polyethylene, polypropylene, ABS copolymer, cellulose acetate. The thermosetting plastics which may be mentioned, among others, are phenol-formaldehyde, urea-formaldehyde, polyurethane, epoxy, and silicone. The methods of manufacture of some of these polymeric foams are given below. 2.18.1 Rigid Foam Blowing Agents There are four types of polymers typically used for rigid foam production, namely, polystyrene, polyurethane, polyoleﬁn, and phenolic. Within the polyoleﬁn segment, rigid foams can be produced using polyethylene or polypropylene. Following the implementation of the Montreal Protocol of 1989, chloroﬂuorocarbons (CFC-11 and CFC-12) which had been the primary blowing agents for both ﬂexible and rigid foams, were no longer available. Hydrochloroﬂuorocarbons (HCFCs) were one of the primary blowing agents that were then adopted, speciﬁcally HCFC-141b, HCFC-142b, and HCFC-22. Insulating foam products (with some exceptions) generally utilize HCFCs due to the superior insulation properties that they impart. The non-ozone depleting (i.e., not CFCs or HCFCs) blowing agents that are currently in use and will be substituted for ozone depleting HCFCs as the latter being still ozone depletants are phased out are: (i) hydroﬂuorocarbons (HFCs); (ii) hydrocarbons (e.g., pentanes, butanes); and (iii) carbon dioxide. Non-insulating foam products typically utilize hydrocarbons, such as isobutane, pentane, isopentane, and hexane. The use of CO2 (either water-based or liquid) is a major identiﬁed option to reduce the emission of non-HCFC blowing agents from polyurethane foam and extruded polystyrene boardstock applications. However, the thermal insulation properties of CO2-blown foam are signiﬁcantly compromised when compared to halocarbon-blown foam. Halocarbons (i.e., HCFCs, HFCs) are thus expected to be used in insulation foam manufacture for several years into the future. The primary HCFC replacements in these sectors are expected to be the liquid HFCs, which may see extensive use once HCFCs can no longer be used. 2.18.2 Polystyrene Foams Polystyrene, widely used in injection and extrusion molding, is also extensively used in the manufacture of plastic foams for a variety of applications. Polystyrene produces light, rigid, closed-cell plastic foams having low thermal conductivity and excellent water resistance, meeting the requirements of low-temperature insulation and buoyancy applications. Two types of low-density polystyrene foams are available to the fabricator, molder, or user: (1) extruded polystyrene foam and (2) expandable polystyrene for molded foam. 2.18.2.1 Extruded Polystyrene Foam This material is manufactured as billets and boards by extruding molten polystyrene containing a blowing agent (nitrogen gas or chemical blowing agent) under elevated temperature and pressure into the atmosphere where the mass expands and solidiﬁes into a rigid foam. Many sizes of extruded foam are available, some as large as 10 in. × 24 in. × 9 ft. The billets and boards can be used directly or cut into different forms. One of the largest markets for extruded polystyrene in the form of boards is in low-temperature insulation 245 Fabrication Processes (e.g., truck bodies, railroad cars, refrigerated pipelines, and low-temperature storage tanks for such things as liqueﬁed natural gas). Another growing market for extruded polystyrene boards is residential insulation. Such boards are also used as the core material for structural sandwich panels, used prominently in the construction of recreational vehicles. 2.18.2.2 Expandable Polystyrene Expandable polystyrene is produced in the form of free-ﬂowing pellets or beads containing a blowing agent. Thus, pellets chopped from an ordinary melt extruder or beads produced by suspension polymerization are impregnated with a hydrocarbon such as pentane. Below 70°F (21°C) the vapor pressure of the pentane dissolved in the polymer is low enough to permit storage of the impregnated material (expandable polystyrene) in a closed container at ordinary temperature and pressure. Even so, manufacturers do not recommend storing for more than a few months. The expandable polystyrene beads may be used in a tabular blow-extrusion process (Figure 2.64) to produce polystyrene foam sheet, which can subsequently be formed into containers, such as egg cartons and cold-drink cups, by thermoforming techniques. Expandable polystyrene beads are often molded in two separate steps: (1) Preexpansion or prefoaming of the expandable beads by heat, and (2) further expansion and fusion of the preexpanded beads by heat in the enclosed space of a shaping mold. Steam heat is used for preexpansion in an agitated drum with a residence time of a few minutes. As the beads expand, a rotating agitator prevents them from fusing together, and the preexpanded beads, being lighter, are forced to the top of the drum and out the discharge chute. They are then collected in storage bins for aging prior to molding. The usual lower limit of bulk density for bead preexpansion is 1.0 lb/ft.3 (0.016 g/cm3), compared to the virgin bead bulk density of about 35 lb/ft.3 (0.56 g/cm3). Molding of preexpanded (prefoamed) beads requires exposing them to heat in a conﬁned space. In a typical operation (Figure 2.65) prefoamed beads are loaded into the mold cavity, the mold is closed, and steam is injected into the mold jacket [60]. The prefoamed beads expand further and fuse together as the temperature exceeds Tg. The mold is cooled by water spray before removing the molded article. Packages shaped to ﬁt their contents (small sailboats, toys, drinking cups, etc.) are made in this way. Special machines have been designed to produce thin-walled polystyrene foam cups. Very small beads at a Pinch rolls Low density foam bubble Tubular die FIGURE 2.64 Expandable polystyrene beads Extruder Tubular blow extrusion for production of low-density polystyrene foam sheet. 246 Plastics Technology Handbook Filling: mold ﬁll Fusion: steam on Steam Drains open Cool: water spray on Steam Drains closed Water Part ejection: mold open Water Drains open Drains open FIGURE 2.65 Molding of preexpanded (prefoamed) polystyrene beads. (Adapted from PELASPAN Expandable Polystyrene, Form 171–414, Dow Chemical Co., 1966.) prefoamed density of approximately 4–5 lb/ft.3 (0.06–0.08 g/cm3) are used, which allow easy ﬂow into the molding cavity and produce a cup having the proper stiffness for handling. 2.18.2.3 Structural Foams Structural foam is the term usually used for foam produced in an injection molding press and made of almost many thermoplastic resin. Structural foam is always produced with a hard integral skin on the outer surfaces and a cellular core in the interior, and is used almost exclusively for production of molded parts. The process is thus ideally suited for fabrication of parts such as business machine housings (commonly for ABS), and similar parts or components in which lightweight and stiffness are required. The structural foam injection molding process (Figure 2.66), by which a product with a cellular core and a solid skin can be molded in a single operation, gets its name from the application of its product rather than the mechanism of the process itself. In a manner directly opposite to the vented extruder (Figure 2.21), a blowing agent, often nitrogen, is injected into the melt in the extruder. The polymer melt, injected with gas, is then forced into the accumulator where it is maintained at a pressure and temperature high enough to prevent foaming (Figure 2.66a). When a sufﬁcient charge has accumulated it is transferred into the mold (Figure 2.66b). The melt foams and ﬁlls the mold at a relatively low pressure (1.3–2.6 MPa) compared to the much higher pressure in the accumulator. The lower operating pressures of the molds make the molds less expensive than those used for conventional injection molding. However, the cycle times are longer because the foam being a good insulator, takes longer time to cool. Structural foams can also be made using a chemical blowing agent (discussed later) rather than an inert gas. In that case, a change in pressure or temperature on entering the mold triggers gas formation. Today structural foam injection molding is a very fast-growing polymer processing technique that can be used to modify the properties of thermoplastics to suit speciﬁc applications. 2.18.3 Polyoleﬁn Foams Polyoleﬁn foams can be produced with closely controlled density and cell structure. Generally the mechanical properties of polyoleﬁns lies between those of a rigid and a ﬂexible foam. Polyoleﬁn foams have a very good chemical and abrasion resistance as well as good thermal insulation properties. Crosslinking improves foam stability and polymer properties. A variety of foams can be produced from various types of polyethylenes and cross-linked systems having a very wide range of physical properties, and foams can be tailor-made to a speciﬁc application. 247 Fabrication Processes Accumulator Blowing agent Mold Valve Extruder Hydraulic press (a) Accumulator Blowing agent Mold Valve (b) Extruder Hydraulic press FIGURE 2.66 Structural foam process. (a) Filling the accumulator. The blowing agent (usually nitrogen) is injected into the melt in the extruder before it is passed into the accumulator. (b) Filling the mold. The accumulator ram injects the melt into the mold where the reduced pressure allows the gas to foam the resin. Polypropylene has a higher thermostability than polyethylene. The production volume of polyoleﬁn foams is not as high as that of polystyrene, polyurethane, or PVC foams. This is due to the higher cost of production and some technical difﬁculties in the production of polyoleﬁn foams. The structural foam injection molding process, described previously for polystyrene, is also used for polyethylene and polypropylene structural foams (see Figure 2.66). Commercial extrusion processes for polyoleﬁn foam products are derived from the original Dow process which basically involves ﬁve steps, namely, extrusion, mixing, cooling, expansion, and aging (see Figure 2.67). These steps of the extrusion process may be performed on equipment of several different conﬁgurations such as single-screw extruders, twin-screw extruders, and tandem-extruder lines. Singlescrew extruders must be equipped with a multistage long screw of high length-to-diameter ratio capable of performing all the aforesaid extrusion steps. Twin-screw extruders, on the other hand, have low shear rate and high mixing ability both of which are desirable in foam extrusion. Except where the extrusion rate is low such as for products having a small cross section, most polyoleﬁn foam products are made with tandem-type extruders, as shown in Figure 2.68. The primary extruder consisting of a two-stage screw melts the resin and then mixes the melt with the solid additives and liquid blowing agent, whereas the second extruder, usually larger than the primary one and designed to provide maximum cooling efﬁciency, cools the molten polymer mixture to the optimum foaming temperature. In some equipment, however, the second extruder is designed to perform both as mixer for the blowing agent and as cooler. An alternative to large tandem extruders is the accumulating extrusion system which provides a high instantaneous extrusion rate. It is commonly employed for producing large plank products. In the simple system, shown in Figure 2.69, the foamable melt is fed into an accumulator by a single screw extruder and 248 Plastics Technology Handbook Aging FIGURE 2.67 Expansion Cooling Blowing agent Polymer, solid additives Mixing Extrusion Block diagram of polyoleﬁn foam manufacture by extrusion process. Blowing agent Primary extruder Cooling extruder FIGURE 2.68 Polymer, solid additives Schematic diagram of tandem extruder. Primary extruder Hydraulic cylinder Accumulator FIGURE 2.69 Sliding gate Schematic of accumulation extrusion system. pushed out by a ram through a die oriﬁce. The process is discontinuous, resulting in a loss of yield, but it has the advantage of low capital requirement. The shape of a polyoleﬁn foam product is determined largely by the shape of the die. Thus a circular die is used for a rod, an annular die for a tube or sheet product, and a slit die for a plank product. To make polyoleﬁn foam sheet, the extruded tubular foam, expanding at the annular die, is guided over a sizing mandrel, slit, laid ﬂat, and wound into a roll [61]. Cool air is blown in at the nose of the mandrel to reduce the friction between the hot expanding foam and the mandrel. The extruded polyoleﬁn foam must be dimensionally stabilized by aging, since the foam deforms according to the internal cell pressure, which changes with time as air and gaseous blowing agent diffuse into and out of the foam at different rates. If the rates are equal, the cell pressure, and hence the foam dimensions, will remain constant, as is found for the LDPE/CFC-114 system. Most blowing agents, however, permeate through polyoleﬁns faster than air and, as a result, the foams shrink. The aging time required for the shrunken foam to recover and stabilize depends on the properties of the polymer and the 249 Fabrication Processes physical attributes of the foam, such as the open-cell content, foam density, and foam thickness. The aging time may range from less than a week for a thin sheet to several weeks for a thick plank. The production line of a typical process to manufacture thin ultra-low-density (ULD) polypropylene foam sheet, consists of tandem extruders, an accumulating vessel, and an annular die. The secondary extruder is designed to mix a large amount of blowing agent into the polymer and then to cool the mixture. The blowing agent consists of a large proportion (90%) of a highly soluble blowing agent which provides the heat sink necessary for foam stabilization and a small proportion (10%) of a low-permeability blowing agent which serves as an inﬂatant. The accumulating extrusion system allows the high extrusion rate required for the production of ULD foam sheet. There are several processes for the production of moldable polyoleﬁn beads. In the BASF process, LDPE foam strands are extruded out of a multi-hole die and granulated to beads by a die-face cutter. Inexpensive butane is used as the blowing agent and the foam beads are then cross-linked by electronbeam. As the beads have atmospheric cell pressure, a special technique is required to develop the necessary cell pressure for molding [62]. In the Kanegafuchi process, most widely used to manufacture LDPE foam beads, dicumyl peroxide is impregnated into ﬁnely pelletized LDPE beads suspended in water in an autoclave with the help of a dispersant such as basic calcium tertiary phosphate and sodium dodecylbenzene sulfonate. The beads are then heated to cross-link. The cross-linked beads are impregnated with a suitable blowing agent (e.g., a non-ozone depleting replacement of CFC-12), cooled, discharged from the autoclave and immediately expanded with steam to make foam beads. For molding, the foam beads are charged into a mold and heated with superheated steam (>140 kPa) to expand and weld. The majority of cross-linked polyoleﬁn foam sheet products are made by one of the four Japanese processes: Sekisui, Toray, Furukawa, and Hitachi, the ﬁrst two of which use the radiation method and the latter two a chemical method for cross-linking. The ﬂow diagram of the radiation cross-linked polyoleﬁn foam sheet process is shown in Figure 2.70. The key steps of the process include a uniform mixing of polymer and blowing agent (powder), manufacturing void-free sheet of uniform thickness, cross-linking the sheet to the desired degree by irradiation with a high-energy ray, and then softening and expanding the sheet in a foaming chamber (oven) using a suitable support mechanism. The Sekisui process employs a vertical air oven like the one shown in Figure 2.71 for expanding the foamable sheet. The oven consists of a horizontal preheating chamber and a vertical foaming chamber. The rapidly expanding sheet supports itself by gravity in the vertical direction, while a specially designed tentering device keeps the sheet spread out. In the Toray process, the foamable sheet is expanded while aﬂoat on the surface of molten salts. The process is suitable for producing crosslinked PP foam sheet as well as PE foam sheet. The ﬂow diagram of the chemically cross-linked polyoleﬁn foam sheet process is shown in Figure 2.72. Unlike in the radiation cross-linking process, a peroxide cross-linking agent is incorporated in the polymer along with the blowing agent. Therefore, a tighter temperature control must be maintained in Oven heating Foaming Electron beam irradiation Crosslinking Polyethylene Sheet forming Mixing Blowing agent FIGURE 2.70 Flow diagram of radiation cross-linked polyoleﬁn foam sheet forming process. 250 Plastics Technology Handbook Infrared heaters Preheating chamber (150°) Foamable sheet Infrared heaters Foaming chamber (200°) Hot air Hot air Sheet tentering device FIGURE 2.71 Schematic of Sekisui vertical foaming oven for cross-linked polyoleﬁn foam sheet. Polyethylene Two-stage oven Foaming Crosslinking Sheet forming Mixing Cross-linking agent Blowing agent FIGURE 2.72 Flow diagram of chemically cross-linked polyoleﬁn foam sheet forming process. the sheet manufacturing steps to prevent premature cross-linking by the peroxide. In the oven, on the other hand, the cross-linking of the polyoleﬁn sheet must be thermally effected without causing the blowing agent to decompose. Consequently, both the oven design and the selection of raw materials are more difﬁcult in the chemical cross-linking process. Both the Furukawa and Hitachi processes employ horizontal air ovens consisting of at least two sections, the preheating section and the foaming/ forming section, and having one or more non-stick conveyors to support the sheet during heating and expansion [61]. Polyoleﬁn foams have many and varied applications due to their unique properties which include buoyancy, resiliency, energy absorption, low thermal conductivity, resistance to chemicals, thermoformability, and ease of fabrication. The major application areas of polyoleﬁn foams are cushion packaging (pads and saddles, encapsulation, case inserts, etc.), construction (expansion joint ﬁller, closure strips, ﬂoor underlayment etc.), automotive (headliner, door trim, instrument panel, trunk liner, air conditioner liner, 251 Fabrication Processes etc.), insulation (insulation of pipe, storage tanks), and sports and leisure (life vests, surfboards, swim aids, ski belts, gym mats, etc.). Thin polyoleﬁn foam sheet products are used primarily as wrapping materials to protect the surfaces of articles from minor dents and abrasion during handling and shipping. 2.18.4 Polyurethane Foams Polyurethane foams, also known as urethane foams or U-foams, are prepared by reacting hydroxylterminated compounds called polyols with an isocyanate (see Figure 1.62). Isocyanates in use today include toluene diisocyanate, known as TDI, crude methylenebis(4-phenyl-isocyanate), known as MDI, and different types of blends, such as TDI/crude MDI. Polyols, the other major ingredient of the urethane foam, are active hydrogen-containing compounds, usually polyester diols and polyether diols. It is possible to prepare many different types of foams by simply changing the molecular weight of the polyol, since it is the molecular backbone formed by the reaction between isocyanate and polyol that supplies the reactive sites for cross-linking (Figure 1.62), which in turn largely determines whether a given foam will be ﬂexible, semirigid, or rigid. In general, high-molecular-weight polyols with low functionality produce a structure with a low amount of cross-linking and, hence, a ﬂexible foam. On the other hand, low-molecular-weight polyols of high functionality produce a structure with a high degree of cross-linking and, consequently, a rigid foam. Of course, the formulation can be varied to produce any degree of ﬂexibility or rigidity within these two extremes. The reactions by which urethane foam are produced can be carried out in a single stage (one-shot process) or in a sequence of several stages (prepolymer process and quasi-prepolymer process.) These variations led to 27 basic types of products or processes, all of which have been used commercially. In the one-shot process, all of the ingredients—isocyanate, polyol, blowing agent, catalyst, additives, etc.—are mixed simultaneously, and the mixture is allowed to foam. In the prepolymer method (Figure 1.62), a portion of the polyol is reacted with an excess of isocyanate to yield a prepolymer having isocyanate end groups. The prepolymer is then mixed with additional polyol, catalyst, and other additives to cause foaming. The quasi-prepolymer process is intermediate between the prepolymer and one-shot processes. 2.18.4.1 Flexible Polyurethane Foams The major interest in ﬂexible polyurethane foams is for cushions and other upholstery materials. Principal market outlets include furniture cushioning, carpet underlay, bedding, automotive seating, crash pads for automobiles, and packaging. The density range of ﬂexible foams is usually 1–6 lb/ft.3 (0.016–0.096 g/cm3). The foam is made in continuous loaves several feet in width and height and then sliced into slabs of desired thickness. 2.18.4.1.1 One-Shot Process The bulk of the ﬂexible polyurethane foam is now being manufactured by the one-shot process using polyether-type polyols because they generally produce foams of better cushioning characteristics. The main components of a one-shot formulation are polyol, isocyanate, catalyst, surfactant, and blowing agent. Today the bulk of the polyether polyols used for ﬂexible foams are propylene oxide polymers. The polymers prepared by polymerizing the oxide in the presence of propylene glycol as an initiator and a caustic catalyst are diols having the general structure CH3 CH OH CH2 O ( CH2 CH CH3 O )n CH2 CH OH CH3 252 Plastics Technology Handbook The polyethers made by polymerizing propylene oxide using trimethylol propane, 1,2,6-hexanetriol, or glycerol as initiator are polymeric triols. For example, glycerol gives HO ( C3H6O )n CH2 CH(OH) CH2 ( C3H6O )n OH The higher hydroxyl content of these polyethers leads to foams of better loadbearing characteristics. Molecular weight in the range 3,000–3,500 is found to give the best combination of properties. The second largest component in the foam formulation is the isocyanate. The most suitable and most commonly used isocyanate is 80:20 TDI—i.e., 80:20 mixture of tolylene-2,4-diisocyanate and tolylene2,6-diisocyanate. One-shot processes require sufﬁciently powerful catalysts to catalyze both the gas evolution and chain extension reaction (Figure 1.62). Use of varying combinations of an organometallic tin catalyst (such as dibutyltin dilaurate and stannous octoate) with a tertiary amine (such as alkyl morpholines and triethylamine), makes it possible to obtain highly active systems in which foaming and cross-linking reactions could be properly balanced. The surface active agent is an essential ingredient in formulations. It facilitates the dispersion of water (see below) in the hydrophobic resin by decreasing the surface tension of the system. In addition, it also aids nucleation, stabilizes the foam, and regulates the cell size and uniformity. A wide range of surfactants, both ionic and nonionic, have been used at various times. Commonly used among them are the watersoluble polyether siloxanes. Water is an important additive in urethane foam formulation. The water reacts with isocyanate to produce carbon dioxide and urea bridges (Figure 1.62). An additional amount of isocyanate corresponding to the water present must therefore be incorporated in the foaming mix. The more water that is present, the more gas that is evolved and the greater number of active urea points for cross-linking. This results in foams of lower density but higher degree of cross-linking, which reduces ﬂexibility. So when soft foams are required, a volatile liquid such a trichloromonoﬂuoromethane (bp 23.8°C) may be incorporated as a blowing agent. This liquid will volatilize during the exothermic urethane reaction and will increase the total gas in the foaming system, thereby decreasing the density, but it will not increase the degree of crosslinking. However, where it is desired to increase the cross-link density independently of the isocyanatewater reaction, polyvalent alcohols, such as glycerol and pentaerythritol, and various amines may be added as additional cross-linking agents. A typical formulation of one-shot urethane foam system is shown in Table 2.4 [63]. Most foam is produced in block form from Henecke-type machines (Figure 2.73) or some modiﬁcation of them. In this process [64], several streams of the ingredients are fed to a mixing head which oscillates in a horizontal plane. In a typical process, four streams may be fed to the mixing head: e.g., polyol and ﬂuorocarbon (if any); isocyanate; water, amine, and silicone; and tin catalyst. The reaction is carried out with slightly warmed components. Foaming is generally complete within a minute of the mixture emerging from the mixing head. The emergent reacting mixture runs into a trough, which is moving backward at right angles to the direction of traverse of the reciprocating mixing head. In this way the whole trough is covered with the foaming mass. Other developments of one-shot ﬂexible foam systems include direct molding, where the mixture is fed into a mold cavity (with or without inserts such as springs, frames, etc.) and cured by heat. In a typical application, molds would be ﬁlled and closed, then heated rapidly to 300°F–400°F (149°C–204°C) to develop maximum properties. A good deal of ﬂexible urethane foam is now being made by the cold-cure technique. This involves more reactive polyols and isocyanates in special foaming formulations which would cure in a reasonable time to their maximum physical properties without the need for additional heat over and above that supplied by the exothermic reaction of the foaming process. 253 Fabrication Processes TABLE 2.4 Urethane One-Shot Foam Formulation Ingredient Parts by Weight Poly(propylene oxide), mol. wt. 2000 and 2OH/molecule 35.5 Poly(propylene oxide) initiated with trifunctional alcohol, mol. wt. 3000 and 3 OH/molecule 35.5 Toluene diisocyanate (80:20 TDI) Dibutyltin dilaurate 26.0 0.3 Triethylamine 0.05 Water Surfactant (silicone) 1.85 0.60 Trichloromonoﬂuoromethane (CCl3F) 12.0 Final density of foam = 1.4 lb/ft.3 or 0.022 g/cm3 (2.0 lb/ft.3 or 0.032 g/cm3 if CCl3F is omitted) Source: One-Step Urethane Foams, 1959. Bull, F40487, Union Carbide Corp. Motor Mixing head Reciprocating motion Deposited foam Direction of motion Trough in motion Stationary frame FIGURE 2.73 shot process. Schematic of a Hanecke-type machine for production of polyurethane foam in block form by one- Cold-cure foaming is used in the production of what is known as high-resilient foams having high sag factor (i.e., ratio of the load needed to compress foam by 65% to the load needed to compress foam by 25%), which is most important to cushioning characteristics. True cold-cure foams will produce a sag factor of 3–3.2, compared to 2–2.6 for hot-cured foams. 2.18.4.1.2 Prepolymer Process In the prepolymer process the polyol is reacted with an excess of isocyanate to give an isocyanateterminated prepolymer which is reasonably stable and has less handling hazards than free isocyanate. If water, catalysts, and other ingredients are added to the product, a foam will result. For better load-bearing and cushioning properties, a low-molecular-weight triol, such as glycerol and trimethylolpropane, is added to the polyol before it reacts with the isocyanate. The triol provides a site for chain branching. Although the two-step prepolymer process is less important than the one-shot process, it has the advantage of low exotherms, greater ﬂexibility in design of compounds, and reduced handling hazards. 254 Plastics Technology Handbook 2.18.4.1.3 Quasi-Prepolymer Process In the quasi-prepolymer process a prepolymer of low molecular weight and hence low viscosity is formed by reacting a polyol with a large excess of isocyanate. This prepolymer, which has a large number of free isocyanate groups, is then reacted at the time of foaming with additional hydroxy compound, water, and catalyst to produce the foam. The additional hydroxy compound, which may be a polyol or a simple molecule such as glycerol or ethylene glycol, also functions as a viscosity depressant. The system thus has the advantage of having low-viscosity components, compared to the prepolymer process, but there are problems with high exotherms and a high free-isocyanate content. Quasi-prepolymer systems based on polyester polyols and polyether polyols are becoming important in shoe soling, the former being most wear resistant and the latter the easiest to process. 2.18.4.2 Rigid and Semirigid Foams The ﬂexible foams discussed in previous sections have polymer structures with low degrees of crosslinking. Semirigid and rigid forms of urethane are products having higher degree of cross-linking. Thus, if polyols of higher functionality—i.e., more hydroxyl groups per molecule—are used, less ﬂexible products may be obtained, and in the case of polyol with a sufﬁciently high functionality, rigid foams will result. The normal density range for rigid and semirigid foams is about 1–3 lb/ft.3 (0.016–0.048 g/cm3). Some packaging applications, however, use densities down to 0.5 lb/ft.3 (0.008 g/cm3); for furniture applications densities can go as high as 20–60 lb/ft.3 (0.32–0.96 g/cm3), thus approaching solids. At densities of from 2 lb/ft.3 (0.032 g/cm3) to 12 lb/ft.3 (0.19 g/cm3) or more, these foams combine the best of structural and insulating properties. Semirigid (or semiﬂexible) foams are characterized by low resilience and high energy-absorbing characteristics. They have thus found prime outlet in the automotive industry for applications like safety padding, arm rests, horn buttons, etc. These foams are cold cured and involve special polymeric isocyanates. They are usually applied behind vinyl or ABS skins. In cold curing, the liquid ingredients are simply poured into a mold in which vinyl or ABS skins and metal inserts for attachments have been laid. The liquid foams and ﬁlls the cavity, bonding to the skin and inserts. Formulations and processing techniques are also available to produce self-skinning semirigid foam in which the foam comes out of the mold with a continuous skin of the same material. Rigid urethane foams have outstanding thermal insulation properties and have proved to be far superior to any other polymeric foam in this respect. Besides, these rigid foams have excellent compressive strength, outstanding buoyancy (ﬂotation) characteristics, good dimensional stability, low water absorption, and the ability to accept nails or screws like wood. Because of these characteristics, rigid foams have found ready acceptance for such applications as insulation, refrigeration, packaging, construction, marine uses, and transportation. For such diverse applications several processes are now available to produce rigid urethane foam. These include foam-in-place (or pour-in-place), spraying, molding, slab, and laminates (i.e., foam cores with integral skins produced as a single unit). One-shot techniques can be used without difﬁculty, although in most systems the reaction is slower than with the ﬂexible foam, and conditions of manufacture are less critical. Prepolymer and quasi-prepolymer systems were also developed in the United States for rigid and semirigid foams, largely to reduce the hazards involved in handling TDI where there are severe ventilation problems. In the foam-in-place process a liquid urethane chemical mixture containing a ﬂuorocarbon blowing agent is simply poured into a cavity or metered in by machine. The liquid ﬂows to the bottom of the cavity and foams up, ﬁlling all cracks and corners and forming a strong seamless core with good adhesion to the inside of the walls that form the cavity. The cavity, of course, can be any space, from the space between two walls of a refrigerator to that between the top and bottom hull of a boat. However, if the cavity is the interior of a closed mold, the process is known as molding. 255 Fabrication Processes Rigid urethane foam can be applied by spraying with a two-component spray gun and a urethane system in which all reactants are incorporated either in the polyol or in the isocyanate. The spraying process can be used for applying rigid foam to the inside of building panels, for insulating cold-storage rooms, for insulating railroad cars, etc. Rigid urethane foam is made in the form of slab stock by the one-shot technique. As in the Henecke process (Figure 2.73), the reactants are metered separately into a mixing head where they are mixed and deposited onto a conveyor. The mixing head oscillates in a horizontal plane to insure an even deposition. Since the foaming urethane can structurally bond itself to most substrates, it is possible (by metering the liquid urethane mixture directly onto the surface skin) to produce board stock with integral skins already attached to the surface of the foam. Sandwich-construction building panels are made by this technique. 2.18.5 Foamed Rubber Although foamed rubber and foamed urethanes have many similar properties, the processes by which they are made differ radically. In a simple process a solution of soap is added to natural rubber (NR) latex so that a froth will result on beating. Antioxidants, cross-linking agents, and a foam stabilizer are added as aqueous dispersions. Foaming is done by combined agitation and aeration with automatic mixing and foaming machines. The stabilizer is usually sodium silicoﬂuoride (Na2SiF6). The salt hydrolyzes, yielding a silica gel which increases the viscosity of the aqueous phase and prevents the foam from collapsing. A typical cross-linking agent is a combination of sulfur and the zinc salt of mercaptobenzothiazole (accelerator). Cross-linking (curing or vulcanization) with this agent takes place in 30 min at 100°C. When making a large article such as a mattress, a metal mold may be ﬁlled with the foamed latex and heated by steam at atmospheric pressure. After removing the foamed rubber article from the mold, it may be dewatered by compressing it between rolls or by centrifuging and by drying with hot air in a tunnel dryer. In foamed rubber formulation a part of the NR latex can be replaced by a synthetic rubber latex. One such combination is shown in Table 2.5. 2.18.6 Epoxy Resins Any epoxy resin can be made foamable by adding to the formulation some agent that is capable of generating a gas at the curing temperature prior to gelation. Such foaming agents may be low-boiling liquids which vaporize on heating (e.g., CFCs such as Freons) or blowing agents that liberate a gas when TABLE 2.5 Foamed Rubber Formulation Ingredient Parts by Weight Styrene-butadiene latex (65% solids) 123 Natural rubber latex (60% solids) Potassium oleate 33 0.75 Sulfur 2.25 Accelerators Zinc diethyldithiocarbamate 0.75 Zinc salt of mercaptobenzothiazole 1.0 Trimene base (reaction product of ethyl chloride, formaldehyde, and ammonia) Antioxidant (phenolic) 0.8 0.75 Zinc oxide 3.0 Na2SiF6 2.5 Source: Stern, H. J. 1967. Rubber: Natural and Synthetic, Palmerton, New York. 256 Plastics Technology Handbook heated above 70°C, such as 2,2′-azobis(isobutyronitrile) or sulfonylhydrazide, which decompose evolving nitrogen. A foaming gas can also be generated in situ by adding a blowing agent that reacts with amine (curing agent) to form the gas. A typical such system consists of an epoxy resin, a primary amine (hardener), and a hydrogen-active siloxane (blowing agent). The siloxane reacts with the amine, evolving hydrogen as a foaming gas. Instead of using amines it is also possible to use other hydrogen-active hardeners such as phenols and carboxylic acids. The reaction of gas evolution occurs immediately after mixing resin and hardeners and before the mixture begins to cure in the mold. This is essential for the formation of closed-cell foam structure during the curing, which takes place under a deﬁnite expansion pressure against the mold wall, leading to formation of dense casting. In the production of expanded laminates of sandwich conﬁguration, this yields very tough and impact-resistant structures. Expanded materials with excellent high-temperature properties are obtained when cresol novolacs are used as hardeners. A typical formulation is based on mixtures of bisphenol A resins and epoxy novolac resins which are cured by cresol novolacs and accelerated by suitable nitrogen-containing agents. Applications of epoxy foams can be categorized in three areas, namely, (1) unreinforced materials, (2) glass-ﬁber-reinforced materials, and (3) sandwich constructions [61]. Because of their light weight and absence of shrinkage, foamed epoxies are used in the production of large-scale patterns. Having excellent dielectric properties, epoxy foams ﬁnd applications in electronics such as for casting and sealing electronic components like small transformers and capacitors, and in insulating cables. The light weight properties of foamed epoxies are utilized in ﬁber-reinforced materials for which glass ﬁber mats and unidirectional rovings are most suitable, with the majority of applications involving sandwich constructions. A few practical examples are foamed epoxy windsurﬁng board, epoxy rotor blades for wind energy generators, and automotive spoilers. 2.18.7 Urea-Formaldehyde Foams Urea-formaldehyde (UF) foams are basically two-component systems as the production of the foams requires mainly a UF resin and a foam stabilizing agent. The UF resin (see Chapter 4) is produced by the condensation of urea and formaldehyde in the range of mole ratios of 2: 3–1: 2 in the presence of alkaline catalysts (pH∼8), which yield only short-chain oligomers, and under weakly acidic conditions (pH 4–6), which result in a higher molecular weight mixture of oligomers (solubilized by attached methylol groups). Though numerous substances have been proposed as foam-stabilizing agents for the commercial foam system, aqueous solutions of the sodium salts of dodecylbenzene sulfonic acid and dibutylnaphthalene sulfonic acid have proven to be of value. Aqueous solutions of strongly dissociating organic and inorganic acids with a pH range of 1–1.5 are used as hardening agents. However, phosphoric acid is preferred because of its negligible corrosive action. The hardening agent is preferentially added to the foam stabilizing agent and the concentration of the hardener is chosen so that the ﬁnal foam gels within 30–90 sec at 20–25°C after mixing the components. In a stationary process for making UF foams, a mixture of water and foam-stabilizing agent is introduced into a vessel equipped with tubular vanes. Foaming takes place upon feeding air and a ureaformaldehyde resin solution (see “Urea-Formaldehyde Resin” in Chapter 5). The generated foam is guided by the action of the tubular vanes to the outlet channel of the vessel from where the foam exits as a rectangular slab and is transported on a conveyor belt until the foam structure hardens sufﬁciently. Blocks are cut from the foam slab, dried at about 40°C for about 2 h, and then pressed mechanically into sheets of the required dimensions [61]. For on-site production of foam, the raw materials are transported by pumps into a foaming machine. A dispersion of foam stabilizing agent is formed in water in the machine and the resin is introduced into the mixing chamber through jets. The ﬁnished foam emerges from the plastic pipe and can be used immediately at the site. One thousand liters (1 m3) of foam can be produced on-site from 20 liters of resin and 18 liters of foam solution. 257 Fabrication Processes The UF foam plastics are open-cell cellular materials with the capability of absorbing oils and solvents. UF foam is non-toxic, nonﬂammable, and stable with respect to almost all organic solvents, light and heavy mineral oils, but is decomposed by dilute and concentrated acids and alkalis. It exhibits extraordinary aging stability, has good sound-absorbing properties, and has the lowest thermal conductivity, despite the open cells. These properties coupled with its light weight (bulk density 11 kg/m3) and low manufacturing cost, make UF foam suitable for a wide spectrum of applications, including industrial ﬁlling materials, insulating in enclosed cavities, plant substrates (soil-free cultivation), and medical applications. UF foam is used on-site to ﬁll cavities of all shapes and sizes, whether of natural origin or resulting from construction wall or other applications. It has been successfully employed in mining for over 50 years. The complete ﬁlling of cavities eliminates the hazard of methane accumulation and reduces the danger of ﬁre and explosion. UF foam ﬁlling is an inexpensive, rapid, and excellent heat-insulated lightweight construction method. When insulation is retroﬁtted, the foam is introduced into the existing cavities through sealable small holes. When UF foam is formed, formaldehyde is released. It is important to make sure that the proper ratio of components is employed and suitable construction measures are taken, as otherwise the problems of formaldehyde release from foam over short term or long term may be encountered. With present day technologies, it is possible to satisfy strict conditions that a formaldehyde level of 0.1 ppm should not be exceeded in the air of a room used continuously for dwelling purposes. Geo- and hydroponics, as well as plastoponics are terms referring to the cultivation and breeding of plants with foam as ﬂakes or in solid form. Beans, potatoes, carrots, tomatoes and ornamental plants have been grown extensively in foam. Its suitability for land recovery in desert and semi-desert regions has been established through extensive testing. 2.18.8 Silicone Foams Silicone foams result from the condensation reaction between ≡SiH and ≡SiOH shown below: ≡ SiH + ≡ SiOH + Catalyst ! ≡ Si – O – Si ≡ +H2 (2.5) When these three components (that is, ≡SiH-containing cross-linker, ≡SiOH-containing polymer, and catalyst) are mixed together, both blowing (generation of hydrogen gas) and curing or cross-linking, that is, formation of siloxane linkage (≡Si–O–Si≡) occur. It should be noted, however, that a cross-linked product forms if the functionality of the ≡SiH-containing component is 3 or greater and the ≡SiOH-containing compound has a functionality of at least 2. These reactions are heat accelerated, but they occur readily at room temperature in the presence of catalyst. These RTV (room temperature vulcanizing) foams are thus two-pack systems. Generally, the ≡SiH (such as methylhydrogen siloxane) and catalyst make up the second component. A variety of catalysts can be used to promote the reaction. Chloroplatinic acid or other soluble platinum compound is most commonly used because it imparts ﬂame retardancy to the formulation. If a vinyl endblocked polymethylsiloxane is used in place of polydimethylsiloxane in the above formulation, then another competing reaction can also occur as shown below: ≡ SiH+ ≡ SiCH = CH2 + Catalyst ! ≡ Si – CH2 – CH2 – Si ≡ (2.6) The reaction is also catalyzed by platinum. The addition of some vinyl-containing polysiloxane can thus improve properties such as density, tensile strength, cure rate, and so forth. Although hydrogen generation [Equation 2.5] is the most prevalent method of blowing silicone foam, there are other approaches. Adding a gas at a high enough pressure (so its volume is low before it expands) is an easy way to make foam. Gases commonly used are N2, CO2, while previously air-pressurized liqueﬁed gases such as CFCs were used. One may also use chemical blowing agents (see later), which decompose generating gas when heat is applied or pH is changed. Nitrogen-liberating organic blowing agents are used extensively for foaming silicone gum. 258 Plastics Technology Handbook Silicones in general are inert to most environmental agents and have many unique properties (see “Silicones” in Chapter 4). When siloxane polymers are processed into foams they carry with them most of their durability characteristics and characteristic properties. Silicone foams are thus used in a wide range of applications. Flexible foam sheet is used in airplanes as the material for ﬁre blocking, insulation of air ducts, gasketing in engine housing compartments, and shock absorbers. Silicone foam is a popular choice in the construction industry, because of its weatherability, thermal insulation, and sealing capability. 2.18.9 Phenolic Foams Phenolic foam is a light weight foam created from phenolic resins. It is used in a large range of applications, such as ﬂower foam blocks, building thermal insulation, ﬁre protection, damping, and civil engineering in a wide variety of shapes: blocks, sheets, and sprayed foams. Phenolic foams are generally made using a resol-type phenolic resin (or a resin blend typically containing 60% of a resol-type resin and 40% of a resorcinol-modiﬁed novolac resin), surfactants, blowing agents, catalysts, and additives. Surfactants and additives are mixed into the resin and the blend is then mixed with the liquid blowing agent(s) and ﬁnally with an acid catalyst, e.g., H3PO4. Surfactants are used to control cell size and structure. The most common surfactants are siloxaneoxyalkylene copolymers, polyoxyethylene sorbitan fatty acid esters, and the condensation products of ethylene oxide with castor oil and alkyl phenols. A commonly added additive is urea which is used as a formaldehyde scavenger. Very ﬁne particle size inorganic ﬁllers can be added to act as nucleating sites and to promote ﬁner, more uniform cell structure, as well as increased compressive strength, but at a cost of higher density. The most common blowing agents used for making phenolic foams are organic liquids that have boiling points approximately in the range 20°C–90°C. Suitable blowing agents include HFCs, HCFCs and others, and hydrocarbons having from about 3–10 carbon atoms such as pentane, hexane and petroleum ether. Hydrocarbons such as isopentane, isobutane and hexane are the preferred blowing agents for ﬂower blocks. In insulating foam sector, however, the non-hydrocarbon blowing agent alternatives are not as viable and Foranil ﬂuorochemical blowing agents have been mainly used for their insulating properties and non-ﬂammability. Phenolic foams can also be prepared without the use of CFC or hydrocarbon blowing agents. In a typical preparation [65], resol 200, ethoxylated castor oil 8, boric anhydride accelerator 36, and SnCl2.2H2O 36 parts are mixed and heated for about 4 minute at 120°C to obtain a foam having density 0.003 g/cm3 and 29% closed cells. In another method [66], 1 mol phenol, 2.6 mol formaldehyde, and 5% dimethylaminoethanol are heated at 70°C–100°C for 4 h to obtain a liquid (70%–80% solids content) which is mixed with 2%–3% silicone foam regulator and 5 parts NaHCO3. Toluenesulfonic acid (20 parts, 80% aq.) is then added to obtain a stable rigid foam. 2.18.10 Poly(Vinyl Chloride) Foams A number of methods have been devised for producing cellular products from PVC, either by a mechanical blowing process or by one of several chemical blowing techniques. PVC foams are produced in rigid or ﬂexible forms. The greatest interest in rigid PVC foam is in applications where low-ﬂammability requirements prevail. It has an almost completely closed cell structure and therefore low water absorption. The rigid PVC foam is used as the cellular layer of some sandwich and multi‐layer panels. Plastisols are the most widely used route to ﬂexible expanded PVC products. Dolls, gaskets, and resilient covers for tool handles, for example, are produced from expandable plastisol compounds by molding, while varied types of upholstery, garment fabrics, and foam layer in coated-fabric ﬂooring are made from coatings with such compounds. Figure 2.74 shows a schematic representation of the German Trovipor process for producing ﬂexible, mainly open cell, and low to medium density (60–270 kg/m3, 3.75–16.87 lb/ft.3) PVC foam. It is normally produced in the form of continuous sheet (Figure 2.74). 259 Fabrication Processes PVC Plasticizer Additives (a) (b) Mixer Autoclave Paste feed container Paste Gas stream Pump Cooler Perforated plate Insert gas Spray tower (c) Paste feed pump To cutter Heating(hf ) After heating Cooling tunnel Conveyor belt FIGURE 2.74 Schematic representation of the Trovipor process. (a) PVC paste preparation. (b) Gasiﬁcation of PVC paste. (c) Spraying and fusion. Many chemical blowing (foaming) agents have been developed for cellular elastomers and plastics, which, generally speaking, are organic nitrogen compounds that are stable at normal storage and mixing temperatures but undergo decomposition with gas evolution at reasonably well-deﬁned temperatures. Three important characteristics of a chemical blowing agent are the decomposition temperature, the volume of gas generated by unit weight (“gas number,” deﬁned as the volume of gas, in cm3, liberated by the transformation of 1 g of the blowing agent per minute), and the nature of the decomposition products. Since nitrogen is an inert, odorless, nontoxic gas, nitrogen-producing organic substances are preferred as blowing agents. Several examples of blowing agents [67] especially recommended for vinyl plastisols are shown in Table 2.6; in each case the gas generated in nitrogen. To produce uniform cells, the blowing agent must be uniformly dispersed or dissolved in the plastisol and uniformly nucleated. It should decompose rapidly and smoothly over a narrow temperature range corresponding to the attainment of a high viscosity or gelation of the plastisol system. The gelation involves solvation of the resin in plasticizer at 300°F–400°F (149°C–204°C), the temperature depending on the ingredients employed in the plastisol. The foam quality is largely determined by the matching of the decomposition of the blowing agent to the gelation of the polymer system. If gelation occurs before gas evolution, large holes or ﬁssures may form. On the other hand, if gas evolution occurs too soon before gelation, the cells may collapse, giving a coarse, weak, and spongy product. Among the blowing agents listed in Table 2.6, azobisformamide (ABFA) is the most widely used for vinyls because it fulﬁlls the requirements efﬁciently. ABFA decomposition can also be adjusted through proper choice of metal organic activators so that the gas evolution occurs over a narrow range within the wide range given in Table 2.6. Though the gas number of ABFA is normally 220–260 cm3/g, it can go up to 420 cm3/g in the presence of catalysts. Azodicarbonamide is recommended for foaming of PVC, polyoleﬁns, polyamides, polysiloxanes, epoxides, polymers and compolymers of acrylonitrile and acrylates, and rubbers. Diazoaminobenzene (DAB) is one of the ﬁrst organic blowing agents to ﬁnd industrial application. Its decomposition point (95°C–150°C) and gas number (115 cm3/g) depend on the pH of the medium; in 260 Plastics Technology Handbook TABLE 2.6 Commercial Blowing (Foaming) Agents Chemical Type Azo compounds Azobisformamide (Azodicarbonamide) (ABFA) Azobisisobutyronitrile (AIBN) Decomposition Temperature in Air (°C) Decomposition Position Range in Plastics (°C) Gas Yield (mL/g) 195–200 160–200 220 115 90–115 130 103 95–100 115 105 90–105 126 195 130–190 265 Benzenesulfonylhydrazide (BSH) Toluene-(4)-sulfonyl hydrazide (TSH) >95 103 95–100 100–106 130 120 Benzene-1, 3-disulfonyl hydrazide (BDH) 146 115–130 85 4, 4′-oxybis (benzenesulfonylhydrazide) (OBSH) 150 120–140 125 Diazoaminobenzene (DAB) N-Nitroso compounds N, N′-Dimethyl-N, N′ -dinitrosoterephthalimide (DMTA) N, N′ -Dinitrosopentamethylenetetramine (DNPA) Sulfonyl hydrazides Source: Lasman, H. R. 1967. Mod Plastics, 45, 1A, Encycl. Issue, 368. acidic media it decomposes at lower temperature and more completely. DAB is used in foaming phenolic and epoxy resins, PVC, rubber and other high poymers. N,N′-Dinitrosopentamethylenetetramine (DNPA) is the cheapest (except for urea oxalate) and most widely used organic blowing agent accounting for 50% of all blowing agents used. It however disperses poorly in mixtures and is sensitive to shock and friction (explosive). Because of the relatively low temperature of decomposition, DTA can be used to make foams with a uniform cellular structure without deterioration of the polymer. The disadvantages of DTA are, however, poor dispersive ability in mixtures and sensitivity to moisture. Nevertheless, DTA is used in foaming PVC (especially for thin walled articles), polyurethane, polystyrene, polyamides, and siloxane rubbers. BSH is used for foaming rubbers, polystyrene, epoxy resins, polyamides, PVC, polyesters, phenolformaldehyde resins, and polyoleﬁns. However, the thermal decomposition of BSH yields not only nitrogen but also a nontoxic residue (disulﬁde and thiosulfone) which may degrade to give thiophenol and thus an unpleasant odor to the foams. OBSH is one of the best blowing agents of the sulfonylhydrazide class. Its gas liberation characteristics (no stepwise change up to 140°C) makes it possible to obtain foams with small, uniform cells. It is nontoxic and does not impart color and smell to articles. OBSH is used for foaming PVC, polyoleﬁns, polysulﬁdes, microporous rubber, or foamed materials based on mixtures of polymers with rubbers. In the last case, OBSH acts additionally as a cross-linking agent. Closed-cell foams result when the decomposition and gelation are carried out in a closed mold almost ﬁlled with plastisol. After the heating cycle, the material is cooled in the mold under pressure until it is dimensionally stable. The mold is then opened, and the free article is again subjected to heat (below the previous molding temperature) for ﬁnal expansion. Protective padding, life jackets, buoys, and ﬂoats are some items made by this process. The blowing agents given in Table 2.6 can be used to make foamed rubber. A stable network in this product results from the cross-linking reaction (vulcanization), which thus corresponds to the step offusion in the case of plastisols. Some thermoplastics also can be foamed by thermal decomposition of blowing agents even though they do not undergo an increase in dimensional stability at an elevated temperature. In Fabrication Processes 261 this case the viscosity of the melt is high enough to slow down the collapse of gas bubbles so that when the polymer is cooled below its Tm a reasonably uniform cell structure can be built in. Cellular polyethylene is made in this way. 2.18.11 Special Foams Some special types of foams are: (1) structural foams; (2) syntactic foams and multifoams; and (3) reinforced foams. Structural foams (Figure 2.63c and d), which possess full-density skins and cellular cores, are similar to structural sandwich constructions or to human bones, which have solid surfaces but cellular cores. Structural foams may be manufactured by high pressure processes or by low-pressure processes (Figure 2.66). The ﬁrst one may provide denser, smoother skins with greater ﬁdelity to ﬁne detail in the mold than may be true of low-pressure processes. Fine wood detail, for example, is used for simulated wood furniture and simulated wood beams. Surfaces made by low-pressure processes may, however, show swirl or other textures, not necessarily detracting from their usefulness. Almost any thermoplastic or thermosetting polymer can be formulated into a structural foam. In the case of syntactic foams (or spheroplastics), instead of employing a blowing agent to form bubbles in the polymer mass, hollow spherical particles, called microspheres, microcapsules, or microballoons, are embedded in a matrix of unblown polymer. (In multifoams, microspheres are combined with a foamed polymer to provide both kinds of cells.) Since the polymer matrix is not foamed, but is ﬁlled mechanically with the hollow spheres, syntactic materials may also be thought of as reinforced or ﬁlled plastics, with the gas-containing particles being the reinforcing component. Synthetic wood, for instance, is provided by a mixture of polyester and small hollow glass spheres (microspheres). The cellular structure of the syntactic foam depends on the size, quantity, and distributive uniformity of the microspheres. Since the microspheres have continuous shells, the ﬁnal material will, as a rule, have completely enclosed cells, and thus can be called absolute foamed plastics or “absolute” closed-cell foams. This, together with the absence of microstructural anisotropy (because the microspheres have practically all the same size and are uniformly distributed in the matrix), gives a syntactic material its valuable properties. They have better strength-to-weight ratios than conventional foamed plastics, absorb less water, and can withstand considerable hydrostatic pressures. Using hollow sphere means that the ﬁnal material is lighter than one containing a compact ﬁller, such as glass powder, talc, kaolin, quartz meal, or asbestos. Figure 2.75a and b are graphical presentations of syntactic foam structure in which the two components, microspheres and resin ﬁll completely the whole volume (no dispersed voids) and the density of the product is thus calculated from the relative proportion of the two. Measured density values often differ from the calculated ones due to the existence of some isolated or interconnected irregularly shaped voids, as shown in Figure 2.75c. The voids are usually an incidental part of the composite, as it is not easy to avoid their formation. Nevertheless, voids are often introduced intentionally to reduce the density below the minimum possible in a close-packed two-phase structure. Syntactic foams exhibit their best mechanical behavior in the compressive mode. The spheres themselves are an extremely strong structure and hence can withstand such stresses very well. Syntactic materials consisting of hollow glass microspheres in epoxy resin are used for sandwich structures and as potting compounds for high-density electronic modules and other units likely to encounter hydrostatic pressures. Hollow glass microspheres and powdered aluminum in resin are used as core materials for sandwich construction and radomes. Hollow glass microspheres in aluminum matrix are used for aerospace and extreme hydrostatic pressure (oceanographic) applications in view of low weight and high compressive strength. Polymer foams may be reinforced, usually with short glass ﬁbers, and also other ﬁbers such as asbestos or metal, and other reinforcements such as carbon black. The reinforcing agent is generally introduced into the basic components and is blown along with them, to form part of and to reinforce the walls of the cells (Figure 2.63e). When this is done, it is not unusual to obtain increases in mechanical properties of 400–500% with ﬁberglass content up to 50% by weight, especially in thermosettings. The principal 262 Plastics Technology Handbook (a) (b) (c) FIGURE 2.75 Graphical representations of syntactic foam structures. (a) Two-phase composite with random dispersion of spheres. (b) Two-phase composite with hexagonal close-packed structure of uniform sized spheres (74% by vol.). (c) Three-phase composite containing packed microspheres, dispersed voids and binding resin. advantages of reinforcement, in addition to increased strength and stiffness, are improved dimensional stability, resistance to extremes of temperature and resistance to creep. Two processes for the manufacture of glass-reinforced foam laminates used as building materials, namely, free-rise process and restrained rise process are presented in Figure 2.76 and Figure 2.77. The glass ﬁber reinforcement is a thin (0.25–1.25 mm) mat supplied in roll form. It consists of layers of relatively long (1.5–4 m) glass ﬁbers, the ﬁbers in one layer being at an acute angle to the ﬁbers in each next adjacent layer. A small amount of silane-modiﬁed polyester, or other binder is present, at a level of 2%–10% by weight. This type of glass mat is relatively porous to the passage of liquids and is also capable of expanding within a mixture of rising foam chemicals to provide a uniform, three-dimensional reinforcing network within the ﬁnal foamed laminate. The glass ﬁber reinforcement is functionally effective when used at levels of 4–24 g per board foot of the laminate [68]. Various types of facing sheets may be Isocyanate, surfactant, blowing agent Top facing sheet Polyol Catalyst Glass ﬁber mat Guillotine knife Mixing head 65 –120°C Matering rolls Oven Bottom facing sheet FIGURE 2.76 Schematic of a free-rise process for manufacture of glass-reinforced foam laminates. 263 Fabrication Processes Isocyanate, surfactant, blowing agent Top facing sheet Polyol Catalyst Glass ﬁber mat Guillotine knife Mixing head Matering rolls Oven Bottom facing sheet FIGURE 2.77 Schematic of a restrained rise process for manufacture of glass-reinforced foam laminates. used, such as aluminum foil for building insulation products, asphalt-saturated felts for roof insulation or any other material, e.g., paper and plastic ﬁlms. 2.19 Rapid Prototyping/3D Printing Rapid prototyping [69,70] refers to the process of fabricating a physical model from 3D digital data by using additive manufacturing (also known as 3D printing) technology. A 3D printer, which is a type of industrial robot, lays successive layers of materials onto a build tray to create a 3D object that can be of almost any shape and geometry. It enables designers to cost-effectively translate their ideas into 3D models for concept evaluation and improvement. Using a multi-material 3D printer, functional parts of machines, vehicles, or equipment can be rapidly prototyped and tested in real working conditions, thereby enabling form, ﬁt, and functional testing to improve design and product quality. Additive manufacturing (AM) is the name to describe technologies like 3D printing that build 3D objects by adding layer upon layer of material, whether the material is plastic, metal, or concrete. The terms 3D printing and additive manufacturing are synonymous umbrella terms for all AM technologies [71]. Common to AM technologies is the use of a computer, 3D modeling software (computer-aided design or CAD), machine equipment, and layering material. In a typical AM process, once a CAD sketch (model) is produced, it is processed by a software called “slicer,” which converts the model into a series of thin layers and produces a G-code ﬁle containing instructions tailored to a speciﬁc type of 3D printer. The AM equipment (3D printer) reads in data from the CAD ﬁle or G-code ﬁle and lays down or adds successive layers of liquid, powder, sheet material, or other, in a layer-upon-layer fashion to fabricate a 3D object [71]. In a 3D printing process involving the use of thermoplastic polymer, the latter is injected through indexing nozzles onto a platform. The nozzles trace the cross-section pattern for each particular layer with the molten polymer hardening before the application of the next layer. The process is repeated until the build or model is completed. The printer resolution describes layer thickness and X–Y resolution in dots per inch (DPI) or micrometers (µm). Typical layer thickness is around 100 µm (250 DPI), although some 264 Plastics Technology Handbook machines can print layers as thin as 16 µm (1,600 DPI). The particles (3D dots) are around 50 to 100 µm (500 to 250 DPI) in diameter [72]. [It may be mentioned that while the term 3D printing originally referred to a process employing standard and custom inkjet print heads, the technology used by most 3D printers to date is fused deposition modeling, a special application of plastic extrusion (see below).] A large number of additive processes are now available. The main differences between processes are in the way layers are deposited to create parts and in the materials that are used. In some methods, the material is softened or melted to produce the layers, for example, selective laser melting (SLM), direct metal laser sintering (DMLS), selective laser sintering (SLS), fused deposition modeling (FDM), or fused ﬁlament fabrication (FFF), while others cure liquid materials into solid using different technologies, such as stereolithography based on photopolymerization [72]. In FDM, the model or part is produced by extruding small beads of material that harden immediately to form layers. A thermoplastic ﬁlament or metal wire is unreeled from a coil to supply material to an extrusion nozzle head (3D printer extruder). The nozzle head heats the material and turns the ﬂow on or off, while stepper or servo motors are used to move the extrusion head and adjust the ﬂow. The printer usually has three axes of motion. Polyjet 3D printing is similar to inkjet printing, but instead of jetting drops of ink onto paper, Polyjet 3D printers jet tiny droplets of liquid photopolymer onto a build tray. The tiny droplets are instantly UV-cured, forming ﬁne layers that accumulate on the build tray to create a precise 3D model or part. Where overhangs or complex shapes require support, the 3D printer jets a removable gel-like support material that can be easily removed by hand or with water. Models, prototypes, parts, or tooling thus produced are ready to use and do not need post-curing. With 16-micron layer resolution and accuracy as high as 0.1 mm, a Polyjet 3D printer can produce thin walls and complex geometries using a wide range of materials. High-resolution plastic prototypes with ﬁne details and moving parts can be produced via 3D printer and many different polymers can be used, including acrylonitrile-butadiene-styrene (ABS), polycarbonate (PC), polylactic acid (PLA), high-density polyethylene (HDPE), PC/ABS, polyphenylenesulfone (PPSU), and high-impact polystyrene (HIPS). In general, the polymer is in the form of a ﬁlament fabricated from virgin resins [72]. The prototypes made by AM or 3D printing can be subjected to rigorous product testing to gain conﬁdence in the design before investigating in an injection mold tool. Plastic prototypes can accurately reﬂect what is achievable with injection molding. The use beneﬁts lie in the design freedoms that the additive approach engenders. Its immense design ﬂexibility gives the potential of multifunctional components and structures. Typical industrial applications are in automotive, aerospace, and prosthetic medical ﬁelds. Developing products frequently require multiple materials that can make the development process and rapid prototyping, in particular, complicated. The Objet process of Rutland Plastics (www.sys-uk.com) that uses multiple materials simultaneously overcomes this problem. The Objet Connex printer offers the unique ability to print parts and assemblies made of multiple materials, with different mechanical properties, all in a single build. With this particular 3D printing and rapid prototyping process it is also possible to undertake low-volume production. Traditional techniques like injection molding are usually less expensive than 3D printing for manufacturing polymer products in high quantities, but AM can be faster and less expensive, in addition to being more ﬂexible, when producing relatively small quantities of parts [72]. Three-dimensional printers, moreover, give designers and concept development teams the ability to produce parts and concept models using a desktop-size printer. The ﬁrst 3D printer designed to operate in zero gravity, Zero-G Printer, was built in 2014 under a joint partnership between NASA and the Marshall Space Flight Center (MSFC). Its applications for space offer the ability to print parts or tools on site in space instead of using rockets to bring pre-manufactured items to the site of use. Three-dimensional printing of nanosized objects can also be performed. It uses microelectronic device fabrication methods. Such printed objects are typically grown on a solid substrate (e.g., silicon carbide). Fabrication Processes 265 2.20 Rubber Compounding and Processing Technology 2.20.1 Compounding Ingredients No rubber becomes technically useful if its molecules are not cross-linked, at least partially, by a process known as curing or vulcanization [73–76]. For NR and many synthetic rubbers, particularly the diene rubbers, the curing agent most commonly used is sulfur. But sulfur curing takes place at technically viable rates only at a relatively high temperature (>140°C) and, moreover, if sulfur alone is used, optimum curing requires use of a fairly high dose of sulfur, typically 8–10 parts per hundred parts of rubber (phr), and heating for nearly 8 h at 140°C. Sulfur dose has been substantially lowered, however, with the advent of organic accelerators. Thus, incorporation of only 0.2–2.0 phr of accelerator allows reduction of sulfur dose from 8–10 to 0.5–3 phr and effective curing is achieved in a time scale of a few minutes to nearly an hour depending on temperature (100°C–140°C) and type of the selected accelerator. The low sulfur dose required in the accelerated sulfur vulcanization has not only eliminated bloom (migration of unreacted sulfur to the surface of the vulcanizate), which was a common feature of the earlier nonaccelerated technology, but also has led to the production of vulcanizates of greatly improved physical properties and good resistance to heat and aging. The selection of the accelerator depends largely on the nature of the rubber taken, the design of the product, and the processing conditions. It is important to adopt a vulcanizing system that not only gives a rapid and effective cross-linking at the desired vulcanizing temperatures but also resists premature vulcanization (scorching) at somewhat lower temperatures encountered in such operations as mixing, extrusion, calendaring, and otherwise shaping the rubber before ﬁnal cross-linking. This may require the use of delayed-action type accelerators as exempliﬁed by sulfenamides. Other principal types of accelerators with different properties are guanidines, thiazoles, dithiocarbamates, thiurams, and xanthates (Table 2.7). Accelerators are more appropriately classiﬁed according to the speed of curing induced in their presence in NR systems. In the order of increasing speed of curing, they are classiﬁed as slow, medium, semiultra, and ultra accelerators. The problem of scorching or premature vulcanization is very acute with ultra or fast accelerators. Rubber stocks are usually bad conductors of heat and therefore ﬂow of heat to the interior of a vulcanizing stock from outside is very slow. As a result, in thick items the outer layers may reach a state of overcuring before the core or interior layers begin to cure. For such thick items, a slow accelerator (Table 2.7) is most suitable. For butyl and EPDM rubbers, which have very limited unsaturations, slow accelerators are, however, unsuitable and, fast accelerators should be used at high temperatures for good curing at convenient rates. Since butyl rubber is characterized by reversion a phenomenon of decrease of tensile strength and modulus with time of cure after reaching a maximum, duration of heating at curing temperatures must be carefully controlled, and prolonged heating must be avoided. For rubbers with higher degree of unsaturations, an ideal accelerator is one that is stable during mixing, processing, and storage of the mix, but that reacts and decomposes sharply at the high vulcanization temperature to effect fast curring. These requirements or demands are closely fulﬁlled by the delayedaction accelerators typical examples of which are given in Table 2.7. The spectacular effects of modern organic accelerators in sulfur vulcanization of rubber are observed only in the presence of some other additives known as accelerator activators. They are usually twocomponent systems comprising a metal oxide and a fatty acid. The primary requirement for satisfactory activation of accelerator is good dispersibility or solubility of the activators in rubber. Oxides of bivalent metals such as zinc, calcium, magnesium, lead, and cadmium act as activators in combination with stearic acid. A combination of zinc oxide and stearic acid is almost universally used. (Where a high degree of transparency is required, the activator may be a fatty acid salt such as zinc stearate.) Besides speeding up the rate of curing, activators also bring about improvements in physical properties of vulcanizates. This is highlighted by the data given in Table 2.8. 266 Plastics Technology Handbook TABLE 2.7 Accelerator for Sulfur Vulcanization of Rubbers Accelerator Type and Formula Chemical Name Accelerator Activity Guanidines Diphenyl guanidine (DPG) Medium accelerator Triphenyl guanidine (TPG) Slow accelerator Mercaptobenzothiazole (MBT) Semi-ultra accelerator Mercaptobenzothiazyl disulﬁde (MBTS)a Semi-ultra (delayed action) accelerator N-Cyclohexyl benzothiazyl sulfenamide (CBS) Semi-ultra (delayed action) accelerator N-Oxydiethylenebenzothiazyl sulfenamide (NOBS) or 2-Morpholinothiobenzothiazole (MBS) Semi-ultra (delayed action) accelerator N-t-Butylbenzothiazyl sulfenamide (TBBS) Semi-ultra (delayed action) accelerator Zinc diethyl Dithiocarbamate (ZDC) Ultra accelerator NH C NH NH NH C N NH Thiazoles N C SH S N N C S S S Sulfenamides C S N C S NH S N C S N O S N C S NH C4H9 S Dithiocarbamates C 2H 5 S N C C 2H 5 Zn2+ S– 2 (Continued) 267 Fabrication Processes TABLE 2.7 (CONTINUED) Accelerator Type and Formula C 2H 5 S N C Accelerator for Sulfur Vulcanization of Rubbers Chemical Name Accelerator Activity Sodium diethyl dithiocarbamate (SDC) Ultra accelerator, water soluble (used for latex) Tetramethyl thiuram disulﬁde (TMTD, TMT)a Ultra accelerator Tetraethyl thiuram disulﬁde (TETD, TET)a Ultra accelerator Tetramethyl thiuram monosulﬁde (TMTM) Ultra accelerator Sodium isopropyl xanthate (SIX) Ultra accelerator, water soluble (suited for latex) Zinc isopropyl xanthate (ZIX) Ultra accelerator S– Na+ C 2H 5 Thiuram sulﬁdes S CH3 N S C S S C CH3 N CH3 CH3 C2H5 S N C S S S C C2H5 N C2H5 C2H5 CH3 N CH3 S S C S C N CH3 CH3 Xanthates S CH3 CH S– Na+ C O CH3 CH3 CH S O C CH3 a Zn2+ S– 2 Sulfur donors. A sulfur-curing system thus has basically four components: a sulfur vulcanizing agent, an accelerator (sometimes combinations of accelerators), a metal oxide, and a fatty acid. In addition, in order to improve the resistance to scorching, a prevulcanization inhibitor such as N-cyclohexylthiophthalimide may be incorporated without adverse effects on either the rate of cure or physical properties of the vulcanizate. The level of accelerator used varies from polymer to polymer. Some typical curing systems for the diene rubbers (NR, SBR, and NBR) and for two oleﬁn rubbers (IIR and EPDM—see Appendix A2 for abbreviations) are given in Table 2.9. In addition to the components of the vulcanization system, several other additives are commonly used with diene rubbers. Rubbers in general, and diene rubbers in particular, are blended with many more additives than is common for most thermoplastics with the possible exceptions of PVC. The major additional classes of additives are 1. Antidegradants (antioxidants and antiozonants) 2. Processing aids (peptizers, plasticizers, softeners, and extenders, tackiﬁers, etc.) 3. Fillers 268 Plastics Technology Handbook TABLE 2.8 Effect of Activator on Vulcanization Tensile Strength, Psi (MPa) ZnO (phr) Time of Cure (min) 0.0 5.0 100 (0.7) 2300 (16) 30 400 (2.8) 2900 (20) 60 90 1050 (7.2) 1300 (9.0) 2900 (20) 2900 (20) Note: Base compound: NR (pale creep) 100; sulfur 3; mercaptobenzothiazole (MBT) 0.5. Temperature of vulcanization 142°C. TABLE 2.9 Components of Sulfur Vulcanization Systems Rubber a Additive (phr) NR SBR NBR IIR EPDM Sulfur 2.5 2.0 1.5 2.0 1.5 Zinc oxide Stearic acid 5.0 2.0 5.0 2.0 5.0 1.0 3.0 2.0 5.0 1.0 TBBS 0.6 1.0 – – – MBTS MBT – – – – 1.0 – 0.5 – – 1.5 TMTD – – 0.1 1.0 0.5 a See Table 2.7 for accelerator abbreviations. 4. Pigments 5. Others (retarders, blowing agents) The use of antioxidants and antiozonants has already been described in Chapter 1. 2.20.1.1 Processing Aids Peptizers are added to rubber at the beginning of mastication (see later) and are used to increase the efﬁciency of mastication. They act chemically and effectively at temperatures greater than 65°C and hasten the rate of breakdown of rubber chains during mastication. Common peptizers are zinc thiobenzoate, zinc-2-benzamidothiophenate, thio-b-naphthol, etc. Processing aids other than the peptizer and compounding ingredients (additives) are added after the rubber attains the desired plasticity on mastication. Common process aids, besides the peptizer, are pine tar, mineral oil, wax, factice, coumarone-indene resins, petroleum resins, rosin derivatives, and polyterpenes. Their main effect is to make rubber soft and tacky to facilitate uniform mixing, particularly when high loading of carbon black or other ﬁllers is to be used. Factice (vulcanized oil) is a soft material made by treating drying or semidrying vegetable oils with sulfur monochloride (cold or white factice) or by heating the oils with sulfur at 140–160°C (hot or brown factice). The use of factice (5–30 phr) allows efﬁcient mixing and dispersion of powdery ingredients and gives a better rubber mix for the purpose of extrusion. Ester plasticizers (phthalates and phosphates) that are used to plasticize PVC (see Chapter 1) are also used as process aids, particularly with NBR and CR. Polymerizable plasticizers such as ethylene glycol dimethacrylate are particularly useful for peroxide curing of rubbers. They act as plasticizers or tackiﬁers during mixing and undergo polymerization by peroxide initiation during cure. Fabrication Processes 269 The diene hydrocarbon rubbers are often blended with hydrocarbon oils. The oils decrease polymer viscosity and reduce hardness and low temperatures brittle point of the cured product. They are thus closely analogous to the plasticizers used with thermoplastics but are generally known as softners. Three main types of softners are distinguished: aliphatic, aromatic, and naphthenic. The naphthenics are preferred for general all-round properties. NRs exhibit the phenomenon known as tack. Thus when two clean surfaces of masticated rubber are brought into contact the two surfaces strongly adhere to each other, which is a consequence of interpenetration of molecular ends followed by crystallization. Amorphous rubbers such as SBR do not display such tack and it is necessary to add tackiﬁers such as rosin derivatives and polyterpenes. 2.20.1.2 Fillers The principles of use of inert ﬁllers, pigments, and blowing agents generally follow those described in Chapter 1. Major ﬁllers used in the rubber industry are classiﬁed as (1) nonblack ﬁllers such as china clay, whiting, magnesium carbonate, hydrated alumina, anhydrous, and hydrated silicas and silicates including those in the form of ground mineral such as slate powder, talc, or French chalk, and (2) carbon blacks. Rather peculiar to the rubber industry is the use of the ﬁne particle size reinforcing ﬁllers, particularly carbon black. Fillers may be used from 50 phr to as high as 100–120 phr or even higher proportions. Their use improves such properties as modulus, tear strength, abrasion resistance, and hardness. They are essential with amorphous rubbers such as SBR and polybutadiene that has little strength without them. They are less essential with strain-crystallizing rubbers such as NR for many applications but are important in the manufacture of tires and related products. Carbon blacks are essentially elemental carbon and are produced by thermal decomposition or partial combustion of liquid or gaseous hydrocarbons to carbon and hydrogen. The principal types, according to their method of production, are channel black, furnace black, and thermal black. Thermal black is made from natural gas by the thermatomic process in which methane is cracked over hot bricks at a temperature of 1,600°F (871°C) to form amorphous carbon and hydrogen. Thermal black consists of relatively coarse particles and is used principally as a pigment. A few grades (FT and MT referring to ﬁne thermal and medium thermal) are also used in the rubber industry. Most of the carbon black used in the rubber industry is made by the furnace process (furnace black), that is, by burning natural gas or vaporized aromatic hydrocarbon oil in a closed furnace with about 50% of the air required for complete combustion. Furnace black produced from natural gas has an intermediate particle size, while that produced from oil can be made in wide range of controlled particle sizes and is particularly suitable for reinforcing rubbers. Quite a variety of grades of furnace blacks are available, e.g., ﬁne furnace black (FF), high modulus (HMF), high elongation (HEF), reinforcing (RF), semireinforcing (SRF), high abrasion (HAF), super abrasion (SAF), intermediate super abrasion (ISAF), fast extruding (FEF), general purpose (GPF), easy processing (EPF), conducting (CF), and super conducting furnace black (SCF). Channel black is characterized by lower pH, higher volatile content, and high surface area. It has the smallest particle size of any industrial material. A few grades of channel blacks (HPC, MPC, or EPC corresponding to hard, medium, or easy processing channel) are used in the rubber industry. For carbon-black ﬁllers, structure, particle size, particle porosity, and overall physico-chemical nature of particle surface are important factors in deciding cure rate and degree of reinforcement attainable. The pH of the carbon black has a profound inﬂuence. Acidic blacks (channel blacks) tend to retard the curing process while alkaline blacks (furnace blacks) produce a rate-enhancing effect in relation to curing, and may even give rise to scorching. Another important factor is the particle size of the carbon black ﬁller. The smaller the particle size, the higher the reinforcement, but the poorer the processability because of the longer time needed for dispersion and the greater heat produced during mixing. Blacks of the smallest particle size are thus unsuitable for use in rubber compounding. 270 Plastics Technology Handbook For carbon black ﬁllers the term structure is used to represent the clustering together and entanglement of ﬁne carbon particles into long chains and three-dimensional aggregates. High-structure blacks produce high-modulus vulcanizates as high shear forces applied during mixing break the agglomerates down to many active free radical sites, which bind the rubber molecules, thereby leading to greater reinforcement. In nonstructure blacks the aggregates are almost nonexistent. Most of the nonblack ﬁllers used in rubber compounds are of nonreinforcing types. They are added for various objectives, the most important being cost reduction. Precipitated silica (hydrated), containing about 10–12% water with average particle size ranging 10–40 nm, produce effective reinforcements and are widely used in translucent and colored products. Finely ground magnesium carbonate and aluminum silicate also induce good reinforcing effects. Precipitated calcium carbonate and activated calcium carbonate (obtained by treating calcium carbonate with a stearate) are used as semireinforcing ﬁllers. Short ﬁbers of cotton, rayon, or nylon may be added to rubber to enhance modulus and tear and abrasion resistance of the vulcanizates. Some resins such as “high styrene resins” and novolac-type phenolic resins mixed with hexamethylene tetramine may also be used as reinforcing ﬁllers or additives. Whereas SBR has a styrene content of about 23.5% and is rubbery, styrene-butadiene copolymers containing about 50% styrene are leatherlike whilst with 70% styrene the materials are more rigid thermoplastics but with low softening points. Both of these copolymers are known in the rubber industry as high styrene resins and are usually blended with a hydrocarbon rubber such as NR and SBR. Such blends have found use in shoe soles, car wash brushes and other moldings, but in recent years have suffered increasing competition from conventional thermoplastics and thermoplastic rubbers. 2.20.2 Mastication and Mixing A deﬁciency of NR, compared with the synthetics, is its very high molecular weight, which makes mixing of compounding ingredients and subsequent processing by extrusion and other shaping operations difﬁcult. For NR it is thus absolutely necessary, while for synthetic rubbers it is helpful, to subject the stock to a process of breakdown of the molecular chains prior to compounding. This is effected by subjecting the rubber to high mechanical work (shearing action), a process commonly known as mastication. Mastication and mixing are conveniently done using two-roll mills and internal mixers. The oxygen in are plays a critical role during mastication. Rubber that has been masticated is more soft and ﬂows more readily than the unmasticated material also allows preparation of solutions of high solids content because of the much lower solution viscosity of the degraded rubber. Rubber is also rendered tacky by mastication, which means that the uncured rubber sticks to itself readily so that articles of suitable thickness can be built up from layers of masticated rubber or rubberized fabric without the use of a solvent. 2.20.2.1 Open Mill The mainstays of the rubber industry for over 80 years has been the two-roll (open) mill and the Banbury (internal) mixer. Roll mills were ﬁrst used for rubber mixing over 120 years ago. The plastics and adhesives industries later adopted these tools. The two-roll mill (Figure 2.78) consists of two opposite-rotating rollers placed close to one another with the roll axes parallel and horizontal, so that relatively small gap or nip (adjustable) between the cylindrical surface exists. The speeds of the two rolls are usually different, the front roll having a slower speeds. For NR mixing, a friction ratio of 1:1.2 for the front to back roll may be used. For some synthetic rubbers or highly ﬁlled NR mixes, friction ratios close to 1.0 produce good results. The nip is adjusted so that when pieces of rubber are placed between the rolls, they are deformed by shearing action and squeezed through the nip to which they are returned by the operator. On repeated passage through the nip, the around the front roll and a moving “bank” above the nip. As the rolls keep on rotating, the operator uses a knife to cut through the band on the front roll, removes the mass in parts from time, and places it is a new position to ensure uniform treatment and mixing through the nip. 271 Fabrication Processes Feed/bank Back roll Front roll Rubber band Nip (adjustable) FIGURE 2.78 Section showing the features of a two-roll open mill. With most rubbers other than NR, addition of different compounding ingredients may be normally started soon after a uniform band is formed on the roll and a bank is obtained. For NR, however, milling is usually continued to masticate the rubber to the desired plasticity, and mixing of compounding ingredients is started only after the adequate mastication. Mixing is effected by adding the different ingredients onto the bank. They are gradually dispersed into the rubber, which is cut at intervals, rolled over, and recycled through the nip of the moving rolls to produce uniform mixing. Since the rate of mastication is a function of temperature, time and temperature of mastication have to be controlled or kept uniform from batch to batch in order to get the desired uniform products from different batches. Roll mills vary greatly in size from very small laboratory machines with rollers of about 1 in. in diameter and driven by fractional horsepower motors to very large mills with rollers of nearly 3 ft. in diameter and 7 or 8 ft. in length and driven by motors over 100 hp. 2.20.2.2 Internal Batch Mixers Internal batch mixers are widely used in the rubber industry. They are also used for processing plastics such as vinyl, polyoleﬁns, ABS, and polystyrene, along with thermosets such melamines and ureas because they can hold materials at a constant temperature. The principle of internal batch mixing was ﬁrst introduced in 1916 with the development of the Banbury mixer (Figure 2.79a). A Banbury-type internal mixer essentially consists of a cylindrical chamber or shall within which materials to be mixed are deformed by rotating blades or rotors with protrusions. The mixer is provided with a feed door and hopper at the top and a discharge door at the bottom. As the rubber or mix is worked and sheared between the two rotors and between each rotor and the body of the casing, mastication takes place over the wide area, unlike in a open mill where it is restricted only in the area of the nip between the two rolls. The rotor blade of the Banbury mixer is pear shaped, but the projection is spiral along the axis and the two spirals interlock and rotate in opposite directions (Figure 2.79b). The interaction of rotor blades between themselves, in addition to producing shearing action, causes folding or “shufﬂing” of the mass, which is further accentuated by the helical arrangement of the blade along the axis of the rotor, thereby imparting motion to the mass in the third, or axial, direction. This combination of intensive working produces a highly homogeneous mix. An important and novel feature of the Banbury mixer is a vertical ram to press the mass into contact with the two rotors. Rubbers, ﬁllers, and other ingredients are charged through the feed hopper and then held in the mixing chamber under the pressure of the hydraulic (or manual) ram. As a result, incorporation of solids is more rapid. Both the cored rotors and the walls of the mixing chamber can be cooled or heated by circulating ﬂuid. Because of the large power consumption of such a machine (up to 500 hp) the cylinder walls are usually water-cooled by sprays (Figure 2.79a). 272 Plastics Technology Handbook Feed hopper door Cored rotors Follower Cooling sprays (a) Sliding discharge door (b) FIGURE 2.79 (a) Cross section of a Banbury mixer; (b) Roll mixing blades in Banbury mixer. Other major machines in use in the rubber industry are the Shaw intermix and the Baker-Perkins shear-mix. Both the Banbury-type and the intermix mixers have passed through modiﬁcations and reﬁnements at different points in time. Mechanical feeding and direct oil injection in measured doses into the mixing chamber through a separate oil injection port are notable features of modern internal mixers. Higher rotor speeds and, in the Banbury type, the higher ram pressures, are used for speedy output. In Banbury-type mixers, the rotors run at different speeds, while in intermix mixers the rotor speeds are equal but the kneading action between the thicker portion of one rotor and the thinner portion of the other produces a frictional effect. The operation of internal mixers is power intensive, and a given job is performed at a much higher speed and in a much shorter time on a two-roll open mill. However, the major vulcanizing agent (such as sulfur) is often added later on a two-roll mill so as to eliminate possibilities of scorching. And even if this is not practiced, the mix, after being discharged from the internal mixer, is usually passed through a two-roll mill in order to convert it from irregular lumps to a sheet form for convenience in subsequent processing. 2.20.3 Reclaimed Rubber The use of reclaimed rubber in a fresh rubber mix not only amounts to waste utilization but offers some processing and economic advantages to make it highly valued in rubber compounding. Though waste Fabrication Processes 273 vulcanized rubber is normally not processable, application of heat and chemical agents to ground vulcanized waste rubber leads to substantial depolymerization whereby conversion of the rubber to a soft, plastic processable state is effected. Rubber so regenerated for reuse is commonly known as reclaimed rubber or simply as reclaim. Reclaimed rubber can be easily revulcanized. Worn-out tires and scraps and trimmings of other vulcanized products constitute the raw material for reclaimed rubber. Therefore a good reclaiming process must not only turn the rubber soft and plastic but also must remove reinforcing cords and fabrics that may be present. There are a number of commercial processes [75] for rubber regeneration: (1) alkali digestion process, (2) neutral or zinc chloride digestion process, (3) heater or pan process, and (4) reclaimator process. Tires are most commonly reclaimed by digestion processes. For processes (1) and (2), debeaded tires and scraps, cut into pieces, are ground with two-roll mills or other devices developed for the purpose. Two-roll mills generally used for grinding tires turn at a ratio of about 1:3, thus providing the shearing action necessary to rip the tire apart. The rubber chunks are screened, and the larger material is recycled until the desired size is reached. The ground rubber is then mixed with a peptizer, softener, and heavy naphtha, and charged into spherical autoclaves with requisite quantities of water containing caustic soda for process (1) or zinc chloride for process (2). The textile is destroyed and mostly lost in the digestion process. Steam pressure and also the amount of air or oxygen in the autoclave greatly inﬂuence the period necessary for rubber reclaiming. On completion of the process, the pressure is released, the contents of the autoclave discharged into water, centrifuged, pressed to squeeze out water, and dried. The material is ﬁnally processed through a two-roll mill during which mineral ﬁllers and oils may be added to give a product of required speciﬁc gravity and oil extension. Butyl and NR tubes and other ﬁber-free scrap rubbers are reclaimed by means of the heater or pan process. Brass tube ﬁttings and other metal are removed from the scrap. The scrap is mechanically ground, mixed with reclaiming agents, loaded into pans or devulcanizing boats, and autoclaved at steam pressures of 10–14 atm (1.03–1.40 MPa) for 3–8 h. The reclaim is ﬁnally processed much the same way as in the digester process. The reclaimator process is more attractive than the above processes. The reclaimator is essentially a high-pressure extruder that devulcanizes ﬁber-free rubber continuously. Ground scrap is mechanically treated in hammer mills to remove the textile material, mixed with reclaiming oils and other materials, and then fed into the reclaimator. High pressure and shear between the rubber mixture and the extruder barrel walls effectively reclaim the rubber mixture. Devulcanization occurs at 175–205°C in a few minutes and turns the rubber into reclaim that issues from the machine continuously. The whole regeneration, which is a dry process, may be completed in about 30 min. The reclaiming oils and chemicals are complex wood and petroleum derivatives that swell the rubber and provide access for breaking the rubber bonds with heat, pressure, chemicals, and mechanical shearing. Approximately 2–4 parts of oil are used per 100 parts of scrap rubber. Some examples of reclaiming oils include monocyclic and mixed terpenes, i.e., pine-tar products, saturated polymerized petroleum hydrocarbons, aryl disulﬁdes in petroleum oil, cycloparaﬁnic hydrocarbons, and alkyl aryl polyether alcohols. Reclaimed rubber contains all the ﬁllers present in the original scrap or waste rubber. It shows very good aging characteristics and is characterized by less heat development during mixing and processing as compared to fresh rubber. The use of reclaim in a fresh rubber mix is advantageous not on consideration of physical and mechanical properties but essentially for smooth processing and reduced cost. Radial tires (see later) do not use reclaimed rubber because they require higher abrasion resistance that cannot be attained by mixing reclaimed rubber. Better processes for the production of higher quality reclaimed rubber are needed in order to use it for radial tires. To improve the quality of reclaimed rubber, cross-links in vulcanizates should be severed selectively during a devulcanization process and no lowmolecular-weight compound such as swelling solvent should remain in the reclaimed rubber after the devulcanization process. 274 Plastics Technology Handbook A devulcanization process that utilizes supercritical CO2 as a devulcanization reaction medium in the presence of diphenyl disulﬁde as a devulcanizing reagent has been reported [77]. The process devulcanizes unﬁlled NR vulcanizates effectively. Further, a comparison of measured sol/gel components as well as dynamic mechanical properties of the devulcanized rubber products of ﬁlled and unﬁlled NR vulcanizates has indicated that the presence of carbon black in the vulcamizate does not disturb the devulcanization in supercritical CO2. 2.20.4 Some Major Rubber Products The most important application of rubbers is in the transport sector, with tires and related products consuming nearly 70% of the rubber produced. Next in importance is the application in belting for making ﬂat conveyor and (power) transmission belts and V-belts (for power transmission), and in the hose industry for making different hoses. Rubber also has a large outlet in cellular and microcellular products. Other important and special applications of rubber are in the areas of adhesives, coated fabrics, rainwear, footwear, pipes and tubing, wire insulation, cables and sheaths, tank lining for chemical plants and oil storage, gaskets and diaphragms, rubber mats, rubber rollers, sports goods, toys and balloons, and a wide variety of molded mechanical and miscellaneous products. Formulations of a few selected rubber compounds are given in Appendix A4. 2.20.4.1 Tires Tire technology is a very specialized area, and a tire designer is faced with the difﬁcult task of trying to satisfy all the needs of the vehicle manufacturer, the prime factors of consideration, however, being safety and tread life. Figure 2.80 shows the constructions of a standard bias (diagonal) ply tire and a radial ply tire. The major components of a tire are: bead, carcass, sidewall, and tread. In terms of material composition, a tire on an average contains nearly 50% of its weight in actual rubber; for oil extended rubbers (typically containing 25 parts of aromatic or cycloparafﬁnic oils to 75 parts of rubber), it is less. The remainder included carbon black, textile cord, and other compounding ingredients plus the beads. The bead is constructed from a number of turns or coils of high tensile steel wire coated with copper and brass to ensure good adhesion of the rubber coating applied on it. The beads function as rigid, practically inextensible units that retain the inﬂated tire on the RIM. The carcass forms the backbone of the tire. The main part of the carcass is the tire cord. The cords consist of textile threads twisted together. Rayon, nylon, and polyester cords are widely used. Steel cords are also used. While the practice of laying rubber-impregnated cords in position is still followed, the use of woven fabric is more widespread. To promote a good bond with rubber, the ﬁber or cord is treated with an adhesive composition such as water-soluble resorcinol-formaldehyde resin and aqueous emulsion of a copolymer of butadiene, styrene, and vinyl pyridine (70:15:15). The resincoated cord or fabric is dried and coated with a rubber compound by calendaring, whereby each cord is isolated from its neighbor. For conventional tires, the rubber-coated fabric is then cut to a predetermined width and bias angle. The biascut plies are joined end-to-end into a continuous length and batched into roll form, interleaved with a textile lining to prevent self-adhesion. The sidewall is a layer of extruded rubber compound that protects the carcass framework from weathering and from damage from chaﬁng. Together with the tread and overlapping it in the buttress region, the sidewall forms the outermost layer of the tire. It is the most highly strained tire component and is susceptible to two types of failure—ﬂex cracking and ozone cracking. The tread is also formed by extrusion, but different rubber compounds are used for sidewall and tread (see Appendix A4). When making the side-wall and the tread in two separate extrusion lines, it is useful to take the extruded sidewall to the second extruder, which would deliver the tread to the sidewall. To simplify the tire-building operation, the sidewall and the tread may also be produced as a single unit by simultaneous extrusion of the two compounds in a single band-the tread over the sidewall-with two 275 Fabrication Processes Casing plies Filler Chafer Casing plies Wall rubber Chafer (a) Casing plies Bead wires Bread wrap Tread pattern Thread bracing layers Radial plies Inner lining Wall rubber Chafer Chafer strip (b) FIGURE 2.80 Bead ﬁller Apex strip Beap wrapping Beap coil Diagram showing. (a) bias (diagonal) ply tire; (b) radial ply tire. extruders arranged head-to-head with a common double die. The band is rapidly cooled to avoid scorching and then cut to the appropriate length. The tread receives its characteristic pattern from the mold when the subsequently built tire is vulcanized. While building the tire, a layer of specially compounded cushion rubber may be used to keep heat development on ﬂexing to a minimum and achieve better adhesion between the tread and the carcass. One or more layers of fabric, known as breakers or bracing layers, may be placed below the cushion. The bracing layers raise the modulus of the tread zone and level out local blows to the tread as it contacts the road. The tire building is carried out on a ﬂat drum, which is rotated at a controlled speed. The plies of rubber-coated cord fabric are placed in position, one over the other, and rolled down as the drum rotates, the inner lining being placed next to the drum and the ends of the drum being ﬂanged to suit the bead conﬁguration of the tire. The plies of rubber-coated textile are assembled in three basic constructions— bias (diagonal), radial, and bias belted. Bias tires have an even number of plies with cords at an angle of 30–38° from the tread center line. Passenger-car bias tires commonly have two or four plies, with six for heavy duty service. Truck tires are often built with six to twelve plies, although the larger earthmover types may contain thirty or more. 276 Plastics Technology Handbook In the radial-ply tire, one or two plies are set at an angle of 90° from the center line and a breaker or belt or rubber-coated wire or textile is added under the tread. This construction gives a different tread-road interaction, resulting in a decreased rate of wear. The sidewall is thin and very ﬂexible. The riding and steering qualities are noticeably different from those of a bias-ply tire and require different suspension systems. The bias-belted tire, on the other hand, has much of the tread wear and traction advantage of the radial tire, but the shift from bias to bias-belted tires requires less radical change in vehicle suspension systems and in tire building machines. These features make the bias-belted tire attractive to both automobile and tire manufacturers. When the tire building on the drum is complete, the drum is collapsed and the uncured (“green”) tire is removed. The cylindrical shape of the uncured tire obtained from the building drum is transformed into a toroidal shape in the mold resulting in a circumferential stretch of the order of 60%. For shaping and curing, the uncured tire is pressed against the inner face of the heated mold by an internal bag or bladder made of a pre-cured heat-resistant rubber and inﬂated by a high-pressure steam or circulating hot water. Different types of molding presses are in use. In the Bag-O-Matic type of press, the uncured tire is placed over a special type of bag or bladder, and the operation of shaping, curing, and ejection of the cured tire are accomplished by an automatic sequence of machine operation. Pneumatic tires require an air container or inner tube of rubber inside the tire. The manufacture of inner tubes is done essentially in three steps: extrusion, cutting into length, and insertion of value and vulcanization. The ends of the cut length of the tube are joined after insertion of the valve prior to vulcanization. Tubeless tires have, instead of an inner tube, an inner liner, which is a layer of rubber cured inside the casing to contain the air, and a chafer around the bead contoured to form an airtight seal with the RIM. 2.20.4.2 Belting and Hoses Uses of ﬂat conveyor and (power) transmission belts and V-belts (for power transmission) are to be found in almost all major industries. V-belts for different types cover applications ranging from fan belts for automobiles, belts for low-power drives for domestic, laboratory, and light industrial applications, to high-power belts for large industrial applications. Textile cords or fabric and even steel cords constitute an important part of all rubber belting and hoses. The various types of cords used in the tire industry are also in use in the belting industry. Essential steps in making conveyor belts are: (1) drying the fabric; (2) frictioning of the hot fabric with a rubber compound and topping to give additional rubber between plies and the outer ply and cover, using a three- or four-roll calender; (3) belt building; and (4) vulcanization. The process of belt building essentially consists or cutting, laying, and folding the frictioned (a) and coated fabric to give the desired number of plies. The construction may be straight-laid or graded-ply types (Figure 2.81) with the joints in neighboring plies being staggered to eliminate weakness and failure. Finally, the cover coat is (b) applied by calendaring. Vulcanization of conveyor belts may be carried out in sections using press cure or continuously by means of a Rotocure equipment. In press cure, vulcanization is done by heating in (c) long presses, the belt being moved between FIGURE 2.81 Construction of conveyor belts: (a) straight successive cures by a length less than the length of the press platens. Since in this process the end laid, (b) folded jacket, and (c) graded ply. 277 Fabrication Processes or overlap zones receive an additional cure, it is desirable to minimize damage or weakness due to overcure by using ﬂat cure mixes and allowing water cooling at each end section of the platen. In each step of vulcanization, the section of the belt to be vulcanized is gripped and stretched hydraulically to minimize or eliminate elongation during use. The difﬁculties of press cure may, however, be avoided by adopting continuous vulcanization with a Rotocure equipment in which the actual curing operation is carried out between an internally steam heated cylinder and a heated steel band. Rotocure is also useful for the vulcanization of transmission belts and rubber sheeting. The most common feature of V-belts is their having a cross section of the shape of a regular trapezium with the unparallel sides at an angle of 40° (Figure 2.82). The V-belts usually consist of ﬁve sections: (1) the top section known as the tension section, (2) the bottom section, called the compression section, (3) the cord section located at the neutral zone, (4) the cushion section on either side of the cord section, and (5) one or two layers of rubberized fabric, called the jacket section covering the whole assembly. Three different rubber compounds are required for use in the above construction of a V-belt: (1) a base compound, which is the major constituent on a weight basis, (2) a soft and resilient cushion compound required for surrounding and protecting the reinforcing cord assembly, and (3) a friction compound used for rubberizing the fabric casing of the belt (see Appendix A4). Relatively short length V-belts are built layer by layer on rotatable collapsible drum formers. The separate belts are then cut out with knives and transferred to a skiving machine that imparts the desired V-shape. The fabric jacket is then applied, and the belts are vulcanized in open steam using multi-cavity ring molds for smaller belts. Long belts are built similarly on V-groove sheave pulleys using weftless cord fabric in place of individually would cord. They are vulcanized endlessly by molding in a hydraulic press under controlled tension. A rubber hose has three concentric layers along the length. While the innermost part consisting of a rubber lining or tube is required to resist the action of the material that would pass through the top layer is meant to play the role of a protective layer to resist weathering, oils, chemicals, abrasion, etc. Between the inner lining and the outer cover is given a layer of reinforcement of textile yarn or steel wire applied by spiraling, knitting, braiding, or circular loom weaving. A cut woven fabric wrapped straight or on the bias may also be used to reinforce the inner lining or tube. Essentially, the process of hose building consists of extruding the lining or tube, braiding or spiraling the textile around the cooled tube, and applying an outer cover of rubber to the reinforced hose using a cross-head extruder. Several methods are employed for vulcanization. In one process, the built hose is passed through a lead press or lead extruder to give a layer of lead cover to the hose. The hose is then wound on a drum, ﬁlled with water or air, and the ends are sealed. The whole assembly is then heated to Width Thickness Tension section Cord section Compression section Cushion section Cover or jacket section 40° FIGURE 2.82 Cross section showing V-belt construction. 278 Plastics Technology Handbook achieve vulcanization. The water or air inside expands, and the vulcanizing hose is pressed against the lead acting as the mold. On completion of cure, the sealed ends are cut open, the lead cover is removed by slitting lengthwise in a stripping machine, and the cured hose is coiled up. 2.20.4.3 Cellular Rubber Products Cellular rubber may be described as an assembly of a multitude of cells distributed in a rubber matrix more or less uniformly. The cells may be interconnected (open cells) as in a sponge or separate (closed cells). Foam rubber made from a liquid starting material such as latex, described earlier, is of open-cell type. Cellular products made from solid rubber are commonly called sponge (open cell structure) and expanded rubber (closed cell structure). The technology of making cellular products from solid rubber is solely dependent on the incorporation of a blowing agent, usually a gas such as nitrogen or a chemical blowing agent, into the rubber compound. The most widely used chemical blowing agent for this application is dinitroso pentamethylene tetramine (DNPT). The curing is carried out either freely using hot air or steam or in a mold that is only partially ﬁlled with the molding compound. Synthetic rubbers, particularly SBR, are preferred as they allow precise control over level of viscosity required for obtaining consistent product quality. The sponge and expanded rubber products include carpet backing, sheets, proﬁles, and molding. The development of microcellular rubber has brought a revolution in footwear technology. Microcellular rubber with an extremely ﬁne noncommunicating cell structure and very comfortable wearing properties, is the lightest form of soling that can be produced. Density of soling as low as 0.3 g/cm3 may be obtained with a high dose (8–10 phr) of DNPT at a curing temperature of 140–150°C. For common solings, the density normally varies between 0.5 and 0.8 g/cm3. High hardness and improved abrasion resistance along with low density, desired in microcellular soling, can be achieved by using SBR and high styrene resins with NR in right proportions. Higher proportions of high styrene resins give products of higher hardness and abrasion resistance and lower density. Silicious ﬁllers such as precipitated silica and aluminum or calcium silicate also give higher hardness, abrasion resistance, and split tear strength. Microcellular crumbs can be used in considerable quantity along with china clay and whiting to reduce the product cost. Higher proportions of stearic acid (5–10 phr) are normally used in microcellular compounds in order to bring down the decomposition temperature of DNPT type blowing agents (see Appendix A4). Post-cure oven stabilization of the microcellular sheets, typically at 100°C for 4 h, reduces the delayed shrinkage after cure to a minimum. 2.21 Miscellaneous Processing Techniques 2.21.1 Coating Processes A coating is thin layer of material used to protect or decorate a substrate. Most often, a coating is intended to remain bonded to the surface permanently, although there are strippable coatings which are used only to afford temporary protection. An example of the latter type is the strippable hotmelt coating with ethyl cellulose as the binder [78], which is used to protect metal pieces such as drill bits or other tools and gears from corrosion and mechanical abrasion during shipping and handling. Two of the principal methods of coating substrates with a polymer, namely extrusion coating and calendaring have already been dealt with in this chapter. Other methods of coating continuous webs include the use of dip, knife, brush, and spray. Dip coating, as applied to PVC, has already been described in previous section on plastisols. In knife coating the coating is applied either by passing the web over a roll running partly immersed in the coating material or by running the coating material onto the face of the web while the thickness of the coating is controlled by a sharply proﬁled bar (or knife). This technique, also referred to as spreading, is Fabrication Processes 279 used extensively for coating fabrics with PVC. The PVC is prepared in the form of a paste, and more than one layer is usually applied, each layer being gelated by means of heat before the next layer is added. Lacquers are a class of coatings in which ﬁlm formation results from mere evaporation of the solvents(s). The term “lacquer” usually connotes a solution of a polymer. Mixtures of solvents and diluents (extenders which may not be good solvents when used alone) are usually needed to achieve a proper balance of volatility and compatibility and a smooth coherent ﬁlm on drying. Some familiar examples of lacquers are the spray cans of touch-up paint sold to the auto owner. These are mostly pigmented acrylic resins in solvents together with a very volatile solvent [usually dichlorodiﬂuoromethane (CCl2F2)] which acts as a propellant. A typical lacquer formulation for coating steel surfaces contains polymer, pigment, plasticizer (nonvolatile solvent), and volatile solvents. Latex paints or emulsion paints are another class of coatings which form ﬁlms by loss of a liquid and deposition of a polymer layer. The paints are composed of two dispersions: (1) a resin dispersion, which is either a latex formed by emulsion polymerization or a resin in emulsion form, and (2) a dispersion of colorants, ﬁllers, extenders, etc., obtained by milling the dry ingredients into water. The two dispersions are blended to produce an emulsion paint. Surfactants and protective colloids are added to stabilize the product. Emulsion paints are characterized by the fact that the binder (polymer) is in a water-dispersed form, whereas in a solvent paint it is in solution form. In emulsion systems the external water phase controls the viscosity, and the molecular weight of the polymer in the internal phase does not affect it, so polymers of high molecular weight are readily utilized in these systems. This is an advantage of emulsion paints. The minimum temperature at which the latex particles will coalesce to form a continuous layer depends mainly on the Tg. The Tg of a latex paint polymer is therefore adjusted by copolymerization or plasticization to a suitable range. The three principal polymer latexes used in emulsion paints are styrenebutadiene copolymer, poly(vinyl acetate), and acrylic resin. Although the term “paint” has been used for latex-based systems as well as many others, traditionally it refers to one of the oldest coating systems known—that of a pigment combined with a drying oil, usually a solvent. Drying oils (e.g., linseed, tung), by virtue of their multiple unsaturation, behave like polyfunctional monomers which can polymerize (“dry”) to produce ﬁlm by a combination of oxidation and free-radical propagation. Oil-soluble metallic soaps are used to catalyze the oxidation process. Combinations of resins with drying oils yield oleoresinous varnishes, whereas addition of a pigment to a varnish yields an enamel. The combination of hard, wear-resistant resin with softer, resilient, drying-oil ﬁlms can be designed to give products with a wide range of durability, gloss, and hardness. Another route to obtaining a balanced combination of these properties is the alkyd resin, formed from alcohols and acids (and hence the “alkyd”). Alkyds are actually a type of polyester resin and are produced by direct fusion of glycerol, phthalic anhydride, and drying oil at 410°F–450°F (210°C–232°C). The process involves an esteriﬁcation reaction of the alcohol and the anhydride and transesteriﬁcation of the drying oil. A common mode of operation today is thus to start with the free fatty acids from the drying oil rather than with the triglycerides. 2.21.1.1 Fluidized Bed Coating Fluidized bed coating is essentially an adaptation of dip coating and designed to be used with plastics in the form of a powder of ﬁne particle size. It is applied for coating metallic objects with plastics. The uniqueness of the process lies in the fact that both thermoplastics and thermosetting resins can be used for the coating. Uniform coating of thicknesses from 0.005 to 0.080 in. (0.13–2.00 mm) can be built on many substrates such as aluminum, carbon steel, brass, and expanded metal. The coating is usually applied for electrical insulation and to enhance the corrosion resistance and chemical resistance of metallic parts. Low-melting polymers are most appropriate for ﬂuidized beds. Highermelting polymers must have a sufﬁciently low melt viscosity that the particles can ﬂow and fuse together to form a continuous coating. Thermoplastic polymers in common use include nylon, PVC, acrylics, polyethylene, and polypropylene. 280 Plastics Technology Handbook Other possible materials are thermoplastic urethane, silicones, EVA, polystyrene, or any other lowmelting, low-viscosity polymer. Thermosetting polymers are limited largely to epoxy and epoxy/polyester hybrids since other thermosets, such as phenolic and urea-formaldehyde resins, give off volatile byproducts that can create voids in the coating. The powder resin particles range in size form about 20–200 microns. Particles larger than 200 microns are difﬁcult to suspend. Particles smaller than 20 microns may create excessive dusting and release of particles from the top of the bed. The actual coating process is uncomplicated; however, achievement of a uniform coating requires considerable skill. The metal object to be coated of powdered resin ﬂuidized by the passage of air through a porous plate. The bed is not heated; only the surface of the object to be coated is hot. As the powder contacts the hot substrate, the particles adhere, melt, and ﬂow together to form a continuously conforming coating. The object is removed from the bed when the desired coating thickness is obtained. On cooling, the coated resin resumes its original characteristics. In the case of a thermosetting resin, additional time at elevated temperature may be required to complete the cure. The ability of the ﬂuidized bed to continuously conform to and coat parts having unusual shapes and sizes permits a high degree of ﬂexibility in application. This is invaluable to the processor who wishes to coat one-of-a-kind products. In general, the coatings are smooth and glossy, with excellent adhesion to the substrate, providing the hermetic seal necessary for proper maintenance. Fluidized bed coating is certainly the most efﬁcient method of applying a thick coating in a single step. Thicknesses of 0.1 in. (2.5 mm) or greater are easily attained. Probably the single biggest advantage of powder coating is the nearly 100% utilization of the coating resin, without the hazard or expense of solvents. Almost all the coatings applied by this process have a deﬁnite function, chieﬂy electrical insulation, but it can be used for applications that simply require a thick coating with powder. Examples of electrical applications in include small motor stators and rotors, electronic components (capacitors or resistors), transformer casings, covers, laminations, and busbar. Other items coated using ﬂuidized beds include valve bodies used in chemical industries, reﬁnery equipment, and appliance and pump parts. 2.21.1.2 Spray Coating Spray coating is especially useful for articles that are too large for dip coating or ﬂuidized bed coating. The process consists of blowing out ﬁne polymer powders through a specially designed burner nozzle, which is usually ﬂame heated by means of acetylene or some similar gas, or it can be heated electrically. Compressed air or oxygen is used as the propelling force for blowing the polymer powder. 2.21.1.3 Electrostatic Spraying The electrostatic spraying of polymer powders utilizes the principle that oppositely charged particles attract, a principle that has been used for many years in spraying solvent-based paints. In electrostatic spraying, polymer powder is ﬁrst ﬂuidized in a bed to separate and suspend the particles. It is then transferred through a hose by air to a specially designed spray gun. As the powder passes through the gun, direct contact with the gun and ionized air applies an electrostatic charge to the particles of powder. The contact area may be a sleeve that extends the length of the gun or merely small pins that extend into the passageway of the powder. For safety the gun is designed for high voltage but low amperage. The part to be coated is electrically grounded, attracting the charged particles. This produces a more even coating and reduces overspray. Parts to be coated may be preheated, thereby forming the coating immediately, or they may be coated cold as the electrostatic charge will hold the particles in place until heat is applied. Once heat is applied, the particles melt and ﬂow together, forming a continuous protective coating. Coatings 50–75 microns thick can be applied electrostatically to cold objects and coatings up to 250 microns thick to hot objects. The polymers used in spraying of powders are the same as those used in ﬂuidized beds. The key characteristic for any polymer, thermoplastic or thermoset, applied as a Fabrication Processes 281 powder is low melt viscosity, which enables polymer particles to ﬂow together and form a continuous coating. The chief characteristic of electrostatically applied powder polymer is the ability to produce a thin coating. It is the preferred method of producing a coating of 0.001–0.002 in. (0.025–0.050 mm) thickness. It is a continuous process and suited to automated assembly-line production. With a well designed recovery system and fully enclosed spray booth, the process permits full utilization of the powder. The particle size of sprayed powders is smaller than that of powders used in ﬂuidized bed coating. The average particles size is 30–60 microns. Much of what is termed decorative powder coating is applied by electrostatic spray. Appliances, laboratory instruments, transformer housings, engine parts, and chain-link fences are among the many types of products so coated for decorative purposes. Coatings that provide electrical insulation are also applied by electrostatic spray. Objects coated include electrical motor armatures and stators, electrical switchgear boxes, and magnet wire. Corrosion-resistant coatings are designed to prevent the corrosion of the underlying substrates. These include pipe, fencing, concrete reinforcing bars, valves, conduit, and pumps. Polymers used for corrosion protection are usually thermosetting, particularly epoxies because of their superior adhesion characteristics. 2.21.1.4 Smart Coatings Whereas traditional coatings only protect the substrate to which they are applied by providing a barrier between the surface and the environment, smart coatings go much further as they are able to sense a change in conditions in the environment and respond to that change in a predictable and noticeable manner to mend or eliminate the problem. Stimuli for smart coatings can be any of a number of changes in environmental conditions, such as heat, pressure, pH, impact, vibrations, presence of pathogens and other organisms, certain chemicals (such as corrosive materials), humidity, electronic and magnetic ﬁelds, sunlight and other radiations, and others. The functional ingredient within the intelligent coating can be the resin itself or a variety of additives including microencapsulated ingredients, pigments, antimicrobial agents, enzymes or other bioactive species, nanomaterials (nanoparticles, nanotubes, nanocapsules, etc.), microelectromechanical devices, and radio-frequency identiﬁcation devices. The potential applications for these numerous types of smart coatings are broad and varied, including corrosion control, camouﬂage, bio-weapon detection and destruction, and other safety applications. The need for smart coatings and functional surfaces exists in diverse industries, including aerospace, marine, automotive, construction, communication, textile, biomedical, electronics, energy, environmental protection, personal safety, and many others. Smart polymer coatings can be broadly classiﬁed into two categories, depending on the type of sensors used, as those based on color response and those based on noncolor response. For smart sensors based on color response, the response may be visible color change, ﬂuorescence, or phosphorescence as a result of a variety of stimuli that include pH change, redox reactions, presence of heavy metals, sorption of chemicals, radiation, mechanical action, temperature variation, and electrical current [79]. A few selected smart coatings based on color response are listed in Table 2.10, indicating types of stimulus, response, sensor used, sensing mechanism, and application. The smart functionality may be imparted to the coatings by additives or colorants that are added separately into the coating before application on a given substrate. The colorants may be simple dyes or pigments as shown for systems 1–5 in Table 2.10. Smart functionality may also be built into the polymeric structure, as shown for systems 6–8 in Table 2.10. Examples of sensors in system 6 are acrylic polymers with long crystallizable, hydrophobic side chains that can act as temperature-activated crystallinity/permeability switches [79]. In a typical application as smart seed coating, germination is prevented at soil temperatures below about 13°C because the seed coating is then crystalline and provides a barrier to moisture, while at higher temperatures, the coating assumes an amorphous structure that enhances moisture penetration and allows seed germination [80]. Examples of sensors in system 7 are conductive ﬁlms of polyaniline (Pani), polypyrrole (Ppy), and polythiophene (Pth), which undergo changes in properties, such as color, electrical conductivity, Color change pH change Oxidation Mechanical force Light, temperature Heavy metal, radioactive contamination Temperature Oxidation UV light, g-rays, electrons Color change 1 2 3 4 5 6 7 8 Side-chain crystallizable polymer Colorimetric dye Polymerization Source: Feng, W., Patel, S. H., Young, M-Y., Zunino III, J. L., and Xanthos, M., 2007. Adv. Polym. Tech., 26, 1. Diacetylenes Moisture barrier (for controlling seed germination) Surface decontamination with strippable coatings Battery tester, printing inks, coatings Radiation detection Switch between charged and neutral Sensing, actuation, corrosion resistance states Crystalline to amorphous transformation Colored metal complex Photochromic dyes, thermochromic Structural transitions pigments, LCP and dyes Film or CaCO3 formation (healing); Self-healing; crack detection dye release (crack sensing) U/F capsules with ﬁlm former, or Ca(OH)2 or marker dye Corrosion detection Corrosion detection Application Redox reaction—ﬂuorescent oxidized form Ionic transition Sensing Mechanism Fluorescein, Schiff bases pH indicator Sensor Effect on optical properties Conducting polymer ﬁlms of Pani, Ppy, Pth Color change Color change Color change Capsule rupture-healing, color change Fluorescence Response System Stimulus TABLE 2.10 Polymeric Smart Coatings Containing Functional Colorants 282 Plastics Technology Handbook Fabrication Processes 283 permeability, density, and charge density, during reversible redox and pH switching reactions. These conducting polymers can be used in smart coatings for corrosion protection on steel, stainless steel, aluminum, and copper. The conductive polymer can respond to oxidants in a corrosive environment and protect the metal anodically by retarding or inhibiting corrosion through the formation of passivating metal oxide ﬁlms. Examples of radiation detection sensor in system 8 are diacetylenes (R–C≡C–C≡C–R, where R is a substituent group). Diacetylenes are colorless solid monomers. They usually form red- or blue-colored polymers [=(R)C–C≡C–C(R)=]n when irradiated with high-energy radiations, such as x-ray, g-ray, electrons, and neutrons. As exposure to radiation increases, the color of the sensing strip composed of diacetylenes intensiﬁes in proportion to the dose. By using a proper diacetylene and thickness of coating, one can monitor doses even lower than 1 rad (e.g., 0.1 rad or even lower). SIRAD dosimeters, for example, are able to monitor as low as 0.01 rad by using more sensitive diacetylenes, thicker sensor and scanners, and CCD camera-type equipment [81]. [Note: A CCD, or charge coupled device, is an integrated circuit etched onto a silicon surface forming light-sensitive elements called pixels. Photons incident on this surface generate charges that can be read electronically and converted into a digital copy of the light pattern falling on the device.] Some examples of smart polymer coatings based on noncolor response are shown in Table 2.11. These are compared on the basis of stimulus, response, sensor type, sensing mechanism, and application. Examples [79] include modiﬁcation of absorptive characteristics of coatings on biomaterials (systems 1 and 2), self-healing of cracks in coatings (system 3), temporary protective coatings that can be removed on demand via dissolution in appropriate reagent (system 4), and monitoring durability of coatings through dielectric sensors (system 5). The smart functionality may be built in the polymeric structure of a coating or may be provided through incorporation of suitable additives to the coating before its application on a given substrate. Since poly(N-isopropylacrylamide) (PNIPAM) has a lower critical solution temperature (LCST) of 31°C in an aqueous environment, below 31°C, the polymer becomes hydrated and assumes a random coil conﬁguration, while above 31°C, the polymer chains take on a much more compact conﬁguration by sudden dehydration and increased hydrophobic interaction between the polymer chains. When grafted onto solid surfaces, the polymer therefore provides a smart coating (system 1) with varying properties that can be controlled by applying an external stimulus—temperature. Below the phase transition temperature, the PNIPAM-grafted surfaces are hydrophilic, swollen, and nonprotein adsorptive (nonfouling), while above the transition temperature, the grafted polymer chains collapse and the surface becomes hydrophobic and protein-retentive. The LCST of PNIPAM being close to the body temperature, PNIPAM-grafted surfaces offer possibilities for a number of novel applications. Examples [82] include smart and thermally responsive coatings as cell culture substrates to control the attachment and detachment of cells, the recovery of cultured cells, a biofouling releasing coating, temperature-responsive membranes, controlled release of drugs, and temperature-responsive chromatography. Different methods have been reported for grafting PNIPAM on surfaces. These include activated substrates and functionalized polymers [83], e-beam irradiation [84], photoinitiated grafting of functionalized polymer [85], and plasma-induced grafting [86]. Microcapsules containing a small amount of “healing agent” that will be released by crack propagation have been incorporated into polymer coatings. This process has been used for self-healing in polymer composites through release of a polymerizable healing agent that would bridge cracks after reaction with appropriate catalysts [79], as shown in Figure 2.83. The microcapsules in self-healing polymers not only store the healing agent in normal conditions of use but also provide a mechanical trigger for the self-healing process when damage occurs in the host material and the capsules rupture. For this to happen, the microcapsules must possess long shelf life, excellent bonding to the host material, and sufﬁcient strength to remain intact during processing of the host polymer and yet rupture when the polymer is damaged. In the example cited (system 2), these Temperature Crack owing to mechanical action Neutralizing agents [NH3, (NH4)2CO3] Aging deterioration Light 1 2 3 4 5 Surface wettability Azobenzene derivatives Frequency dependent dielectric measurement Carboxylated copolymer Dissolution in water Changes in molecular mobility of ions and dipoles U/F microcapsules containing DCPD Poly(N-isopropyl acrylamide) Sensor Capsule rupture and healing Hydrophilicity, hydrophobicity Response Source: Feng, W., Patel, S. H., Young, M-Y., Zunino III, J. L., and Xanthos, M., 2007. Adv. Polym. Tech., 26, 1. Stimulus System TABLE 2.11 Polymeric Smart Coatings Based on Noncolor Response Azobenzene cis–trans photoisomerization Dielectric sensing of changes in mobility at molecular level Coating for photo-control of cell adhesion Monitor durability of coatings Temporary protective coating Self-healing of cracks Catalyzed polymerization of DCPD Neutralization of carboxylic groups Coating on biomaterials Application Transition from hydrophobic (protein retentive) to hydrophilic (nonprotein adsorptive) below LCST (31°C) Sensing Mechanism 284 Plastics Technology Handbook 285 Fabrication Processes Catalyst Microcapsule (a) Crack (b) (c) FIGURE 2.83 Autonomic healing of a polymer composite containing an embedded, microencapsulated healing agent and a catalytic chemical trigger (see text for functioning mechanism). (After White, S. R., Sottos, N. R., Geubelle, P. H., Moore, J. S., Kessler, M. R., Sriram, S. R., Brown, E. N., and Viswanathan, S. 2001. Nature, 409, 794.) combined characteristics are achieved by microencapsulation of dicyclopentadiene (DCPD) healing agent utilizing acid-catalyzed in situ polymerization of urea with formaldehyde in an oil-in-water emulsion to form capsule wall [87,88]. The addition of the resulting free-ﬂowing microcapsules (50–200 mm) to an epoxy matrix also provides signiﬁcant toughening to the composite system in addition to self-healing property. For a structural polymeric material (such as epoxy) that has the ability to autonomically heal cracks, the material incorporates a microencapsulated healing agent and a catalytic chemical trigger within the polymer matrix (Figure 2.83a). The microcapsule shell provides a protective barrier between the catalyst and DCPD to prevent polymerization during the preparation of the composite. An approaching crack ruptures embedded microcapsules, releasing the healing agent into the crack plane through capillary action (Figure 2.83b). Polymerization of the healing agent is triggered by contact with the embedded catalyst, resulting in bonding of the crack faces (Figure 2.83c). The damage-induced triggering mechanism provides site-speciﬁc autonomic control of repair [87]. On being released into the crack plane, DCPD undergoes ring-opening metathesis polymerization (ROMP) in the presence of a transition metal catalyst (Grubbs’ catalyst) at room temperature to yield a tough and highly cross-linked polymer network in several minutes. The use of living (i.e., having unterminated chain ends) polymerization catalysts, as above, also enables multiple healing events. 2.21.1.5 Electrografted Coatings Achieving permanent adhesion between very dissimilar materials such as polymers and metals is a very challenging task. However, synthetic polymers can now be chemisorbed on a variety of conducting surfaces by cathodic electrografting of acrylic monomers [89]. Electrografting is merely implemented in a standard electrochemical assembly, as shown in Figure 2.84. The conducting surface to be coated is cathodically polarized in an oxygen- and water-free solution of the monomer and a conducting salt in an organic solvent under a dry and inert atmosphere. Upon application of the appropriate cathodic 286 Plastics Technology Handbook ΔE O CH2 CH O n O R O Counter electrode (anode) (a) Substrate (cathode) R Reference (b) FIGURE 2.84 (a) Electrochemical setup used for electrografting of acrylates. (b) Electrografted acrylate on the cathodic substrate. (After Gabriel, S., Jérôme, R., and Jérôme, C. 2010. Prog. Polym. Sci., 35, 113.) potential to the working electrode for a few seconds, a strongly adhering thin polymer ﬁlm is deposited on the cathode surface, which is accordingly passivated. A proposed mechanism [89] consists of the transfer of one electron from the substrate (cathode) to the monomer, with the bonding of the radical-anion species to the substrate. Chain initiation is thus an electrochemical event, in contrast to the chain propagation that proceeds through the repeated addition of the monomer to the chemisorbed anionic species, for example, S–CH2CH(COOR)− for S/CH2=CH(COOR) system, where S is a conductive substrate. Common metals such as Fe, Ni, and Cu have been successfully coated by cathodic electrografting. These must be pretreated, however, for removing any oxide formed at the surface. Semiconductors, such as n- and p-doped silicon, chemically pretreated with hydroﬂuoric acid to remove their oxide layer, are well suited to electrografting that results in chemisorption of polymer chains through quite stable Si−C bonds. Electrografting has also been performed on electrically conducting powders and (nano)particles, which are placed in a zinc container immersed in (and ﬁlled by) the electrochemical bath and used as a cathode. (Zinc fulﬁlls the critical requirement of the process, i.e., the inability of the container to be electrografted.) While the electrografting process, as initially developed, commonly used readily available monomers, namely, acrylonitrile and (meth)acrylates, new monomers were later synthesized that contained in the ester a large variety of substituents, which do not interfere with electro-polymerization. These substituents may be an initiator of radical or ring-opening polymerization, a monomer that is anodically polymerizable, a preformed polymer, and so on. Since electrooxidation of aromatic monomers, such as pyrrole and thiophene (or their derivatives), which is a classical technique to prepare intrinsically conducting polymers, has the drawback of poor adhesion of the ﬁlm to the substrate, to tackle this problem, cathodic electrografting has been combined with anodic synthesis of conjugated polymers. In a ﬁrst strategy, dual monomers, that is, acrylates that contain a pyrrole or thiophene unit in the ester group, for example, N-(2-acryloyloxyethyl)pyrrole (PyA) [formula: CH2=CHCO2CH2CH2N(C4H4)] and 3-(2-acryloyloxyethyl)thiophene (ThiA) [formula: CH2=CHCO2CH2CH2(C4H3S)], have been synthesized. The parent polyacrylates are easily chemisorbed under cathodic polarization, followed by a polarization inversion that results in the polymerization of the pyrrole (Figure 2.85) or thiophene substituents, thus making the electrically conducting ﬁlm strongly adhering. Pyrrole or thiophene can also be added to the electrochemical bath with the purpose of increasing the thickness of the electrically conducting ﬁlm. 287 Fabrication Processes n O Ec O N O O n Ea N N O N O H N m (PyA) FIGURE 2.85 Electrografting of N-(2-acryloyloxyethyl)pyrrole (PyA) followed by electrooxidation of pyrrole. (After Gabriel, S., Jérôme, R., and Jérôme, C. 2010. Prog. Polym. Sci., 35, 113.) As an alternative, PyA or ThiA can be polymerized by conventional or controlled radical polymerization and—before electrooxidation of pyrrole or thiophene—the polyacrylates (polyPyA or polyThiA) so obtained can be dissolved in the electrochemical bath or cast onto an electrode that has been previously modiﬁed by electrografting of PyA or ThiA. Electroactive ﬁlms of a controlled thickness (up to several microns) can thus be obtained, depending on the anodic polarization time. This process can be instrumental in improving the performance of various high-tech devices, such as light-emitting diodes, anticorrosion coatings, electrochromic windows, and electrochemical sensors that depend on the stability and adherence of a conjugated polymer coating on a conducting solid surface [89]. A very large variety of organic ﬁlms have been successfully deposited with strong adhesion onto various substrates by cathodic electrografting with the possibility of having a second layer of chains grown from the ﬁrst one under controlled/living conditions. In this case, the electrografted ﬁlm is thin and plays the role of an anchoring “primer” [89], whereas the covalently bonded top layer can be designed to have desired molecular structures, composition, and thickness. The unique ﬂexibility of this strategy of coating and multilayered ﬁlm deposition has generated high value-added practical applications, such as biocompatibilization of medical implants, bactericidal coatings, surface functionalization of electronic devices, and stimuli-responsive coatings (smart coatings). The evolving shift from micro- to nanoelectronics needs the ability to perform very local surface modiﬁcation, molecular handling of polymer chains, and sensing at the nanometric scale with respect to spatial detection and spatial deposition. Electrografting can prove to be very instrumental for successfully facing such challenges generated by the emerging nanotechnologies. 2.21.2 Powder Molding of Thermoplastics 2.21.2.1 Static (Sinter) Molding The process is often used with polyethylene and is limited to making open-ended containers. The mold which represents the exterior shape of the product is ﬁlled with powder. The ﬁlled mold is heated in an oven, causing the powder to melt and thus creating a wall of plastics on the inner surface of the mold. After a speciﬁc time to build the required wall thickness, the excess powder is dumped from the mold and the mold is returned to the oven to smooth the inner wall. The mold is then cooled, and the product is removed. The product is strain free, unlike pressure-molded products. In polyethylene this is especially signiﬁcant if the product is used to contain oxidizing acids. 2.21.2.2 Rotational Molding Rotational molding (popularly known as rotomolding) is best suited for large, hollow products, requiring complicated curves, uniform wall thickness, a good ﬁnish, and stress-free strength. It has been used for a variety of products such as car and truck body components (including an entire car body), industrial 288 Plastics Technology Handbook containers, telephone booths, portable outhouses, garbage cans, ice buckets, appliance housing, light globes, toys, and boat hulls. The process is applicable to most thermoplastics and has also been adapted for possible use with thermosets. Essentially four steps are involved in rotational molding: loading, melting and shaping, cooling, and unloading. In the loading stage a predetermined weight of powdered plastic is charged into a hollow mold. The mold halves are closed, and the loaded mold is moved into a hot oven where it is caused to simultaneously rotate in two perpendicular planes. A 4:1 ratio of rotation speeds on minor and major axes is generally used for symmetrically shaped objects, but wide variability of ratios is necessary for objects having complicated conﬁgurations. In most units the heating is done by air or by a liquid of high speciﬁc heat. The temperature in the oven may range from 500°F–900°F (260°C–482°C), depending on the material and the product. As the molds continue to rotate in the oven, the polymer melts and forms a homogenous layer of molten plastic, distributed evenly on the mold cavity walls through gravitational force (centrifugal force is not a factor). The molds are then moved, while still rotating, from the oven into the cooling chamber, where the molds and contents are cooled by forced cold air, water fog, or water spray. Finally, the molds are opened, and the parts removed. (Rotational molding of plastisol, described earlier, is similar to that described here. However, in this case the plastic is charged in the form of liquid dispersion, which gels in the cavity of the rotating mold during the heating cycle in the oven.) The most common rotational molding machine in use today is the carousel-type machine. There are three-arm machines consisting of three cantilevered arms or mold spindles extending from a rotating hub in the center of the unit. In operation, individual arms are simultaneously involved in different phases (loading, heating, and cooling) in three stations so that no arms are idle at any time (Figure 2.86). Modern rotational-molding machines enable large parts to be molded (e.g., 500-lb car bodies and 500-gal industrial containers). For producing small parts an arm of a carousel-type machine may hold as many as 96 cavities. 2.21.2.3 Centrifugal Casting Centrifugal casting is generally used for making large tube forms. It consists of rotating a heated tube mold which is charged uniformly with powdered thermoplastic along its length. When a tubular molten layer of the desired thickness builds up on the mold surface, the heat source is removed and the mold is cooled. The mold, however, continues to rotate during cooling, thus maintaining uniform wall thickness of the tube. Upon completion of the cooling cycle, the plastic tube, which has shrunk away from the mold surface, is removed, and the process is repeated. Usual tube sizes molded by the process range from 6–30 in. in diameter and up to 96 in. in length. 2.21.3 Adhesive Bonding of Plastics Adhestives are widely used for joining and assembling of plastics by virtue of low cost and adaptability to high-speed production. They can be subdivided into solvent or dope cements, which are suitable for most thermoplastics (not thermosets), and monomeric or polymerizable cements which can be used for most thermoplastics and thermosets. Solvent cements and dope cements function by attacking the surfaces of the adherends so that they soften and, on evaporation of the solvent, will join together. The dope cements, or bodied cements, differ from the straight solvents in that they contain, in solution, quantity of the same plastic which is being bonded. On drying, these cements leave a ﬁlm of plastic that contributes to the bond between the surfaces to be joined. Monomeric or polymerizable cements consist of a reactive monomer, identical with or compatible with the plastic to be bonded, together with a suitable catalyst system and accelerator. The mixture polymerizes either at room temperature or at a temperature below the softening point of the thermoplastic being joined. In order to hasten the bonding and to reduce shrinkage, some amount of the solid plastic may also 289 Fabrication Processes Charging area Powder Mold halves Oven Cooling chamber FIGURE 2.86 Basics of continuous-type three-arm rotational molding machine. be initially dissolved in the monomer. Adhesives of this type may be of an entirely different chemical type than the plastic being joined. A typical example is the liquid mixture of epoxy resins and hardener which by virtue of the chemical reactivity and hydrogen bonding available from the epoxy adhesive provide excellent adhesion to many materials. In joining with solvents or adhesives, it is very important that the surfaces of the joint be clean and well matched since poor contact of mating surfaces can cause many troubles. The problem of getting proper contact is aggravated by shrinkage, warpage, ﬂash, marks from ejector pins, and nonﬂat surfaces. 2.21.3.1 Solvent Cementing Solvent cementing or solvent welding basically involves softening the bonding area with a solvent or a solvent containing small quantities of the parent plastic, referred to as dope or bodied cement, generally containing less than 15% resin. The solvent or cement must be of such composition that it will dry completely without blushing. Light pressures must be applied to the cemented joint until it has hardened to the extent that there is no movement when released. Structural bonds of up to 100% of the strength of the parent material are possible with this type of bonding. Table 2.12 gives a list of typical solvents selected as useful for cementing of plastics. A key to selection of solvents in this table is how fast they evaporate: a fast-evaporating solvent may not last long enough for some assemblies, while evaporation that is too slow could hold up production. It may be noted that solvent bonding of dissimilar materials is possible were the materials can be bonded with the same solvents (see Table 2.12). The solvent is usually applied by the soak method in which pieces are immersed in the solvent, softened, removed, quickly brought together, and held under light pressure for some time. Areas adjoining the joint area should be masked to prevent them from being etched. For some applications where the surfaces to be cemented ﬁt very closely, it is possible to introduce the cementing solvent by brush, eyedropper, or hypodermic needle into the edges of the joint. The solvent is allowed to spread to the rest of the joint by capillary action. 2.21.3.2 Adhesive Bonding Adhesive bonding is a process in which the adhesive acts as an agent to hold two substrates together (as opposed to solvent cementing where the parent materials actually become an integral part of the bond) and the adhesion is chemical. 290 Plastics Technology Handbook TABLE 2.12 Typical Solvents for Solvent Cementing of Plastics Plastics Solvent ABS Methylene chloride, tetrahydrofuran, methyl ethyl ketone, methyl isobutyl ketone Acetate Methylene chloride, chloroform, acetone, ethyl acetate, methyl ethyl ketone Acrylic Cellulosics Methylene chloride, ethylene dichloride Acetone, methyl ethyl ketone Nylon Aqueous phenol, solutions of resorcinol in alcohol, solutions of calcium chloride in alcohol PPO PVC Methylene chloride, chloroform, ethylene dichloride, trichloroethylene Cyclohexane, tetrahydrofuran, dichlorobenzene Polycarbonate Methylene chloride, ethylene dichloride Polystyrene Polysulfone Methylene chloride, ethylene dichloride, trichloroethylene, methyl ethyl ketone, xylene Methylene chloride Table 2.13 gives a list of various adhesives and typical applications in bonding of plastics [90]. This table is not complete, but it does give a general idea of what types of adhesives are used and where. It should be remembered, however, that thousands and thousands of variations of standard adhesives are available off the shelf. The computer may shape up as an excellent selection aid for adhesives. Selection is made according to the combination of properties desired, tack time, strength, method of application, and economics (performance/cost ratio). The form that the adhesive takes (liquids, mastics, hot melts, etc.) can have a bearing on how and where they are used. The anaerobics, which can give some very high bond strength and are usable with all materials except polyethylene and ﬂuorocarbons, are dispensed by the drop. A thin application of the anaerobics is said to give better bond strength than a thick application. The more viscous, mastic-type cements include some of the epoxies, urethanes and silicones. Epoxies adhere well to both thermosets and thermoplastics. But epoxies are not recommended for most polyoleﬁn bonding. Urethane adhesives have made inroads into ﬂexible packaging, the shoe industry, and vinyl bonding. Polyester-based polyurethanes are often preferred over polyether systems because of their higher cohesive and adhesive properties. Silicones are especially recommended where both bonding and sealing are desired. Hot melts—100% solids adhesives that are heated to produce a workable material—are based on polyethylene, saturated polyester or polyamide in chunk, granule, pellet, rope, or slug form. Saturated polyesters are the primary hot melts for plastics; the polyethylenes are largely used in packaging; polyamides are used most widely in the shoe industry. Application speeds of hot melts are high and it pays to consider hot melts if production requirements are correspondingly high. Pressure sensitives are contact-bond adhesives. Usually rubber based, they provide a low-strength, permanently tacky bond. They have a number of consumer applications (e.g., cellophane tape), but they are also used in industrial applications where a permanent bond is not desirable or where a strong bond may not be necessary. The adhesive itself is applied rapidly by spray. Assembly is merely a matter of pressing the parts together. Film adhesives require an outside means such as heat, water, or solvent to reactivate them to a tacky state. Among the ﬁlm types are some hot melts, epoxies, phenolics, elastomers, and polyamides. Film adhesives can be die cut into complicated shapes to ensure precision bonding of unusual shapes. Applications for this type of adhesive include bonding plastic bezels onto automobiles, attaching trim to both interiors and exteriors, and attaching nameplates on luggage. While most plastics bond without trouble once the proper adhesive has been selected, a few, notably polyoleﬁns, ﬂuorocarbons and acetals, require special treatment prior to bonding. Untreated polyoleﬁns adhere to very few substrates (a reason that polyethylene is such a popular material for packaging adhesives). Their treating methods include corona discharge, ﬂame treating (especially for large, irregularshaped articles), and surface oxidation by dipping the articles in a solution of potassium dichromate and 2,3 1–4 – – MF 14 – – 10,14 – – – 4,10 4,8,10 10 – 10 10 – – – – 4,8,10 10 – – – 10 10 – – – 10 10 – – – 2,9,10,13 – 10 – 2,3,7,9 – – – 4,7 10 – – – – – 3,4,6,13 3,4,6,13 – 10 10 3,4,6,10 10 10 3,4 3,4,7 – 3,10,13 2,3,7,10,13,14 PVC – – – – – – 2,3,7,9 – 2,3,7,9 8,10 10 3,10 2 3 2,3 2,10 10 9,10 3,4,13 – 10 10 9,10 10 9,10 3 1–4 – 3 2,3 Nylon – – – – – – – – 10,12 – – – – 10 – 10 – 2 2 – 2 2,12 PE – – – – – – 15 – – – – 4,10 – 10,13 4,13 13 10,13 10 PC – – – – – 10 – – – – 5 4,12,13 12,13 12 PS – – – – – 4,10 – – 10,12 – – 4,5,10,12,13 – – – – – – – – – 1 1 1 1 1 PP – 3,8,10,11 – 2,3,10,12,13 – – – – – 8,10 – – – – – 2 2,3,14 – 2,3 3,14 MF – – – – 10 – 8,10 – – 10 – – 2 1,4 – 2 4 Polyesters – – 2–4,10,12,14 – – 3,10,12,13 – – 4,10,14 – 4,7 – 8,10,14 10 – 4,8,10 – 10 2 2,3 – 2 2 PF 3,4,10,13 – – – – – – – – – – 10 4,10 – – 3 4,13 4,13 13 3,4 PU Source: Adapted from O'Rinda Trauernicht, J. 1970. Plastics Technology, Reinhold Publishing, New York. Note: Elastomeric: 1, Natural rubber. 2, Neoprene. 3, Nitrile. 4, Urethane. 5, Styrene-butadiene. Thermoplastic: 6, Poly(vinyl acetate). 7, Polyamide. Thermosetting: 8, Phenolformaldehyde. 9, Resorcinol, Phenol-resorcinol/formaldehyde. 10, Epoxy. 11, urea-formaldehyde. Resin: 12, Phenolic-poly(vinyl butyral). 13, Polyester. Other: 14, Cyanoacrylate. 15, Solvent. PU PF Polyester – – – – – – – – PP PS 10 PE 3,4,6,10 – 2,3,7,9 – – 15 10 – 10 10 8,10 8,10 – 10 PVC PC – – Nylon Cellulosic Fluorocarbons 2,4,6,10 2,4,6,10 – Ceramics Acrylic 3 10 10 Rubber – 2,3 2,3 10,14 2,4,6,10 2,4,6,10 2,4,6,10 3,10 10 3,10 10 – 10 Acrylic Cellulosics Fluorocarbons ABS Acetal 3,10 10 Paper Wood Acetal Metals ABS TABLE 2.13 Typical Adhesives for Bonding Plastics Fabrication Processes 291 292 Plastics Technology Handbook sulfuric acid. Fluorocarbons can be prepared for bonding by cleaning with a solvent such as acetone and then treating with a special etching solution. For acetals, one pretreatment method involves immersing articles in a special solution composed mainly of perchloroethylene, drying at 120°C (250°F), rinsing, and then air drying. 2.21.3.3 Joining of Speciﬁc Plastics The same basic handling techniques apply to almost all thermoplastic materials. In the following section, however, a few thermoplastics will be treated separately, with mention of the speciﬁc cements most suitable for each. The section on acrylics should be read in connection with the cementing of any other thermoplastics. 2.21.3.3.1 Cast Acrylic Sheeting Articles of considerable size and complexity can be fabricated from methyl methacrylate plastics by joining sections together by solvent welding. The technique described here applies to cast sheeting. With articles made from methyl methacrylate molding powders or extruded rod, tubing or other shapes, joining is generally not as satisfactory as with cast sheeting. With care and practice, the transparency of acrylic resin can be retained in joints with the formation of a complete union of the two surfaces brought into contact. Usually one of the two surfaces to be joined is soaked in the cementing solvent until a soft, swollen layer (cushion) has been formed upon it. This soft surface is then pressed against the surface to be attached and held in contact with it so that the excess solvent contained in the soaked area softens it also. For some purposes, it may be desirable to dissolve clean savings of methyl methacrylate resin in the solvent in order to raise its viscosity so that it can be handled like glue. The most universally applicable type of solvent cement is the polymerizable type, comprising a mixture of solvent and catalyzed monomer. These are mobile liquids, volatile, rapid in action, and capable of yielding strong sound bonds. An example of these is a 40–60 mixture of methyl methacrylate monomer and methylene chloride. Before using this cement. 1.2 grams of benzoyl peroxide per pint of solvent, should be added. Heat treatment or annealing of joints made with solvent cements is highly desirable because it greatly increases the strength of the joint. 2.21.3.3.2 Cellulosics The cements used with cellulosic plastics are of two types: (1) solvent type, consisting only of a solvent or a mixture of solvents; (2) dope type, consisting of a solution of the cellulosic plastic in a solvent or mixture of solvents. Acetone and mixture of acetone and methyl “cellosolve” are commonly used as solvent cements for cellulose acetate. Acetone is a strong solvent for the plastic, but evaporates rapidly. The addition of methyl “cellosolve” retards the evaporation, prevents blushing, and permits more time for handling the parts after application of the cement. A cement of the dope type leaves upon drying a ﬁlm of plastic that forms the bond between the surfaces to be joined. These cements are generally used when an imperfect of the parts requires ﬁlling. A typical composition of the dope-type cement for cellulose acetate is Parts by Weight Cellulose acetate Acetone 130 400 Methyl “cellosolve” 150 Methyl “cellosolve” acetate 50 Fabrication Processes 293 Other cellulosics, cellulose acetate butyrate and propionate are cemented in accordance with the technique described for cellulose acetate. In the case of dope cements, the plastic to be dissolved in solvents is cellulose propionate. Similarly for ethyl cellulose plastic, the strongest bonds are made by solvents or by solvents bodied with ethyl cellulose plastic. 2.21.3.3.3 Nylon The recommended cements for nylon-to-nylon bonding are generally solvents, such as aqueous phenol, solutions of resorcinol in alcohol, and solutions of calcium chloride in alcohol, sometimes “bodied” by the inclusion of nylon in small percentages. Aqueous phenol containing 10–15% water is the most generally used cement for bonding nylon to itself. The bond achieved by use of this cement is water resistant, ﬂexible, and has strength approaching that of the nylon. Calcium–chloride–ethanol solution bodied with nylon is recommended for nylon-to-nylon joints where there is possibility of contact with foods or where phenol or resorcinol would be otherwise objectionable. For bonding nylon to metal and other materials, various commercial adhesives, especially those based on phenol-formaldehyde and epoxy resins, are sometimes used. Epoxy adhesives (in two-part systems), for example, have been used to produce satisfactory joints between nylon and metal, wood, glass and leather. 2.21.3.3.4 Polycarbonate Solvent cementing of parts of polycarbonate may be effected by the use of a variety of solvents or light solutions of polycarbonate in solvents. Methylene chloride, a 1–5% solution of polycarbonate in methylene chloride, and a mixture of methylene chloride and ethylene dichloride (with a maximum of 40% ethylene dichloride) are commonly recommended. Solvent should be applied to only one of the bonding surfaces while the other half remains dry and ready in the clamping ﬁxture. As soon as the two parts have been put together, pressure should be applied immediately. Pressure between 200 and 600 psi is suggested for best results. Holding time in the pressure ﬁxture is approximately 1–5 min, depending on the size of the bonding area. For bonding molded parts of polycarbonate to other plastics, glass, wood, aluminum, brass, steel, and other materials, a wide variety of adhesives can be used. Generally, the best results are obtained with solventless materials, such as epoxies and urethanes. 2.21.3.3.5 Polyethylene The good solvent resistance of polyethylene and other oleﬁns precludes the use of solvent-type cements. Several commercial rubber-type adhesives produce moderate adhesion with polyethylene that has been surface treated. One technique for surface treatment is to dip polyethylene in a chromic acid bath (made up of concentrated sulfuric acid 150 parts by weight, water 12 parts, and potassium dichromate 7.5 parts) for about 30 sec at 70°C. The parts are rinsed with water after this treatment. Still another effective surface treatment for producing cementable surfaces on polyethylene is electrical discharge. The open oxidizing ﬂame method is also used extensively for this purpose. 2.21.3.3.6 Polystyrene Complex assemblies of polystyrene, usually molded in section, may be joined by means of solvents and adhesives. Polystyrene is soluble in a wide variety of solvents. According to their relative volatilities they may be divided into three groups—fast drying, medium drying, and slow drying. Methylene chloride, ethylene dichloride, and trichloroethylene are some of the fast-drying solvents that produce strong joints. However, they are unsatisfactory for transparent articles of polystyrene because they cause rapid crazing. “Medium-drying” solvents such as toluene, perchloroethylene, and ethyl benzene that have higher boiling temperature are less apt to cause crazing. 294 Plastics Technology Handbook High-boiling or “slow-drying” solvents such as amylbenzol and 2-ethylnaphthalene often require excessive time for development of sufﬁcient bond strength, but they will not cause crazing to appear so quickly. Up to 65% of a fast- or medium-drying solvent may be added to a slow-drying solvent to speed up the development of initial tack without greatly reducing the time before crazing appears. A bodied, or more viscous, solvent may be required by certain joint designs and for producing airtight or watertight seals. These are made by dissolving usually 5%–15% of polystyrene by weight in a solvent. Solvent-based contact cements provide the strongest bond between polystyrene and wood. These adhesives all have a neoprene (polychloroprene) base and a ketonic-aromatic solvent system. 2.21.3.3.7 Poly(Vinyl Chloride) and Copolymers On account of the relative insolubility of PVC and the markedly increased effect of solvents with the increasing content of vinyl acetate in the copolymer resins, there exists among the vinyl chloride-acetate copolymer system a great diversity of composition and of ability to be cemented by solvents. The copolymer resins are most rapidly dissolved by the ketone solvents, such as acetone, methyl ethyl ketone, methyl isobutyl ketone, and cyclohexanone. Propylene oxide also is a very useful solvent in hastening solution of copolymer resins, especially those of high molecular weight, and of straight PVC. This solvent penetrates the resins very rapidly and, in amounts up to about 20–25%, improves the “bite” into the resin. The chlorinated hydrocarbons also are excellent solvents for the vinyl chloride-vinyl acetate copolymer resins, and are suitable for use in cements. 2.21.3.3.8 Thermosetting Plastics Adhesive bonding is, for various reasons, the logical method of fastening or joining cross-linked and reinforced thermoset plastics of themselves, or to other materials. Since most thermoset plastics are quite resistant to solvents and heat, heat-curing solvent-dispersed adhesives may be used. Such adhesives consist of reactive or thermosetting resins (e.g., phenolics, epoxies, urea-formaldehydes, alkyds, and combinations of these), together with compatible ﬁlm-formers such as elastomers or vinyl-aldehyde condensation resins. Isocyanates are frequently added as modiﬁers to improve speciﬁc adhesion to surfaces that are difﬁcult to bond. These adhesives may be applied not only in solvent-dispersed form, but also in the form of ﬁlm, either unsupported, or supported on fabric, glass-mat, and so forth. A great many of outstanding adhesive formulations are based on epoxy resins. A broad spectrum of adhesive formulations with a wide range of available properties have resulted from the use of polymeric hardeners such as polyamides and polyamines, phenolics, isocyanates, alkyds, and combinations of amines with polysulﬁde elastomers, and the “alloying” of the epoxy with compatible polymeric ﬁlmformers, such as poly(vinyl acetate) and certain elastomers. In cemented assemblies of thermoset plastics and metals, where structural strength is generally desired, the adhesive must be more rigid than those used for bonding plastic to plastic, i.e., one with modulus, strength, and coefﬁcient of thermal expansion between those of the plastic and the metal. In many cases, such adhesives are stronger than the plastic itself. 2.21.4 Welding Often it is necessary to join two or more components of plastics to produce a particular setup or to repair a broken part. For some thermoplastics solvent welding is applicable. The process uses solvents which dissolve the plastic to provide molecular interlocking and then evaporate. Normally it requires closeﬁtting joints. The more common method of joining plastics, however, is to use heat, with or without pressure. Various heat welding processes are available. Those processes in common commercial use are described here. 295 Fabrication Processes 2.21.4.1 Hot-Gas Welding Hot-gas welding, which bears a superﬁcial resemblance to welding of metals with an oxyacetylene ﬂame, is particularly useful for joining thermoplastic sheets in the fabrication of chemical plant items, such as tanks and ducting. The sheets to be joined are cleaned, beveled, and placed side by side so that the two beveled edges form a V-shaped channel. The tip of a ﬁller rod (of the same plastic) is placed in the channel, and both it and the adjacent area of the sheets are heated with a hot-gas stream (200–400°C) directed from an electrically heated hot-gas nozzle (Figure 2.87a), which melts the plastics. The plastics then fuse and unite the two sheets. The hot gas may be air in PVC welding, but for polyethylene an inert gas such as nitrogen must be used to prevent oxidation of the plastics during welding. 2.21.4.2 Fusion Welding Fusion or hot-tool welding is accomplished with an electrically heated hot plate or a heated tool (usually of metal), which is used to bring the two plastic surfaces to be joined to the required temperature. The polyfusion process for joining plastic pipes by means of injection-molded couplings is an example of this type of welding. The tool for this process is so shaped that one side of it ﬁts over the pipe while the other side ﬁts into the coupling. The tool is heated and used to soften the outside wall of the pipe and the inside wall of the coupling. The pipe and coupling are ﬁrmly pressed together and held until the joint cools to achieve the maximum strength of the weld. The tool is chrome plated to prevent the plastic sticking to its surfaces. 2.21.4.3 Friction Welding In friction or spin-welding of thermoplastics, one of the two pieces to be jointed is ﬁxed in the chuck of a modiﬁed lathe and rotated at high speed while the other piece is held against it until frictional heat causes the polymer to ﬂow. The chuck is stopped, and the two pieces are allowed to cool under pressure. The process is limited to objects having a circular conﬁguration. Typical examples are dual-colored knobs, molded hemispheres, and injection-molded bottle halves. 2.21.4.4 High-Frequency Welding Dielectric or high-frequency welding can be used for joining those thermoplastics which have high dielectric-loss characteristics, including cellulose acetate, ABS, and PVC. Obviously, polyethylene, polypropylene, and polystyrene cannot be welded by this method. The device used for high-frequency welding is essentially a radio transmitter operated at frequencies between 27 and 40 MHz. The energy obtained from the transmitter is directed to electrodes of the welding apparatus. The high-frequency ﬁeld causes the molecules in the plastic to vibrate and rub against each other very fast, which creates frictional heat sufﬁcient to melt the interfaces and produce a weld. Filler rod pressed into weld bed Welding gun nozzle Ultrasonic tool Welded joint (a) FIGURE 2.87 Shaded surface on rod and material must be molten Fixed anvil (b) Welding of plastics: (a) hot-gas welding; (b) ultrasonic contact welding. 296 Plastics Technology Handbook 2.21.4.5 Ultrasonic Welding In ultrasonic welding the molecules of the plastic to be welded are sufﬁciently disturbed by the application of ultrahigh-frequency mechanical energy to create frictional heat, thereby causing the plastics to melt and join quickly and ﬁrmly. The machinery for ultrasonic welding consists of an electronic device which generates electrical energy at 20/50 kHz/sec and a transducer (either magnetostrictive or piezoelectric) to convert the electrical energy to mechanical energy. In the contact-welding method (Figure 2.87b) the ultrasonic force from the transducer is transmitted to the objects (to be welded) through a tool or “horn,” generally made of titanium. The amplitude of the motion of the horn is from 0.0005 to 0.005 in. (0.013–0.13 mm) depending on the design. The method is generally used for welding thin or less rigid thermoplastics, such as ﬁlms, or sheets of polyethylene, plasticized PVC, and others having low stiffness. 2.21.5 Joining Polymer–Metal Hybrids Polymer–metal hybrid (PMH) structures are increasingly used in industrial components, especially because of associated weight savings. Compared to polymers and polymeric composites, PMH structures are more difﬁcult to join by traditional joining methods, mostly because of the strong dissimilar physical– chemical features of the joining partners. However, constant efforts on developing improved alternative joining techniques for these hybrid structures, such as FricRiveting and injection over molding, have contributed to increasing use of such structures in industrial applications. FricRiveting is a recently patented technique for joining PMH structures. In the simplest process variant, a rotating cylindrical metallic rivet is inserted in a thermoplastic base plate. As the high rotation speed and pressure increase friction, heat is generated and it builds up owing to low polymer thermal conductivity. The local increase in temperature induces the formation of a softened/molten polymeric layer around the tip of the rotating rivet. When the desired penetration depth is achieved, rotation is decelerated and the forging pressure is applied so that the plasticized rivet’s tip is deformed into a paraboloidal shape by the opposite reactive forces of the colder polymer and becomes anchored to the polymeric base after cooling (Figure 2.88). Several other joint conﬁgurations that allow fabrication of hermetic sealed joints by FricRiveting are seen in Figure 2.89. Metal–polymer overlap joints can also be produced by creating friction-welded seams between the metallic part and the rivet (Figures 2.89a and 2.89b), in addition to the mechanical anchoring Metallic rivet Plasticized material Plastic (a) (b) (c) Plastic FIGURE 2.88 Schematic of friction-riveting technique. (a) Positioning of the components; (b) insertion of the threaded rivet (e.g., aluminum 6060-T6) in the plastic component; (c) forging of the plasticized tip of the rivet, a joint being formed after consolidation of the molten plastic. (After Amancio-Filho, S. T. and dos Santos, J. F. 2009. Polym. Eng. Sci., 49, 1461.) 297 Fabrication Processes Weld seam Metal Metal Plastic Plastic Metal Metal (a) Weld seam (b) Plastic (c) FIGURE 2.89 Examples of polymer–metal joints made by friction riveting: (a) Lap joint of plastic and metallic (e.g., polyetherimide/aluminum 2024-T351) components; (b) hermetic sealed sandwich joint; and (c) hermetic sealed lap joint of metallic and plastic components. Polyetherimide/aluminum 2024-T351 joints with plain cylindrical rivets have displayed only very little amount of thermally degraded polymer and the mechanical performance of the joint has been found to be as high as 95% of the tensile strength of the metallic partner. (After Amancio-Filho, S. T. and dos Santos, J. F. 2009. Polym. Eng. Sci., 49, 1461.) effect of the deformed tip of the rivet into the polymeric part. In the process, only a very little amount of thermomechanical degradation of the polymer is observed, although high temperatures within 50%–95% of the metallic rivet melting temperature are observed [91]. The accompanying thermomechanical ﬂaws are small and do not strongly inﬂuence the overall mechanical strength of these joints. The main advantages of FricRiveting, as compared to traditional joining techniques, are as follows: (a) little or no surface cleaning or preparation of the joining partners is required; (b) only single side accessibility is required; (c) hermetic sealed joints can be obtained by choosing adequate joint geometry; (d) a wide range of materials can be joined; (e) joints have good mechanical strength and performance; and (f) robotic applications are possible. On the other hand, FricRiveting has a few limitations: (a) allows only spot-like joints; (b) not generally applicable to thermoset polymers; and (c) joints are not reopenable. There are three main PMH technologies currently being employed by the automotive original equipment manufacturers and their tier one and tier two suppliers: (a) injection over-molding technologies, (b) metal over-molding technologies combined with secondary joining operations, and (c) adhesively bonded PMH technologies [91]. These are brieﬂy described below. In injection over-molding technologies, the polymer (typically glass ﬁber reinforced nylon) is injected around a metal stamping proﬁle that has been placed in an injection mold so that the plastic wraps around the edges of the sheet metal and/or through carefully designed extruded holes or buttons. No secondary operations are required in the process and the drawing oils/greases do not need to be removed from the metal stamping. In metal over-molding technologies, a steel stamping is placed in an injection mold in order to coat its underside typically with a thin layer of reinforced nylon. The polymer-coated surface of the metal insert is then ultrasonically welded, in a secondary operation, to an injection-molded nylon subcomponent. In this process, a closed-section structure with continuous bond lines is produced that affords a high loadbearing capability. The hollow core of the part permits functional integration like cable housings and air or water channels. In adhesively bonded technologies, glass-ﬁber-reinforced polypropylene is typically joined to a metal stamping using an adhesive that is applied by high-speed robots. Adhesive bonding minimizes stress 298 Plastics Technology Handbook concentrations and acts as a buffer to absorb contact stresses between the metal and polymer components of the joint. Moreover, the joining process enables creation of closed-section structures that offer high load-bearing capabilities and permit functional integration like direct mounting of air bags in instrumentpanel beams or incorporation of air or water channels inside door modules. The aforesaid PMH technologies, though used widely in the manufacturing of various nonstructural and load-bearing automotive body-in-white (BIW) components, are known to display some signiﬁcant shortcomings. [Note: BIW refers to the stage in automotive manufacturing in which a car body’s sheet metal components have been welded together, but before doors, hoods, motor, trim, etc. have been added and before painting has been done.] For example, to achieve polymer-to-metal interlocking, the injection over-molding technologies rely on the presence of holes and free edges in the metal stamping, but the holes may compromise the structural integrity of the stamping and edge over-molding may often be restricted. The disadvantages of adhesively bonded technologies are the adhesive cost, long curing time, and limited ability of the adhesive to withstand aggressive chemical and thermal environments encountered in BIW pretreatment in paint shops. Moreover, the secondary (joining) operations needed in both metal over-molding and adhesively bonded PMH technologies are associated with additional cost. To overcome the aforesaid shortcomings of the PMH technologies and to help meet the needs of automotive equipment manufacturers for a cost-effective, robust, reliable PMH technology that can be used for the manufacturing of load-bearing BIW components, a new approach, the so-called direct adhesion PMH technology, was proposed [92]. The approach uses direct adhesion and mechanical interlocking, and is named “clinch-lock PMH technology.” No holes and free edges are required in the process and a structural adhesive is not used. In the clinch-lock PMH technology, ﬁrst, shallow millimeter-size impressions/indentations within the metallic stamping are produced using a simple stamping process and the impressions are used to anchor the subsequently injection-molded plastic ribs to the metal stamping [93]. To ensure an effective metal/ polymer interlocking, the indentations should have a “dove tail” shape (Figure 2.90a). The joint provides (a) (b) FIGURE 2.90 Schematic of two types of polymer–metal clinch-locking. (a) Indentations have a “dove tail” shape; (b) plastic is injected around metal-stamping protrusion. (After Grujicic, M., Sellappan, V., Arakere, G., Ochterbeck, J. M., Seyr, N., Obieglo, A., Erdmann, M., and Holzleitner, J. 2010. Multidiscipline Modeling in Materials and Structures, 6(1), 23.) Fabrication Processes 299 effective metal/polymer connectivity by at least two distinct mechanisms, namely, mechanical interlocking and enhanced adhesion owing to an increased metal/polymer contact surface area. In Figure 2.90b, an alternative method of polymer–metal clinch-locking is shown in which the plastic is injected around the metal-stamping protrusion instead of into the metal-stamping impression. When compared with the existing competing PMH technologies, the new clinch-lock PMH technology offers two main advantages, namely, no holes in the metal stamping or over-molding of stamping ﬂanges are required and no expensive adhesive is needed to obtain the required level of polymer–metal mechanical interconnectivity. 2.22 Decoration of Plastics Commercial techniques for decorating plastics are almost as varied as plastics themselves. Depending on end-use applications or market demands, virtually any desired effect or combination of effects, shading of tone, and degree of brightness can be imparted to ﬂexible or rigid plastics products. The primary decorating technique is raw-materials coloring achieved at the compounding stage. Although most thermoplastics are produced in natural white or colorless transparent form, color is usually added by directly blending colorants into the base resin prior to the processing stage. These colorants (or color concentrates) are available in a wide range of stock shades with precise tinctorial values. Colors can also be matched to exact customer speciﬁcations and these speciﬁcations kept in computer memory to ensure batch-to-batch or order-to-order consistency. Color blending can also be utilitarian, as in color-coded wire- and cable-sheathing. Besides basic raw-materials coloring, mentioned above, designers have a large palette of decorating media at their disposal. Plastics can be decorated in various ways, which include painting processes, direct printing, transfer decoration, in-mold decoration, embossing, vacuum metallizing, sputtering, and electroplating. Most of these processes require bonding other media, such as inks, enamels, and other materials to the plastics to be decorated. Some plastics, notably polyoleﬁns and acetals, are, however, highly resistant to bonding and need separate treatment to activate the surface. Commonly used treatment processes are ﬂame treatment, electronic treatments such as corona discharge and plasma discharge, and chemical treatment. In ﬂame treatment, plastic objects such as bottles and ﬁlm are passed through an oxidizing gas ﬂame. Momentary contact with the ﬁlm causes oxidation of the surface, which makes it receptive to material used in decorating the product. In the corona discharge process the plastic ﬁlm to be treated is allowed to pass over an insulated metal drum beneath conductors charged with a high voltage. When the electron discharge (“corona”) between the charged conductors and the drum strikes the intervening ﬁlm surface, oxidation occurs and makes the surface receptive to coatings. Molded products are also treated in a similar manner, often by fully automatic machinery. In the plasma process [94], air at low pressure is passed through an electric discharge, where it is partially dissociated into the plasma state and then expanded into a closed vacuum chamber containing the plastic object to be treated. The plasma reacting with the surfaces of the plastic alters their physicochemical characteristics in a manner that affords excellent adhesion to surface coatings. The process can be used for batch processing of plastics products, including ﬁlms which may be unreeled in the vacuum chamber for treatment. Acetal resin products are surface treated by a chemical process consisting of subjecting the product to a short acid dip that results in an etched surface receptive to paint. 2.22.1 Painting Virtually all plastics, both thermoplastic and thermosetting, can be pained, with or without priming or other preliminary preparation procedures. The process, however, requires special consideration of the 300 Plastics Technology Handbook resin-solvent system to achieve adhesion, adequate covering, and chemical resistance. Painting operations have the advantage of being as simple or as sophisticated as the application may dictate. Plastics parts or materials can be coated manually by brushing, dipping, hand-spray painting, ﬂow coating or roller coating; they can be automatically spray-painted with rotating or reciprocating spray guns, and electro-statically painted using a conductive precoating procedure. Painting operations have the advantage of offering almost unlimited color options as well as great variety of surface ﬁnishes and ﬁnal surface properties to meet such needs as gloss, UV resistance, abrasion resistance, and chemical resistance. 2.22.2 Printing The primary printing presses used in plastics are gravure printing, ﬂexography, silk-screen printing, and pad printing. 2.22.2.1 Gravure Printing Gravure printing is a process that requires the use of an engraved metal cylinder or roller. Rotogravure is thus an appropriate title for this printing process. The engraving or etching process on the surfaces of the metal cylinder results in recessed areas that pick up ink or liquid coatings from a reservoir. With proper formulation of printing ink, the gravure process can be applied to a great variety of plastic substrates. Virtually all thermoplastic ﬁlm or sheet applications are printable by this process. A good example of the capabilities of the gravure process is the printing of woodgrain patterns on carrier foil for use in hot-stamping applications (described later). Woodgrain patterns may require the application of several coatings to achieve the proper effect. Several engraved cylinders can be used in sequence for continuous printing. 2.22.2.2 Flexography Flexography uses a ﬂexible printing plate, typically a metal-silicone rubber-bonded combination with the rubber surface processed to leave the printing surface raised over the back-ground area. The raised and recessed areas on the surface can be fabricated through photographic etching and/or engraving. After transferring the ink from a reservoir through a roller-doctor blade system onto the curved ﬂexible plate, the ink is transferred off the raised portions to the material to be printed. The process is suitable for a variety of applications, ranging from simple label ﬁlm to decoration on molded parts such as plaques, medallions or wall tile. However, the ﬂexible printing plates used in ﬂexography do not permit the very ﬁne detail that can be achieved on metal surfaces such as used in gravure printing. There are also limitations to the size and shape of articles that can be printed. 2.22.2.3 Screen Process Printing The process derived its name from the use of silk cloth or silk screen in the transfer of printing ink to articles to be printed. Integral to the process is the use of a suitable open-weave cloth or screen (silk is still commonly but not exclusively used) stretched over a framework. Screens made of nylon or other synthetic material are often employed, as also stainless steel or other metallic screens. The stretched screen is selectively coated through the use of a stencil corresponding to an art copy of the image to be printed (see “Silkscreen Printing” in Chapter 5); this coated (closed) area resists the passage of printing ink, which can only penetrate through the uncoated (open) areas of the screen. There are various ways to prepare the screen for printing, other than stenciling. Polyoleﬁns such as polyethylene and polypropylene must be surface treated before being printed. The most effective way is in an integrated machine where surface treatment takes place right before printing. A time lapse will mean that the treatment will lose some of its effect. Three methods are used; ﬂame treatment, corona discharge, and chemical treatment. Flame treatment is considered the most practical and most widely used. Fabrication Processes 301 2.22.2.4 Pad Printing Pad printing uses printing principles and techniques from letterpress and ﬂexography. The uniqueness of the process has to do with the use of a smooth silicone pad that picks up ink impression from an engraved or etched plate and transfers it to the product to be decorated. The engraved plate, known as a cliche, is produced in a manner similar to that of printing plates for offset or gravure roller printing. The silicone pickup pad can be designed to meet almost any shape and conﬁguration of the product part. This ability has prompted tremendous growth in pad printing. An additional capability of the process is that it can print several colors and impressions within one cycle of operation. Coatings can be layered when wet to accomplish multicolor designs with very accurate registration and impression quality. 2.22.2.5 Flex Printing Flex is a sheet of poly(vinyl chloride) (PVC), or simply “vinyl,” widely used to produce high-quality digital print for outdoor hoardings and banners. The weights of banner substrate may range from as light as 9 oz to as heavy as 22 oz per square yard, and may be double sided or single sided. A vinyl banner can also be reinforced with nylon for added strength and durability. Large banners (which can be so large as to cover the side of a building) are printed on a special mesh material to allow passage of air. Such mesh banners are a great solution for long-term use in windy outdoor locations. Usually, matte vinyl is chosen for indoors and glossy is chosen for outdoors. Hoardings and banners are commonly printed with large format inkjet printers (manufactured by companies such as HP, EFi Vutek, Mimaki, Roland, Mutoh, or one of many Chinese and Korean manufacturers) in CMYK mode. CMYK refers to the four inks typically used in color printing, namely, cyan, magenta, yellow, and black. (The last letter in “black” is chosen because “B” means blue.) The ink is typically applied in the order of the abbreviations. In CMYK printing with color halftoning (also called screening), which allows for less than full saturation of the printing colors, a full continuous range of colors can be produced. However, without halftoning, the three primary colors could be printed only as solid blocks of color, and therefore only seven colors can be produced, namely, the three primaries themselves plus three secondary colors produced by layering two of the primaries (cyan and yellow producing green, cyan and magenta producing blue, and yellow and magenta producing red) plus layering all three of them resulting in black. Very large vinyl banners may be produced using large-format inkjet printers of >2.5 m width, or computer-controlled airbrush devices that print the ink directly onto the banner material. Readily available banner templates may be used or a color graphics ﬁle can be created with standard tools for highresolution digital printing. Banners are also produced by applying individual elements cut from selfadhesive vinyl by a computer-driven vinyl cutter. A vinyl cutter (or vinyl plotter) is a computer-controlled plotting device with a blade instead of a pen. When a vector-based design created in a software program is sent to the vinyl cutter, the latter cuts along the vector paths laid out in the design from vinyl sheet or other material. The design may then be removed from the sheet and used for scrapbooking, card making, sign making, sewing, or many other crafts. Thus, computer-designed vector ﬁles with patterns and letters can be directly cut on a roll of vinyl that has been mounted and fed into the vinyl cutter. Large-format vinyl cutters are available for those using the machine to make large signs or other large-format designs. These machines tend to cut vinyl only. 2.22.3 Hot Stamping Hot stamping is one of the original methods of decorating plastics materials. Though familiarly known as hot stamping, the terminology “coated foil transferring” might be more appropriate since in this process the printed coating on a carrier ﬁlm is transferred onto a plastic surface. Secure adhesion is accomplished with the use of heat, pressure, and time under controlled conditions. 302 Plastics Technology Handbook The key to the process is the use of a carrier ﬁlm (usually a polycarbonate, polyester, or cellophane) upon which various coatings provide the desired decorative effect. The coated foil is placed over the plastic to be decorated, and a heated die forces the foil onto the plastic. The proper control of heat, pressure, and time transfers the coating off the carrier foil onto the plastic. The hot-stamping process is a versatile tool for plastics decoration. A wide variety of coatings can be deposited on the carrier ﬁlm which allows the process to be used on almost any thermoplastic material and many thermosets. Metallic effects can be imparted by depositing microthin coatings of gold or silver or chrome; multiple coatings can be applied to the carrier ﬁlm to achieve such effects as woodgraining, marbleizing, or multicolored designs, Three-dimensional decorative effects can also be achieved by embossing the surface of the carrier medium coating. 2.22.4 In-Mold Decorating As the name implies, in-mold decorating is a process in which a predecorated overlay (ﬁlm), or decal, is placed in the mold, where the decorated element is fused to the molded part during the heating/cooling cycles of the molding operation. Since the decorated coating is bonded between the plastic part and the ﬁlm (which will be the exterior surface), thus forming an integral part of the product, it produces one of the most durable and permanent decorations. High-quality melamine dinnerware is decorated by this method, and so are a host of other household and hardware plastics goods. In-mold decoration can be done with either injection molding of thermoplastics or compression molding of thermosets. Thermosetting plastics are decorated with a two-stage process. For melamine products, for example, the mold is loaded with the molding powder in the usual manner and closed. It is opened after a partial cure, and the decorative “foil” or overlay is placed in position. The mold is then closed again, and the curing cycle is completed. The overlay consists of a cellulose sheet having printed decoration and covered with a thin layer of partially cured clear melamine resin. During the molding cycle the overlay is fused to the product and becomes a part of the molding. The process is relatively inexpensive, especially when a multicolor decoration is required. For in-mold decoration of thermoplastic products, single-stage process is used. The foil or overlay is thus placed in the mold cavity prior to the injection of the polymer. It is held in place in the mold by its inherent static charge. Shifting is prevented during molding by inducing an additional charge by passing the wand of an electronic static charging unit over the foil after it is properly positioned. The overlay, in all cases, is a printed or decorated ﬁlm (0.003–0.005 in. thick) of the same polymer. Thus, polystyrene ﬁlm is used for a polystyrene product, and polypropylene ﬁlm for a polypropylene product. A similar procedure may also be used for decorating blow-molded products. 2.22.5 Embossing Embossing is used for producing a tactile texture or pattern on plastics sheet or ﬁlm. As the process involves the use of heat and pressure to texture a semiﬁnished substrate, embossing is largely limited to thermoplastic materials. However, it can be adapted to thermoset composites, such as melamineimpregnated sheet stock. Embossing is most commonly done with a two-roller system, in which one roller carries the embossing pattern and the other provides the essential pressure backup and feeding actions. Texture or pattern can be applied to the embossing or surface roller through a variety of processes, including conventional engraving, chemical engraving, etching, and laser cutting. Embossing can also be performed without rollers, e.g., using textured aluminum foil in one-time use, or stainless steel plates with engraved textures can be used in the press cycle time and again, offering multiple impression economies. Fabrication Processes 303 Embossing is most frequently used as a method of decorating nonslip packaging materials, vinyl wall coverings, furniture laminates, building-panel laminates, textured foil for hot stamping, and other applications where the innate quality of three-dimensional printing is of value. 2.22.6 Electroplating Electroplating is a chemical process for depositing heavy metals on plastics to achieve decorative effects and/or upgraded functionality. Since plastics are nonconductors of electricity, electroplating requires that the surface be properly conditioned and sensitized to receive metallic coatings. The principle of electroplating is to electrically conduct metal atoms such as copper, nickel and chrome off anodes placed within the plating baths through the plating solutions and onto the plastic production part. The target, i.e., the production part, acts as a cathode via connection to conductive plating racks, the part being attached to the plating rack with metal holding devices, spring-loaded contacts or prongs. The point of contact between the plating rack and the plastic part forms the continuity of the current ﬂow from anode through the solution onto the plastic part. The process of electroplating begins with the plastic part attached to the plating rack being subjected to preplate procedure, which is designed to create a surface on the plastic parts that will develop a bond between the plastic and the ﬁrst nickel or copper deposit. These initial deposits are extremely thin, in the micron (10−6 mm) range. This ﬁrst deposit is designed to increase conductivity uniformly over the plastic surface. When preplating is completed (and the plastic articles have a conductive coating), it is possible to proceed to the electroplating operation, which is very similar to conventional electroplating on metal. Electroplating of plastic products provides the high-quality appearance and wear resistance of metal combined with the light weight and corrosion resistance of plastics. Plating is done on many plastics, including phenolic, urea, ABS, acetal, and polycarbonate. Many automotive, appliance, and hardware uses of plated plastics include knobs, instrument cluster panels, bezel, speaker grilles, and nameplates. In marine searchlights zinc has been replaced by chrome-plated ABS plastics to gain lighter weight, greater corrosion resistance, and lower cost. An advantage of plastics plating is that, unlike metal die castings, which require bufﬁng in most cases after plating, plastics do not ordinarily require this extra expensive operation. The use of plated plastics also affords the possibility of obtaining attractive texture contrasts. 2.22.7 Vacuum Metallizing Vacuum metallizing is a process whereby a bright thin ﬁlm of metal is deposited on the surface of a molded product or ﬁlm under high vacuum. The metal may be gold, solver, or most generally, aluminum. The process produces a somewhat delicate surface compared to electroplating. The metallizing process can be used on virtually all properly (surface) prepared thermoplastic and thermosetting materials. Small clips of the metal to be deposited are attached to a ﬁlament. When the ﬁlament is heated electrically, the clips melt and, through capillary action, coat the ﬁlament. An increased supply of electrical energy then causes vaporization of this metal coating, and plating of the plastic product takes place. To minimize surface defects and enhance the adhesion of the metal coating, manufacturers initially give the plastics parts a lacquer base coat and dry in an oven. The lacquered parts are secured to a rack ﬁtted with ﬁlaments, to which are fastened clips of metal to be vaporized. The vaporization and deposition are accomplished at high vacuum (about 1/2 micron). The axles supporting the part holding the ﬁxtures are moved so as to rotate the parts during the plating cycle to promote uniform deposition. The thickness of the coating produced is about 5 × 10−6 in. (127 nm). After the deposition is completed, the parts are removed and dipped or sprayed with a top-coat lacquer to protect the metal from abrasion. Color tones, such as gold, copper, and brass may be added to this coating if desired. 304 Plastics Technology Handbook Vacuum metallizing of polymer ﬁlms, such as cellulose acetate, butyrate, and Mylar, is performed in essentially the same way. Film rolls are unreeled and rewound during the deposition process to metallize the desired surface. A protective abrasion-resistant coating is then applied to the metallized surface in an automatic coating machine. Vacuum metallizing is a versatile process used in a great variety of applications. Examples range from highly decorative cosmetic closures to automotive grilles and instrument clusters. Vacuum metallized plastic parts can replace metal parts with large saving in manufacturing costs and weight. The process can also serve functional needs, such as lamp reﬂectors or diffusion grids for overhead ﬂuorescent lighting. Vacuum metallizing on interior surfaces of computer or communication equipment provides a degree of radio frequency interference shielding. References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. Modern Plastics Encyclopedia, Vols. 42–45, McGraw-Hill, New York, 1964–1967. Bikales, N. M., ed. 1971. Molding of Plastics, p. 737, Interscience, New York. Butler, J. A. 1964. Compression Transfer Molding, Plastics Institute Monograph, Iliffe, London. Vaill, E. W. 1962. Modern Plastics, 40, 1A, Encycl. Issue, 767. Maiocco, A. L. 1964. Transfer molding, past, present, and future, SPE Tech. Pap., 10, XIV-4. Frados, J., ed. 1976. Plastics Engineering Handbook, 4th Ed., Van Nostrand Reinhold, New York. Rubin, I. I. 1973. Injection Molding of Plastics, Wiley, New York. Brown, J. 1979. Injection Molding of Plastic Components, McGraw-Hill, New York. Dyn, J. B. 1979. Injection Molds and Molding, Van Nostrand Reinhold, New York. Lukov, L. J. 1963. Injection molding of thermosets, SPE. J., 13, 10, 1057. Morita, Y. 1966. Screw injection molding of thermosets, SPE. Tech. Pap., 12, XIV-5. O’Brien, J. C. 1976. Business injection molding thermosets, Plast. Engl., 32, 2, 23. Schenkel, G. 1966. Plastics Extrusion Technology and Theory, Elsevier, New York. Bikales, N. M. 1971. Extrusion and Other Plastics Operations, Interscience, New York. Fischer, E. G. 1976. Extrusion of Plastics, Newness-Butterworth, London. Van, R. T., De Hoff, G. R., and Bonner, R. M. 1968. Mod. Plastics, 45, 14A, Encycl. Issue, 672. Fisher, E. G. 1971. Blow Molding of Plastics, Iliffe, London. Elden, R. A. and Swan, A. D. 1971. Calendering of Plastics, Plastics Institute Monograph, Iliffe, London. Mark, H. F., Atlas, S. M., and Cernia, E., eds. 1967. Man-Made Fibers: Science and Technology, Vol. 3, p. 91. Interscience, New York. Moncrieff, R. W. 1963. Man-Made Fibers, Wiley, New York. Riley, J. L. 1956. In Polymer Processes, C.E. Schildknecht, ed., p. 91. Interscience, New York, Chap. XVIII. Carraher, C. E. 2002. Polym. News, 27, 3, 91. Ondarcuhu, T and Joachim, C. 1998. Europhys. Lett., 42, 2, 215. Martin, C. P. 1996. Chem. Mater., 8, 1739. Whitesides, G. M. and Grzybowski, B. 2002. Science, 295, 2418. Deitzel, J. M., Kleinmeyer, J., Hirvonen, J. K., and Beek, T. N. C. 2001. Polymer, 42, 8163. Formalas, A. 1934. US Patent 1,975,504; Formalas, A. 1944. US Patent, 2,349,950. Fakirov, S., ed. 2016. Nano-size Polymers: Preparation, Properties, Applications, Springer International Publishing AG Switzerland. Fong, H. and Renekar, D. H. 1999. J. Polym. Sci.: Part B Polym. Phys., 37, 24, 3488. Doshi, J. and Renekar, D. H. 1995. J. Electrostatics, 35, 2–3, 151. Liu, H. Q. and Hsieh, Y. L. 2002. J. Polym. Sci. Part B: Polym. Physics, 40, 2119. Zussman, E., Yarin, A. L., and Weihs, D. 2002. Experiments in Fluids, 33, 315. Gibson, P. W., Schreuder-Gibson, H. L., and Riven, D. 1999. AIChE J., 45, 1, 190. Fabrication Processes 305 34. Fertala, A., Han, W. B., and Ko, F. K. 2001. J. Biomed. Mater. Res., 57, 48. 35. Buchko, C. J., Chen, L. C., Shen, Y., and Martin, D. C. 1999. Polymer, 40, 7397. 36. Jin, H. J., Fridrikh, S., Rutledge, G. C., and Kaplan, D. 2002. Abstracts of Papers Am. Chem. Soc., 224, 1–2, 408. 37. Boland, E. D., Simpson, D. G., Polanski, K. J., and Bowlin, G. L. 2001. J. Macromol. Sci. Pur. Appl. Chem., A38, 12, 1231. 38. Mathews, J. A., Wnek, G. E., Simpson, D. G., and Bowlin, G. L. 2002. Biomolecules, 3, 2, 232. 39. Li, D. and Xia, Y. 2004. Adv. Mater., 16, 14, 1151. 40. Jalili, R., Morshed, M., Abdol Karim, S., and Ravandi, H. 2006. J. Appl. Polym. Sci., 101, 4350. 41. Zhou, F.-L., Gong, R.-H., and Porat, I. 2009. Polym. Int., 58, 331. 42. Kim, G., Choo, Y. S., and Kim, W. D. 2006. Europ. Polym. J., 42, 9, 2031. 43. Yanin, A. L. and Zussman, E. 2007. Polymer, 45, 9, 2977. 44. Moghe, A. K. and Gupta, B. S. 2008. Polymer Rev., 48, 353. 45. Butzko, R. L. 1958. Plastics Sheet Forming, Van Nostrand Reinhold, New York. 46. Sarvetnick, H. A., ed. 1972. Plastisols and Organosols, p. 58, Van Nostrand Reinhold, New York. 47. Lubin, G., ed. 1982. Handbook of Composites, p. 58, Van Nostrand Reinhold, New York. 48. https://en.wikipedia.org/wiki/Pultrusion. 49. Lee, W. J., Seferis, J. C., and Bonner, D. C. 1986. Prepreg processing science, SAMPE Q., 17, 2, 58. 50. Tsunoda, Y. 1966. U.S. Patent 3,286,969. 51. Otani, S. 1965. Carbon, 3, 213. 52. Daumit, G. P. 1987. Latest in carbon ﬁbers for advanced composites, Performance Plastics ‘87 First International Ryder Conference on Special Performance Plastics and Markets, Feb., pp. 11–13. Atlanta, Georgia. 53. www.sollercomposites.com/fabricchoice.html 54. http://encyclopedia2.thefreedictionary.com/computer+numerical+control 55. http://www.tapplastics.com 56. http://www.performancecomposites.com/about-composites-technical-info/124-designing-with -carbon-ﬁber.html 57. https://en.wikipedia.org/wiki/Reinforced_carbon-carbon 58. Becker, W. E., ed. 1979. Reaction Injection Molding, p. 368, Van Nostrand Reinhold, New York. 59. Kresta, J. E., ed. 1985. Reaction Injection Molding, ACS Symp. Ser., Vol. 270, p. 368, American Chemical Society, Washington, DC. 60. PELASPAN Expandable Polystyrene, Form 171–414, Dow Chemical Co., 1966. 61. Klempner, D. and Frisch, K. C., eds. 1991. Handbook of Polymeric Foams and Foam Technology, p. 368, Hanser, Munich. 62. Zizlsperger, J., Statny, F., Beck, G., and Tatzel, H. 1970. U.S. Patent 3504 068 (to BASF). 63. One-Step Urethane Foams, Bull. F40487, Union Carbide Corp., 1959. 64. Phillips, L. N. and Parker, D. B. V. 1964. Polyurethanes: Chemistry, Technology, and Properties, Iliffe, London. 65. Harris, T. G. 1981. U.S. Patent 4281069 1981 10728 (to Armstrong World Industries, U.S.A.). 66. Chandra, P. and Kumar, A. 1979. Indian Patent 147107 19791117 (to Bakelite Hylam Ltd., India). 67. Lasman, H. R. 1967. Mod. Plastics, 45, 1A, Encycl. Issue, 368. 68. Gluck, D. G., Hagan, J. R., and Hipchen, D. E. 1980. In Advances in Urethane Science and Technology, Vol. 7, K.C. Frisch and D. Klempner, eds., p. 137, Technomic Publ. Co., Lancaster, Pennsylvania. 69. Noorani, R. I. 2005. Rapid Prototyping: Principles and Applications. Wiley, New York. 70. Chua, C. K., Leong, K. F., and Lim, C. S. 2010. Rapid Prototyping: Principles and Applications, World Scientiﬁc. 71. http://www.eos.info/additive_manufacturing/for_technology-interested 72. https://en.wikipedia.org/wiki/3D_printing 73. Schidrowitz, P. and Dawson, T. R., eds. 1952. History of Rubber Industry, p. 137, IRI, London. 306 Plastics Technology Handbook 74. Hoffman, W. 1967. Vulcanization and Vulcanizing Agents, Maclaren, London. 75. Nourry, A., ed. 1962. Reclaimed Rubber, Its Developments, Applications and Future, p. 137, Maclaren, London. 76. Ghosh, P. 1990. Polymer Science and Technology of Plastics and Rubbers, Tata McGraw-Hill, New Delhi. 77. Kojima, M., Tosaka, M., Ikeda, Y., and Kohjiya, S. 2005. J. Appl. Polym. Sci., 95, 137. 78. Parker, D. H. 1965. Principles of Surface Coating Technology, Wiley, New York. 79. Feng W., Patel S. H., Young M.-Y., Jurino III J. L., and Xanthos M. 2007. Advances in Polym. Tech., 26, 1, 1. 80. Intelligent Technology, www.ﬁelderschoicedirect.com 81. Patel, G. N. 2009. U.S. Patent 2009/0224176 A1, Sep 10. 82. Pan, Y. V., Wesley, R. A., Luginbuhl, R., Denton, D. D., and Ratner, B. D. 2001. Biomacromolecules, 2, 32. 83. Ito, Y., Chan, G., Guam, Y. Q., Imanishi, Y. 1997. Langmuir, 13, 2756. 84. Takei, Y. G., Aoki, T., Sanui, K., Ogata, N., Sakurai, Y., and Okano, T. 1994. Macromolecules, 27, 6163. 85. Peng, T. and Chen, Y. L. 1998. J. Appl. Polym. Sci. 70, 2133. 86. Lee, Y. M. and Shim, J. K. 1997. Polymer, 38, 1227. 87. White, S. R., Sottos, N. R., Geubelle, P. H., Moore, J. S., Kessler, M. R., Sriram, S. R., Brown, E. N., and Viswanathan, S. 2001. Nature, 409, 794. 88. Brown, E. N., Kessler, M. R., Sottos, N. R., and White, S. R. 2003. J. Microencapsulation, 20, 6, 719. 89. Gabriel, S., Jérôme, R., and Jérôme, C. 2010. Prog. Polym. Sci., 35, 113. 90. O’Rinda Trauernicht, J. 1970. Bonding and joining plastics, Plastics Technology, Reinhold Publishing, New York. 91. Amancio-Filho, S. T. and dos Santos, J. F. 2009. Polym. Eng. Sci., 49, 1461. 92. Grujicic, M., Sellappan, V., Omar, M. A., Seyr, N., and Erdmann, M. 2008. J. Matls. Processing Tech., 197, 363. 93. Grujicic, M., Sellappan, V., Arakere, G., Ochterbeck, J. M., Seyr, N., Obieglo, A., Erdmann, M., and Holzleitner, J. 2010. Multidiscipline Modeling in Materials and Structures, 6(1), 23. 94. Harris, R. M., ed. 1999. Coloring Technology for Plastics, Chemtec Publishing, Toronto. 3 Plastics Properties and Testing 3.1 Introduction There are two stages in the process of becoming familiar with plastics. The ﬁrst is rather general and involves an introduction to the unique molecular structures of polymers, their physical states, and transitions which have marked inﬂuence on their behavior. These have been dealt with in Chapter 1. The second stage, which will be treated in this chapter, is more speciﬁc in that it involves a study of the speciﬁc properties of plastics which dictate their applications. Besides the relative ease of molding and fabrication, many plastics offer a range of important advantages in terms of high strength/weight ratio, toughness, corrosion and abrasion resistance, low friction, and excellent electrical resistance. These qualities have made plastics acceptable as materials for a wide variety of engineering applications. It is important therefore that an engineer be aware of the performance characteristics and signiﬁcant properties of plastics. In this chapter plastics have been generally dealt with in respect to broad categories of properties, namely, mechanical, electrical, thermal, and optical. In this treatment the most characteristic features of plastic materials have been highlighted. An important facet of materials development and proper materials selection is testing and standardization. The latter part of this chapter is therefore devoted to this aspect. It presents schematically (in simpliﬁed form) a number of standard test methods for plastics, highlighting the principles of the tests and the properties measured by them. 3.2 Mechanical Properties Several unfamiliar aspects of material behavior of plastic need to be appreciated, the most important probably being that, in contrast to most metals at room temperature, the properties of plastics are time dependent [1–4]. Then superimposed on this aspect are the effects of the level of stress, the temperature of the material, and its structure (such as molecular weight, molecular orientation, and density). For example, with polypropylene an increase in temperature from 20 to 60°C may typically cause a 50% decrease in the allowable design stress. In addition, for each 0.001 g/cm3 change in density of this material there is a corresponding 4% change in design stress. The material, moreover, will have enhanced strength in the direction of molecular alignment (that is, in the direction of ﬂow in the mold) and less in the transverse direction. Because of the inﬂuence of so many additional factors on the behavior of plastics, properties (such as modulus) quoted as a single value will be applicable only for the conditions at which they are measured. Properties measured as single values following standard test procedures are therefore useful only as a means of quality control. They would be useless as far as design in concerned, because to design a plastic component it is necessary to have complete information, at the relevant service temperature, on the 307 308 Plastics Technology Handbook time-dependent behavior (viscoelastic behavior) of the material over the whole range of stresses to be experienced by the component. 3.2.1 Stress and Strain Any force or load acting on a body results in stress and strain in the body. Stress represents the intensity of the force at any point in the body and is measured as the force acting per unit area of a plane. The deformation or alteration in shape or dimensions of the body resulting from the stress is called strain. Strain is expressed in dimensionless units, such as cm/cm, in./in., or in percentage. Corresponding to the three main types of stress—tensile, compressive, and shear—three types of strain can be distinguished. Thus, tensile strain is expressed as elongation per unit length (Figure 3.1a), e = D‘=‘0 = (‘ − ‘0 )=‘0 (3.1) and compressive strain as contraction per unit length (Figure 3.1b), e = D‘=‘0 = (‘0 − ‘)=‘0 (3.2) If the applied force or load, F, is tensile or compressive, the resulting tensile or compressive stress, s, is deﬁned by s = F=A (3.3) where A is the cross-sectional area perpendicular to the direction in which the force acts (Figure 3.1a). The shearing stress is deﬁned by a similar equation t = Fs =A (3.4) where Fs is the shearing force acting on an area A, which is parallel to the direction of the applied force (Figure 3.1c). Shear strain is measured by the magnitude of the angle representing the displacement of a certain plane relative to the other, due to the application of a pure shear stress, such as a in Figure 3.1c. The corresponding shear strain g may be taken equal to the ratio aa′/ab (=tan a). A shear strain is produced in torsion, when, for example, a circular rod is twisted by tangential forces, as shown in Figure 3.1d. For small deformations the shear strain, g, can be calculated from the triangle ABC g = BC=AB = rq=‘ (3.5) where r is the radius and q is the angle of twist. An ideal elastic material is one which exhibits no time effects. When a stress is applied the body deforms immediately, and it recovers its original dimensions completely and instantaneously when the F F Fs a a´ d´ lo d l d´ lo d α F (b) F (c) d´ α l l r r θ Fs c b (a) A d Fs (d) Fs B C FIGURE 3.1 (a) Tensile or longitudinal strain, e = (‘ − ‘0 )=‘0 . (b) Compressive strain, e = (‘0 − ‘0 )=‘0 . (c) Shear strain, g = aa′/ab. (d) Shear strain in torsion g = rq=‘. 309 Plastics Properties and Testing stress is removed. When the ideal elastic body is subjected to tensile (or compressive) stress, the proportionality is expressed as s=E·e (3.6) where s is the applied stress (tensile or compressive) in lbf/in.2, kgf/cm2 or other appropriate units of force per unit cross-sectional area (Equation 3.3), e is the axial strain (Equation 3.1 and Equation 3.2), and E is the modulus of elasticity, commonly known as the Young’s modulus. The proportionality law as deﬁned above is known as Hooke’s law. Likewise, if the ideal solid is subjected to a shear stress (t), then the shear strain (g ) developed as a function of stress applied is given by the expression t=G·g (3.7) Here, the proportionality constant G is known as the shear modulus, also called the modulus of rigidity. The elastic constants in tensile deformation and shear deformation are summarized and compared below: Tensile (Figure 3.1a) Shear (Figure 3.1c) Stress s = F/A t = F/A Strain Modulus e = (‘ − ‘0 )=‘ E = s/e g = tan a G = t/g Compliance D = e/s J = g /t It may be noted that for an ideal elastic body compliance is the inverse of modulus. The modulus of elasticity, E, and the modulus of rigidity, G, as deﬁned above, apply under longitudinal and shear forces, respectively. When a hydrostatic force is applied, a third elastic modulus, the modulus of compressibility or bulk modulus, K, is used. It is the reciprocal of compressibility, b, and is deﬁned as the ratio of the hydrostatic pressure, sh, to the volumetric strain, DV/V0: K= sh 1 = DV=V0 b (3.8) As indicated in Figure 3.1, an elongation (or compression) in one direction, due to an axial force, produces a contraction (or expansion) in the lateral direction, i.e., at right angles to the direction of the force. The ratio of the lateral strain to the longitudinal strain is called Poisson’s ratio v. It is an important elastic constant, For instance, a tensile stress, sx, which produces a tensile strain, ex in the x-direction will also produce a contractive strain, ey, in the y-direction, the two being related by v = −ey =ex (3.9) Combining Equation 3.9 with Equation 3.6 and rearranging yields ey = −(v=E)sx (3.10) Equation 3.10 thus deﬁnes the contribution (ey) of the stress sx in the x-direction to the total strain in the y-direction. Poisson’s ratio, v, as deﬁned above, is a fourth elastic constant. For small deformations, the bulk modulus and modulus of rigidity can be calculated from the modulus of elasticity and Poisson’s ratio by the following equations: E K= (3.11) 3(1 − 2v) G= E 2(1 + v) (3.12) 310 Plastics Technology Handbook The elastic modulus can also be calculated from the bulk modulus and the modulus of rigidity by the relation 1 1 1 = + E 9K 3G (3.13) For soft materials such as gels, pastes, putties and many colloidal systems, which do not compress to the extent to which they are deformed under stress, K is very large compared to G, and therefore from Equation 3.13, E = 3G. For other materials as metals, ﬁbers, and some plastics, however, K must be considered. 3.2.2 Stress–Strain Behavior The stress–strain behavior of plastics measured at a constant rate of loading provides a basis for quality control and comparative evaluation of various plastics. The diagram shown in Figure 3.2a is most typical of that obtained in tension for a constant rate of loading. For compression and shear the behavior is quite similar except that the magnitude and the extent to which the curve is followed are different. In the diagram, load per unit cross section (stress) is plotted against deformation expressed as a fraction of the original dimension (strain). Even for different materials the nature of the curves will be similar, but they will differ in (1) the numerical values obtained and (2) how far the course of the typical curve is followed before failure occurs. Cellulose acetate and many other thermoplastics may follow the typical curve for almost its entire course. Thermosets like phenolics, on the other hand, have cross-linked molecules, and only a limited amount of intermolecular slippage can occur. As a result, they undergo Elongation at failure E = σ/ 2 Ultimate strength Stress (σ) ε x 1 (a) Strain Tangent modulus Initial tangent modulus 2% secant modulus 0 (b) FIGURE 3.2 1 2 3 Strain (%) 4 5 (a) Nominal stress–strain diagram. (b) Typical moduli values quoted for plastics. 311 Plastics Properties and Testing fracture at low strains, and the stress–strain curve is followed no further than to some point below the knee, such as point 1. Ultimate strength, elongation, and elastic modulus (Young’s modulus) can be obtained from the stressstrain study (Figure 3.2a). For determining the Young’s modulus (E) the slope of the initial tangent, i.e., the steepest portion of the curve, is measured. Other moduli values are also used for plastics (see Figure 3.2b). The appearance of a permanent set is said to mark a yield point, which indicates the upper limit of usefulness for any material. Unlike some metals, in particular, the ferrous alloys, the drop-of-beam effect and a sharp knee in the stress–strain diagram are not exhibited by plastics. An arbitrary yield point is usually assigned to them. Typical of these arbitrary values is the 0.2% or the 1% offset yield stress (Figure 3.3a). Alternatively, a yield stress can be deﬁned as that at which the ratio of total stress to total strain is some selected amount, say 50% or 70% of the elastic modulus (Figure 3.3b). In the ﬁrst case the yield stress is conveniently located graphically by offsetting to the right the stated amount of 0.2% (or 1%) and drawing a line paralleling that drawn for the elastic modulus. The point at which this line intersects the observed stress–strain line deﬁned the yield stress. In the second case also the point of intersection of the line drawn with a slope of 0.7E, for instance, with the observed stress–strain line determines the yield stress. Up to point 1 in Figure 3.2a, the material behaves as an elastic solid, and the deformation is recoverable. This deformation, which is small, is associated with the bending or stretching of the inter-atomic bonds between atoms of the polymer molecules (see Figure 3.4a). This type of deformation is nearly instantaneous and recoverable. There is no permanent displacement of the molecules relative to each other. Strain 0 .7E e=0 Slop Stress Slope =E =E Slope Stress Yield Strain Oﬀset (a) FIGURE 3.3 (b) Location of a yield value. (a) (b) (c) FIGURE 3.4 Deformation in plastics. (a) Stretching of polymer molecule. (b) Staightening out of a coiled molecular chain. (c) Intermolecular slippage. 312 Plastics Technology Handbook 0 (a) Strain (ε) 0 (b) ε 0 (c) σ σ Stress (σ) σ Between points 1 and 2 in Figure 3.2a, deformations have been associated with a straightening out of a kinked or coiled portion of the molecular chains (see Figure 3.4b), if loaded in tension. (For compression the reverse is true.) This can occur without intermolecular slippage. The deformation is recoverable ultimately but not instantaneously and hence is analogous to that of a nonlinear spring. Although the deformation occurs at stresses exceeding the stress at the proportional limit, there is no permanent change in intermolecular arrangement. This kind of deformation, characterized by recoverability and nonlinearity, is very pronounced in the rubber state. The greatest extension that is recoverable marks the elastic limit for the material. Beyond this point extensions occur by displacement of molecules with respect to each other (Figure 3.4c), as in Newtonian ﬂow of a liquid. The displaced molecules have no tendency to slip back to their original positions, therefore these deformations are permanent and not recoverable. Poisson’s ratio is a measure of the reduction in the cross section accompanying stretching and is the ratio of the transverse strain (a contraction for tensile stress) to longitudinal strain (elongation). Poisson’s ratio for many of the more brittle plastics such as polystyrene, the acrylics, and the thermoset materials is about 0.3; for the more ﬂexible plasticized materials, such as cellulose acetate, the value is somewhat higher, about 0.45. Poisson’s ratio for rubber is 0.5 (characteristic of a liquid); it decrease to 0.4 for vulcanized rubber and to about 0.3 for ebonite. Poisson’s ratio varies not only with the nature of the material but also with the magnitude of the strain for a given material. All values cited here are for zero strain. Strain energy per unit volume is represented as the area under the stress–strain curve. It is another property that measures the ability of a material to withstand rough treatment and is related to toughness of the material. The stress–strain diagram thus serves as a basis for classiﬁcation of plastics. Strong materials have higher ultimate strength than weak materials. Hard or unyielding materials have a higher modulus of elasticity (steeper initial slope) than soft materials. Tough materials have high elongations with large strain energy per unit volume. Stress–strain curves for type cases are shown in Figure 3.5. It must be emphasized that the type behavior shown in Figure 3.5 depends not only on the material but also very deﬁnitely on conditions under which the test is made. For example, the bouncing putty silicone is putty-like under slow rates of loading (type curve a) but behaves as an elastic solid under high rates of impact (type curve b or d). Figure 3.6a shows that at high extension rates (>1 mm/sec) unplasticized PVC is almost brittle with a relatively high modulus and strength. At low extension rates (<0.05 mm/sec) the same material exhibits a lower modulus and a high ductility, because at low extension rates the polymer molecular chains have time to align themselves under the inﬂuence of the applied load. Thus the material is able to ﬂow at the same rate as it is being strained. (This interesting phenomenon observed in some plastics is known as cold drawing.) Further examples of the effect of conditions on the behavior of plastics are illustrated by the stressstrain curves for plasticized cellulose acetate when determined at different temperatures (Figure 3.6b). Thus the material is hard and strong at low temperatures, relatively tough at ordinary temperatures, and soft and weak at higher temperatures. This behavior may be attributed to variable molecular slippage effects associated with plasticizer action. ε 0 ε (d) FIGURE 3.5 Classiﬁcation of plastics on the basis of stress–strain diagram. (a) Soft and weak. (b) Weak and brittle. (c) Strong and tough. (d) Hard and strong. 313 Plastics Properties and Testing Cold drawing Slow extension rate (a) –65°C x-Fracture 69 0°C 350 34 75°C 0 Strain 25°C (b) 20 40 Elongation (%) Stress (MPa) 700 x-Fracture Stress (kgf/cm2) Stress Fast extension rate 0 60 FIGURE 3.6 (a) Typical tensile behavior of unplasticized PVC. (b) Stress–strain curves of cellulose acetate at different temperatures. 3.2.3 Viscoelastic Behavior of Plastics In a perfectly elastic (Hookean) material the stress, s, is directly proportional to the strain, e. For uniaxial stress and strain the relationship may be written as s = constant e (3.14) where the constant is referred to as the modulus of elasticity. In a perfectly viscous (Newtonian) liquid the shear stress, t, is directly proportional to the rate of strain, _ and the relationship may be written as g, t = constant g_ (3.15) where the constant is referred to as the viscosity. Polymeric materials exhibit stress–strain behavior which falls somewhere between these two ideal cases; hence, they are termed viscoelastic. In a viscoelastic material the stress is a function of both strain and time and so may be described by an equation of the form s = f (e,t) (3.16) This equation represents nonlinear viscoelastic behavior. For simplicity of analysis it is often reduced to the form s = ef (t) (3.17) which represents linear viscoelasticity. It means that in a tensile test on linear viscoelastic material, for a ﬁxed value of elapsed time, the stress will be directly proportional to strain. The most characteristics features of viscoelastic materials are that they exhibit time-dependent deformation or strain when subjected to a constant stress (creep) and a time-dependent stress when subjected to a constant strain (relaxation). Viscoelastic materials also have the ability to recover when the applied stress is removed. To a ﬁrst approximation, this recovery can be considered as a reversal of creep. 3.2.3.1 Modulus and Compliance Consider the tensile experiment of Figure 3.1a as a stress relaxation study in which the deformation is imposed suddenly and held ﬁxed while the resulting stress, s(t), is followed with time. The tensile relaxation modulus, E(t), is then obtained as E(t) = s(t)=e0 (3.18) 314 Plastics Technology Handbook with e0 being the constant strain and the parenthesis showing the functional dependence. Similarly a shear relaxation experiment measures the shear relaxation modulus G(t): G(t) = t(t)=g 0 (3.19) where g0 is the constant strain. Consider now the tensile experiment of Figure 3.1a as a creep study in which a steady stress s0 is suddenly applied to the polymer specimen and held constant. In general, the resulting strain e(t) will be a function of time starting from the imposition of load. The results of creep experiments are often expressed in terms of compliances rather than moduli. The tensile creep compliance D(t) is D(t) = e(t)=s0 (3.20) The shear creep compliance J(t) is similarly deﬁned as J(t) = g(t)=t0 (3.21) where t0 is the constant shear stress and g(t) is the resulting time-dependent strain. A compliance is the inverse of a modulus for an ideal elastic body, but this is not true for viscoelastic materials. Consider, for example, two experiments carried out with identical samples of a viscoelastic material. In experiment (a) the sample is subjected to a tensile stress s1 for time t. The resulting tensile strain at t is e1, and the creep compliance measured at that time is D1 (t) = e1 =s1 . In experiment (b) a sample is stressed to a level s2 such that strain e1 is achieved immediately. The stress is then gradually decreased so that the strain remains at ϵ1 for time t (i.e., the sample does not creep further). Let the stress on the material at time t be s3; the corresponding relaxation modulus will be E2 (t) = s3 =e1 . In measurements of this type, it can be expected that s2 > s1 > s3 and E(t)≠1/D(t). E(t) and G(t) are obtained directly only from stress relaxation measurements, while D(t) and J(t) require creep experiments for their direct observation. 3.2.4 Stress–Strain–Time Behavior When a mass of polymer is stressed the deformation produced may be considered as a sum of the following three deformations (see Figure 3.4): 1. A deformation due to bond bending and stretching which is instantaneous and independent of temperature (ordinary elastic deformation, eoe). 2. A deformation due to chain uncoiling which is not instantaneous and whose rate depends on temperature (high elastic deformation, ehe). 3. A deformation due to slippage of polymer molecules past one another (viscous deformation ev). It is often assumed that the rates of such viscous deformation do not change with time if the applied stress is constant. Figure 3.7 shows schematically the above types of deformational response as a result of a ﬁxed stress imposed on a body showing ordinary elastic deformation only (Figure 3.7b), a second body showing high elastic deformation only (Figure 3.7c), and a third body showing viscous deformation only (Figure 3.7d). In each case, the stress is imposed at time t0 and held at a constant value until time t1, when it is removed. Real polymers exhibit deformation-time curves which are a combination of the three basic responses, and a simple relationship for a combined or total strain e e = eoe + ehe + ev can be used to analyze the deformation under a given stress. 315 Load Strain Plastics Properties and Testing t0 t0 (b) t0 (c) t1 Time Strain Time Strain (a) t1 t0 t1 Time (d) t1 Time FIGURE 3.7 Types of deformational response as a result of (a) a ﬁxed load being imposed between times t0 and t1: (b) ordinary elastic material; (c) highly elastic material; (d) viscous material. The combined response, however, differs widely among polymers. Figure 3.8 shows typical deformationtime curves. It will be noted that, given sufﬁcient time, ehe will reach a constant value while ev continues to increase with time. On release of stress, ehe will eventually disappear but ev will remain constant. An important conclusion resulting from this is that since both the high elastic and the viscous components of strain depend on both time and temperature the total strain also will depend on time and temperature. This has been shown to be an important factor affecting many polymer properties. It is therefore proposed to consider the background to this fact in greater detail in the following section. Recoverable Total strain (ε) Strain εhe εoe Irrecoverable εv t0 t1 Time (a) Recoverable Strain Total strain (ε) εhe εoe Irrecoverable t0 (b) t1 Time FIGURE 3.8 Strain-time curves: (a) material showing substantial ordinary elastic, high elastic, and viscous components of strain; (b) material in which high elastic deformation predominates. 316 Plastics Technology Handbook 3.2.4.1 The WLF Equations For a polymeric segment to move from its position to its adjacent site there must be some holes in the mass of the material into which the segment can move—and simultaneously leave a vacant space into which another segment may move. The important point is that molecular motion cannot take place without the presence of holes. These holes, collectively, are called free volume. One interpretation of the glass transition temperature (Tg) is that it is a temperature below which the free volume is really too small for much molecular movement. However, at or above Tg the molecules have sufﬁcient energy for movement, jostling occurs and the free volume increases quite sharply with an increase in temperature. It is usual to express the temperature coefﬁcient of the free volume as being the difference between the thermal expansion coefﬁcients above and below Tg. This may be expressed mathematically by the equation f = fg + (aa − ab )(T − Tg ) = fg + Da(T − Tg ) (3.22) where f is the fractional free volume at temperature T, fg is the fractional free volume at Tg, and aa and ab are the coefﬁcients of thermal expansion above and below the Tg, respectively. The value of Da is simply (aa − ab). Now it has been shown that the viscosity is related to the fractional free volume by an expression of the form 1 = Ke−A=f hT (3.23) 1 = Ke−A=fg hTg (3.24) so that where K and A are constants. Combining these one may write ! hT 1 1 loge = − hTg f fg Substituting for f from Equation 3.22, this expression yields ! −(T − Tg ) hT log10 = 2:303fg ½(fg =Da) + (T − Tg ) hTg (3.25) (3.26) Experimental data on a large range of polymers have demonstrated the approximate general validity of the equation ! −17:44(T − Tg ) hT log10 (3.27) = 51:6 + (T − Tg ) h Tg known as the Williams, Landel, and Ferry Equation (WLF equation). Solving Equation 3.26 and Equation 3.27 one obtains fg = 0.025 and Da = 4.8 × 10−4 deg−1. Equation 3.27 implies that if we know the viscosity at some temperature T we can estimate the viscosity at Tg, and from this estimate the viscosity at another temperature T1. The WLF equation thus gives the effect of temperature on viscosity. There are also other applications of the WLF equation. In essence, if the value of material property changes with temperature, and if this change arises from changes in the viscosity of the system, then it may well be possible to apply the WLF equation to the property change. 317 Plastics Properties and Testing One example of this is in relation to stress relaxation. If a polymer is deformed to a ﬁxed strain at constant temperature, the force required to maintain that strain will decay with time due to viscous slippage of molecules. One measure of this rate of decay or stress relaxation is the relaxation time l, which may be deﬁned as the time taken for the stress to decrease to 1/e of its initial value on application of strain (discussed later). In this case, it is found that ! −17:44(T − Tg ) lT log10 (3.28) = 51:6 + (T − Tg ) lTg which is of the same form as Equation 3.27. For experiments performed in shear, there is a rather complicated relation between the time-dependent stress relaxation shear modulus G(t) deﬁned by Equation 3.19 and the time-dependent creep compliance J(t) deﬁned by Equation 3.21. But if the slope of log G(t) versus log t is −m, then, to a good approximation, G(t) · J(t) = sin mp mp (3.29) for m < 0.8. Not only are G(t) and J(t) related, but the former in turn is related to the tensile modulus which itself is related to the stress relaxation time l. It is therefore possible in theory to predict creep-temperature relationships from WLF data, though in practice these are best determined by experiments. 3.2.5 Creep Behavior Except for a few exceptions like lead, metals generally exhibit creep at higher temperatures. Plastics and rubbers, however, possess very temperature-sensitive creep behavior; they exhibit signiﬁcant creep even at room temperature. In creep tests a constant load or stress is applied to the material, and the variation of deformation or strain with time is recorded. A typical creep curve plotted from such a creep test is shown in Figure 3.9a. The ﬁgure shows that there is typically an almost instantaneous elastic strain AB followed by a time-dependent strain, which occurs in three stages: primary or transient creep BC (stage 1), secondary or steady-state creep CD (stage II), and tertiary or accelerated creep DE (stage III). III 2.8 m2 E B m2 6 2.0 f/c 2.4 Stress = σ1 kg C 28 Strain D gf/ c II 35 7k I Strain (a) Strain (%) A Time Stress = σ1 1.6 1.2 21 0.8 143 0.4 71.4 k 0 10–1 (b) Log time 2 /cm gf 4k 2 cm kgf/ 2 gf/cm 100 101 102 103 Log time (h) 104 (c) FIGURE 3.9 Typical creep curve (a) with linear time scale and (b) with logarithmic time scale. (c) Family of creep curves for poly(methyl methacrylate) at 20°C (1 kgf/cm2 = 0.098 MPa). 318 Plastics Technology Handbook The primary creep has a rapidly decreasing strain rate. It is essentially similar in mechanism to retarded elasticity and, as such, is recoverable if the stress is removed. The secondary or steady-state creep is essentially viscous in character and is therefore nonrecoverable. The strain rate during this state is commonly referred to as the creep rate. It determines the useful life of the material. Tertiary creep occurs at an accelerated rate because of an increase in the true stress due to necking of the specimen. Normally a logarithmic time scale is used to plot the creep curve, as shown in Figure 3.9b, so that the time dependence of strain after long periods can be included. If a material is linearly viscoelastic (Equation 3.17), then at any selected time each line in a family of creep curves (with equally spaced stress levels) should be offset along the strain axis by the same amount. Although this type of behavior may be observed for plastics at low strains and short times, in most cases the behavior is nonlinear, as indicated in Figure 3.9c. Plastics generally exhibit high rates of creep under relatively low stresses and temperatures, which limits their use for structural purposes. Creep behavior varies widely from one polymer to another; thermoset polymers, in general, are much more creep resistant than thermoplastic polymers. The steadystate creep in plastics and rubbers (often referred to as cold ﬂow) is due to viscous ﬂow, and increases continuously. Clearly, the material cannot continue to get larger indeﬁnitely, and eventually fracture will occur. This behavior is referred to as creep rupture. The creep strength of these materials is deﬁned as the maximum stress which may be applied for a speciﬁed time without causing fracture. The creep strength of plastics is considerably increased by adding ﬁllers and other reinforcing materials, such as glass ﬁbers and glass cloth, since they reduce the rate of ﬂow. The creep and recovery of plastics can be simulated by an appropriate combination of elementary mechanical models for ideal elastic and ideal viscous deformations. Although there are no discrete molecular structures which behave like individual elements of the models, they nevertheless aid in understanding the response of plastic materials. 3.2.6 Maxwell Model The Maxwell model consists of a spring and dashpot connected in series (Figure 3.10a). When a load is applied, the elastic displacement of the spring occurs immediately and is followed by the viscous ﬂow of liquid in the dashpot which requires time. After the load is removed, the elastic displacement is recovered immediately, but the viscous displacement is not recovered. 3.2.6.1 Stress–Strain Relation The spring is the elastic component of the response and obeys the relation s1 = Ee1 (3.30) s1 and e1 are the stress and strain, respectively, and E is a constant. Strain ε σ2, ε2 η FIGURE 3.10 εo = σo/E Recovery Relaxation (ε constant) σo εo = σo/E Time t Stress σ (a) ep Cre Strain σ Unloaded σ1, ε1 E (b) Time t (c) (a) The Maxwell model. (b), (c) Responses of the model under time-dependent modes of deformation. 319 Plastics Properties and Testing The dashpot (consisting of a piston loosely ﬁtting in a cylindrical vessel containing a liquid) accounts for the viscous component of the response. In this case the stress s2, is proportional to the rate of strain e_ 2 ; i.e., s2 = he_ 2 (3.31) where h is a material constant called the coefﬁcient of viscous traction. The total strain, e, of the model under a given stress, s, is distributed between the spring and the dashpot elements: e = e1 + e2 (3.32) From Equation 3.32 the rate of total displacement with time is e_ = e_ 1 + e_ 2 (3.33) and from Equation 3.30 through Equation 3.32, e_ = 1 1 s_ + s E 1 h 2 (3.34) But both elements are subjected to the entire stress, s, s = s1 = s2 (3.35) Therefore Equation 3.34 can be written as e_ = 1 1 s_ + s E h (3.36) which is the governing equation of the Maxwell model. It is interesting to consider the responses that this model predicts under three common time-dependent modes of deformation. Equation 3.36 is commonly rearranged as follows: s_ = Ee_ − 1 s l (3.37) where l = h/E is the ratio of the viscosity h of the dashpot and the tensile modulus E of the spring. Note that l has the units of time and that it characterizes the viscoelastic nature of the element very concisely, as the ratio of the viscous portion of the response to the elastic portion. This naturally occurring parameter is taken to be the response time or the relaxation time of the model. Equation 3.37 is the governing equation of the Maxwell model. It is interesting to consider the responses that this model predicts under three common time-dependent modes of deformation. 1. Creep. If a constant stress s0 is applied, then Equation 3.37 becomes s0 1 e0 = E l l (3.38) e = e0 (1 + t=l) (3.39) e_ = Integration yields which indicates a constant rate of increase of strain with time—i.e., steady-state creep (Figure 3.10b). Equation 3.39 describes the creep response of the Maxwell element. 2. Relaxation. If the strain is held constant, then Equation 3.37 becomes s_ + s=l = 0 (3.40) 320 Plastics Technology Handbook Solving this differential equation with the initial condition s = s0 at t = t0 gives s = s0 expð−t=l) (3.41) This indicates an exponential decay of stress with time (Figure 3.10c). The stress will relax and approach zero monotonically. The relaxation time l is thus the time required for the stress to decay to 1/e, or 0.37, of its initial value. Since the strain remains constant in a relaxation experiment, Equation 3.41 can also be written as E(t) = E0 expð−t=l) (3.42) where E(t) is the tensile modulus of the Maxwell element at time t, and E0 the modulus at the initial time of deformation. The corresponding equation for the Maxwell element in shear is G(t) = G0 expð−t=l) (3.43) where G(t) is the shear modulus of the Maxwell element at time t, and G0 is the modulus at t = 0; the relaxation time l is now the ratio of the viscosity h of the viscous component and the shear modulus G of the elastic component of the model. 3. Recovery. When the initial stress, s0, is removed, there is an instantaneous recovery of the elastic strain, e0, and then, as shown by Equation 3.36, the strain rate is zero, so there is no further recovery (Figure 3.10b). 3.2.6.2 Generalized Maxwell Model The behavior of a polymer system is so complicated that we cannot represent it with the response time of a single Maxwell element. In other words, the simple model described above cannot approach the behavior of a real system. In 1893, Weichert showed that stress–relaxation experiments could be represented as a generalization of Maxwell’s equation. The mechanical model according to Weichert’s formulation is shown in Figure 3.11; it consists of a large number of Maxwell elements coupled in parallel. Since the strain in each element is common, we sum the forces acting on the individual elements to obtain the total stress as a function of time, i.e., X s(t) = si (3.44) For relaxation with constant strain e0 we combine Equation 3.41 and Equation 3.44 to obtain X t s(t) = e0 Ei exp − li (3.45) where li = hi/Ei is the relaxation time of the ith element. The overall modulus as a function of time, E(t), is thus E(t) = s(t) X t = Ei exp − e0 li (3.46) The synthesis of E(t) from known values of Ei and li is simpliﬁed by the use of semilog paper. σ (t), εo E1 η1 FIGURE 3.11 E2 η2 E3 η3 Generalized Maxwell model (Weichert’s formulation). Ei ηi λi = ηi/Ei 321 Plastics Properties and Testing Example 3.1: Derive the overall time-dependent modulus E(t) for 0 < t < 200 sec when i E0 [(dynes/cm2) × 10−8] li (sec) 1 1.000 100 2 0.667 50 3 0.333 25 Answer: For each element, Ei(t) is given by a straight line on semilog paper (Equation 3.42) with intercept (Ei)0 and a negative slope of li (Figure 3.12). Adding the curves arithmetically gives E(t) directly: t Ei (t) = (Ei )0 exp − li E(t) = 3 X Ei (t) i=1 Note that E(t) is not a straight line in Figure 3.12. As can be seen, the Maxwell–Weichert model possesses many relaxation times. For real materials we postulate the existence of a continuous spectrum of relaxation times (li). A spectrum-skewed toward lower times would be characteristic of a viscoelastic ﬂuid, whereas a spectrum skewed toward longer times would be characteristic of a viscoelastic solid. For a real system containing crosslinks the spectrum would be skewed heavily toward very long or inﬁnite relaxation times. In generalizing, l may thus be allowed to range from zero to inﬁnity. The concept that a continuous distribution of relaxation times should be required to represent the behavior of real systems would seem to follow naturally from the fact that real polymeric systems also exhibit distributions in conformational size, molecular weight, and distance between crosslinks. If the number of units is allowed to become inﬁnite, the summation over the differential units of the model (Equation 3.44) can be explained by an integration over all relaxation times. Thus as i!∞ in Figure 3.11, the range of allowable relaxation times becomes zero to inﬁnity. From the notion that the stresses in the individual elements, si, are functions of time and relaxation times, si = si(t, li), we ΣEi (t) 109 1.00 exp(– t ) 100 Ei 0.667 exp(– t ) 50 108 0.333 exp(– t ) 25 107 0 50 Time (sec) 100 FIGURE 3.12 Time-dependent modulus for individual Maxwell elements and for the sum of three elements in parallel ∑Ei(t). 322 Plastics Technology Handbook deﬁne a continuous function s (t, l) such that the total stress, s (t), is given by the following (compare Equation 3.44): ð∞ s(t) = s(t, l)dl (3.47) 0 and for relaxation with constant strain, ϵ0, it is given by the following (compare Equation 3.45): ð∞ (3.48) s(t) = e0 E(l)e−t=l dl 0 Since E(t) = s(t)/e0, we ﬁnd that we have developed an expression suitable for representing the time dependence of the relaxation modulus, i.e., ð∞ E(t) = E(l)e−t=l dl (3.49) 0 The function E(l) is referred to as the distribution of relaxation times or the relaxation spectrum. In principle, once E(l) is known, the result of any other type of mechanical experiment can be predicted. In practice E(l) is determined from experimental data on E(t). Since the distribution of relaxation times is so broad, it is more convenient to consider ln l. Hence we introduce the function H(ln l), where the parenthesis denotes functional dependence, to replace E(l) as E(l) = H(ln l) l (3.50) Then Equation 3.49 becomes (note the change of limits): +∞ ð H(ln l)e−t=l d(ln l) E(t) = (3.51) −∞ and all relaxation times are considered as ln l. What we desire now is a means to determine H(ln l) from data obtained as E(t) versus ln t. This is virtually impossible to do directly, and a number of approximate methods have been devised. These approximations are discussed in the advanced reference works of Ferry [5] and Tobolsky [6]. 3.2.7 Kelvin or Voigt Model In the Kelvin or Voigt model the spring and dashpot elements are connected in parallel, as shown in Figure 3.13a. This model roughly approximates the behavior of rubber. When the load is applied at zero time, the elastic deformation cannot occur immediately because the rate of ﬂow is limited by the dashpot. Displacements continue until the strain equals the elastic deformation of the spring and it resists further movement. On removal of the load the spring recovers the displacement by reversing the ﬂow through the dashpot, and ultimately there is no strain. The mathematical relations are derived next. 3.2.7.1 Stress–Strain Relation Since the two elements are connected in parallel, the total stress will be distributed between them, but any deformation will take place equally and simultaneously in both of them; that is, s = s1 + s2 (3.52) e = e1 + e2 (3.53) 323 Plastics Properties and Testing σ1, ε1 σ2, ε2 η E Stress (σ) (a) Unloaded Relaxation (b) ee Cr p Recovery Time Stress (σ) Strains (ε) ε = σ0/ E ε constant σ0 (c) Time FIGURE 3.13 (a) The Kelvin or Voigt model. (b), (c) Responses of the model under time-dependent modes of deformation. From Equation 3.30, Equation 3.31, and Equation 3.52, s = Ee1 + he_ 2 or, using Equation 3.53, s = Ee + he_ (3.54) which is the governing equation for the Kelvin (or Voigt) model. Its predictions for the common timedependent deformations are derived next. 1. Creep. If a constant stress, s0, is applied, Equation 3.54 becomes s0 = Ee + he_ The differential equation may be solved for the total strain, e, to give s E e = 0 1 − exp − t E h (3.55) (3.56) This equation indicates that the deformation does not appear instantaneously on application of stress, but it increases gradually, attaining asymptotically its maximum value e = s0/E at inﬁnite time (Figure 3.13b). The Voigt model is thus said to exhibit retarded elastic deformation in creep experiments. The quantity h/E = l is called a retardation time. It is the time (t = l) at which the deformation is retarded by 1/e of its maximum value. (The physical meaning of h/E for Maxwell and Voigt models should not be confused.) By comparison with Equation 3.56 the creep equation under a constant shear stress t0 may be written as g= t0 ½1 − exp( − t=l) G (3.57) where g is the time-dependent shear strain of the Voigt element and G is the shear modulus of its elastic component; l is the retardation time (=h/G). 324 Plastics Technology Handbook Equation 3.57 is conveniently written as J = J0 ½1 − exp( − t=l) (3.58) where J is the shear creep compliance (=g /t0) at time t, and J0 is the shear creep compliance at the time of stress application. 2. Relaxation. If the strain is held constant; then Equation 3.54 becomes s = Ee 3. Recovery. If the stress is removed, Equation 3.54 becomes 0 = Ee + he_ This differential equation may be solved with the initial condition e = e0 at the time of stress removal to give E e = e0 exp − t (3.59) h This equation represents an exponential recovery of strain which, as a comparison with Equation 3.56 shows, is a reversal of the predicted creep. The Kelvin (or Voigt) model therefore gives an acceptable ﬁrst approximation to creep and recovery behavior but does not predict relaxation. By comparison, the previous model (Maxwell model) could account for relaxation but was poor in relation to creep and recovery. It is evident therefore that a better simulation of viscoelastic materials may be achieved by combining the two models. 3.2.8 Four-Element Model A combination of the Maxwell and Kelvin models comprising four elements is shown in Figure 3.14a. The total strain is e = e1 + e2 + e_ k (3.60) σ1, ε1 E1 σ2, ε2 η1 η2 E2 Stress (σ) (a) e ep Cr ε0 = σ0 /E Recovery ε0 = σ0 /E Relaxation (ε constant) Stress (σ) Strain (ε) Unloaded Time (b) σ0 Time (c) FIGURE 3.14 (a) Four-element model. (b), (c) Responses of the model under time-dependent modes of deformation. 325 Plastics Properties and Testing where ek is the strain response of the Kelvin model. From Equation 3.30, Equation 3.31, and Equation 3.56, s st s E e = 0 + 0 + 0 1 − exp − 2 t (3.61) E1 h1 E2 h2 Thus the strain rate is e_ = s0 s0 E2 t + exp − h1 h2 h2 (3.62) The response of this model to creep, relaxation, and recovery situations is thus the sum of the effects described previously for the Maxwell and Kelvin models and is illustrated in Figure 3.14b. Though the model is not a true representation of the complex viscoelastic response of polymeric materials, it is nonetheless an acceptable approximation to the actual behavior. The simulation becomes better as more and more elements are added to the model, but the mathematics also becomes more complex. 3.2.9 Zener Model Another model, attributed to Zener, consists of three elements connected in series and parallel, as illustrated in Figure 3.15, and known as the standard linear solid. Following the procedure already given, we derive the governing equation of this model: h3 s_ + E1 s = h3 (E1 + E2 )e_ + E1 E2 e (3.63) This equation may be written in the form a1 s_ + a0 s = b1 e_ + b0 e (3.64) where a1, a0, b1, and b0 are all material constants. A more general form of Equation 3.64 is an ∂n s ∂n−1 s + ⋯ +a0 s n + an−1 ∂t ∂ t n−1 = bm σ1, ε1 E1 σ2, ε2 E2 σ3, ε3 ∂m e + ⋯ +b0 e ∂ tm The modern theory of viscoelasticity favors this type of equation. The models described earlier are special cases of this equation. Hookean body. All constants a and b except a0 and b0 are zero. Equation 3.65 becomes a0 s = b 0 e η3 (3.65) (3.65a) Maxwell element. All constants a and b except a0, a1, and b1 are zero. Equation 3.65 becomes FIGURE 3.15 The standard linear solid. a0 s + a1 ∂s ∂e = b1 ∂t ∂t (3.65b) 326 Plastics Technology Handbook This is the spring and dashpot in series and applies to stress relaxation at constant strain. Voigt element. All constants a and b except a0, b0, and b1 are zero. Equation 3.65 becomes ∂e (3.65c) ∂t This is the spring and dashpot in parallel and applies to strain retardation at constant stress. a0 s = b0 e + b1 3.2.10 Superposition Principle Each of the creep curves in Figure 3.9c depicts the strain response of a material under a constant stress. However, in service, materials are often subjected to a complex sequence of stresses or stress histories, and obviously it is not practical to obtain experimental creep data for all combinations of loading. In such cases a theoretical model can be very useful for describing the response of a material to a given loading pattern. The most commonly used model is the Boltzmann superposition principle, which proposes that for a linear viscoelastic material the entire loading history contributes to the strain response, and the latter is simply given by the algebraic sum of the strains due to each step in the load. The principle may be expressed as follows. If an equation for the strain is obtained as a function of time under a constant stress, then the modulus as a function of time may be expressed as E(t) = s e(t) (3.66) Thus if the applied stress is s0 at zero time, the creep strain at any time, t, will be given by e(t) = 1 s E(t) 0 (3.67) On the other hand, if the stress, s0, was applied at zero time and an additional stress, s1, at time u, the Boltzmann superposition principle says that the total strain at time t is the algebraic sum of two independent responses; that is, e(t) = 1 1 s0 + s E(t) E(t − u) 1 (3.68) For any series of stress increments this equation can be generalized to e(t) = u=t X si u=−∞ 1 E(t − u) (3.69) The lower limit of the summation is taken as −∞ since the entire stress history contributes to the response. As an illustration, for a series of step changes in stress as in Figure 3.16a, the strain response predicted by the model is shown schematically in Figure 3.16b. The time-dependent strain response (creep curve) due to the stress s0 applied at zero time is predicted by Equation 3.66 with s = s0. When a second stress, s1, is added to s0, the new curve will be obtained, as illustrated in Figure 3.16b, by adding the creep due to s1 to the anticipated creep due to s0. Removal of all stress at a subsequent time u2 is then equivalent to removing the creep stain due to s0 and s1, independently, as shown in Figure 3.16b. The procedure is repeated in a similar way for other stress changes. To take into account a continuous loading cycle, we can further generalize Equation 3.69 to ðt e(t) = −∞ 1 ds(u) du E(t − u) du (3.70) 327 Stress (σ) Plastics Properties and Testing σ1 σ0 σ3 0 u2 u1 u3 B Creep due to σ0 C Creep due to σ1 0 (b) FIGURE 3.16 Time A+B Strain (ε) (a) u1 t1 u2 t1 t1 Creep due to σ3 u3 Time t1 (a) Stress history. (b) Predicted strain response using Boltzmann’s superposition principle. In the same way the stress response to a complex strain history may be derived as ðt de(u) du s(t) = E(t − u) du (3.71) −∞ When the stress history has been deﬁned mathematically, substitution in Equation 3.70 and integration within limits gives the strain at the given time. The stress at a given time is similarly obtained from Equation 3.71. 3.2.11 Isometric and Isochronous Curves Isometric curves are obtained by plotting stress vs. time for a constant strain; isochronous curves are obtained by plotting stress vs. strain for a constant time of loading. These curves may be obtained from the creep curves by taking a constant-strain section and a constant-time section, respectively, through the creep curves and replotting the data, as shown in Figure 3.17. An isometric curve provides an indication of the relaxation of stress in the material when the strain is kept constant. Since stress relaxation is a less common experimental procedure than creep testing, an isometric curve, derived like the preceding curves from creep curves, is often used as a good approximation of this property. Isochronous curves, on the other hand, are more advantageously obtained by direct experiments because they are less time consuming and require less specimen preparation than creep testing. The experiments actually involve a series of mini creep and recovery tests on the material. Thus a stress is applied to a specimen of the material, and the strain is recorded after a time t (typically 100 sec). The stress 328 Plastics Technology Handbook Creep curves Strain σ5 σ4 σ3 ε' σ2 σ1 Log time σ4 σ5 Stress Stress t´ σ4 σ3 σ3 σ2 σ1 Time = t´ Strain = ε (a) FIGURE 3.17 Log time (b) Strain (a) Isometric and (b) isochronous curves from creep curves. is then removed and the material is allowed to recover. This procedure is repeated unit there are sufﬁcient points to plot the isochronous curve. Note that the isochronous test method is quite similar to that of a conventional incremental loading tensile test and differs only in that the presence of creep is recognized and the “memory” of the material for its stress history is overcome by the recovery periods. Isochronous data are often presented on log–log scales because this provides a more precise indication of the nonlinearity of the data by yielding a straightline plot of slope less than unity. 3.2.12 Pseudoelastic Design Method Due to the viscoelastic nature of plastics, deformations depend on such factors as the time under load and the temperature. Therefore the classical equations available for the design of structural components, such as springs, beams, plates, and cylinders, and derived under the assumptions that (1) the modulus is constant and (2) the strains are small and independent of loading rate or history and are immediately reversible, cannot be used indiscriminately. For example, classical equations are derived using the relation Stress = constant strain where the constant is the modulus. From the nature of the creep curves shown in Figure 3.17a, it is clear that the modulus of a plastic is not constant. Several approaches have been developed to allow for this fact, and some of them also give very accurate results; but mathematically they are quite complex, and this has limited their use. However, one method that has been widely accepted is the pseudoelastic design method. In this method appropriate values are chosen for the time-dependent properties, such as modulus, and substituted into the classical equations. 329 Plastics Properties and Testing The method has been found to give sufﬁciently accurate results, provided that the value of the modulus is chosen judiciously, taking into account the service life of the component and the limiting strain of the plastic. Unfortunately, however, there is no straightforward method for ﬁnding the limiting strain of a plastic. The value may differ for various plastics and even for the same plastic in different applications. The value is often arbitrarily chosen, although several methods have been suggested for arriving at an appropriate value. One method is to draw a secant modulus which is 0.85 of the initial tangent modulus and to note the strain at which this intersects the stress–strain curve (see Figure 3.2b). But this method may be too restrictive for many plastics, particularly those which are highly crystalline. In most situations the maximum allowable strain is therefore decided in consultations between designer and product manufacturer. Once an appropriate value for the maximum strain is chosen, design methods based on creep curves and the classical equations are quite straightforward, as shown in the following examples. Example 3.2: A plastic beam, 200 mm long and simply supported at each end, is subjected to a point load of 10 kg at its mid-span. If the width of the beam is 14 mm, calculate a suitable depth so that the central deﬂection does not exceed 5 mm in a service life of 20,000 h. The creep curves for the material at the service temperature of 20°C are shown in Figure 3.18a. The maximum permissible strain in this material is assumed to be 1%. Answer: The linear elastic equation for the central deﬂection, d, of the beam is d= PL3 48EI where P, load at mid-span; L, length of beam; E, modulus of beam material; I, second moment of area of beam cross section The second moment of area is I= bd3 14d 3 = mm4 12 12 m2 gf /c 8k 2 17 4k 1.5 3 14 200 m /c f kg Stress (kgf/cm2) Strain (%) 21 25 0k 2.0 gf/ gf/ cm 2 cm 2 2.5 2 m f/c g 7k 1.0 2 10 m gf/c 71 k Initial modulus = 9490 kgf/cm2 160 120 80 1% secant modulus = 9285 kgf/cm2 0.5 40 0 10–1 (a) 10 101 102 103 Log time (h) 104 105 0 (b) 1.0 2.0 Strain (%) 3.0 4.0 FIGURE 3.18 (a) Creep curves for material used in illustrative examples. (b) Isochronus curve at 20,000 h service life (1 kgf/cm2 = 0.098 MPa). 330 Plastics Technology Handbook So from the expression for d, d3 = PL3 56Ed The only unknown on the right side is E. For plastic this is time dependent, but a suitable value corresponding to the maximum permissible strain may be obtained by referring to the creep curves in Figure 3.18a. A constant-time section across these curves at 20,000 h gives the isochronous curve shown in Figure 3.18b. Since the maximum strain is recommended as 1%, a secant modulus may be taken at this value. It is 9285 kgf/cm2 (=92.85 kgf/mm2). Using this value in the above equation gives d3 = 10(200)3 56 92:85 5 d = 14:5 mm Example 3.3: A thin-wall plastic pipe of diameter 150 mm is subjected to an internal pressure of 8 kgf/cm2 at 20°C. It is suggested that the service life of the pipe should be 20,000 h with a maximum strain of 2%. The creep curves for the plastic material are shown in Figure 3.18a. Calculate a suitable wall thickness for the pipe. Answer: The hoop stress, s, in a thin-wall pipe of diameter d and thickness h, subjected to an internal pressure, P, is given by s= Pd 2h so h = Pd 2s A suitable design stress may be obtained from the creep curves in Figure 3.18a. By referring to the 20,000-h isochronous curve (Figure 3.18b) derived from these curves, the design stress at 2% strain is obtained as 167.7 kg/cm2. (Note that a similar result could have been obtained by plotting a 2% isometric curve from the creep curves and reading the design stress at a service life of 20,000 h.) Substituting the design stress into the equation for h gives h= 8 150 = 3:58 mm 2 167:7 It may be seen from the creep curves (Figure 3.18a) that when the pipe is ﬁrst pressurized, the strain is less than 1%. Then as the material creeps, the strain increases steadily to reach its limit of 2% at 20,000 h. In both examples it has been assumed that the service temperature is 20°C. If this is not the case, then creep curves at the appropriate temperature should be used. However, if none are available, a linear extrapolation between available temperatures may be sufﬁcient for most purposes. Again, for some materials like nylon the moisture content of the material has a signiﬁcant effect on its creep behavior. In such a case creep curves are normally available for the material in both wet and dry states, and appropriate data should be used, depending on the service conditions. 3.2.13 Effect of Temperature Many attempts have been made to obtain mathematical expressions which describe the time and temperature dependence of the strength of plastics. Since for many plastics at constant temperature a plot of 331 Plastics Properties and Testing stress, s, against the logarithm of time to failure (creep rupture), t, is approximately linear, one of the expressions most commonly used is t = Ae−Bs (3.72) where A and B are constants. In reality, however, they depend on factors such as material structure and on temperature. The most successful attempts to include the effects of temperature in a relatively simple expression have been made by Zhurkov and Bueche, who used an equation of the form [7] −gs U t = t0 exp 0 RT (3.73) where t0 is a constant which has approximately the same value for most plastics, U0 is the activation energy of the fracture process, g is a coefﬁcient which depends on the structure of the material, R is the molar gas constant, and T is the absolute temperature. A series of creep rupture tests on a given material at a ﬁxed temperature would permit the values for U0 and g for the material to be determined from this expression. The times to failure at other stresses and temperatures could then be predicted. The relative effects of temperature rises on different plastic materials depend on the structure of each material and, particularly, whether it is crystalline or amorphous. If a plastic is largely amorphous (e.g., polymethyl methacrylate, polystyrene), then it is the glass transition temperature (Tg) which will determine the maximum service temperature, since above Tg the material passes into the rubbery region (see Figure 1.19). On the other hand, in plastics which have a high degree of crystallinity (e.g., polyethylene, polypropylene), the amorphous regions are small, so Tg is only of secondary importance. For them it is the melting temperature which will limit the maximum service temperature. The lowest service temperatures which can be used are normally limited by the brittleness introduced into the material. The behavior of plastics materials at room temperature is related to their respective Tg values. This aspect has been dealt with in Chapter 1. 3.2.14 Time–Temperature Superposition In engineering practice, it is often necessary to design for the use of a material over a long period of time— many years, for example. A common parameter to use in design work is the elastic modulus. We know, however, that for polymers the modulus decreases with increasing time under load. Accumulation of long-term data for design with plastics can be very inconvenient and expensive. A method is thus needed to extrapolate data from shorter time studies at higher temperature to longer times over several decades of time scale at the desired temperature so that a lower limit of the modulus can be determined for use in design. On the other hand, it is sometimes difﬁcult to obtain data over a very short time scale. One must then extrapolate data obtained under practicable experimental conditions to these short time scales. An empirical method for such extrapolations is available for amorphous polymer systems and, in general, for polymer systems where structure does not change during the period of testing. The aforesaid extrapolations make use of a time-temperature superposition principle which is based on the fact that time and temperature have essentially equivalent effects on the modulus values of amorphous polymers. Figure 3.19 shows modulus data taken at several temperatures for poly(methyl methacrylate) [8]. Because of the equivalent effect of time and temperature, data at different temperatures can be superposed on data taken at a speciﬁed reference temperature merely by shifting individual curves one at a time and consecutively along the log t axis about the reference temperature. 332 Plastics Technology Handbook 40°C Log E(t ), (dynes/cm2) 10 80°C 9 100°C 110°C 115°C Stress relaxation of PMMA 8 135°C 7 0.001 0.0001 0.1 1 10 Time (hr) 100 1000 FIGURE 3.19 Logarithm of tensile relaxation modulus versus logarithm of time for unfractionated poly(methyl v = 3:6 106 . (After McLoughlin, J. R. and Tobolsky, A. V. 1952. J. Colloid Sci., 7, 555.) methacrylate) of M This time–temperature superposition procedure has the effect of producing a single continuous curve of modulus values extending over many decades of log t at the reference temperature. A curve constructed in this way, as shown in Figure 3.20 (with a reference temperature 115°C), is known as the master curve. The time-temperature superposition can be expressed mathematically as E(T1 ,t) = E(T2 ,t=aT ) (3.74) for a tensile stress relaxation experiment (T2 > T1). The procedure asserts that the effect of changing the test temperature on viscoelastic properties is the same as that of multiplying or dividing the time scale by a constant quantity (aT) at each temperature. The quantity aT is called the shift factor, and it must be obtained directly from the experimental curve by measuring the amount of shift along the log t scale 40° Log E(t) (dynes/cm2) 10 100° 110° 9 115° 8 Master curve 135° 7 –10 –5 0 +5 Log t (h at 115°C) FIGURE 3.20 Modulus-time master curve based on time–temperature superposition of data in Figure 3.19. Times referred to temperature of 115°C. 333 Plastics Properties and Testing necessary to match the curve. The parameter aT is chosen as unity at the reference temperature and is a function of the temperature alone, decreasing with increasing temperature. It is common practice now to use the glass transition temperature (Tg) as the reference temperature for master curve construction. For most amorphous polymers, the shift factor at any other temperature T is then given fairly well by log10 aT = log t(T) −17:44(T − Tg ) = t(Tg ) 51:6 + (T − Tg ) (3.75) Equation 3.75 is known as the WLF equation (see Equation 3.27) after the initials of the researchers who proposed it [9]. The expression given holds between Tg and Tg + 100°C. However, if a different reference temperature is chosen an equation with the same form as Equation 3.75 can be used, but the constants on the right hand side must be re-evaluated. The signiﬁcance of the WLF generalization (Equation 3.75) cannot be over-emphasized. Again and again, one ﬁnds in the literature methods of superposing time and temperature for mechanical and other properties in amorphous and partially amorphous materials. Whatever modiﬁcations are introduced usually reduce the behavior back in the direction of Equation 3.75. 3.2.15 Dynamic Mechanical Properties A complete description of the viscoelastic properties of a material requires information over very long times. To supplement creep and stress relaxation measurements which are limited by experimental limitations, experiments are therefore performed in which an oscillating stress or strain is applied to the specimen. These constitute an important class of experiments for studying the viscoelastic behavior of polymeric solids. In addition to elastic modulus, it is possible to measure by these methods the viscous behavior of the material in terms of characteristic damping parameters. Damping is an engineering material property and the observed response is much more sensitive to the polymer constitution than in step-function experiments. Oscillatory experiments (also referred to as dynamic mechanical experiments) thus offer a powerful technique to study molecular structure and morphology. A signiﬁcant feature is the breadth of the time-scale spectrum available with these methods, e.g., 10−5–108 cycles/sec. In a dynamic experiment, the stress will be directly proportional to the strain if the magnitude of the strain is small enough. Then, if the stress is applied sinusoidally the resulting strain will also vary sinusoidally. (The same holds true if the strain is the input and the stress the output.) At sufﬁciently low frequencies, the strain will follow the stress in phase. However, in the general case the strain will be out of phase. In the last instance, the strain can be factored into two components—one of which is in phase with the stress and the other which lags behind the stress by p/2 rad. Alternatively, the stress can be decomposed into a component in phase with the strain and one which leads the strain by p/2 rad. This is accomplished by use of a rotating vector scheme, as shown in Figure 3.21. The magnitude of the stress at any time is represented by the projection OC of the vector OA on the vertical axis. Vector OA rotates with a frequency w equal to that of the sinusoidally varying stress. The length of OA is the stress amplitude (maximum stress) involved in the experiment. The strain is represented by the projection OD of vector OB on the vertical axis. The strain vector OB rotates in the same direction as OA with frequency w but it lags OA by an angle d. The loss tangent (discussed later) is deﬁned as tan d. The strain vector OB can be resolved into vector OE along the direction of OA and OF perpendicular to OA. Then the projection OH of OE on the vertical axis is the magnitude of the strain which is in phase with the stress at any time. Similarly, projection OI of vector OF is the magnitude of the strain which is p/2 rad (one quarter cycle) out of phase with the stress. The stress can be similarly resolved into two components with one along the direction of OB and one leading the strain vector by p/2 rad. The ratio of the in-phase stress to the strain amplitude (maximum 334 Plastics Technology Handbook C E A H B D δ ω O I FIGURE 3.21 F Decomposition of strain vector into two components in a dynamic experiment. strain) is called the storage modulus. In a shear deformation experiment this quantity is labeled G′(w). The ratio of the out-of-phase stress to the strain is the loss modulus G″(w). If, on the other hand, the strain vector is resolved into its components, the ratio of the in-phase strain to stress amplitude (maximum stress) is the storage compliance J′(w) and the ratio of the out-of-phase strain to the stress amplitude is the loss compliance J″(w). It is evident from the above description that G′(w) and J′(w) are associated with the periodic storage and complete release of energy in the sinusoidal deformation process. The loss parameters G″(w) and J″w), on the other hand, reﬂect the nonrecoverable use of applied mechanical energy to cause viscous ﬂow in the material. At a speciﬁed frequency and temperature, the dynamic response of a polymer in shear deformation can be summarized by any one of the following pairs of parameters: G′(w) and G″(w), J′(w) and J″(w), or absolute modulus |G| and tan d. 3.2.16 Rheological Behavior Rheology is the science of deformation and ﬂow of matter. Essentially, all thermoplastic resins (and many thermosetting resins) are required to undergo ﬂow in the molten state during the course of product manufacture. Important fabrication processes such as injection, extrusion, and calendering all involve the ﬂow of molten polymers. In plastics fabrication, it is important to understand the effect, on melt viscosity, of such factors as temperature, pressure, rate of shear, molecular weight, and structure. It is also equally important to have reliable means of measuring viscous properties of materials. The ﬂow behavior of polymeric melts cannot be considered to be purely viscous in character. The response of such materials is more complex, involving characteristics that are both viscous and elastic. This is only to be expected when one is trying to deform variously entangled long-chain molecules with a distribution of molecular weights. During ﬂow, polymer molecules not only slide past each other, but also tend to uncoil—or at least they are deformed from their equilibrium, random coiled-up conﬁguration. On release of the deforming stresses, these molecules tend to revert to random coiled-up forms. Since molecular entanglements cause the molecules to act in a cooperative manner, some recovery of shape corresponding to the recoiling 335 Plastics Properties and Testing occurs. In phenomenological terms, we say that the melt shows elasticity in addition to viscous ﬂow. The elastic—that is to say, time-dependent—effects play a most important part in die swell, extrusion defects, and melt fracture, which will be considered later in this section. 3.2.16.1 Classiﬁcation of Fluid Behavior Although one can measure deformation in a solid, one cannot normally do this in a liquid since it undergoes a continuously increasing amount of deformation when a shear stress is applied. But one can determine the deformation rate (the shear rate) caused by an applied shear stress, or vice versa, and ﬂuid behavior can be classiﬁed on this basis. We begin by making a reference to Figure 3.22, which schematically illustrates two parallel plates of very large area A separated by a distance r with the space in between ﬁlled with a liquid. The lower plate is ﬁxed and a shear force Fs is applied to the top plate of area A producing a shear stress (t = Fs/A) that causes the plate to move at a uniform velocity v in a direction parallel to the direction of the force. It may be assumed that the liquid wets the plates and that the molecular layer of liquid adjacent to the stationary plate is stationary while the layer adjacent to the top plate moves at the same velocity as the plate. Intermediate layers of liquid move at intermediate velocities, and at steady state in laminar ﬂow a velocity distribution is established as indicated by the arrows in the diagram. The velocity gradient between the two plates is dv/dr. It is deﬁned as the shear rate and is commonly given the symbol g_ i.e., g_ = dv=dr (3.76) If the liquid is ideal and it is maintained at a constant temperature, the shear stress is linearly and directly proportional to the shear rate such that one may write t = h(dv=dr) = hg_ or (3.77) h = t=g_ (3.78) where h is the coefﬁcient of viscosity or simply the viscosity or internal friction of the liquid. The linear relationship between t and g_ given by Equation 3.77 or Equation 3.78 is known as Newton’s law and liquids which behave in this manner are called Newtonian ﬂuids or ideal ﬂuids. Other ﬂuids which deviate from Newton’s law are described as non-Newtonian. For such ﬂuids, the viscosity deﬁned by Equation 3.78 is also known as the apparent viscosity. In practice, the Newtonian behavior is conﬁned to low molecular weight liquids. Polymer melts obey Newton’s law only at shear rates close to zero and polymer solutions only at concentrations close to zero. The most general rheological equation is _ h = f (g,T,t,P,c, … …) (3.79) where the variables are g_ = shear rate (itself a function of the shear stress), T = temperature, t = time, P = pressure (itself a function of volume), c = conMoving plate of area A centration, and the multiple dots which follow and velocity v include, for example, molecular parameters (such v Shear as molecular weight and molecular weight distridr force (F5) bution), compositional variables (crystallinity and r the presence of additives), and factors that relate to processing history. Such an equation is clearly dv unrealistic, so we shall consider here some of the Stationary plate principal variables, one at a time, assuming that the others remain constant. Several common types of rheological behavior FIGURE 3.22 Velocity distribution of a liquid between are shown in Figure 3.23 based upon t vs. g_ curves. two parallel plates, one stationary and the other moving. 336 Plastics Technology Handbook Pseudoplastic τ Shear stress τ Newtonian . Shear rate γ (a) (b) τ Bingham τ Dilatant . γ (c) FIGURE 3.23 . γ (d) . γ _ for different types of ﬂuid material. Flow curves (t versus g) These ﬂow phenotypes are named Newtonian, pseudoplastic, dilatant, and Bingham. In Newtonian liquids, the viscosity is constant and independent of shear rate. In pseudoplastic and dilatant liquids the viscosity is no longer constant. In the former it decreases and in the latter it increases with increasing shear rate; that is to say, the shear stress increases with increasing shear rate less than proportionately in a pseudoplastic and more than proportionately in a dilatant. Pseudoplastics are thus described as shear-thinning and dilatants as shear-thickening ﬂuid systems. These two ﬂow phenotypes can be described by an Equation the “power law”: t = hN g_ n n>1 Dilatant Newtonian Log τ n=1 n<1 Pseudoplastic Non-power-law ﬂuid (3.80) where hN is the zero shear (Newtonian) viscosity. The exponent n is greater than unity for a dilatant and less than unity for a pseudoplastic. A Newtonian is then seen to be a special case, with n = 1. Equation 3.80 gives a linear relationship between log t and log g_ and the slope of the experimental plot (Figure 3.24) gives the value of n. For the analysis of ﬂow behavior of many systems, the power law relationship has been useful as from a plot of data (Figure 3.24), and measuring the slope, one can readily get an idea of just how non-Newtonian the ﬂuid is. A Bingham body would be described by the equation t − ty = hg_ (3.81) Log (dγ/dt) FIGURE 3.24 Power-law plot showing log t versus log (dg/dt) for different types of ﬂuid material (schematic). where ty is the yield stress or yield value. Below ty the material will not ﬂow at all (hence g_ = 0 and h = ∞) so that the material, to all intents and 337 Plastics Properties and Testing purposes, is a solid. However, as soon as t exceeds ty the material suddenly behaves like a liquid with a viscosity that remains constant with increasing shear rate. Materials which exhibit this type of behavior include drilling muds, sewage sludge, toothpaste, greases and fats, as well as the clay slurries originally observed by Bingham. Lenk [10] has shown that the ﬂow phenotypes form part of a general response pattern which may be summarized in a general ﬂow curve. The generalized ﬂow curve is shown in Figure 3.25 alongside a fully developed stress–strain curve for a typical tough solid in tension after conversion of the conventional stress (i.e., force per unit of original cross-sectional area) to true stress (i.e., force per unit of actual crosssectional area after deformation). This conversion can be effected if a continuous record of the changes in cross-sectional area of the specimen under test is kept. It may be seen that the shapes of the two curves are absolutely identical and the regions into which the two curves divide have also analogous physical signiﬁcance. It is well known that at extremely low shear rates the slope of the t=g_ curve (Figure 3.23) is constant and that there exists some very low but ﬁnite threshold shear rate beyond which deviation from linearity commences. The slope of the initial linear portion of the curve is known as the “limiting viscosity,” the zero shear viscosity, or the Newtonian viscosity. Beyond this low shear rate region (initial Newtonian regime) the material is shear-softened (i.e., becomes pseudoplastic), a phenomenon which has its counterpart in the solid state where it is known as strain-softening. Continuing in the pseudoplastic region it is often found that an upper threshold can be reached beyond which no further reduction in viscosity occurs. The curve then enters a second linear region of proportionality the slope of which is the second Newtonian viscosity. Polymer melts are almost invariably of the pseudoplastic type, and the existence of ﬁrst and second Newtonian regions has long been recognized. The pseudoplastic behavior appears to arise from the elastic nature of the melt and from the fact that under shear, polymers tend to be oriented. At low shear rates Brownian motion of the segments occurs so polymers can coil up (re-entangle) at a faster rate than they are oriented. At higher shear rates such re-entangling rates are slower than the orientation rates and the polymer is hence apparently less viscous. However at very high shear rates (beyond the range of usual interest in the polymer processing industries) the degree of orientation reaches a maximum and so a further decrease in effective viscosity cannot occur; the polymer is this range again becomes Newtonian. Generally speaking, the larger the polymer molecule the longer the recoiling (re-entangling, relaxation) time so that high-molecular-weight materials tend to be more non-Newtonian at lower shear rates than lower-molecular-weight polymers. Turbulence/melt fracture Second Newtonian region Strain-hardening region Tensile stress Shear stress Dilatant region Rupture Second Hookean region Strain softening region Pseudoplastic region Initial Newtonian region (a) Shear rate Initial Hookean region (b) Strain FIGURE 3.25 (a) Generalized ﬂow curve. (b) Typical fully developed stress–strain curve as found in tough plastics under appropriate conditions. The conventional stress has been converted to the true stress. 338 Plastics Technology Handbook The dilatant type is less common among plastics under ordinary conditions, but it can be found in heavily ﬁlled systems and in some PVC pastes. The dilatant type (Figure 3.23c) represents the entire generalized ﬂow curve in which the initial Newtonian and pseudoplastic regions have degenerated to a vanishingly small portion of the curve as a whole. In some cases of dilatancy a linear region may be distinguished before the curvature appears. In other cases no distinctly linear portion can be seen at the low-shear-rate end. The Bingham type is not common among plastics. 3.2.16.2 Effect of Shear Rate on Viscosity Most polymer melts exhibit non-Newtonian behavior with the apparent viscosity decreasing with increasing shear rate (shear thinning). Viscosity of polymer at high shear rates may be several orders of magnitude smaller than the viscosity at low shear rates. Typical shear rate vs. viscosity curves are shown in Figure 3.26. The polymer melt has a Newtonian viscosity which is high at very low shear rates. Viscosity decreases nearly linearly with shear rate when plotted on a log–log scale (Figure 3.26). In this linear range, the power law equation (Equation 3.80) with n<1 is applicable. The reduction of viscosity with increasing rate of shear [11,12] is taken advantage of in achieving desirable and optimum viscosity for polymers in processing machines an d equipment without raising the temperature to detrimental levels, simply by raising the shear rate to as high a level as economically and otherwise possible. A reduction in viscosity with increasing shear rate is also taken advantage of in brushing and spraying of paints which are polymer solutions/suspensions containing pigments. 3.2.16.3 Effect of Molecular Weight on Viscosity The molecular weight of a polymer is the most important factor affecting rheology. For most polymers the zero-shear viscosity is approximately proportional to the weight-average molecular weight (Mw) below a critical value (Mc) and depends on Mw to a power equal to 3.5 at molecular weights above Mc: h = K1 Mw for Mw < Mc (3.82) h = K2 Mw3:5 for Mw > Mc (3.83) and 106 5 Viscosity, η 105 4 3 2 104 1 103 10–3 10–2 10–1 . Shear rate, γ 100 0 FIGURE 3.26 Typical log viscosity–log shear rate curves at ﬁve different temperatures. Curve 1 is for the highest temperature and curve 5 is for the lowest temperature. For a typical polymer, the temperature difference between each curve is approximately 10°C. 339 Plastics Properties and Testing The relationship between viscosity and molecular weight shown by a graphical logarithmic plot in Figure 3.27 is characterized by a sharp change at the critical molecular weight (Mc). The value of Mc varies from one polymer to another. For most polymers Mc is between 5,000 and 15,000. The critical molecular weight corresponding to the transition in the viscosity behavior at Mc points to additional hindrance to ﬂow from this point onwards due to chain entanglements. Below this point the molecules usually move independently as in low-molecular-weight liquids, but above this point mutual entanglements of chain molecules is so prominent that movement of one involves dragging of others along with it; as a consequence, increasing molecular weight is associated with a very high rate of viscosity increase. Considering the fact that the viscosity is increasing logarithmically with molecular weight in Figure 3.27, it should be clear why molecular weight control is important in polymer processing. One needs a molecular weight high enough to attain good mechanical properties but not so high that the molten polymer is too viscous to be processed economically. The distribution of molecular weights in a polymer also inﬂuences its rheology. In general, the broader the range, the lower the shear rate at which shear thinning (decrease in viscosity) develops. Thus polymers with broad molecular-weight distribution are easier to extrude than those with narrow distribution. Chain branching is another factor that inﬂuences ﬂow. The more highly branched a given polymer, the lower will be its hydrodynamic volume and the lower its degree of entanglement at a given molecular weight. One can make the general observation, therefore, that viscosity is higher with linear than with branched polymers at a given shear rate and molecular weight. This does not mean that chain branching is necessarily desirable. In fact, branching results in weaker secondary bonding forces and possibly poorer mechanical properties. 3.2.16.4 Effect of Temperature on Polymer Viscosity The viscosity of most polymers changes with temperature. An Arrhenius equation of the form h = AeE=RT where A is a constant and E is the activation energy, has often been used to relate viscosity and temperature. Constants in the Arrhenius equation can be evaluated by plotting the logarithm of viscosity against the reciprocal of absolute temperature, using shear stress or shear rate as a parameter. The data for most materials give straight line over reasonably large range of temperature. Whilst the Arrhenius equation can be made to ﬁt experimental data quite well it does nothing to explain the difference between polymers. In this regard, the WLF equation (see Equation 3.27): 6 Slope: 3.4 –3.5 Log viscosity poise (3.84) 4 2 Slope~1 0 hT log hTg Mc –2 3 4 5 ! = −17:44(T − Tg ) 51:6 + (T − Tg ) 6 Log M FIGURE 3.27 Dependence of polymer (melt) viscosity on molecular weight (M): a typical plot of log viscosity against log M. is more useful. Melt viscosity, according to this equation, is a function of (T − Tg). Thus, for example, a major cause of the difference between the viscosity of poly(methacrylate) at its processing temperature (where T − Tg = 100°C) and the 340 Plastics Technology Handbook viscosity of polyethylene at its processing temperature (where T − Tg = 200°C) is explicable by this relationship. The WLF equation also explains why viscosity is more temperature sensitive with materials processed closer to their Tg, for example, poly(methyl methacrylate), compared with nylon 6. 3.2.16.5 Effect of Pressure on Viscosity While temperature rises at constant pressure cause a decrease in viscosity, pressure rises at constant temperature cause an increase in viscosity since this causes a decrease in free volume. It is commonly found that ∂h ∂T =0 (3.85) v In other words, if the volume and hence free volume are made constant by increasing pressure as temperature is increased then the viscosity also remains constant. It is in fact found that within the normal processing temperature range for a polymer it is possible to consider an increase in pressure as equivalent, in its effect on viscosity, to a decrease in temperature. For most polymers an increase in pressure of 100 atm is equivalent to a drop of temperature in the range 30–50°C. It is also found that those polymers most sensitive to temperature changes in their normal processing range are the most sensitive to pressure. 3.2.16.6 Weissenberg Effects An elastic aftereffect is generally found in high-molecular-weight ﬂuid materials after extrusion under high shear stress through an oriﬁce or die, and this is seen to happen within a fraction of a second after extrusion. Herzog and Weissenberg [13] observed the existence of a “normal” force in polymers subjected to shear stress. In the polymeric melt systems, the entangled polymer chains get deformed elastically during ﬂow and a different kind of force is generated within the ﬂowing melt in addition to the force applied. The additional force generated is manifested as tensile force perpendicular to the shear plane. This is visualized when a cone with vertical channels and with its axis normal to a plate is rotated with a viscous liquid placed between cone and plate. Liquid will climb into these channels (Figure 3.28a). The arrows in Figure 3.28a indicate that a force exists which acts on the liquid, normal to the shear plane. This force can be measured, without drilling channels into the cone, by placing pressure transducers in contact with the liquid at the cone face and by measuring the pressure exerted on the cone by the liquid. The effect, known as the Weissenberg effect, is more easily demonstrated by subjecting a viscous liquid mass to shear in a coaxial cylinder system by rotating one while keeping the other ﬁxed (Figure 3.28b). It is most easily demonstrated by rotating the inner cylinder. On rotation, the liquid climbs up the rotating inner rod or cylinder to a signiﬁcant height. The Weissenberg effect is clearly manifested in the increase in diameter of extruded proﬁles of a variety of molten polymers. The extrusion swelling (Figure 3.29), more commonly known as die swell, arises probably due to a combination of normal stress effects and a possible elastic recovery consequent to prior compression before the melt or liquid enters the die. A practical aspect of the Weissenberg effect or die swell is that, on extrusion or calendaring, the melt coming out is larger than the die or nip size. The control of size is done by partly using a smaller aperture and by partly having an increased draw down. The extrusion, the swelling may be largely minimized by having a conical die design with the narrow end towards the interior and by lowering of melt temperatures, extrusion rates, or molecular weight of the polymer. 3.2.16.7 Irregular Flow or Melt Fracture Above a certain shear stress considered as the critical shear stress, ﬂow instabilities of many polymer melts exhibited by an abrupt change in the shape of the molten extrudate are found to occur. The irregularities 341 Plastics Properties and Testing Liquid being sheared Plate (a) (b) FIGURE 3.28 Experiments demonstrating the normal force (Weissenberg) effect. Liquid climb on rotation (a) in channels drilled into a cone and (b) in a coaxial cylinder. Die swell Pressure Die Molten polymer FIGURE 3.29 Schematic representation of polymer ﬂow through a die oriﬁce. in ﬂow may be due to Reynolds turbulence or structural turbulence and thermo-mechanical breakdown of the polymer [14]. Above the critical shear stress, the material near the wall relaxes very much faster than the core material leading to the ﬂow irregularities, which, according to Tordella [15], is caused by a fracture of the melt before its entry into the die. Polymers that show ﬂow irregularities at low output rates are those that have comparatively long relaxation times. The origin of ﬂow irregularities and the site of melt fracture is near the entrance to the die since this is the zone of greatest shear stress. The stress enhancement is so great compared to the relaxation time that the polymer melt fractures much like a solid. Melt fracture depends on die geometry, molecular weight, molecular-weight distribution, and chain branching. A linear polymer such as high-density polyethylene is characterized by a higher critical shear stress than the corresponding branched polymer (low-density) polyethylene) of comparable molecular weight. 3.2.17 Measurement of Viscosity Perhaps the most important factor to a process engineer in predicting extrusion or molding behavior is melt viscosity. Several methods are used to obtain the viscosity of polymer solutions and melts experimentally as a function of shear rate [16]. Instruments for making such measurements must necessarily accomplish two things: (1) the ﬂuid must be sheared at measurable rates, and (2) the stress developed must be known. Two kinds of instruments having simple geometry and wide use a rotational viscometer and capillary or extrusion rheometer. 342 Plastics Technology Handbook 3.2.17.1 Rotational Viscometers In a rotational viscometer (using cylinders, cones, spheres, and discs) the ﬂuid is sheared at a given temperature in the ﬂuid is sheared at a given temperature in the annular or enclosed space due to rotation of the inner cylinder or the like device while the outer cylinder or device is kept stationary or vice versa. In either case the torque required for the rotation is a measure of the shearing stress and the speed of rotation gives a measure of the rate of shear. Rotational viscometers using coaxial cylinders (Figure 3.30a) measure relatively low viscosity liquids. Typical is the Haake-Rotovisco. In this device, the cup is stationary and the bob is driven a through a torsion spring. In a cone-and-plate rotational viscometer (Figure 3.30b), the molten polymer is contained between the bottom plate and the cone, which is rotated at a constant velocity (W). Shear stress (t) is deﬁned as t= 3F 2pR3c (3.86) where F is the torque in dynes per centimeter (CGS) or in Newtons per meter (SI), and Rc is the cone _ is given by radius in centimeters or meters. Shear rate (g) g_ = W a (3.87) where W is the angular velocity in degrees per second (CGS) or in radians per second (SI) and a is the cone angle in degrees or radians. Viscosity is then h = t=g_ = 3aF kF = 2pR3c W W (3.88) where (K = 3a=2pR3c ) is a constant deﬁned by viscometer design. An analogous result is obtained if the plate rotates and the cone and plate viscometer is the Weissenberg Rheogoniometer. It consists of a plate that can be rotated at different speeds by means of a constant speed motor-cum-gear assembly (Figure 3.31). The speed or rotation is measured accurately by means of a transducer. A cone is placed concentrically above the plate, the cone angle being around 1–5°. The cone is supported vertically by a frictionless air bearing and is attached to a ﬁrm support through a calibrated torsional spring. Any torque experienced by the cone leads to an equilibrium deﬂection of this spring which is measured by means of a transducer. The polymer sample is placed in the space between the cone Ω h Cone α Polymer Rc R1 (a) FIGURE 3.30 R2 (b) (a) Coaxial cylinder viscometer. (b) Cone and plate rheometer. 343 Plastics Properties and Testing Torsional spring Transducer Ampliﬁers Recorders Air bearing Transducer Constant speed motor FIGURE 3.31 Gear assembly Bearing Scheme of a Weissenberg Rheogoniometer. and plate and the torque experienced by the stationary cone is measured for different rotational speeds of the plate. Relating the shear stress at the cone surface to the measured torque and the shear rate to the angular velocity of the plate, the expression for the viscosity (h) is obtained as h= 3Kqsin a 2pR3p w (3.89) where K is the torsional constant and q is the deﬂection of the spring; Rp is the radius and w is the angular velocity of the plate; and a is the angle of the cone. While q and w are experimentally determined quantities, K and a are obtained by calibration on other materials. The cone and plate viscometer gives reliable experimental data over an extensive range of shear rates (10–4–104 sec−1). Not only can it be used to measure viscosities in simple shear, but it can also be used to determine the dynamic properties of viscoelastic materials. The unit is also set up to measure the normal stresses exhibited by viscoelastics, i.e., those perpendicular to the plane of shear. 3.2.17.2 Capillary Rheometers These rheometers are widely used to study the rheological behavior of molten polymers. As shown in Figure 3.32 the ﬂuid is forced from a reservoir into and through a ﬁne-bore tube, or capillary, by either mechanical or pneumatic means. The ﬂuid is maintained at isothermal conditions by electrical temperature control methods. Either the extrusion pressure or volumetric ﬂow rate can be controlled as the independent variable with the other being the measured dependent variable. Under steady ﬂow and isothermal conditions for an incompressible ﬂuid (assuming only axial ﬂow and no slip at the wall), the viscous force resisting the motion of a column of ﬂuid in the capillary is equal to the applied force tending to move the column in the direction of ﬂow. Thus, t= RDP 2L (3.90) where R and L are the radius and length of the column and DP is the pressure drop across the capillary. The shear stress t is therefore zero at the center of the capillary and increases to a maximum value at the capillary wall. This maximum value is the one generally used for the shear stress in capillary ﬂow. In normal capillary rheometry for polymer melts, the ﬂowing stream exits into the atmosphere, and the driving static pressure in the reservoir is taken to be DP. In such cases, end effects involving viscous and elastic deformations at the entrance and exit of the capillary should be taken into account when calculating the true shear stress at the capillary wall, particularly if the ratio of capillary length to radius (L/R) is small. 344 Plastics Technology Handbook For a ﬂuid showing Newtonian behavior the shear _ at the wall is given by rate (g) P g_ = 4Q=pR3 (3.91) where Q is the volumetric ﬂow rate through the capillary under a pressure drop DP. Melt viscosity is expressed as pR4 DP h = t=g_ = (3.92) 8LQ ΔP L 2R FIGURE 3.32 The measured values of polymer ﬂow taken by capillary rheometers are often presented as plots of shear stress versus shear rate at certain temperatures. These values are called apparent shear stress and apparent shear rate at the tube wall. Corrections must be applied to these values in order to obtain true values. The corrected value of shear stress is determined by the Bagley correction [17] tc = Capillary rheometer (schematic). RDP 2(L + e) (3.93) where tc is the corrected value and e is the length correction expressed as a function of radius. Correction to the shear rate is necessitated by the fact that unlike in isothermal Newtonian ﬂow where the velocity distribution from wall to wall in a tube is parabolic, nonparabolic velocity proﬁle develops in nonNewtonian ﬂow. The Rabinowitsch correction [18] is applied to shear rate to eliminate this error as follows: 3n + 1 g_ (3.94) 4n a where the subscript c stands for corrected value and subscript a stands for apparent value of shear rate at tube wall; the correction term n is given by d log ta =d g_ a and is 1 for Newtonian ﬂow. There are three main reasons why the capillary rheometer is widely used in the plastics industry: (1) shear rate and ﬂow geometry in capillary rheometer are very similar to conditions actually encountered in extrusion and injection modeling; (2) a capillary rheometer typically covers the widest shear rate ranges (10−6 sec−1 to 106 sec−1); and (3) a capillary rheometer provides good practical data and information on the die swell, melt instability, and extrudate defects. g_ c = 3.2.18 Plastics Fractures The principal causes of fracture of a plastic part are the prolonged action of a steady stress (creep rupture), the application of a stress in a very short period of time (impact), and the continuous application of a cyclically varying stress (fatigue). In all cases the process of failure will be accelerated if the plastic is in a aggressive environment. Two basic types of fracture under mechanical stresses are recognized; brittle fracture and ductile fracture. These terms refer to the type of deformation that precedes fracture. Brittle fractures and potentially more dangerous because there occurs no observable deformation of the material. In a ductile failure, on the other hand, large nonrecoverable deformations occur before rupture actually takes place and serve as a valuable warning. A material thus absorbs more energy when it fractures in a ductile fashion than in a brittle fashion. In polymeric materials fracture may be ductile or brittle, depending on several variables, the most important of which are the straining rate, the stress system, and the temperature. Both types of failures may thus be observed in the one material, depending on the service conditions. 345 Plastics Properties and Testing 3.2.19 Impact Behavior of Plastics Tests of brittleness make use of impact tests. The main causes of brittle failure in materials have been found to be (1) triaxiality of stress, (2) high strain rates, and (3) low temperatures. Test methods developed for determining the impact behavior of materials thus involve striking a notched bar with a pendulum. This is the most convenient way of subjecting the material to triaxiality of stress (at the notch tip) and a high strain rate so as to promote brittle failures. The standard test methods are the Charpy and Izod tests, which employ the pendulum principle (Figure 3.33a). The test procedures are illustrated in Figure 3.33b and c. The specimen has a standard notch on the tension side. In the Charpy test the specimen is supported as a simple beam and is loaded at the midpoint (Figure 3.33b). In the Izod test it is supported as a cantilever and is loaded at its end (Figure 3.33c). The standard energy absorbed in breaking the specimen is recorded. The results of impact tests are often scattered, even with the most careful test procedure. A normal practice in such cases is to quote the median strength rather than the average strength, because the median is more representative of the bulk of the results if there is a wide scatter with a few very high or very low results. Impact strengths are normally expressed as Impact = Energy abosorbed to break Area at notch section (ft-lbf/in2, cm-kgf/cm2, or J/m2). Occasionally, the less satisfactory term of energy to break per unit width may be quoted in units of ft-lbf/in, cm-kgf/cm or J/m. The choice of notch depth and tip radius will affect the impact strength observed. A sharp notch is usually taken as a 0.25-mm radius, a blunt notch as a 2-mm radius. The typical variation of impact strength with notch-tip variation for several thermoplastics is presented in Figure 3.34. It is evident that the use of a sharp notch may even rank plastic materials in an order different from that obtained by using a blunt notch. This fact may be explained by considering the total energy absorbed to break the specimen as consisting of energy necessary for crack initiation and for crack propagation. Scale Impact Pendulum Pointer 45° Specimen 0.25–3 mm Radius 0.25–2 mm d = 5–13 mm >d /4 (b) Impact (a) (c) FIGURE 3.33 Impact test. (a) Schematic diagram of Charpy impact testing machine. (b) Arrangement of Charpy impact specimen. (c) Mounting of Izod impact specimen. 346 Plastics Technology Handbook 20 40 High density polyethylene 15 Impact strength (ft-lbf/in2) Dry nylon 10 20 Acetal Impact strength (kJ/m2) 30 PVC ABS 5 n Polystyre e 10 Acr ylic 0 0.5 1.0 1.5 2.0 Notch tip radius (mm) FIGURE 3.34 Variation of impact strength with notch radius for several thermoplastics. When the sharp notch (0.25-mm radius) is used, it may be assumed that the energy necessary to initiate the crack is small, and the main contribution to the impact strength is the propagation energy. On this basis Figure 3.34 would suggest that high-density polyethylene and ABS have relatively high crackpropagation energies, whereas materials such as PVC, nylon, polystyrene, and acrylics have low values. The large improvement in impact strength observed for PVC and nylon when a blunt notch is used would imply that their crack-initiation energies are high. On the other hand, the smaller improvement in the impact strength of ABS with a blunt notch would suggest that the crack-initiation energy is low. Thus the beneﬁt derived from using rounded corners would be much less for ABS than for materials such as nylon or PVC. Temperature has a pronounced effect on the impact strength of plastics. In common with metals, many plastic materials exhibit a transition from ductile behavior to brittle as the temperature is reduced. The variation of impact strength with temperature for several common thermoplastics is shown in Figure 3.35. The ranking of the materials with regard to impact strength is seen to be inﬂuenced by the test temperature. Thus, at room temperature (approximately 20°C) polypropylene is superior to acetal; at subzero temperatures (e.g., −20°C) polypropylene does not perform as well as acetal. This comparison pertains to impact behavior measured with a sharp (0.25-mm) notch. Note that notch sharpness can inﬂuence the impact strength variation with temperature quite signiﬁcantly. Figure 3.36 shows that when a blunt (2-mm) notch is used, there is indeed very little difference between acetal and polypropylene at 20°C, whereas at −20°C acetal is much superior to polypropylene. It may be seen from Figure 3.35 and Figure 3.36 that some plastics undergo a change from ductile or tough (high impact strength) to brittle (low impact strength) behavior over a relatively narrow temperature change. This allows a temperature for ductile-brittle transition to be cited. In other plastic materials this transition is much more gradual, so it is not possible to cite a single value for transition temperature. It is common to quote in such cases a brittleness temperature, TB(1/4). 347 Plastics Properties and Testing 8 0.25-mm notch 15 Nylon (wet) ABS 10 Polypropylene 4 Impact strength (kJ/m2) Impact strength (ft-lbf/in2) 6 Acetal 5 PVC 2 Acrylic 0 –40 –20 0 20 40 Test temperature (°C) FIGURE 3.35 Variation of impact strength with temperature for several thermoplastics with sharp notch. This temperature is deﬁned as the value at which the impact strength of the material with a sharp notch (1/4-mm tip radius) is 10 kJ/m2 (4.7 ft-lbf/in2). When quoted, it provides an indication of the temperature above which there should be no problem of brittle failure. However, it does not mean that a material should never be used below its TB (1/4), because this temperature, by deﬁnition, refers only to the impact behavior with a sharp notch. When the material is unnotched or has a blunt notch, it may still have satisfactory impact behavior well below TB (1/4). Other environmental factors besides temperature may also affect impact behavior. For example, if the material is in the vicinity of a ﬂuid which attacks it, then the crack-initiation energies may be reduced, resulting in lower impact strength. Some materials, particularly nylon, are signiﬁcantly affected by water, as illustrated in Figure 3.37. The absorption of water produces a spectacular improvement in the impact behavior of nylon. Note that the method of making the plastic sample and the test specimen can have signiﬁcant effect on the measured values of the properties of the material. Test specimens may be molded directly or machined from samples which have been compression molded, injection molded, or extruded. Each processing method involves a range of variables, such as melt temperature, mold or die temperature, and shear rate, which inﬂuence the properties of the material. Fabrication defects can affect impact behavior for example, internal voids, inclusion, and additives, such as pigments, which can produce stress concentrations within the material. The surface ﬁnish of the specimen may also affect impact behavior. All these account for the large variation usually observed in the results of testing one material processed and/or fabricated in different ways. It also emphasizes the point that if design data are needed for a particular application, then the test specimen must match as closely as possible the component to be designed. In some applications impact properties of plastics may not be critical, and only a general knowledge of their impact behavior is needed. In these circumstances the information provided in Table 3.1 would be 348 Plastics Technology Handbook 10 2-mm notch 8 20 l 15 6 PVC 10 4 Polypropylene Impact strength (kJ/m2) Impact strength (ft-lbf/in2) eta Ac 5 c Acr yli 2 0 –40 –20 0 20 0 40 Test temperature (°C) FIGURE 3.36 Variation of impact strength with temperature for several thermoplastics with blunt notch. adequate. The table lists the impact behavior of a number of commonly used thermoplastics over a range of temperatures in three broad categories [19]. 3.2.20 Fatigue of Plastics A material subject to alternating stresses over long periods may fracture at stresses much below its maximum strength under static loading (tensile strength) due to the phenomenon called fatigue. Fatigue has been recognized as one of the major causes of fracture in metals. Although plastics are susceptible to a wider range of failure mechanisms, it is likely that fatigue still plays an important part in plastics failure. For metals the fatigue process is generally well understood and is divided into three stages crack initiation, crack growth, and fracture. Fatigue theory of metals is well developed, but the fatigue theory of polymers is not. The completely different molecular structure of polymers means that three is unlikely to be a similar type of crack initiation process as in metals, though it is possible that once a crack is initiated the subsequent phase of propagation and failure may be similar. Fatigue cracks may develop in plastics in several ways. If the plastic article has been machined, surface ﬂaws capable of propagation may be introduced. However, if the article has been molded, it is more probable that fatigue cracks will develop from within the bulk of the material. In a crystalline polymer the initiation of cracks capable of propagation may occur through slip of molecules. In addition to acting as a path for crack propagation, the boundaries of spherulites (see Chapter 1), being areas of weakness, may thus develop cracks during straining. In amorphous polymers cracks may develop at the voids formed during viscous ﬂow. 349 Plastics Properties and Testing 8 0.25-mm notch 4-week immersion 15 10 1-week immersion 4 2-week immersion Impact strength (kJ/tm2) Impact strength (ft-tbf/in2) 6 5 2 Dry –40 FIGURE 3.37 –20 0 20 Test temperature (°C) 40 Effect of water content on impact strength of nylon. TABLE 3.1 Impact Behavior of Common Thermoplastics over a Range of Temperatures Temperature (°C) −20 −10 0 10 20 30 40 50 Polyethylene (low density) A A A A A A A A Polyethylene (high density) Polypropylene B C B C B C B C B B B B B B B B Polystyrene C C C C C C C C Poly(methyl methacrylate) ABS C B C B C B C B C B C B C A C A Plastic Material Acetal B B B B B B B B Teﬂon PVC (rigid) B B A B A B A B A B A B A A A A Polycarbonate B B B B A A A A Poly(phenylene oxide) Poly(ethylene terephthalate) B B B B B B B B B B B B A B A B Nylon (dry) B B B B B B B B Nylon (wet) Glass-ﬁlled nylon (dry) B C B C B C A C A C A C A C A B Polysulfone B B B B B B B B Note: A, tough (specimens do not break completely even when sharply notched); B, notch brittle; C, brittle even when unnotched. 350 Plastics Technology Handbook A number of features are peculiar to plastics, which make their fatigue behavior a complex subject not simply analyzed. Included are viscoelastic behavior, inherent damping, and low thermal conductivity. Consider, for example, a sample of plastic subjected to a cyclic stress of ﬁxed amplitude. Because of the high damping and low thermal conductivity of the material, some of the input energy will be dissipated in each cycle and appear as heat. The temperature of the material will therefore rise, and eventually a stage will be reached when the heat transfer to the surroundings equals the heat generation in the material. The temperature of the material will stabilize at this point until a conventional metal-type fatigue failure occurs. If, in the next test, the stress amplitude is increased to a higher value, the material temperature will rise further and stabilize, followed again by a metal-type fatigue failure. In Figure 3.38, where the stress amplitude has been plotted against the logarithm of the number of cycles to failure, failures of this type have been labeled as fatigue failures. This pattern will be repeated at higher stress amplitudes until a point is reached when the temperature rise no longer stabilizes but continues to rise, resulting in a short-term thermal softening failure in the material. At stress amplitudes above this crossover point there will be thermal failures in an even shorter time. Failures of this type have been labeled as thermal failures in Figure 3.38. The fatigue curves in Figure 3.38 thus have two distinct regimes—one for the long-term conventional fatigue failures, and one for the relatively short-term thermal softening failures. The frequency of the cyclic stress would be expected to have a pronounced effect on the fatigue behavior of plastics, a lower frequency promoting the conventional-type fatigue failure rather than thermal softening failure. Thus it is evident from Figure 3.38 that if the frequency of cycling is reduced, then stress amplitudes which would have produced thermal softening failures at a higher frequency may now result in temperature stabilization and eventually fatigue failure. Therefore, if fatigue failures are required at relatively high stresses, the frequency of cycling must be reduced. Normally, fatigue failures at one frequency on the extrapolated curve fall from the fatigue failures at the previous frequency (Figure 3.38). As the cyclic stress amplitude is further reduced in some plastics, the frequency remaining constant, the fatigue failure curve becomes almost horizontal at large values of the number of stress cycles (N). The stress amplitude at which this leveling off occurs is clearly important for design purposes and is known as the fatigue limit. For plastics in which fatigue failure continues to occur even at relatively low stress amplitudes, it is necessary to deﬁne an endurance limit—that is, the stress amplitude which would not cause fatigue failure up to an acceptably large value of N. 400 35 T F T 300 30 F T T 25 F T T TT F 20 F 200 F F Stress amplitude (Mpa) Stress amplitude (kgf/cm2) T 15 100 103 104 105 106 Log cycles to failure 107 FIGURE 3.38 Typical fatigue behavior of a thermoplastic at several frequencies, F, fatigue failure; T, thermal failure, ○, 5.0 Hz; D, 1.67 Hz; □, 0.5 Hz. (Adapted from Crawford, R. J. 1981. Plastics Engineering, Pergamon, London.) 351 Plastics Properties and Testing 3.2.21 Hardness Hardness of a material may be determined in several ways: (1) resistance to indentation, (2) rebound efﬁciency, and (3) resistance to scratching. The ﬁrst method is the most commonly used technique for plastics. Numerous test methods are available for measuring the resistance of a material to indentation, but they differ only in detail. Basically they all use the size of an indent produced by a hardened steel or diamond indentor in the material as an indication of its hardness—the smaller the indent produced, the harder the material, and so the greater the hardness number. The measured hardness is deﬁned as macroor micro-hardness according to the load applied on the indenter, the load being more than 1 kg for macrohardness and 1 kg or less for microhardness tests. Hardness tests are simple, quick, and nondestructive, which account for their wide use for quality control purposes. 3.2.22 Indentation Hardness The test methods used for plastics are similar to those used for metals. The main difference is that because plastics are viscoelastic allowance must be made for the creep and the time-dependent recovery which occurs as a result of the applied indentation load. 3.2.22.1 Brinell Hardness Number A hardened steel ball 10 mm in diameter is pressed into the ﬂat surface of the test specimen under load of 500 kg for 30 sec. The load is then removed, and the diameter of the indent produced is measured (Figure 3.39). The Brinell hardness number (BHN) for macrohardness is deﬁned as BHN = Load applied to indentor (kgf ) 2P pﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ = Contact area of indentation (mm2 ) pD(D − D2 − d 2 ) (3.95) where D is the diameter of the ball and d is the diameter of the indent. Tables are available to convert the diameter of the indent into BHN. Although the units of Brinell hardness are kgf/mm2, it is quoted only as a number. A disadvantage of the Brinell hardness test when used for plastics is that the edge of the indent is usually not well deﬁned. This problem is overcome in the following test. 3.2.22.2 Vickers Hardness Number The Vickers hardness test differs from the Brinell test in that the indentor is a diamond (square-based) pyramid (Figure 3.39) having an apex angle of 136°. If the average diagonal of the indent is d, the hardness number is calculated from Load applied to indentor (kgf ) P Vickers hardness number = (3.96) = 1:854 2 Contact area of indentation (mm ) d2 Tables are available to convert the average diagonal into Vickers number used for both macro- and microhardnesses. Load P 172°30' 136° D d (a) FIGURE 3.39 (b) d (c) D Indentation hardness tests. (a) Brinell test. (b) Vickers test. (c) Knoop test. 130° 352 Plastics Technology Handbook 3.2.22.3 Knoop Hardness Number The indentor used in the Knoop hardness test for microhardness measurement is a diamond pyramid, but the lengths of the two diagonals, as shown in Figure 3.39, are different. If the long diagonal of the indent is measured as D, the hardness number is obtained from P Knoop hardness number = 14:23 2 (3.97) D Time-dependent recovery of the indentation in plastics is a problem common to all three tests. To overcome this problem, allow a ﬁxed time before making measurements on the indent. 3.2.22.4 Rockwell Hardness Number The Rockwell test used to measure macrohardness differs from the other three tests because the depth of the indent rather than its surface area is taken as a measure of hardness. A hardened steel ball is used as the indentor. A major advantage of the Rockwell test is that no visual measurement of the indentation is necessary, and the depth of the indent is read directly as a hardness value on the scale. The test involves three steps, as shown in Figure 3.40. A minor load of 10 kg is applied on the steel ball, and the scale pointer is set to zero within 10 sec of applying the load. In addition to this minor load, a major load is applied for 15 sec. A further 15 sec after removal of the major load (with the minor load still on the ball), the hardness value is read off the scale. Since creep and recovery effects can inﬂuence readings, it is essential to follow a deﬁned time cycle for the test. Several Rockwell scales (Table 3.2) are used, depending on the hardness of the material under test (Table 3.3). The scale letter is quoted along with the hardness number e.g., Rockwell R60. Scales R and L are used for low-hardness number, e.g., Rockwell R60. Scales R and L are used for low-hardness materials, and scales M and E when the hardness value is high. When the hardness number exceeds 115 on any scale, the sensitivity is lost, so another scale should be used. 3.2.22.5 Barcol Hardness The Barcol hardness tester is a hand-operated hardness measuring device. Its general construction is shown in Figure 3.41. With the back support leg placed on the surface, the instrument is gripped in the hand and its measuring head is pressed ﬁrmly and steadily onto the surface until the instrument rests on the stop ring. The depth of penetration of the spring-loaded indentor is transferred by a lever system to an indicating dial, which is calibrated from 0 to 100 to indicate increasing hardness. To allow for creep, one normally takes readings after 10 sec. The indentor in the Barcol Tester Model No. 934-1 is a truncated steel cone having an included angle of 26° with a ﬂat tip of 0.157 mm (0.0062 in.) in diameter. The values obtained using this instrument are 1 Minor load 2 Minor + major loads 3 Minor load only Measurement taken as indication of hardness FIGURE 3.40 Stages in Rockwell hardness test: 1, minor load; 2, minor and major loads; 3, minor load only. 353 Plastics Properties and Testing TABLE 3.2 Rockwell Hardness Scales Scale Major Load (kg) Dia. of Indentor (in.) R 60 1/2 L 60 1/4 M E 100 100 1/4 1/8 TABLE 3.3 Choice of Hardness Test Methods Based on Modulus Range of Plastics Material Low modulus " # Test Method Rubber Shore A or BS 903 Plasticized PVC Shore A or BS 2782 Low-density polyethylene Medium-density polyethylene Shore D Shore D High-density polyethylene Shore D Polypropylene Toughened polystyrene Rockwell R Rockwell R ABS Rockwell R Polystyrene Poly(methyl methacrylate) Rockwell M Rockwell M High modulus Indicating dial Spring 26° Indentor Lever Support leg 0.006 in. dia. (0.156 mm) Plunger tip FIGURE 3.41 General construction of Barcol hardness tester. found to correlate well to Rockwell values on the M scale. This instrument is used for metals and plastics. Two other models, No. 935 and No. 936, are used for plastics and very soft materials, respectively. 3.2.22.6 Durometer Hardness A durometer is an instrument for measuring hardness by pressing a needle-like indentor into the specimen. Operationally, a durometer resembles the Barcol tester in that the instrument is pressed onto the sample surface until it reaches a stop ring. This forces the indentor into the material, and a system of levers transforms the depth of penetration into a pointer movement on an indicating dial, which is calibrated from 0–100. 354 Plastics Technology Handbook Type A Type B 822 g 10 lbf 1.27 1.27 3 35° 3 2.5 0.8 30° Stop ring FIGURE 3.42 Two types of Shore durometer. TABLE 3.4 Some Typical Hardness Values for Plastics Material Brinell Vickers 4 7 2 6 Polystyrene 25 7 17 M83 76 74 Poly(methyl methacrylate) Poly(vinyl chloride) 20 11 5 9 16 M102 M60 80 90 80 Poly(vinyl chloride-co-vinyl acetate) 20 5 70 60 80 High-density Polyethylene Polypropylene Polycarbonate Nylon Cellulose acetate 12 Knoop Rockwell Barcol R40 R100 14 M75 7 5 15 M70 M75 4 12 M64 Shore D 70 74 70 The two most common types of durometers used for plastics are the Shore Type A and Shore Type D. They differ in the spring force and the geometry of the indentor, as shown in Figure 3.42. Due to creep, readings should be taken after a ﬁxed time interval, often chosen as 10 sec. Typical hardness values of some of the common plastics measured by different test methods are shown in Table 3.4. The indentationbased techniques cannot be applied to soft, rubber-like polymers, for which, particularly when dealing with blends, copolymers, etc., one can use the Fakirov equation: H = 1.97Tg – 571 with H (microhardness) in MPa and Tg (glass transition temperature) in °K. Combining the rule of mixtures (H = SHi ji) and this equation, it is possible to calculate the H-value of materials comprising soft component and/or phase [20]. 3.2.23 Rebound Hardness The energy absorbed when an object strikes a surface is related to the hardness of the surface: the harder the surface, the less the energy absorbed, and the greater the rebound height of the object after impact. Several methods have been developed to measure hardness in this way. The most common method uses a Shore scleroscope, in which the hardness is determined from the rebound height after the impact of a diamond cone dropped onto the surface of the test piece. Typical values of Scleroscope hardness together with the Rockwell M values (in parentheses) for some common plastics are as follows: PMMA 99 (M 102), LDPE 45 (M 25), polystyrene 70 (M 83), and PVC 75 (M 60). 355 Plastics Properties and Testing 3.2.24 Scratch Hardness Basically, scratch hardness is a measure of the resistance the test sample has to being scratched by other materials. The most common way of qualifying this property is by means of the Mohs scale. On this scale various materials are classiﬁed from 1 to 10. The materials used, as shown in Figure 3.43, range from talc (1) to diamond (10). Each material on the scale can scratch the materials that have a lower Mohs number; however, the Mohs scale is not of much value for classifying plastic materials, because most common plastics fall in the 2–3 Mohs range. However, the basic technique of scratch hardness may be used to establish the relative merits of different plastic materials from their ability to scratch one another. 10,000 Diamond 2000 1000 80 Corundum or sapphire 9 Nitrided steels Topaz 8 60 Cutting tools File hard 500 110 100 200 80 100 50 60 40 20 0 Rockwell B 40 20 0 Rockwell C 140 120 Easily machined steels Brinell hardness Orthoclase Apatite 6 5 Fluorite Calcite 4 3 Gypsum 2 Talc 1 Most plastics 40 120 100 5 7 100 60 130 10 Quartz Brasses and aluminum alloys 80 20 10 80 60 40 Rockwell 20 Rockwell M Mohs hardness FIGURE 3.43 Comparison of hardness scales (approximate). 356 Plastics Technology Handbook Scratch hardness is particularly important in plastics used for their optical properties and is usually determined by some of mar-resistance test. In one type of test a specimen is subjected to an abrasive treatment by allowing exposure to a controlled stream of abrasive, and its gloss (specular reﬂection) is measured before and after the treatment. In some tests the light transmission property of the plastic is measured before and after marring. 3.2.25 Stress Corrosion Cracking of Polymers Stress corrosion cracking of polymers occurs in a corrosive environment and also under stress [21,22]. This kind of crack starts at the surface and proceeds at right angles to the direction of stress. The amount of stress necessary to cause stress corrosion cracking is much lower than the normal fracture stress, although there is a minimum stress below which no stress corrosion cracking occurs. The stress corrosion resistance of polymers depends on the magnitude of the stress, the nature of the environment, the temperature, and the molecular weight of the specimen. Ozone cracking is a typical example of stress corrosion cracking of polymers. The critical energy for crack propagation (tc) in ozone cracking varies very little from one polymer to another and is about 100 erg/cm2 (0.1 J/m2). This value is much lower that the tc values for mechanical fracture, which are about 107 erg/cm2 (104 J/m2). In ozone cracking very little energy is dissipated in plastic or viscoelastic deformations at the propagating crack, and that is why tc is about the same as the true surface energy. The only energy supplied to the crack is that necessary to provide for the fresh surfaces due to propagation of the crack, because in ozone cracking chemical bonds at the crack tip are broken by chemical reaction, so no high stress is necessary at the tip. The critical energy tc is about 4,000 erg/cm2 for PMMA in methylated spirits at room temperature, but the value is lower in benzene and higher in petroleum ether. Thus tc in this case is much higher than the true surface energy but still much lower than that for mechanical crack propagation. 3.3 Reinforced Plastics The modulus and strength of plastics can be increased signiﬁcantly by means of reinforcement [23–25]. A reinforced plastic consists of two main components—a matrix, which may be either a thermoplastic or thermosetting resin, and a reinforcing ﬁller, which is mostly used in the form of ﬁbers (but particles, for example glass spheres, are also used). The greater tensile strength and stiffness of ﬁbers as compared with the polymer matrix is utilized in producing such composites. In general, the ﬁbers are the load-carrying members, and the main role of the matrix is to transmit the load to the ﬁbers, to protect their surface, and to raise the energy for crack propagation, thus preventing a brittle-type fracture. The strength of the ﬁber-reinforced plastics is determined by the strength of the ﬁber and by the nature and strength of the bond between the ﬁbers and the matrix. 3.3.1 Types of Reinforcement The reinforcing ﬁller usually takes the form ofﬁbers, since it is in this form that the maximum strengthening of the composite is attained. A wide range of amorphous and crystalline materials can be used as reinforcing ﬁbers, including glass, carbon, asbestos, boron, silicon carbide, and more recently, synthetic polymers (e.g., Kevlar ﬁbers from aromatic polyamides). Some typical properties of these reinforcing ﬁbers are given in Table 3.5. Glass is relatively inexpensive, and in ﬁber form it is the principal form of reinforcement used in plastics. The earliest successful glass reinforcement had a low-alkali calcium–alumina borosilicate composition (E glass) developed speciﬁcally for electrical insulation systems. Although glasses of other compositions were developed subsequently for other applications, no commercial glass better than E glass 357 Plastics Properties and Testing TABLE 3.5 Typical Properties of Reinforcing Fibers Tensile Strength Fiber 3 Density (g/cm ) 4 10 kgf/cm 2 Tensile Modulus GPa 5 10 kgf/cm2 GPa E Glass 2.54–2.56 3.5–3.7 3.4–3.6 7.1–7.7 70–76 Carbon 1.75–2.0 2.1–2.8 2.1–2.8 24.5–40.8 240–400 Asbestos Boron 2.5–3.3 2.6 2.1–3.6 3.0–3.6 2.1–3.5 3.0–3.5 14.3–19.4 40.8–45.9 140–190 400–450 Silicon carbide 3.2–3.4 3.0–3.7 3.0–3.6 46.9–50.0 460–490 Kevlar-49 1.45 3.0–3.6 3.0–3.6 13.2 130 Source: Crawford, R. J. 1981. Plastic Engineering, Pergamon, London. has been found for plastics reinforcement. However, certain special glasses having extra high-strength properties or modulus have been produced in small quantities for speciﬁc applications (e.g., aerospace technology). Glass ﬁbers are usually treated with ﬁnishes. The function of a ﬁnish is to secure good wetting and to provide a bond between the inorganic glass and the organic resin. The most important ﬁnishes are based on silane compounds—e.g., vinyltrichlorosilane or vinyltriethoxysilane. 3.3.2 Types of Matrix The matrix in reinforced plastics may be either a thermosetting or thermoplastic resin. The major thermosetting resins used in conjunction with glass-ﬁber reinforcement are unsaturated polyester resins and, to a lesser extent, epoxy resins. These resins have the advantage that they can be cured (cross-linked) at room temperature, and no volatiles are liberated during curing. Among thermoplastic resins used as the matrix in reinforced plastics, the largest tonnage group is the polyoleﬁns, followed by nylon, polystyrene, thermoplastic polyesters, acetal, polycarbonate, and polysulfone. The choice of any thermoplastic is dictated by the type of application, the service environment, and the cost. 3.3.3 Analysis of Reinforced Plastics Fibers exert their effect by restraining the deformation of the matrix while the latter transfers the external loading to the ﬁbers by shear at the interface. The resultant stress distributions in the ﬁber and the matrix tend to be complex. Theoretical analysis becomes further complicated because ﬁber length, diameter, and orientation are all factors. A simpliﬁed analysis follows for two types of ﬁber reinforcement commonly used, namely, (1) continuous ﬁbers and (2) discontinuous ﬁbers. 3.3.3.1 Continuous Fibers We will examine what happens when a load is applied to an ideal ﬁber composite in which the matrix material is reinforced by ﬁbers which are uniform, continuous, and arranged uniaxially, as shown in Figure 3.44a. Let us assume that the ﬁbers are gripped ﬁrmly by the matrix so that there is no slippage at the ﬁbermatrix interface and both phases act as a unit. Under these conditions the strains in the matrix and in the ﬁber under a load are the same (Figure 3.44b), and the total load is shared by the ﬁber and the matrix: Pc = Pm + Pf (3.98) where P is the load and the subscripts c, m, and f refer, respectively, to composite, matrix and ﬁber. 358 Plastics Technology Handbook Fib e r Load ΔL Fiber L σf Stress Matrix Matr ix σm (a) Strain, (ΔL/L) (b) (c) FIGURE 3.44 (a) Continuous-ﬁber reinforced composite under tensile load. (b) Iso-strain assumption in a composite. (c) Arrangement of ﬁbers in a cross-plied laminate. Since the load P = sA, Equation 3.98, expressed in terms of stresses (s) and cross-sectional areas (A), becomes sc Ac = sm Am + sf Af Rearranging gives sc = sm Am Ac + sf Af Ac (3.99) (3.100) Since the ﬁbers run throughout the length of the specimen, the ratio Am/Ac can be replaced by the volume fraction Fm = Vm/Vc, and similarly Af/Ac by Ff. Equation 3.100 thus becomes sc = sm Fm + sf Ff (3.101) Equation 3.101 represents the rule of mixture for stresses. It is valid only for the linear elastic region of the stress–strain curve (see Figure 3.2). Since Fm + Ff = 1, we can write sc = sm (1 − Ff ) + sf Ff (3.102) Since the strains on the components are equal, ec = em = ef (3.103) Equation 3.101 can now be rewritten to give the rule of mixture for moduli Ec ec = Em em Fm + Ef ef Ff i.e., Ec = Em Fm + Ef Ff (3.104) Equation 3.103 also affords a comparison of loads carried by the ﬁber and the matrix. Thus for elastic deformation Equation 3.103 can be rewritten as sc sm sf = = Ec Em Ef or Pf E = r Pm Em Af Am = Ef Em Ff Fm (3.105) Because the modulus of ﬁbers is usually much higher than that of the matrix, the load on a composite will therefore be carried mostly by its ﬁber component (see Example 3.4). However, a critical volume 359 Plastics Properties and Testing fraction of ﬁbers (Fcrit) is required to realize matrix reinforcement. Thus for Equation 3.102 and Equation 3.105 to be valid, Ff > Fcrit. The efﬁciency of reinforcement is related to the ﬁber direction in the composite and to the direction of the applied stress. The maximum strength and modulus are realized in a composite along the direction of the ﬁber. However, if the load is applied at 90° to the ﬁlament direction, tensile failure occurs at very low stresses, and this transverse strength is not much different than the matrix strength. To counteract this situation, one uses cross-plied laminates having alternate layers of unidirectional ﬁbers rotated at 90°, as shown in Figure 3.44c. (A more isotropic composite results if 45° plies are also inserted.) The stress–strain behavior for several types of ﬁber reinforcement is compared in Figure 3.45. As already noted, if the load is applied perpendicularly to the longitudinal direction of the ﬁbers, the ﬁbers exert a relatively small effect. The strains in the ﬁbers and the matrix are then different, because they act independently, and the total deformation is thus equal to the sum of the deformations of the two phases. Vc ec = Vm em + Vf ef (3.106) Dividing Equation 3.106 by Vc and applying Hooke’s law, since the stress is constant, give s sFm sFf = + Em Ef Ec (3.107) Dividing by s and rearranging, we get Ec = Em Ef Em Ff + Ef Fm (3.108) The ﬁber composite thus has a lower modulus in transverse loading than in longitudinal loading. 0° 45° 90° Unidirectional 0° Bidirectional Random 0° 90° Stress 0°,45°,90° 45° 45° 45° 0 FIGURE 3.45 Strain Stress–strain behavior for several types of ﬁber reinforcement. 360 Plastics Technology Handbook Example 4: A unidirectional ﬁber composite is made by using 75% by weight of E glass continuous ﬁbers (sp. gr. 2.4) having a Young’s modulus of 7 × 105 kg/cm2 (68.6 GPa), practical fracture strength of 2 × 104 kg/cm2 (1.9 GPa), and an epoxy resin (sp. gr. 1.2) whose modulus and tensile strength, on curing, are found to be 105 kg/cm2 (9.8 GPa) and 6 × 102 kg/cm2 (58.8 MPa), respectively. Estimate the modulus of the composite, its tensile strength, and the fractional load carried by the ﬁber under tensile loading. What will be the value of the modulus of the composite under transverse loading? Answer: Volume fraction of glass ﬁbers (Ff) = 0:75=2:4 = 0:60 0:75=2:4 + 0:25=1:2 Fm = 1 − 0:6 = 0:4 From Equation 3.104, Ec = 0:4(105 ) + 0:6(7 105 ) = 4:6 105 kg=cm2 (45 GPa) From Equation 3.101, sc = 0:4(6 102 ) + 0:6(2 104 ) = 1:22 104 kg=cm2 (1:2 GPa) Equation 3.105, on rearranging, gives Pf Et 7 105 = Ff = 0:6 = 0:91 Pc Ec 4:6 105 Thus, nearly 90% of the load is carried by the ﬁber, and the weakness of the plastic matrix is relatively unimportant. For transverse loading, from Equation 3.108, Ec = 105 (7 105 ) = 2 105 kg=cm2 (19:6 GPa) 0:6 105 + 0:4(7 105 ) Equation 3.102 and Equation 3.104 apply to ideal ﬁber composites having uni-axial arrangement of ﬁbers. In practice, however, not all the ﬁbers are aligned in the direction of the load. This practice reduces the efﬁciency of the reinforcement, so Equation 3.102 and Equation 3.104 are modiﬁed to the forms sc = sm (1 − ff ) + k1 sf ff (3.109) Ec = Em (1 − ff ) + k2 Ef ff (3.110) If the ﬁbers are bi-directional (see Figure 3.45), then the strength and modulus factors, k1 and k2, are about 0.3 and 0.65, respectively. 3.3.3.2 Discontinuous Fibers If the ﬁbers are discontinuous, the bond between the ﬁber and the matrix is broken at the ﬁbers ends, which thus carry less stress than the middle part of the ﬁber. The stress in a discontinuous ﬁber therefore varies along its length. A useful approximation pictures the stress as being zero at the end of the ﬁller and as reaching the maximum stress in the ﬁber at a distance from the end (Figure 3.46a). 361 Plastics Properties and Testing Fiber d l σMax Stress in ﬁber F3 F1 F2 dx (a) lc/2 lc/2 τ σ Fiber Tensile stress σ (b) σ Matrix d l FIGURE 3.46 Composite reinforced with discontinuous ﬁbers. (a) A total length lc, at the two ends of a ﬁber carries less than the maximum stress. (b) Interfacial strength of the matrix ﬁber. The length over which the load is transferred to the ﬁber is called the transfer length. As the stress on the composite is increased, the maximum ﬁber stress as well as the transfer length increase, as shown in Figure 3.46a, until a limit is reached, because the transfer regions from the two ends meet at the middle of the ﬁber (and so no further transfer of stress can take place), or because the ﬁber fractures. For the latter objective to be reached, so as to attain the maximum strength of the composite, the ﬁber length must be greater than a minimum value called the critical ﬁber length, lc. Consider a ﬁber of length l embedded in a polymer matrix, as shown in Figure 3.46b. One can then write, equating the tensile load on the ﬁber with the shear load on the interface, spd 2 = tpdl 4 (3.111) where s is the applied stress, d is the ﬁber diameter, and t is the shear stress at the interface. The critical ﬁber length, lc, can be derived from a similar force balance for an embedded length of lc/2. Thus, lc = sff d 2ti and lc sff = d 2ti (3.112) where sff is the ﬁber strength and ti is the shear strength of the interface or the matrix, whichever is smaller. So if the composite is to fail through tensile fracture of the ﬁber rather than shear failure due to matrix ﬂow at the interface between the ﬁber and the matrix, the ratio lc/d, known as the critical aspect ratio, must be exceeded, or, in other words, for a given diameter of ﬁber, d, the critical ﬁber length, lc, must be exceeded. 362 Plastics Technology Handbook If the ﬁber length is less than lc, the matrix will ﬂow around the ﬁber, and maximum transfer of stress from matrix to ﬁber will not occur. Using Equation 3.112, we can estimate the value of lc/d from the values of sff and ti, and vice versa. Typical values of lc/d for glass ﬁber and carbon ﬁber in an epoxy resin matrix are 30–100 and 70, respectively. If the ﬁbers are discontinuous, then, since the stress is zero at the end of the ﬁber, the average stress in the ﬁbers will be less than the value sfmax, which it would have achieved if the ﬁbers had been continuous over the whole length of the matrix. The value of the average stress will depend on the stress distribution in the end portions of the ﬁbers and also on their lengths. If the stress distributions are assumed to be as shown in Figure 3.46a, then the average stress in the ﬁbers may be obtained as follows. Considering a differential section of the ﬁber as shown in Figure 3.46a, we obtain pd 2 F1 = sf 4 dsf pd 2 F2 = sf + dx dx 4 F3 = tpd dx For equilibrium, F1 = F2 + F3 so sf pd 2 = 4 2 ds pd sf + f dx + tpd dx dx 4 d ds = −t dx 4 f Integrating gives d 4 s ð1 (3.113) ðx dsf = − 0 t dx l=2 sf = 4t(l=2 − x) d (3.114) Three cases may now be considered. 3.3.3.3 Fiber Length Less than lc In this case the peak stress occurs at x = 0 (Figure 3.47a). So from Equation 3.114, 2tl d The average ﬁber stress is obtained by dividing the area of the stress-ﬁber length diagram by the ﬁber length; that is, sf = f = s (l=2)2tl=d tl = l d The stress, sc, in the composite is now obtained from Equation 3.109 sc = sm (1 − Ff ) + tlk1 f d f (3.115) 363 Plastics Properties and Testing l < lc l = lc C Stress Stress σf max σf max C Stress σf max σf 2τlc d σf 2τl d σf l > lc C lc /2 lc /2 (a) FIGURE 3.47 l l (b) (c) l Stress variation for short and long ﬁbers. 3.3.3.4 Fiber Length Equal to lc In this case the peak value of stress occurs at x = 0 and is equal to the maximum ﬁber stress (Figure 3.47b). So 2tl sf = sfmax = c (3.116) d f = Average fiber stress = s 1 lc (2tlc =d) 2 lc i.e., f = s tlc d So from Equation 3.109, sc = sm (1 − ff ) + k1 tlc ff d (3.117) 3.3.3.5 Fiber Length Greater than lc 1. For l/2 > x > (l − lc)/2 (Figure 3.47c), 4t 1 l−x sf = d 2 2. For (l − lc)/2 > x > 0 (Figure 3.47c), sf = constant = sfmax = 2tlc d The average ﬁber stress, from the area under the stress-ﬁber length graph is f = s (lc =2)sfmax + (l − lc )sfmax = l lc 1− s 2l fmax (3.118) 364 Plastics Technology Handbook So from Equation 3.109, l sc = sm (1 − ff ) + k1 ff 1 − c ffmax : 2l (3.119) It is evident from Equation 3.118 that to get the average ﬁber stress as close as possible to the maximum ﬁber stress, the ﬁbers must be considerably longer than the critical length. At the critical length the average ﬁber stress is only half of the maximum ﬁber stress, i.e., the value achieved in continuous ﬁbers (Figure 3.47c). Equations such as Equation 3.119 give satisfactory agreement with the measured values of strength and modulus for polyester composites reinforced with chopped strands of glass ﬁbers. These strength and modulus values are only about 20%–25% of those achieved by reinforcement with continuous ﬁbers. Example 5: Calculate the maximum and average ﬁber stresses for glass ﬁbers of diameter 15 mm and length 2 mm embedded in a polymer matrix. The matrix exerts a shear stress of 40 kgf/cm2 (3.9 MPa) at the interface, and the critical aspect ratio of the ﬁber is 50. Answer: lc = 50 15 10−3 = 0:75 mm Since l > lc, then smax = 2tlc = 2 40 50 = 4 103 kgf =cm2 ( = 392 MPa) d Also, lc 0:75 f = 1 − s (4 103 ) = 3:25 103 kgf =cm2 ( = 318 MPa) s = 1− 22 2l fmax 3.3.4 Deformation Behavior of Fiber-Reinforced Plastic As we have seen, the presence of ﬁbers in the matrix has the effect of stiffening and strengthening it. The tensile deformation behavior of ﬁber-reinforced composites depends largely on the direction of the applied stress in relation to the orientation of the ﬁbers, as illustrated in Figure 3.45. The maximum strength and modulus are achieved with unidirectional ﬁber reinforcement when the stress is aligned with the ﬁbers (0°), but there is no enhancement of matrix properties when the stress is applied perpendicular to the ﬁbers. With random orientation of ﬁbers the properties of the composite are approximately the same in all directions, but the strength and modulus are somewhat less than for the continuous-ﬁber reinforcement. In many applications the stiffness of a material is just as important as its strength. In tension the stiffness per unit length is given the product EA, where E is the modulus and A is the cross-sectional area. When the material is subjected to ﬂexure, the stiffness per unit length is a function of the product EI, where l is the second moment of area of cross section (see Example 3.2). Therefore the stiffness in both tension and ﬂexure increases as the modulus of the material increases, and the advantages of ﬁber reinforcement thus become immediately apparent, considering the very high modulus values for ﬁbers. 365 Plastics Properties and Testing 3.3.5 Fracture of Fiber-Reinforced Plastics Although the presence of the reinforcing ﬁbers enhances the strength and modulus properties of the base material, they also cause a complex distribution of stress in the materials. For example, even under simple tensile loading, a triaxial stress system is set up since the presence of the ﬁber restricts the lateral contraction of the matrix. This system increases the possibility of brittle failure in the material. The type of fracture which occurs depends on the loading conditions and ﬁber matrix bonding. 3.3.5.1 Tension With continuous-ﬁber reinforcement it is necessary to break the ﬁbers before overall fracture can occur. The two different of fracture which can occur in tension are shown in Figure 3.48. It is interesting to note that when an individual ﬁber in a continuous-ﬁber composite breaks, it does not cease to contribute to the strength of the material, because the broken ﬁber then behaves like a long short ﬁber and will still be supporting part of the external load at sections remote from the broken end. In short-ﬁber composites, however, ﬁber breakage is not an essential prerequisite to complete composite fracture, especially when the interfacial bond is weak, because the ﬁbers may then be simply pulled out of the matrix as the crack propagates through the latter. 3.3.5.2 Compression In compression the strength of glass-ﬁber reinforced plastics is usually less than in tension. Under compressive loading, shear stresses are set up in the matrix parallel to the ﬁbers. The ﬁber aligned in the loading direction thus promote shear deformation. Short-ﬁber reinforcement may therefore have advantages over continuous ﬁbers in compressive loading because in the former not all the ﬁbers can be aligned, so the ﬁbers which are inclined to the loading plane will resist shear deformation. If the matrixﬁber bond is weak, debonding may occur, causing longitudinal cracks in the composite and buckling failure of the continuous ﬁbers. Matrix Fiber (a) (b) FIGURE 3.48 Typical fracture modes in ﬁber-reinforced plastics. (a) Fracture due to strong interfacial bond. (b) Jagged fracture due to weak interfacial bond. 366 Plastics Technology Handbook 3.3.5.3 Flexure or Shear In ﬂexure or shear, as in the previous case of compression, plastics reinforced with short ﬁbers are probably better than those with continuous ﬁbers, because in the former with random orientation of ﬁbers at least some of the ﬁbers will be correctly aligned to resist the shear deformation. However, with continuousﬁber reinforcement if the shear stresses are on planes perpendicular to the continuous ﬁbers, then the ﬁbers will offer resistance to shear deformation. Since high volume fraction (ff) can be achieved with continuous ﬁbers, this resistance can be substantial. 3.3.6 Fatigue Behavior of Reinforced Plastics Like unreinforced plastics, reinforced plastics are also susceptible to fatigue. There is, however, no general rule concerning whether glass reinforcement enhances the fatigue endurance of the base material. In some cases the unreinforced plastic exhibits greater fatigue endurance than the reinforced material; in other cases, the converse is true. In short-ﬁber glass-reinforced plastics, cracks may develop relatively easily at the interface close to the ends of the ﬁbers. The cracks may then propagate through the matrix and destroy its integrity long before fracture of the composite takes place. In many glass-reinforced plastics subjected to cyclic tensile stresses, debonding many occur after a few cycles even at modes stress levels. Debonding may be followed by resin cracks at higher stresses, leading eventually to the separation of the component due to intensive localized damage. In other modes of loading, e.g., ﬂexure or torsion, the fatigue endurance of glass-reinforced plastics is even worse than in tension. In most cases the fatigue endurance is reduced by the presence of moisture. Plastics reinforced with carbon and boron, which have higher tensile moduli than glass, are stiffer than glass-reinforced plastics and are found to be less vulnerable to fatigue. 3.3.7 Impact Behavior of Reinforced Plastics Although it might be expected that a combination of brittle reinforcing ﬁbers and a brittle matrix (e.g., epoxy or polyester resins) would have low impact strength, this is not the case, and the impact strengths of the ﬁbers or the matrix. For example, polyester composites with chopped-strand mat have impact strengths from 45 to 70 ft-lbf/in2 (94–147 kJ/m2), whereas a typical impact strength for polyster resin is only about 1 ft-lbf/in2 (2.1 kJ/m2). The signiﬁcant improvement in impact strength by reinforcement is explained by the energy required to cause debonding and to overcome friction in pulling the ﬁbers out of the matrix. It follows from this that impact strengths would be higher if the bond between the ﬁber and the matrix is relatively weak, because if the interfacial bond is very strong, impact failure will occur by propagation of cracks across the matrix and ﬁbers requiring very little energy. It is also found that in short-ﬁber-reinforced plastics the impact strength is maximum when the ﬁber length has the critical value. The requirements for maximum impact strength (i.e., short ﬁber and relatively weak interfacial bond) are thus seen to be contrary to those for maximum tensile strength (long ﬁbers and strong bond). The structure of a reinforced plastic material should therefore be tailored in accordance with the service conditions to be encountered by the material. 3.4 Electrical Properties The usefulness of an insulator or dielectric ultimately depends on its ability to act as a separator for points across which a potential difference exists. This ability depends on the dielectric strength of the material, which is deﬁned as the maximum voltage gradient that the material withstands before failure or loss of the material’s insulating properties occurs. Plastics Properties and Testing 367 Besides permittivity (dielectric constant), dielectric losses, and dielectric strength, another property used to deﬁne the dielectric behavior of a material is the insulation resistance, i.e., the resistance offered by the material to the passage of electric current. This property may be important in almost all applications of insulators. This resistivity (i.e., reciprocal of conductivity) of a plastic material with a perfect structure would tend to be inﬁnite at low electric ﬁelds. However, the various types of defects which occur in plastics may acts as sources of electrons or ions which are free to contribute to the conductivity or that can be thermally activated to do so. These defects may be impurities, discontinuities in the structure, and interfaces between crystallites and between crystalline and amorphous phases. Common plastics therefore have ﬁnite, though very high, resistivities from 108 to 1020 ohm-cm. These resistivity values qualify them as electrical insulators. Polymeric materials have also been produced which have relatively large conductivities and behave in some cases like semiconductors and even photoconductors [26]. For example, polyphenylacetylene, polyaminoquinones, and polyacenequinone radical polymers have been reported with resistivities from 103 to 108 ohm-cm. It has been suggested that the conductivity in these organic semiconductors is due to the existence of large number of unpaired electrons, which are free within a given molecule and contribute to the conduction current by hopping (tunneling) from one molecule to an adjacent one (see “Electroactive Polymers” in Chapter 5). 3.4.1 Dielectric Strength Dielectric strength is calculated as the maximum voltage gradient that an insulator can withstand before puncture or failure occurs. It is expressed as volts (V) per unit of thickness, usually per mil (1 mil = 1/1,000 in.). Puncture of an insulator under an applied voltage gradient results from small electric leakage currents which pass through the insulator due to the presence of various types of defects in the material. (Note that only a perfect insulator would be completely free from such leakage currents.) The leakage currents warm the material locally, causing the passage of a greater current and greater localized warming of the material, eventually leading to the failure of the material. The failure may be a simple puncture in the area where material has volatilized and escaped, or it may be a conducting carbonized path (tracking) that short circuits the electrodes. It is obvious from the cause of dielectric failure that the measured values of dielectric strength will depend on the magnitude of the applied electric ﬁeld and on the time of exposure to the ﬁeld. Since the probability of a ﬂaw and a local leakage current leading ultimately to failure increases with the thickness of the sample, dielectric strength will also be expected to depend on the sample thickness. The measurement of dielectric strength (Figure 3.49a) is usually carried out either by the short-time method or by the step-by-step method. In the former method the voltage is increased continuously at a uniform rate (500 V/sec) until failure occurs. Typically, a 1/8-in. thick specimen requiring a voltage of about 50,000 V for dielectric failure will thus involve a testing period of 100 sec or so. In step-by-step testing, deﬁnite voltages are applied to the sample for a deﬁnite time (1 min), starting with a value that is half of that obtained by short-time testing, with equal increments of 2,000 V until failure occurs. Since step-by-step testing provides longer exposure to the electric ﬁeld, dielectric strength values obtained by this method are lower than those obtained by the short-time test. Conditions to stepby-step testing correspond more nearly with those met in service. Even so, service failure almost invariably occurs at voltages below the measure dielectric strength. It is thus necessary to employ a proper safety factor to provide for the discrepancy between test and service conditions. Increase in thickness increases the voltage required to give the same voltage gradient, but the probability of a ﬂaw and a local leakage current leading ultimately to failure also increases. The breakdown voltage increases proportionally less than thickness increases, and as a result the dielectric strength of a material decreases with the thickness of specimen (Figure 3.49b). For this reason, testing of insulation plastic should be done with approximately the thickness in which it is to be used in service. 368 Electrodes Test piece Breakdown voltage (v) 15,000 FIGURE 3.49 (b) 4000 2000 5000 Dielectric strength 0 (a) down Break To 450 v per mil at 1"/8 in 10,000 2 4 6 8 10 Thickness (mils) 0 12 Dielectric strength (volts/mil) Plastics Technology Handbook (a) Dielectric strength test. (b) Dependence of dielectric strength on thickness of sample. It is seen from Figure 3.49b that the dielectric strength increases rapidly with decreasing thickness of the sample. A rule of thumb is that the dielectric strength varies inversely with the 0.4 power of the thickness. For example, if the dielectric strength of poly(vinyl chloride) plastic is 375 V/mil in a thickness of 0.075 in., it would be 375(75/15)0.4 or about 700 V/mil in foils only 15 mils thick. The fact that thin foils may have proportionally higher dielectric strength is utilized in the insulation between layers of transformer turns. The dielectric strength of an insulation material usually decreases with increase in temperature and is approximately inversely proportional to the absolute temperature. But the converse is not observed, and below room temperature dielectric strength is substantially independent of temperature change. Mechanical loading has a pronounced effect on dielectric strength. Since a mechanical stress may introduce internal ﬂaws which serve as leakage paths, mechanically loaded insulators may show substantially reduced values of dielectric strength. Reductions up to 90% have been observed. Dielectric strength of an insulating material is inﬂuenced by the fabrication detail. For example, ﬂow lines in a compression molding or weld lines in an injection molding may serve as paths of least resistance for leakage currents, thus reducing the dielectric strength. Even nearly invisible minute ﬂaws in a plastic insulator may reduce the dielectric strength to one-third its normal value. 3.4.2 Insulation Resistance Test piece Guard Galv. Electrode Applied voltage FIGURE 3.50 Insulation resistance test. The resistance offered by an insulating material to the electric current is the composite effect of volume and surface resistances, which always act in parallel. Volume resistance is the resistance to leakage of the electric current through the body of the material. It depends largely on the nature of the material. But surface resistance, which is the resistance to leakage along the surface of a material, is largely a function of surface ﬁnish and cleanliness. Surface resistance is reduced by oil or moisture on the surface and by surface roughness. On the other hand, a very smooth or polished surface gives greater surface resistance. A three-electrode system, as shown in Figure 3.50, is used for measurement of insulation resistance. In this way the surface and volume leakage currents are separated. The applied voltage must be well below the dielectric strength of the material. Thus, in practice, a voltage gradient less than 30 V/mil is applied. From the applied voltage and the leakage current, the leakage resistance is computed. Since 369 10–5 FIGURE 3.51 1 105 1010 Speciﬁc resistance (ohm/cm) Porcelain Polyethylene Mica Polytetraﬂuoroethylene Amber Sulfur Polystyrene Fused quartz Glass Cellulose acetate Pure water Rubber Sea water Graphite Gold Copper Iron Nichrome Plastics Properties and Testing 1015 The resistivity spectrum. the measured value depends, among other things, on the time during which the voltage is applied, it is essential to follow a standardized technique, including preconditioning of the specimen to obtain consistent results. The insulation resistance of a dielectric is represented by its volume resistivity and surface resistivity. The volume resistivity (also known as speciﬁc volume resistance) is deﬁned as the resistance between two electrodes covering opposite faces of a centimeter cube. The range of volume resistivities of different materials including plastics is shown in Figure 3.51. Values for plastics range from approximately 1010 ohm-cm for a typical cellulose acetate to abut 1019 ohm-cm for a high-performance polystyrene. The surface resistivity (also known as speciﬁc surface resistance) is deﬁned as the resistance measured between the opposite edges of the surface of a material having an area of 1 cm2 It ranges from 1010 ohm for cellulose acetate to 1014 ohm for polystyrene. The insulation resistance of most plastic insulating materials is affected by temperature and the relative humidity of the atmosphere. The insulation resistance falls off appreciably with an increase in temperature or humidity. Even polystyrene, which has very high insulation resistance at room temperature, becomes generally unsatisfactory above 80°C (176°F). Under these conditions polymers like polytetraﬂuoroethylene and polychlorotriﬂuoroethylene are more suitable. Plastics that have high water resistance are relatively less affected by high humidities. 3.4.3 Arc Resistance The arc resistance of a plastic is its ability to withstand the action of an electric are tending to form a conducting path across the surface. In applications where the material is subject to arcing, such as switches, contact bushes, and circuit breakers, resistance to arc is an important requirement. Arcing tends to produce a conducting carbonized path on the surface. The arc resistance of an insulator may be deﬁned as the time in seconds that an arc may play across the surface without burning a conducting path. A schematic of an arc-resistance test is shown in Figure 3.52. Plastics that carbonize easily (such as phenolics) have relatively poor arc resistance. On the other hand, there are plastics (such as methacrylates) that do not carbonize, although they would decompose and give off combustible gases. There would thus be no failure in the usual sense. Special arc-resistant formulations involving noncarbonizing mineral ﬁllers are useful for certain applications. But when service conditions are severe in this respect, ceramics ought to be used, because they generally have much better arc resistance than organic plastics. Related to arc resistance is ozone resistance. This gas is found in the atmosphere around high-voltage equipment. Ignition cable insulation, for example, should be ozone resistant. Natural rubber is easily 370 Plastics Technology Handbook Electrodes Arc between electrodes Test piece FIGURE 3.52 Arc-resistance test. attacked and deteriorated by ozone. Fortunately, most synthetic resins have good ozone resistance and are satisfactory from this point of view. 3.4.4 Dielectric Constant The effect of a dielectric material in increasing the charge storing capacity of a capacitor can be understood by considering the parallel-plate type sketched in Figure 3.53. If a voltage V is applied across two metal plates, each of area A m2, separated by a distance, d m, and held parallel to each other in vacuum, the electric ﬁeld established between the plates (Figure 3.53a) is E= −V d (3.120) The charge density, Q0/A, where Q0 is the total charge produced on the surface area A of each plate, is directly proportional to the electric ﬁeld. Q0 V = −e0 E = e0 A d (3.121) Ae0 V = C0 V d (3.122) or Q0 = Bound charge + + V – Q = Q0 + Q ' Area of plate (A) Charge (Q0) + + + + + + + + + + E = –V/d d (Vacuum) – (a) – Free charge – + + + + + + + + + + + + + + + + + + Net negative charge, –Q' at surface Region of no net charge (b) FIGURE 3.53 Schematic illustration of the effect of dielectric material in increasing the charge storing capacity of a capacitor. 371 Plastics Properties and Testing The proportionality constant, e0, is called the dielectric constant (or permittivity) of a vacuum. It has units of Q =A coul=m2 coul=V farad e0 = 0 = = = V=m V=d m m and a value of 8.854 × 10−12 farad/m. The quantity C0 in Equation 3.122 is the capacitance of a capacitor (condenser) with a vacuum between its plates. It can be deﬁned as the ratio of the charge on either of the plates to the potential differences between the plates. Now if a sheet of a dielectric material is inserted between the plates of a capacitor (Figure 3.53b), an increased charge appears on the plates for the some voltage, due to polarization of the dielectric. The applied ﬁeld E causes polarization of the entire volume of the dielectric and thus gives rise to induced charges, or bound charges, Q′, at its surface, represented by the ends of the dipole chains. These induced charges may be pictured as neutralizing equal charges of opposite signs on the metal plates. If one assumes, for instance, that the induced charge −Q′ neutralizes an equal positive charge in the upper plate of the capacitor (Figure 3.53b), the total charge stored in the presence of the dielectric is Q = Q0 + Q′. The ratio of the total charge Q to the free charge Q0 (which is not neutralized by polarization) is called the relative dielectric constant or relative permittivity, er, and is characteristic of the dielectric material. er = Total charge Q = Free charge Q0 (3.123) Obviously, er is always greater than unity and has no dimensions. For most materials er exceeds 2 (Table 3.6). Dividing both the numerator and denominator of Equation 3.123 by the applied voltage V and applying the deﬁnition of C from Equation 3.122, we obtain er = e C = e0 C0 (3.124) The relative dielectric constant or relative permittivity is thus deﬁned as the ratio of the capacitance of a condense with the given material as the dielectric to that of the same condenser without the dielectric. TABLE 3.6 Dielectric Properties of Electrical Insulators tan d Material er at 60 Hz 60 Hz 106 Hz Dielectirc Strengtha (V/mil) Ceramics Porcelain 6 0.010 – Alumina Zircon 9.6 9.2 – 0.035 <0.0005 0.001 – Soda-lime 7 0.1 0.01 – Fused silica Mica 4 7 0.001 – 0.0001 0.0002 – 3,000–6,000 (1–3 mil specimen) <0.0005 0.0002 – 0.0003 450–1,000 300–1,000 – 300–1,000 200–300 60–290 Polymers Polyethylene Polystyrene 2.3 2.5 Polyvinyl chloride 7 Nylon-6, 6 Teﬂon 4 2.1 a Specimen thickness 1/8 in. 0.1 0.02 <0.0001 0.03 – 300–400 400 372 Plastics Technology Handbook (The dielectric constant for air is 1.0006. It is usually taken as unity—that is, the same as a vacuum—and the relative dielectric constant is referred to simply as the dielectric constant). The dielectric susceptibility, c, is deﬁned as c = er − 1 = e − e0 e0 (3.125) It thus represents the part of the total dielectric constant which is a consequence of the material. From Equation 3.123, c= Bound charge Free charge The magnitude of the induced or bound charge Q′ per unit area is the polarization, P, which has the same units as charge density. Therefore, P= Q0 Q − Q0 = A A (3.126) Substituting from Equation 3.121, Equation 3.123, and Equation 3.125, we obtain P = ce0 E = e0 (er − 1)E (3.127) This equation, as we shall see later, provides a link between the permittivity, which is a macroscopic, measurable property of a dielectric, and the atomic or molecular mechanisms in the dielectric which give rise to this property. 3.4.4.1 Polarization and Dipole Moment In terms of the wave-mechanical picture, an atom may be looked upon as consisting of a positively charged nucleus surrounded by a negatively charged cloud, which is made up of contribution from electrons in various orbitals. Since the centers of positive and negative charges are coincident (see Figure 3.54a), the net dipole moment of the atom is zero. If an electric ﬁeld is applied, however, the electron cloud will be attracted by the positive plate and the nucleus by the negative plate, with the result that there will occur a small displacement of the center of gravity of the negative charge relative to that of the positive charge (Figure 3.54a). The phenomenon is described by the statement that the ﬁled has induced an electric dipole in the atom; and the atom is said to have suffered electronic polarization—electronic because it arises from the displacement of the electron cloud relative to the nucleus. The electric dipole moment of two equal but opposite charges, +q and −q, at a distance r apart is deﬁned as qr. For the atomic model of Figure 3.54a, it can be shown by balancing the opposite forces of the electric ﬁeld and the coulombic attraction between the nucleus and the center of the electron cloud that the dipole moment, m, induced in an atom by the ﬁeld E is m = (4pe0 R3 )E = ae E (3.128) where ae, is a constant, called the electronic polarizability of the atom. It is proportional to the volume of the electron cloud, R3. Thus the polarizability increases as the atoms become larger. On the macroscopic scale we have earlier deﬁned the polarization, P, to represent the bound charges induced per unit area on the surface of the material. Therefore, if we take unit areas on opposite faces of a cube separated by a distance d, the dipole moment due to unit area will be m = Pd 373 Plastics Properties and Testing External ﬁeld applied + + + + + + No external ﬁeld (a) + + + + + + + + + + + + + (b) + + + + + + + + + (c) + + + – + – – + + + – + + + + + + + + + ++ + + + + (d) FIGURE 3.54 Schematic illustrations of polarization mechanisms. (a) Electronic displacement. (b) Ionic displacement. (c) Dipole orientation. (d) Space charge. For d = 1 (i.e., for unit volume) m = P. The polarization P is thus identical with the dipole moment per unit volume (check the units: coul/m2 = coul m/m3). A dipole moment may rise through a variety of mechanisms, any or all of which may thus contribute to the value of P. The total polarization may be represented as a sum of individual polarizations, each arising from one particular mechanism (Figure 3.54) or, more appropriately, as an integrated sum of all the individual dipole moments per unit volume. X X X X mi + m0 + ms me + P = Pe + Pi + P0 + Ps = (3.129) V Electronic polarization, Pe, as discussed previously, arises from electron displacement within atoms or ions in an electric ﬁeld; it occurs in all dielectrics. Similarly, displacements of ions and atoms within molecules and crystal structures (Figure 3.54b) under an applied electric ﬁeld give rise to ionic (or atomic) polarization, Pi. Orientation polarization, P0, arises when asymmetric (polar) molecules having permanent dipole moments are present, since they become preferentially oriented by an electric ﬁeld (Figure 3.54c). Interfacial (or space charge) polarization, Ps, is the result of the presence of higher conductivity phases in the insulating matrix of a dielectric, causing localized accumulation of charge under the inﬂuence of an electric ﬁeld (Figure 3.54d). Any or all of these mechanisms may be operative in any material to contribute to its polarization. A question to be discussed now is, which of the mechanisms are important in any given dielectric? The answer lies in studying the frequency dependence of the dielectric constant. 374 Plastics Technology Handbook 3.4.4.2 Dielectric Constant versus Frequency Let us consider ﬁrst a single dipole in an electric ﬁeld. Given time, the dipole will line up with its axis parallel to the ﬁeld (Figure 3.54c). If now the ﬁeld is reversed, the dipole will turn 180° to again lie parallel to the ﬁeld, but it will take a ﬁnite time; so if the frequency of the ﬁeld reversal increases, a point will be reached when the dipole cannot keep up with the ﬁeld, and the alteration of the dipole direction lags behind that of the ﬁeld. For an assembly of dipoles in a dielectric, this condition results in an apparent reduction in the dielectric constant of the material. As the frequency of the ﬁeld continues to increase, at some stage the dipoles will barely have started to move before the ﬁeld reverses. Beyond this frequency, called the relaxation frequency, the dipoles make virtually no contribution to the polarization of the dielectric. We may now consider the various mechanisms and predict, in a general way, the relaxation frequency for each one. Electrons with their extremely small mass have little inertia and can follow alterations of the electric ﬁeld up to very high frequencies. Relaxation of electronic polarization is thus not observed until about 106 Hz (ultraviolet region). Atoms or ions vibrate with thermal energy, and the frequencies of these vibrations correspond to the infrared frequencies of the electromagnetic spectrum. The relaxation frequencies for ionic polarization are thus in the infrared range. Molecules or groups of atoms (ions) behaving as permanent dipoles may have considerable inertia, so relaxation frequencies for orientation polarization may be expected to occur at relatively smaller frequencies, as in the radio-frequency range. Since the alternation of interfacial polarization requires a whole body of charge to be moved through a resistive material, the process may be slow. The relaxation frequency for this mechanism is thus low, occurring at about 103 Hz. Figure 3.55 shows a curve of the variation of the dielectric constant (relative permittivity) with frequency for a hypothetical solid dielectric having all four mechanisms of polarization. Note that except at high frequencies the electronic mechanism makes a relatively low contribution to permittivity. However, in the optical range of frequencies, only this mechanism and the ionic mechanism operate; they therefore strongly inﬂuence the optical properties of materials. 3.4.4.3 Dielectric Constant versus Temperature Dielectric constant (εr) Liquids have higher dielectric constants than solids because dipole orientation is easier in the former. The effect is shown schematically in Figure 3.56a. After the abrupt change due to melting, the dielectric constant decreases as the temperature is increased, which is due to the higher atomic or molecular mobility and thermal collisions tending to destroy the orientation of dipoles. Interfacial polarization Dipole orientation polarization Ionic (or atomic) polarization Electronic polarization Power Radio, TV, radar Frequency FIGURE 3.55 Dielectric constant versus frequency. Optical 375 Solid Dielectric constant (εr) Dielectric constant (εr) Plastics Properties and Testing Liquid Amorphous Crystalline Tm (a) Tg Temperature (b) Tm Temperature FIGURE 3.56 Variation of dielectric constant with temperature (schematic). (a) Crystalline material. (b) Amorphous polymer. A crystalline polymer containing polar group would behave as shown by dashed lines. Figure 3.56b shows the schematic variations of dielectric constants with temperature for amorphous solids, such as glasses and many polymers. Above the glass transition temperature (Tg), atoms and molecules have some freedom of movement, which allows orientation of permanent dipoles with the ﬁeld, thereby increasing the dielectric constant. Since the polar groups which contribute to orientation polarization are not identically situated in an amorphous matrix, the dielectric constant changes over a temperature range rather than abruptly at a single temperature as in a crystalline material (cf. Figure 3.56a). The decrease in the dielectric constant after melting is again due to greater molecular mobility and thermal collisions. 3.4.4.4 Dielectric Losses The behavior of a dielectric under an applied ﬁeld has much in common with that of a material subjected to mechanical loading. The displacements of atoms and molecules within a material, when a mechanical force is applied, do not occur instantaneously but lag behind the force, resulting in elastic aftereffect and energy dissipation by mechanical hysteresis under an alternating force. Similarly, in dielectrics the lag of polarization behind the applied ﬁeld produces energy dissipation by electrical hysteresis in an alternating ﬁeld (Figure 3.57b). Such energy losses are related to the internal dipole friction. The rotation of dipoles with the ﬁeld is opposed by the internal friction of the material, and the energy required to maintain this rotation contributes to the power loss in dielectrics. Charge density Charge density Electric ﬁeld Electric ﬁeld (a) (b) I (c) Loss angle V (d) δ I Phase angle (θ) V FIGURE 3.57 Charge density versus electric ﬁeld. (a) Loss-free cycle. (b) High-loss cycle. (c) Phase shift in a perfect capacitor. (d) Phase shift in a real capacitor. 376 Plastics Technology Handbook Besides electrical hysteresis, leakage currents also contribute to dielectric losses. Leakage currents occur mainly by ionic conduction through the dielectric material and are usually negligible except at high temperatures. There are various ways of measuring energy losses by a dielectric. A fundamental property of a capacitor is that if an alternating voltage is applied across it in a vacuum, the current that ﬂows to and from it due to its successive charging and discharging is 90° out of phase with the voltage (Figure 3.57c), and no energy is lost. However, in real capacitors containing a dielectric, the lag of polarization causes a phase shift of the current (Figure 3.57d). The phase shift angle, d, is called the loss angle, and its tangent (tan d) is referred to as the loss tangent or dissipation factor. (An ideal dielectric would have a phase angle of 90°, and hence the loss angle would be zero.) The sine of the loss angle (sin d) or the cosine of the phase angle (cos q) is termed the power factor. In electrical applications the power loss (PL) is deﬁned as the rate of energy loss per unit volume and is derived to be wE2 e0 PL = er tan d (3.130) 2 where E is the electric ﬁeld and w is the angular velocity; er tan d is called the loss factor. For most materials d is very small; consequently, the three measures of energy dissipation—d, tan d, and sin d (= cos q)—are all approximately the same. Since these values vary somewhat with frequency, the frequency must be speciﬁed (see Table 3.6). It is evident from this discussion that the power loss and heat dissipation in a dielectric will be aided by a high dielectric constant, high dissipation factor, and high frequency. Therefore, for satisfactory performance electrical insulating materials should have a low dielectric constant and a low dissipation factor but a high dielectric strength (Table 3.6) and a high insulation resistance. Polyethylene and polystyrene with their exceptionally low dissipation factors (<0.0005) and low dielectric constant (2.3–2.5) are the most suitable for high-frequency applications, as in television and radar. For dielectrics used in capacitors, however, a high dielectric constant is desirable. 3.4.4.5 Dielectric Losses of Polar Polymers + + + – – – – – + – + – + – – + + + + + – No ﬁeld Field applied (a) Low response at high frequency Loss factor + High response at low frequency – + – + – + – – – + + When a polymer having polar groups (e.g., polymethyl methacrylate, polyvinyl chloride) is placed in an electric ﬁeld, the polar groups behaving as dipoles tend to orient themselves in response to the ﬁeld (Figure 3.58a). In an alternating ﬁeld the friction of the dipoles rotating with the ﬁeld contributes to the dielectric loss. This loss is small at low frequencies where the polar groups are able to respond easily to the ﬁeld, and also at high frequencies where the polar groups are unable to change their alignment with Frequency of ﬁeld (Log scale) (b) FIGURE 3.58 (a) Orientation of a polar polymer molecule in an electric ﬁeld. (b) Dielectric response of a polar polymer in an alternating electric ﬁeld. 377 Plastics Properties and Testing the ﬁeld. The loss is maximum in the transitional region where the polymer is passing from high response at low frequency to low response at high frequency (Figure 3.58b). Since the friction of dipoles in an alternating ﬁeld produces heat, polar polymers can be heated by the application of radio-frequency ﬁeld in fabrication processes (see high-frequency welding, Chapter 2). 3.5 Optical Properties Optical characteristics of plastics include color, clarity, general appearance, and more directly measurable properties, such as index of refraction [27]. For optical applications, however, other properties, including dimensional stability, scratch resistance, temperature limitation, weatherability, and water absorption, must be considered. 3.5.1 Optical Clarity Most resins by nature are clear and transparent. They can be colored by dyes and will become opaque as pigments or ﬁllers are added. Polystyrene and poly (methyl methacrylate) are well known for their optical clarity, which even exceeds that of most glass. Optical clarity is a measure of the light transmitting ability of the material. It depends, among other things, on the length of the light path, which can be quantitatively expressed by the Lambert–Beer law, or log (I/I0) = −AL, where I/I0 is the fraction of light transmitted, L is the path length, and A is the absorptivity of the material at that wavelength. The absorptivity describes the effect of path length and has the dimension of reciprocal length. Transparent colored materials are obtained by adding a dye to a water white resin. A color results when a dye removes part of the visible light traveling through the piece. The red color, for example, is produced by a dye which absorbs the blue, green, and yellow components of the light and transmits the red unchanged (Figure 3.59). However, for any dye to be effective it must be soluble in the plastic, and it is best incorporated in the plastic before molding. Fluorescent dyes absorb radiant energy at one wavelength, perhaps in the ultraviolet, and emit it as less energetic but more visible radiation. Cast phenolics, allyls, cellulosics, and many other clear plastics show a natural tendency to absorb in the blue and to be yellowish. Ultraviolet photodecomposition is largely responsible for the development of yellowish color in plastics exposed to sunlight. Incorporation of an invisible ultraviolet-absorbing material such as phenyl salicylate or dihydroxybenzophenone in the plastic greatly reduces photodecomposition. Addition of a blue or green tint into the plastic can also mask the yellow color. In addition to yellowing, plastics show darkening due to outdoor exposure as the transmission curve shifts downward, and there is pronounced absorption at wavelengths shorter than about 5,000 A (Figure 3.60). Violet Clear pla Blue Green Yellow Orange Red Infrared 100 ye red d With stic 10 1 4,000 5,000 6,000 Wave length (A) FIGURE 3.59 Light transmission diagram for plastic. 7,000 Transmission (%) (Log scale) Ultraviolet 378 Plastics Technology Handbook 100 Transmission (%) (A) Clear pla stic ure (B) r Afte 10 1 4,000 os exp 5,000 6,000 7,000 Wave length (A) FIGURE 3.60 Yellowing of plastics on exposure. The optical clarity of a plastics specimen is measured by the lack of optical haze. Haze is arbitrarily deﬁned as the fraction of transmitted light which is deviated 2½° or more from an incident beam by forward scattering. When the haze value is greater than 30%, the material is considered to be translucent. The forward scattering of the light beam, which is responsible for haze, occurs at internal interfaces such as caused by a dust particle, bubble, particles of a ﬁller or pigment, or by density changes. Due to scattering at interfaces, a crystalline plastic with myriads of crystallite regions bounded by interfaces is translucent. Crystalline polyethylene is thus translucent at room temperature, but, on warming, the crystallites disappear and the material becomes transparent. It can thus be inferred that plastics which are transparent at room temperature, such as polystyrene or poly(methyl methacrylate), are of the noncrystalline type and without ﬁllers. The effect of interfaces due to ﬁllers depends to a large extent on the difference in the indices of refraction of plastic and ﬁller. Thus transparent glass-ﬁlled polyester panels are obtained if the indices of refraction of glass and resin are identical and the glass is surface treated to enable the resin to wet the glass completely. 3.5.2 Index of Refraction The index of refraction for any transparent material is the ratio of the sine of the angle of an incident ray to the sine of the angle of refraction (Figure 3.61). It also corresponds to the ration of the speed of light in a vacuum (or air, closely) to its speed in the material. The refractive index values of several common plastics are compared with those of other materials in Table 3.7. The refraction property makes possible the focusing action of a lens. Plastic lenses have the Incident advantage that they are light in weight (about half angle as heavy as glass) and practically nonshattering. But i they have the disadvantage of low scratch resistance and comparatively poor dimensional stability for temperature changes. When tolerances are less critical, plastic lenses can be mass produced by virtue of the moldability of plastics; these lenses are quite satisfactory for r inexpensive camera view ﬁnders, for example, but Refraction in applications where exacting tolerances are angle required, such as cameras, periscopes, or similar high-resolution devices, molded plastic lenses have FIGURE 3.61 Refraction of light. Index of refraction: not been suitable. If plastic lenses are to be used in n = (sin i)/(sin r). 379 Plastics Properties and Testing TABLE 3.7 Index of Refraction Material Index of Refraction (n) Air 1.00 Water 1.33 Cellulose acetate Poly(methyl methacrylate) 1.48 1.49 Common glass 1.52 Poly(vinyl chloride) copolymer Polystyrene 1.53 1.59 Flint glass 1.65 Diamond 2.42 these applications, they should be ground and polished in much the same manner as in glass, though, of course, it is easier to do so (Figure 3.62). The refractive index, n, of an isotropic material is given by the Lorentz–Lorenz relation n2 − 1 4 = pP n2 + 2 3 where P is the optical polarizability per unit volume of material. For a pure substance it is more convenient to write this equation in the form (n2 − 1)M 4 = pN a = R (n2 + 2)r 3 A (3.131) where M is the molecular weight of the substance, r is the density, NA is Avogadro’s number, a is the optical polarizability of a single molecule of the substance, and R is the molar refraction. The quantity a represents the amount of polarization of the molecule per unit electric ﬁeld caused by the alternating electric ﬁeld associated with the light ray passing through the material. The polarization may be regarded as due to the shift of the center of charge of the more loosely bound electrons relative to the nucleus (see Figure 3.54). The refractive index varies with wavelength of light and is measured by the optical dispersion—that is, the difference in the refractive indexes for different wavelengths. It is responsible for the spectrum-separating ability of a prism in a spectroscope. Most plastics have relatively low optical dispersion. This property makes them more suitable for eyeglasses and large lenses for projection television. Radius of bend greater than d n–1 Angle less than 90° – sin–1(1/n) d Internal reﬂection d Radius of bend less than d n–1 FIGURE 3.62 Principle of piped light. Angle greater than 90° – sin–1(1/n) Ray escapes 380 Plastics Technology Handbook 3.5.3 Piped Lighting Effect The difference in indexes of refraction for air and for a transparent solid (plastic or glass) is responsible for the ability of the latter, used as a rod or plate, to bend or pipe light around a curve. Bending can be explained as follows. It is evident from Figure 3.61 that a light beam in air cannot enter a material with i > 90°. Since sin 90° = 1, the maximum angle of refraction is that for which the sine is 1/n, where n is the index of refraction. For poly(methyl methacrylate) n = 1.49, and the maximum angle of refraction becomes sin−1 (1/1.49) or 42° approximately. Obviously a light beam refracted at a greater angle cannot have come from the air, but it must have been reﬂected internally. A plastic plate with small curvature therefore gives internal reﬂections and can thus bend a light beam, as shown in Figure 3.62. Consideration of geometric optics shows the minimum radius of curvature for this piping of light to correspond to 1/(n − 1) times the thickness. Polymer has numerous advantages over glass [28]. First and foremost, it is more pliable. It is also nonshattering, lightweight, and easy to fabricate. Polymer optical ﬁber (POF) has a shorter bend radius (i.e., more ﬂexibility) and is more resilient to damage and abuse than glass due to its intrinsic material characteristics. The bending radius depends largely on the diameter of the ﬁber, large ﬁber giving a larger bend radius. Thus, for an outside diameter of one millimeter (1,000 mm), the bend radius (damage threshold) is approximately one centimeter, while for single-mode POF with outside diameter of 125 mm, the bend radius is as low as 0.125 cm. The level of intensity loss from bending depends on the individual characteristics of the ﬁber, but intensity loss values have been observed at 10%–15% for bends close to the minimum (damage threshold) radius. Attenuation is a measurement, in decibels (dB) per unit length, of how much reduction in light (due to absorption) is experienced per unit length of the ﬁber. Absorption varies largely with the wavelength of light. The attenuation of PMMA, generally used as POF core material, is approximately 100 dB/km in the visible region and increases rapidly at higher wavelength (Figure 3.63). This high attenuation of POF compared to the silica-based ﬁber limits the data link length. However, the ﬁeld witnessed real progress in 1980s when low loss polymers were developed and drawn into ﬁbers (see Chapter 5). The light bending ability of POF is made use of in a wide variety of applications ranging from lighting scheme for pools and fountains to telecommunications and consumer electronics. In these applications (see more details in Chapter 5), the POF serves as a medium for transmitting light for illumination or imaging, or for transporting and/or controlling information that is encoded on a beam of light [29]. 0.5 Loss (dB/km) 10 Wavelength (μm) 0.7 1.0 2.0 3.5.4 Stress-Optical Characteristics 6 104 PMMA (POF) 102 Silica ﬁber 100 20 16 12 8 Wavenumber (103 cm–1) 4 FIGURE 3.63 Variation of attenuation of PMMA and silica-based ﬁbers with wavelength. An important phenomenon observed in amorphous plastics (also observed in optical glass) is the development of optical anisotropy due to stress. The stress-optical characteristic of transparent plastics is the basis of the important technique of photoelasticity by which stress and strain in complicated shapes, for which no analytical solution is readily available, can be determined experimentally and simply. The amount of strain in various parts of a transparent plastic model of a machine part, subjected to loads simulating those in actual operation, is determined by measuring the anisotropy with polarized light; therefore from this measurement the distribution of stress and strain in the actual metal part can be deduced. Very complicated Plastics Properties and Testing 381 shapes, including complete structures such as bridges or aircraft wings, have been successfully analyzed in this manner, employing plastic models. The same stress-optical characteristic also permits examination of locked-in stresses in a molded plastic part. The part is examined under polarized light, and the amount of stress is indicated by the number of fringes or rings that become visible. Illumination with white light gives colorful patterns involving all the colors of the spectrum. Monochromatic light is, however, used for stress analysis because it permits more precise measurements. In general, a single ray of light entering an anisotropic transparent material, such as crystals of sufﬁciently low symmetry and strained or drawn polymers, is propagated as two separate rays which travel with different velocities and are polarized differently. Both the velocities and the state of polarization vary with the direction of propagation. This phenomenon is known as birefringence or double refraction, and the material is said to be birefringent. The stress-optical characteristic of plastics arises from this phenomenon of birefringence, induced by strains due to applied stress. The applied stress produces different densities along different axes. The stress-optical coefﬁcient of most plastics is about 1,000 psi (70 kgf/cm2 or 6.9 Mpa) per inch (2.54 cm) thickness per fringe. This means that a 1-in. thick part when illuminated with monochromatic light and viewed through a polarizing ﬁlter will show a dark fringe or ring for a stress of 1,000 psi. This sensitivity is many times higher than that shown by glass and makes transparent plastics useful in photoelastic stress analysis. In one variation of the process the deforming stress is applied to a warm, soft plastic model which is then quenched to room temperature. The resulting locked-in stress may then be analyzed at leisure or, perhaps more conveniently, by cutting sections from the model and examining them separately. For this application, however, the plastic should have stress-optical stability. In this regard, unplasticized transparent plastics such as poly(methyl methacrylate), polystyrene, and cast poly(allyl phthalate) are superior to plasticized materials such as cellulose acetate. 3.6 Thermal Properties The useful thermal properties of plastics include speciﬁc heat, thermal expansion, thermal conductivity, and thermal softening [30]. 3.6.1 Speciﬁc Heat For most plastics the speciﬁc heat value (calories per gram per °C) lies between 0.3 and 0.4. On a weight basis this value is much higher than that of most metals. Both iron and copper, for example, have speciﬁc heats of about 0.1 at ordinary temperatures. However, along a volume basis, the speciﬁc heats of plastics are lower than those of common metals, because of the substantially lower density of plastics. Knowledge of speciﬁc heat of a material helps to determine the energy requirement for increasing its temperature. In compression molding and injection molding the theoretical heat requirement can be calculated as the sum of this direct heat and any latent heat of melting minus any energy released by chemical reaction. But this heat requirement is not large compared to the heat loss by radiation and conduction from the press. The proportion of losses is substantially lower for injection molding than for compression molding. It is nevertheless customary in both cases to establish the heating requirements as well as mold cooling requirements by a process of trial and error. In dielectric preheating, however, the speciﬁc heat of a molding powder is of more direct concern; this knowledge along with the amount of material and time enables one to calculate the required amount of radio-frequency power. For example, to raise the temperature of a 1-kg perform of speciﬁc heat 0.35 through 80°C in 1 min requires 1 × 0.35 × 80 or 28 kcal/min or about 2 kW. 382 Plastics Technology Handbook 3.6.2 Thermal Expansion Linear (thermoplastic) polymers have very high thermal expansion coefﬁcients since they are weakly bonded materials and need less input of thermal energy to expand the structure. This applies to all polymers of the vinyl type which have expansion coefﬁcients of about 90 × 10−6/°C. Network (thermosetting) polymers having a three-dimensional framework of strong covalent bonds exhibit less thermal expansion and have expansion coefﬁcients in the range of 30 – 70 × 10−6/°C. These may be compared with values of 11 × 10−6/°C for mild steel, 17 × 10−6/°C for ordinary brass, and less than 10 × 10−6/°C for ceramics. In spite of high thermal expansion, plastics do not easily undergo thermal cracking, because they also have very low elastic moduli and large strains do not induce high stresses. One area where the high expansion of polymers plays a signiﬁcant role is the molded dimensions of plastic parts. The linear mold shrinkage due to thermal contraction from molding to room temperature is usually about 1/2%–1%. Polyethylene and certain other materials exhibit even a higher shrinkage. As a result, plastic parts with close tolerances are difﬁcult to make. It is also possible during the molding operation that different parts are not all at a uniform temperature. This may lead to differential shrinkage during cooling and produce warping, locked-up internal stresses, and weakening of the part. Warping may be prevented by a suitable mounting. Internal stresses in plastics that are more subject to creep and cold ﬂow tend to relieve themselves slowly; warming the part accelerates the process. Internal stresses in polystyrene and other brittle plastics can be removed by an annealing process similar to the one used in glass manufacture. It involves a controlled heating-cooling cycle, and for plastics it is obtained by immersing the parts in a liquid held at the proper temperature, followed by very slow cooling. Polystyrene for example can be heated in water at 80°C and then slowly cooled to 65°C. It is then cooled in undisturbed air. The relatively high thermal expansion of plastics poses a problem in the use of molded metal inserts which are sometimes required for electrical contacts, screw thread mountings, or increased strength. To minimize stresses due to inserts, manufacturers usually use plastics with low coefﬁcient of expansion. Phenolics and ureas are commonly used because their coefﬁcient of expansion are among the lowest of the common plastics. For many plastics the use of molded metal inserts is not satisfactory because of the excessive stress produced. As an example, polystyrene with brass inserts on cooling from the molding temperature of 160 to 20°C produces a strain of (70 − 17) × 10−6 × (160−20) or 0.0074. For an elastic modulus of 0.46 × 106 psi (3.2 Gpa), the internal stresses become 3,400 psi (23 Mpa). The presence of this much internal stress renders the part useless, since the tensile strength of polystyrene is only 3,600 psi (25 Mpa). However, by use of an appropriate ﬁller the thermal expansion of plastics and, hence, the internal stresses can be reduced. Addition of 11% by weight of aluminum oxides in polystyrene gives a mixture whose thermal expansion is identical to that of brass. Brass inserts in this alumina-polystyrene composite show no evidence of internal stresses and are quite satisfactory. When synthetic resins are used as cements, the thermal expansion needs to be matched; this matching is done by judiciously choosing the ﬁller. In the production of panels of aluminum alloy overlaid on phenolic laminate, a ﬁller consisting of a glass ﬁber–starch mixture is used to reduce the expansion characteristic of the resin and to match its expansion exactly with that of aluminum. The aluminum panel then adheres completely and cannot be removed without tearing the metal or rupturing the phenolic core. There are cases where the high thermal expansion of plastics is used to advantage. One example is the shrink ﬁtting of handles of cellulose nitrate on screwdrivers. Another example is in mold design where the shrinkage on cooling is sufﬁcient to permit a small undercut. 3.6.3 Thermal Conductivity Thermal conductivities of plastics are relatively low and approximate 0.0004 (cal-cm)/(°C-cm2-sec). The corresponding values are 0.95 for copper, 0.12 for cast iron, 0.002 for asbestos, 0.0008 for wood, and 0.0001 for cork (SI values in J/s-m-k are obtained by multiplying by 418.7). Because of their low thermal Plastics Properties and Testing 383 conductivities, plastics are used for handles to cooking utensils and for automobile steering wheels. The low thermal conductivity is also responsible for the pleasant feel of plastic parts. Quite hot or quite cold objects can be handled with less difﬁculty if they are made of plastic, since the thermal insulation afforded by the plastic prevents a continuous rush of heat energy to (or from) the hand. Both thermal conductivity and temperature resistance of plastics have to be considered for their use at high temperatures. As an example consider a teapot handle; it must not deform even at 100°C. Therefore, common thermoplastics such as cellulose acetate are ruled out, but both a wood-ﬂour-ﬁlled and an asbestos-ﬁlled phenolic might be considered. The thermal conductivity of a mixture is nearly proportional to the volume percentage of each component. Wood-ﬂour-ﬁlled phenolic has a higher thermal conductivity than the pure resin, but the conductivity of this composite is still low enough to justify its use as the handle to a teapot. This composite can also withstand temperatures up to 100°C sufﬁciently well to give the handle a reasonable service life. For parts subjected to higher temperatures, asbestos-ﬁlled phenolic is a better choice. It can be used as the insulating connection to an electric iron, for example. Design of a handle determines to a great extent its service life. Quite often handles are found fastened directly to the hot object without regard for any temperature limitation; at the junction the plastic becomes brittle because of high temperature, and failure occurs. If the handle can be separated from the heated part and some cooling arrangement is included as part of the design, improved performance is to be expected. 3.6.4 Transition Temperatures and Temperature Limitations Both ﬁrst- and second-order transitions are observed in polymers. Melting and allotropic transformations are accompanied by latent-heat effects and are known as ﬁrst-order transitions. During second-order transitions, changes in properties occur without any latent-heat effects. Below the second-order-transition temperature (glass transition temperature) a rubberlike material acts like a true solid (see Chapter 1). Above this temperature the ﬁxed molecular structure is broken down partially by a combination of thermal expansion and thermal agitation. The glass transition temperature of polystyrene is 100°C; below 100°C polystyrene is hard and brittle, and above 100°C it is rubberlike and becomes easily deformed. The design engineer often requires to know the maximum temperature for which a polymer can be used in a given application. This depends largely on two independent factors: (1) the thermal stability of the polymer, particularly in air; and (2) the softening behavior of the polymer. Thermal stability testing requires the study of the change in properties on aging at various service temperatures. The relationship between structure and stability was considered brieﬂy in Chapter 1 and will be considered again in Chapter 5. The use of additives for improving such stability is discussed in Chapter 1. The rigidity of a polymer is determined by the ease with which polymer molecules are deformed under load. Young’s modulus is a fundamental measure of the rigidity or stiffness of a material. It is thus clear from Figure 1.52 that the softening point is associated with the Tg for amorphous polymers and with the Tm for highly crystalline polymers. In addition, there are many polymers that soften progressively between Tg and Tm and the value determined for the softening point can depend signiﬁcantly on the test method used. Two particular test methods have become very widely used. These are the Vicat softening point test and the test widely known as the heat distortion temperature test (also called the deﬂection temperature under load test). In the Vicat softening point test a sample of polymer is heated at a speciﬁed rate temperature increase and the temperature is noted at which a needle of speciﬁed dimensions indents into the polymer a speciﬁed distance under a speciﬁed load. The heat-distortion temperature is deﬁned as the temperature at which the midpoint of a beam 1/2 by 1/2 by 5 in. supported 1/2-in. from each end, shows a net deﬂection of 0.01 in. when loaded centrally with 2.5 kg and heated at the speciﬁed rate of 2°C/min. Testing is also done at one-quarter of this load. For most materials the two heat-distortion temperatures are within 10°. 384 Plastics Technology Handbook A large difference in these two heat distortion temperature indicates a material sensitive to temperature change, and for such materials at any elevated temperature stress should be minimum up to the heatdistortion temperature. A plastic is expected to maintain its shape under load, and hence this temperature represents an upper limiting point at which a plastic may be used. Recommendations of upper temperature limits for plastics are usually based on general experience, although some consideration is given to high-load and low-load heat-distortion temperatures. The size and shape of the part as well as its molding conditions govern to a certain extent the maximum permissible temperature; service conditions such as temperature variations and humidity are also important. 3.6.5 Standard Properties of Plastics Figure 3.64 through Figure 3.90 illustrate schematically (in simpliﬁed form) the bases of some of the standard properties of plastics [31]. Where standard test methods have been developed, these have been included. The principles of the different tests are shown, and the properties measured are indicated. Note that the property values of plastics are highly dependent on the specimen preparation, equipment, and testing techniques used. For this reason it is essential to refer to the appropriate ofﬁcial standard test method when executing the work. 3.7 Identiﬁcation of Common Plastics The ﬁrst step in the identiﬁcation of polymers is a critical visual examination. While the appearance of the sample may indicate whether it is essentially a raw polymer or a compounded and processed item, learning about its form, feel, odor, color, transparency or opacity, softness, stiffness, brittleness, bounce, and surface texture may be important in the process of the identiﬁcation of the polymer. For example, polystyrene, the general purpose polymer, is transparent and brittle, and produces a characteristic metallic tinkle when objects molded from it are dropped or struck. Identiﬁcation of plastics [32] is carried out by a systematic procedure: preliminary test, detection of elements, determination of characteristic values, and, ﬁnally, speciﬁc tests. For an exact identiﬁcation, however, the test sample should ﬁrst be puriﬁed so that it contains no additives (plasticizers, ﬁllers, pigments, etc.) that may affect the results of an analysis. Puriﬁcation is achieved by solvent extraction; Load at constant speed L1 L2 a b Load FIGURE 3.64 Tests for tensile strength (stress at fracture of the specimen) and elongation (extension of materials under load) of plastics. The ﬁrst provides a measure of the breaking strength of the material but is radically affected by the rate of loading and the ambient temperature. Tensile strength in lbf/in.2 or kgf/cm2 = load (lbf or kgf)/a × b (in.2 or cm2). Standard test methods: BS 2782 method 301, ASTM D638, ISO R527. Elongation % = (L2 − L1) × 100/L1. Standard test methods: BS 2782 method 301, ASTM D638. 385 Plastics Properties and Testing Load Load / extension curve (exaggerated) Load for tangent modulus Load for secant modulus Extensions Speciﬁed extensions (1% of original length) FIGURE 3.65 Test for tangent and secant moduli of plastics. For tangent modulus, load on tangent to loadextension curve at speciﬁed extension is used for calculating the stress value, while for secant modulus load on secant to load-extension curve is used. The modulus is given by stress/strain. Standard test methods: BS 2782 method 302, ASTM D638, ISO R527. Load (w) at constant speed Specimen t Support s Sp an (s) d (Deﬂection) a FIGURE 3.66 Tests for ﬂexural properties of plastics. Flexural strength = 3ws/2at (lbf/in.2 or kgf/cm2). Standard test method: ASTM D790. Modulus in ﬂexure = s3w/4at3d (lbf/in.2 or kgf/cm2). Standard test methods: BS 2782 method 302D, ASTM D790, ISO R178. Load Specimen a b FIGURE 3.67 Test for compressive strength of plastics. Compressive strength (lbf/in.2 or kgf/cm2) = load (lbf or kgf)/a · b (in.2 or cm2). Standard test methods: ASTM D795, BS 2782 method 303, ISO R604. 386 Plastics Technology Handbook Specimen Load (gf ) FIGURE 3.68 Test for tenacity of ﬁlaments, cords, twines, etc. Tenacity (gf/denier) = breaking load (gf)/denier of specimen. (Denier of ﬁlaments, cords, twines, etc., is equal to the weight in grams of 9,000 meters of the sample). Calibrated scale in units of force Pivot Pendulum weight (initial position) Clamp Force absorbed in tearing sample (gt/cm) Specimen Clamp Clamp Film sample standard width and length Standard slit Clamp FIGURE 3.69 Test for tear propagation resistance of plastic ﬁlm and thin sheeting. Tear resistance (gf/cm/mil) = force required to tear sample (gf/cm)/thickness of ﬁlm (mil). Standard test methods: ASTM D1922, BS 2782 method 308B. Film samples FIGURE 3.70 Tests for blocking of plastic ﬁlm. Blocking force (lbf/in.2 or kgf/cm2) = load (lbf or kgf)/initial area of ﬁlms in contact (in.2 or cm2). Standard test method: ASTM D1893. 387 Plastics Properties and Testing Residual energy, E2 (ft.-lbf or cm-kgf ) Pendulum Standard notch Impact energy E1 (ft.-lbf or cm-kgf ) Notched specimen Specimen width, D (in. or cm) Clamp W Adjustable weight or eﬀect fracture of specimen (lbf or kgf ) H = Height of fall (ft. or m) Specimen Support FIGURE 3.71 Tests for impact resistance of plastics. Izod impact strength = (E1−E2)/D ft-lbf/in. of notch or cm-kgf/ cm of notch. Standard test methods: ASTM D256; BS 2782 method 306A, ISO R180. Falling weight impact strength = W· H ft-lbf or m-kgf. (F50 is the energy required to fracture 50% of the specimens.) Standard test method: BS 2782 method 306 B. L B W a W S FIGURE 3.72 Test for fracture toughness of plastics. Specimen in three-point bending conﬁguration, as shown, is loaded in a testing machine that has provision for autographic recording of load applied to the specimen. Specimen dimensions: L, length (mm); W, width (mm); B, thickness (mm); S, span length (mm); a, initial crack length (mm). Plane strain fracture toughness, Klc = PSf (a/W)/BW3/2, where P = load (N) determined from load displacement record; f (a/W) = geometric factor. Example: L = 64 mm, W = 10 mm B = 3 mm, S = 40 mm, a = 3.2 mm, P = 50.88 N, f (a/W) = 1.61 (from ASTM standard), Klc = 34.3 N mm−3/2 = 1.09 MPa m1/2. Standard test method: ASTM E399-74. 388 Plastics Technology Handbook 1.6 mm Adhesive 25.4 mm 25.4 mm Area in test grips L 63.5 mm 63.5 mm 25.4 mm Area in test grips 127 + L mm 177.8 + L mm FIGURE 3.73 Form and dimensions of test specimen for single-lap-joint shear test to determine shear strengths of adhesives for bonding metals (standard test method: ASTM D1002-94). Recommended length of overlap (L) for most metals of 1.62 mm in thickness is 12.7 mm. Specimen is placed in grips of a testing machine so that outer 25.4 mm (1 in.) of each end is in contact with the jaws. Load is applied at the rate of 80–100 kg/cm2 (1,200–1,400 psi) of shear area per min and continued till failure occurs. T-peel test panel 76 mm (unbonded) Pull 76 mm ed) 152 m m 241 mm 25 (b d on 241 mm 76 mm Test specimen mm Pull FIGURE 3.74 Test panel and T-type test (T-peel test) specimen for peel resistance of adhesives (standard test method: ASTM D1876-95). The bent, unbonded ends of test specimen are clamped in test grips of tensile testing machine and load applied at a constant head speed of 254 mm (10 in.). Average peeling load (in pounds per inch of specimen width) is determined for the ﬁrst 127 mm (5 in.) of peeling after the initial peak. Balance Water (23 ± 2°C) Wire Specimen (Tied to a sinker for plastics lighter than water) FIGURE 3.75 Test for speciﬁc gravity and density of plastics. Sp. gr. = a/(a−w−b) where a = wt. of specimen without wire, b = wt. of specimen completely immersed and of the wire partially immersed in water, and w = wt. of partially immersed wire. Standard test method: ASTM D792. 389 Plastics Properties and Testing Clamp Sample Heat transfer liquid 30° Heater to raise the temperature at constant rate, i.e., 1°C/min for polystyrene Standard weight i.e., 20 g for polystyrene Standard weight i.e., 1 kgf for polyethylene Heat transfer liquid Heater to raise the temperature at a constant rate, i.e., 50°C/h for polyethylene Standard needle Sample Support FIGURE 3.76 Tests for softening temperatures of plastics. Cantilever softening point is the temperature at which the sample bends through 30°. Standard test method: BS 2782 method 102 C. Vicat softening point is the temperature at which the needle penetrates 1 mm into the sample. Standard test methods: ASTM D1525, BS 2782 method 102D, ISO R306. Standard weight Standard sample Heater to raise the temperature at a constant rate, i.e., 2°C/min for polyethylene Heat transfer liquid Supports FIGURE 3.77 Test for deﬂection temperature of plastics under ﬂexural load. Heat distortion temperature is the temperature at which a sample deﬂects by 0.1 in. (2.5 mm). Two measurements are made and quoted: (a) with a stress of 66 lbf/in.2 (4.6 kgf/cm2) and (b) with a stress of 264 lbf/in.2 (18.5 kgf/cm2). Standard test methods: ASTM D648, BS 2782 method 102, ISO R75. 390 Plastics Technology Handbook Standard weight, i.e., 2.16 kgf for polyethylene Melt index = weight of material in grams extruded in 10 min Polymer melt at standard temperature i.e.,190°C for polyethylene FIGURE 3.78 Melt index of plastics. The test measures the rate of ﬂow of polymer melt. It provides an indication of the ease of processing. Standard test methods: ASTM D 1238, BS 2782 method 105C, ISO R292. Standard weight Polymer melt at standard temperature Diameter A Diameter B FIGURE 3.79 Measurement of swelling of die extrudate. Die swell ratio = B/A. Glass column (45 cm/hr × 7.5 cm i.d.) Igniter Burning specimen Gas ﬂow in column 4 ± 1 cm/sec at STP Clamp with rod support Wire screen Glass beads in a bed O2 N2 FIGURE 3.80 Determination of oxygen index. Oxygen index, n% = 100 × O2/(O2 + N2), where O2, volumetric ﬂow rate of oxygen, cm3/sec, at the minimum concentration necessary to support ﬂaming combustion of a material initially at room temperature and N2, corresponding ﬂow rate of nitrogen, cm3/sec. Standard test method: ASTM D2863. 391 Plastics Properties and Testing Specimen 38 cm FIGURE 3.81 Test for rate of burning. Burning rate = 38 (cm)/t (min). Standard test method: ASTM D568. Traveling microscope Light source Focus Focus on top surface of specimen (microscope reading = D1) Specimen Base Focus Specimen Focus on apparent bottom surface of specimen (microscope reading = D2) Apparent Real depth depth Base FIGURE 3.82 Test for index of refraction of transparent plastics. Refractive index = real depth (measured with vernier)/apparent depth (D2 − D1). Standard test method: ASTM D542, ISO R489. 392 Plastics Technology Handbook Object Film Wide-angle light scatter Film sample Light source Reﬂecting sphere Photocell collecting all transmitted light reﬂected (T1) Light trap After T1 is determined the sphere is rotated to measure T2 Film sample Light source Photocell collecting all wideangle transmitted light scattered by ﬁlm more than 2½° (T2) Light trap absorbs all light scattered by ﬁlm less than 2½° Reﬂecting sphere FIGURE 3.83 Test for haze of transparent plastics. Haze, % = 100 × T2/T1. A low haze value is important for good short distance vision. Standard test method: ASTM D1003. Object Film Narrow-angle light scatter Less than ½° to normal = T2 ½° Light source Light stop Annular photocell collecting light greater than that at ½° to normal = T1 FIGURE 3.84 Measurement of narrow-angle light-scattering property of plastic ﬁlm. Clarity, % = 100 × T1/ (T1 + T2). 393 Plastics Properties and Testing Upper electrode Variable voltage Specimen V Voltage measurement Lower electrode Oil bath FIGURE 3.85 Test for dielectric strength of solid insulating materials. Dielectric strength (V/mil) = maximum voltage before failure (V)/thickness of specimen (mil). Standard test methods: ASTM D149, BS 2782 method 201. Oscillator 1MHz Resistance Electrode gap with air = D2 Electrode gap = specimen thickness = D1 V Specimen V Variable condenser (a) (b) FIGURE 3.86 Test for permittivity (dielectric constant) of insulating materials. (a) Position of maximum voltage obtained with sample by adjusting variable capacitor. Electrode gap = specimen thickness = D1; (b) position of maximum voltage obtained with air by adjusting electrode gap to D2 [variable capacitor remains as set in (a)]. Dielectric constant = D1 (in. or mm)/D2 (in. or mm). Standard test methods: ASTM D150, BS 2782 method 207A. Current measurement A (–) Upper electrode Standard specimen Standard voltage Lower electrode (+) FIGURE 3.87 Test for DC resistance of insulating materials. Electrical resistance of specimen (ohm) = applied voltage (V)/current measured (A). Volume resistivity (ohm-cm) = resistance of specimen (ohm) × arc of upper electrode (cm2)/specimen thickness (cm). Standard test methods: ASTM D257, BS 2782 method 202. 394 Plastics Technology Handbook Polymer dielectric Applied alternating current Heat loss due to dipole reversal (w) FIGURE 3.88 Test for AC loss characteristics of solid insulating materials. Power factor = W/V × I, where W = power loss in watts and V × I = effective sinusoidal voltage × current in volt-amperes. Standard test methods: ASTM D150, BS 2782 method 207A. Bath Standard notch Standard stressed specimen Support Environment (eg., detergent soln.) FIGURE 3.89 Test for environmental stress cracking of ethylene plastics. Stress cracking resistance (F50) = time taken for 50% of the specimens to fail (h). Standard test method: ASTM D1693. either the material is dissolved out and polymer is obtained by reprecipitation or evaporation of the solvent, or the pure polymer remains as the insoluble residue. The solvent varies, and a general method cannot be given. However, for many materials particularly for those in which additives do not interfere, the unpuriﬁed material can be investigated and qualitative preliminary tests used. In general, the following series of investigations should be carried out for preliminary tests of the material: (1) behavior on heating in an ignition tube and in the ﬂame; (2) qualitative detection of heteroelements such as N, S, and halogens; (3) determination of properties such as density, refractive index, and melting point or range; (4) determination of solubility as an aid to polymer identiﬁcation; and (5) ash or sulfated ash determination as a test for inorganic additives (ﬁllers, pigments, stabilizers). 3.7.1 Behaviors on Heating and Ignition Many polymers can be roughly identiﬁed by their behavior when carefully heated and ignited. Nitrocellulose and plastics containing this (e.g., celluloid) burn with explosive violence and other materials such as 395 Plastics Properties and Testing Stop Cylinder 1 Position 2 Position 1 Valve Diﬀerential pressure indicator Valve Position 3 Sample Cylinder 2 V1 FIGURE 3.90 Determination of open cell content of rigid cellular plastics. Open cell content (approx.) = 100 (V − V1)/ V, where Vt, displacement volume of specimen, cm3 and V, geometric volume of specimen, cm3. Standard test method: ASTM D2856. poly(vinyl chloride) or ﬂuoro-hydrocarbons decompose with the evolution of poisonous or irritating vapors. Only small quantities of material should therefore be taken for heating tests. The heating should be done gently, because if the heating is too rapid or intense, the characteristic changes may not be observed. In a typical procedure, a small piece or amount (0.1 g) of the test sample is placed on a cleaned glass or stainless steel spatula, previously heated to remove any traces of combustible or volatile materials, and then warmed gently over a small colorless Bunsen ﬂame until the sample begins to fume. The decomposing sample is removed from the ﬂame and the nature of the fume or gas is examined with respect to color, odor, inﬂammability, and chemical identity including acidity, alkalinity, etc. The sample is next moved to the hottest zone of the small Bunsen ﬂame and note is taken of the following: (1) if the material burns and if so, how readily; (2) the nature and color of the ﬂame as the material burns; (3) whether the material is self-extinguishing or continues to burn after removal from the ﬂame, and (4) the nature of the residue. Observations on heating and ignition of some common polymers are listed in Table 3.8. 3.7.2 Tests for Characteristic Elements The results of tests for characteristic elements such as nitrogen, sulfur and halogens may serve to roughly indicate the nature of the unknown material, including the nature of the base polymer and additives, if present. The following tests may be performed for qualitative detection of N, S, and halogens. 1. Nitrogen. Above 50 mg of the test material is heated carefully in an ignition tube with twice the amount of freshly cut sodium, until the sodium melts. A further small amount of material is added and the tube is heated to red heat. It is then dropped carefully into water (20 ml) in a mortar. The solid is powdered and the solution ﬁltered. A little freshly prepared ferrous sulfate is added to the ﬁltrate and the latter boiled for 1 min. It is acidiﬁed with dilute hydrochloric acid, and 1–2 drops of ferric chloride solution added. A deep blue coloration or precipitate (Prussian blue) indicates nitrogen (if very little nitrogen is present, a green– blue color is formed initially). This test sometimes fails with substances containing nitro-groups and with nitrogen-containing heterocyclic compounds. 2. Sulfur. The fusion is carried out as for nitrogen. The ﬁltrate (2 ml) is acidiﬁed with acetic acid and a few drops of an aqueous solution of lead acetate is added. A black precipitate of lead sulﬁde indicates the presence of sulfur. 396 Plastics Technology Handbook TABLE 3.8 Heating Tests of Some Common Polymers Polymer Color of Flame Odor of Vapor Other Notable Points The material burns but self-extinguishes on removal from ﬂame Poly(vinyl chloride) Yellow–orange, green bordered Resembles hydrochloric acid and plasticizer (usually ester like) Strongly acidic fumes (HCl), black residue Poly(vinylidene chloride) As above Resembles hydrochloric acid As above Polychloroprene Yellow, smoky As above As above Phenol-formaldehyde resin Yellow, smoky Phenol, formaldehyde Very difﬁcult to ignite, vapor reaction neutral Melamine-formaldehyde resin Pale yellow, light Ammonia, amines (typically ﬁsh like), formaldehyde As above Very difﬁcult to ignite, vapor reaction alkaline Melts sharply to clear, ﬂowing liquid; melt can be drawn into a ﬁber; vapor reaction alkaline Melts and chars Urea-formaldehyde resin As above Nylons Yellow–orange, blue edge Resembling burnt hair Polycarbonate Luminous, sooty Phenolic Chlorinated rubber Yellow, green bordered Acrid As above Strongly acidic fumes, liberation of HCl; swollen, black residue The material burns and continues burning on removal from ﬂame Polybutadiene (BR) Yellow, blue base, smoky Disagreeable, sweet Chars readily; vapor reaction neutral Polyisoprene (NR, gutta percha, synthetic) Yellow, sooty Pungent, disagreeable, like burnt rubber As above Styrene–butadiene rubber (SBR) Yellow, sooty Pungent, fruity smell of styrene As above Nitrile rubber (NBR) Yellow, sooty Like burnt rubber/burnt hair As above Butyl rubber (IIR) Practically smoke free candle like Slightly like burnt paper Melt does not char readily Polysulﬁde rubber (polymer itself emits unpleasant, mercaptan like odor) Smoke-free, bluish Pungent; smell of H2S Yellow, acidic (SO2) fumes Cellulose (cotton, cellophane, viscose rayon, etc.) Yellow Burnt paper Chars, burns without melting Cellulose acetate Yellow–green, sparks Acetic acid, burnt paper Melts, drips, burns rapidly, chars, acidic fumes Cellulose acetate butyrate Dark yellow (edges slightly blue), somewhat sooty, sparks Yellow Acetic acid/butyric acid, burnt paper Melts and forms drops which continue burning Camphor Burns very fast, often with explosion Yellow, luminous Pale yellow with blue– green base Yellow Burnt paper Slightly sweet, burnt paper Resembling burnt hair Melts and chars Melts and chars Yellow, luminous, sooty Acetic acid Sticky residue, acidic vapor Cellulose nitrate (plasticized with camphor) Methyl cellulose Ehtyl cellulose Polyacrylonitrile Poly(vinyl acetate) Dark residue; vapor reaction alkaline (Continued) 397 Plastics Properties and Testing TABLE 3.8 (CONTINUED) Polymer Heating Tests of Some Common Polymers Color of Flame Odor of Vapor Other Notable Points Poly(vinyl alcohol) Luminous, limited smoky Unpleasant, charry smell Burns in ﬂame, self extinguishing slowly on removal; black residue Poly(vinyl butyral) Poly(vinyl acetal) Bluish (yellow edge) Purple edge Like rancid butter Acetic acid Poly(vinyl formal) Yellow–white Slightly sweet Melts, forms drops Does not drip like poly (vinyl butyral) Does not drip like poly(vinyl butyral) Polyethylene Luminous (blue center) Like parafﬁn wax (extinguished candles) Poly(a-oleﬁns)(PP, EPR, etc.) As above As above Melts, forms drops; droplets continue burning As above Polystyrene Luminous, very sooty Sweet (styrene) Softens, easily ignited Poly(methyl metyhacrylate) Luminous, yellow (blue, edge), slightly sooty, crackling Sweet, fruity Softens, chars slightly Source: From Krause, A. and Lange, A. 1969. Introduction to Chemical Analysis of Plastics. Iliffe Books Ltd, London; Ghosh, P. 1990. Polymer Science and Technology of Plastics and Rubber. Tata McGraw-Hill, New Delhi, India. 3. Halogens. The fusion is carried out as for nitrogen. The ﬁltrate (2 ml) is acidiﬁed with dilute nitric acid, boiled in the fume cupboard to remove H2S and HCN, and then a few drops of silver nitrate solution is added. A while precipitate, turning grey–blue, indicates chlorine, a yellowish precipitate indicates bromine and a yellow precipitate indicates iodine. In Beilstein’s test, which provides a simple means of detecting halogens, a minute quantity of the test material is placed on a copper wire (initially heated in a nonluminous bunsen ﬂame until the ﬂame appears colorless and then cooled) and heated in the outer part of the ﬂame. Carbon burns ﬁrst with a luminous ﬂame. A subsequent green to blue–green color, produced by volatilized copper halide, indicates halogen (chlorine, bromine, iodine). The different polymers may be classiﬁed into several groups according to the element present as shown in Table 3.9. The focus of identiﬁcation may be further narrowed down on the basis of other preliminary observations, e.g., fusibility or otherwise, melting point or range, heat distortion temperature, ﬂame tests and tests for thermal degradation, and solubility or extractability in water or different organic solvents. The solubility behaviors of common polymers are compared in Table 3.10. 3.7.3 Speciﬁc Tests When the observations and results of preliminary tests have been considered and most of the possible structures for the polymer base eliminated, an exact identiﬁcation can be made by carrying out speciﬁc tests [33]. Some speciﬁc tests for ready identiﬁcation of speciﬁc polymers are described below. 1. Tests for polystyrene and styrenic copolymers. The test depends on the aromatic rings of the styrene units in the polymer chain. The polymer sample (0.1 g) is heated under reﬂux in concentrated nitric acid and the clear mixture is poured into water (25 ml) to yield a pale yellow precipitate which is then extracted with ether (2 × 25 ml). The ether extracts are combined and washed thoroughly with water (2 × 5 ml), then extracted with dilute sodium hydroxide solution (2 × 5 ml) and ﬁnally with water (5 ml). The alkaline extracts and the aqueous extract are combined and the nitro compounds present in the mixture are reduced by adding granulated zinc (1 g) and concentrated hydrochloric acid and warming gently. The solution is cooled, ﬁltered and then diazotized with a dilute solution (5 ml) of sodium nitrite under ice-cooled condition. The solution is ﬁnally poured into excess of alkaline b-naphthol solution producing a deep red color. 398 Plastics Technology Handbook TABLE 3.9 Classiﬁcation of Most Common Polymers According to Elements Present Group Element Found Rubbersa Plastics/Fibers Group 1 N Nitrile (NBR), polyurethane (ester/ ether urethanes) Cellulose nitrate, silk, polyamides, polyimides, polyurethanes, polyacrylonitrile, SAN and ABS resins, urea-formaldehyde resins, melamine-formaldehyde resins, etc. Group 2 S S-vulcanized diene rubbers (NR, IR, SBR, BR, CR, IIR, EPDM, NBR), polysulﬁde rubbers and chlorosulfonated polyethylenes Wool, polysulfones, ebonite, etc. Group 3 Cl Polychloroprene, chlorosulfonated polyethylenes, etc. Group 4 Absence of N, S, Cl, etc. Peroxide cross-linked or unvulcanized hydrocarbon rubbers (NR, IR, SBR, BR, IIR, EPDM, EPR, etc.) Poly(vinyl chloride), poly(vinylidene chloride) and related copolymers, polychlorotriﬂuoroethylene, chlorinated or hydrochlorinated rubber, etc. Petroleum resins, coumaroneindene resins, cellulose and cellulosics other than cellulose nitrate, polyesters, polyethers or acetal resins, polyoleﬁns, polystyrene, poly(methyl methacrylate), poly(vinyl acetate/alcohol), etc. a See Appendix A3 for abbreviations. 2. Test for poly(vinyl chloride). Speciﬁc tests for chlorine containing polymers are performed only when the presence of chlorine is conﬁrmed by preliminary test. The simplest method of chlorine determination is the Beilstein test, described previously. Plasticizers are removed from the test material by ether extraction and the Beilstein test for chlorine is repeated to make certain that chlorine is still present. The material is then dissolved in tetrahydrofuran, ﬁltered and the polymeric product reprecipitated by adding methanol. On addition of a few drops of a ∼10% methanolic solution of sodium hydroxide to 2–5 ml of ∼5% solution of the plasticizer free material, the mixture changes with time from colorless to light yellow–brown, dark brown, and ﬁnally to black. In another test, poly(vinyl chloride) and vinyl chloride polymers readily form brown coloration and eventually dark brown precipitates when their pyridine solutions are boiled and treated with a few drops of methanolic sodium hydroxide (5%). 3. Test for poly(vinylidene chloride). Poly(vinylidene chloride), when immersed in morpholine, develops a brown color both in the liquid and the partially swollen polymer. The change takes place faster if the mixture is warmed in a water bath. 4. Test for poly(vinyl alcohol) and poly(vinyl acetate). When a small volume of iodine solution (0.2 g iodine and 1.0 g KI dissolved in 20 ml of a 1:1 alcohol–water mixture and diluted to 100 ml using 2N hydrochloric acid solution) is added to an equal volume of 0.25% neutral or acidic solution of poly(vinyl alcohol), a blue color develops almost immediately or on addition of a pinch of borax into the solution. Poly(vinyl acetate), however, turns deep brown on contact with the dilute iodine solution. 5. Test for polyacrylonitrile and acrylonitrile copolymers. When a strip of cupric acetate paper that is freshly moistened with a dilute solution of benzidine in dilute acetic acid is held in the pyrolytic vapors of polyacrylonitrole or its copolymers, a bright blue color develops readily. In another test, if the condensed pyrolyzate of polyacrylonitrile or its copolymers is made alkaline, boiled with a trace of ferrous sulfate, and then acidiﬁed, Prussian blue precipitate is readily produced. 6. Test for phenolic resins. The test material (dry) is heated in an ignition tube over a small ﬂame. The mouth of the tube is covered with a ﬁlter paper, prepared by soaking it in an ethereal solution of 2,6-dibromoquinone-4-chloro-imide and then drying it in air. After the material has been pyrolyzed for about a minute, the paper is removed and moistened with 1–2 drops of dilute ammonia solution. A blue color indicates phenols (care must be taken with plastics that contain 399 Plastics Properties and Testing TABLE 3.10 Solubility Behavior of Some Common Plastics Resin Soluble In Insoluble In Polyoleﬁns Polyethylene Dichloroethylene, xylene, tetralin, decalin (boiling) Polypropylene Chloroform, trichloroethylene, xylene, tetralin, decalin (boiling) Alcohols, esters, halogenated hydrocarbons Alcohols, esters Polyisobutylene Ethers, petroleum ether Alcohols, esters Poly(vinyl chloride) Dimethyl formamide, tetrahydrofuran, cyclohexanone Alcohols, hydrocarbons, butyl acetate Poly(vinylidene chloride) Polytetraﬂuoroethylene Butyl acetate, dioxane, ketones, tetrahydrofuran Insoluble Alcohols, hydrocarbons All solvents Polychlorotriﬂuoroethylene o-Chlorobenzotriﬂuoride (above 120°C) All solvents Polystyrene Benzene, methylene chloride, ethyl acetate Alcohols, water ABS Polybutadiene Chlorinated hydrocarbons, eg., p-dichlorobenzene Benzene Alcohols, water Aliphatic hydrocarbons, alcohols, esters, ketones Polyisoprene Benzene Alcohols, esters, ketones Acrylics Polyacrylonitrile Dimethylformamide and nitrophenol Alcohols, esters, ketones, hydrocarbons Polyacrylamide Water Alcohols, esters, hydrocarbons Esters of polyacrylic acid Aromatic hydrocarbons, esters, chlorinated hydrocarbons, ketones, tetrahydrofuran Aliphatic hydrocarbons Esters of polymethacrylic acid Aromatic hydrocarbons, chlorinated hydrocarbons, esters, ketones, dioxane Aliphatic hydrocarbons alcohols, ethers Poly(vinyl acetate) Aromatic hydrocarbons, chlorinated hydrocarbons, ketones, methanol Aliphatic hydrocarbons Poly(vinyl alcohol) Formamide, water Aliphatic and aromatic hydrocarbons, alcohols, ethers, esters, ketones Poly(vinyl acetals) Esters, ketones, tetrahydrofuran, (butyrals in 9:1 chloroformmethanol mixture) Aliphatic hydrocarbons, methanol Water, alcohol, benzene, chlorinated hydrocarbons, ethers, esters Petroleum ether, benzene, chlorinated hydrocarbons, alcohols, ethers, esters, ketones Benzyl alcohol, nitrated hydrocarbons, phenols Petroleum ether Water Poly(vinyl ethers) Methyl ether Ethyl ether Polyesters Polycarbonate Nylon Molded phenolic resins Molded amino resins (urea, melamine) Chlorinated hydrocarbons, cyclohexanone, dimethyl formamide, cresol Formic acid, phenols, triﬂuoroethanol Benzylamine (at 200°C) Benzylamine (at 160°C) Alcohols, esters, hydrocarbons (Only swelling in usual solvents) Alcohols, esters, hydrocarbons All common solvents All common solvents (Continued) 400 Plastics Technology Handbook TABLE 3.10 (CONTINUED) Solubility Behavior of Some Common Plastics Resin Soluble In Insoluble In Polyurethanes Noncross-linked Methylene chloride, hot phenol, dimethylformamide Petroleum ether, benzene, alcohols, ethers Cross-linked Polyoxymethylene Dimethylformamide Insoluble Common solvents All solvents Poly(ethylene oxide) Alcohols, chlorinated hydrocarbons water Petroleum ether Epoxy resins Uncured Alcohols, ketones, esters, dioxane, benzene, methylene chloride Aliphatic hydrocarbons, water Cured Cellulose, regenerated Practically insoluble Schweizer’s reagent Organic solvents Cellulose ethers Methyl Ethyl Water, dil. Sodium hydroxide Methanol, methylene chloride Organic solvents Water, aliphatic and aromatic hydrocarbons Cellulose esters (acetate, propionate) Ketons, esters Aliphatic hydrocarbons Cellulose nitrate Esters (ethyl acetate, butyl acetate, etc.), ketones (acetone, methyl ethyl ketone, etc.), mixtures (eg. 80% methyl isobutyl ketone + 20% isopropyl alcohol, 80% butyl acetate + 20% isopropyl alcohol) Aliphatic hydrocarbons (hexane, heptane, etc.) methyl alcohol, water Natural rubber Aromatic hydrocarbons, chlorinated hydrocarbons Petroleum ether, alcohols ketones, esters Chlorinated rubber Esters, ketones, linseed oil (80–100°C), carbon tetrachloride, tetrahydrofuran Aliphatic hydrocarbons Styrene–butadiene rubber Ethyl acetate, benzene, methylene chloride Alcohols, water Source: From Krause, A. and Lange, A. 1969. Introduction to Chemical Analysis of Plastics. Iliffe Books Ltd., London. substances yielding phenols on pyrolysis, e.g., phenyl and cresyl phosphate, cross-linked epoxide resins, etc.). 7. Test for formaldehyde condensate resins. Formaldehyde enters into the composition of several resins and polymers. Common examples are phenol-, urea-, and melamine-formaldehyde resins, polyoxymethylene and poly(vinyl formal). Formaldehyde is evolved when these are thermally treated or boiled with water in the presence of an acid (H2SO4). The aqueous extracts or acid distillates are treated with chromotropic acid (1,8dihydroxynaphthalene-3, 6-disulfonic acid). A few drops of 5% aqueous chromotropic acid solution are added to the aqueous test solution and then an excess of concentrated sulfuric acid is added, and the mixture preferably warmed to nearly 100°C for a few minutes. In the presence of formaldehyde, the solution turns violet/dark violet. Phenol, urea, and melamine are also obtained as intermediates on acid hydrolysis of the corresponding formaldehyde condensate resins and appropriate tests for phenol, urea, and melamine may be employed for identiﬁcation purposes. a. Phenol. On addition to the extract (10 ml), approximately 0.5N potassium hydroxide (8–10 ml) and 2 ml of diazotized p-nitroaniline (5% sodium nitrite solution added to an ice-cold solution of 2–5 mg of p-nitroaniline dissolved in 500 ml of approximately 3% hydrochloric acid, until the Plastics Properties and Testing 401 solution becomes just colorless), a red or violet color develops indicating phenol. No differentiation between phenol and its homologs is possible by this test. b. Urea and melamine. The aqueous extract is divided into two parts. (solutions 1 and 2). Solution 1 is made alkaline with dilute sodium hydroxide, and 1 ml of sodium hypochlorite solution is added. Solution 2 is treated with freshly prepared furfural reagent (5 drops of pure, freshly distilled furfural, 2 ml of acetone, 1 ml of concentrated hydrochloric acid, and 2 ml of water). For urea, solution 1 remains colorless and solution 2 becomes orange to red after 3–5 h. For melamine, solution 1 slowly becomes orange and solution 2 remains colorless. Detection of urea with urease also provides a useful differentiation between urea and melamine resins. For this test, the powdered sample (0.25 g) is placed in a 100 ml Erlenmeyer ﬂask and boiled with 5% sulphuric acid until the smell of formaldehyde has disappeared. The mixture is neutralized with sodium hydroxide (phenolphthalein as indicator). Then 1 drop of 1 N sulphuric acid and 10 ml of 10% urease solution is added, a strip of red litmus paper is placed in the vapors and the ﬂask is stoppered. The appearance of a blue color in the litmus paper after a short time indicates urea and thus the presence of urea resin. 8. Test for cellulose esters. Cellulose esters respond to the Molisch test for carbohydrates. The sample is dissolved in acetone and treated with 2–3 drops of 2% ethanolic solution of a-naphthol; a volume of 2–2.5 ml of concentrated H2SO4 is so added as to form a lower layer. A red to red– brown ring at the interface of the liquids indicates cellulose (glucose). A green ring at the interface indicates nitrocellulose and differentiates it from other cellulose esters. A more sensitive test for nitrocellulose is provided by an intense blue color reaction when a drop of a solution of diphenylamine in concentrated H2SO4 (5% w/v) is added to the sample in the absence of other oxidizing agents. 9. Test for polyamides. When a strip of ﬁlter paper soaked in a fresh saturated solution of o-nitrobenzaldehyde in dilute sodium hydroxide is held over the pyrolytic vapors of polyamides containing adipic acid, a mauve-black color is readily developed. Pyrolytic vapors of polyamides from diacids other than adipic acid produce a grey color when tested similarly. 10. Tests for natural rubber and synthetic rubbers. Rubbers may be identiﬁed by testing the pyrolytic vapors from test samples. A strip of ﬁlter paper soaked in an ethereal solution containing p-dimethylaminobenzaldehyde (3%) and hydroquinone (0.05%) and then moistened with a 30% solution of trichloroacetic acid in isopropanol produces different color reactions in the presence of pyrolytic vapors of different rubbers. Vapors from natural rubber (NR) produce a deep blue or blue–violet color, and those from styrenebutadiene rubber (SBR) pyrolysis turn the paper green or blue with a distinct green tinge. Polyisobutylenes and butyl rubber resemble NR, and silicone rubbers resemble SBR, in this color reaction. Pyrolytic vapors from nitrile rubbers, on the other hand, give a brown or brown–yellow color and those from polychloroprenes (neoprene) turn the test paper grey with a yellow tinge. In another test, pyrolytic vapors from polyisobutylenes and butyl rubbers produce a bright yellow color on ﬁlter paper freshly soaked in a solution obtained by dissolving yellow mercuric oxide (5 g) in concentrated sulfuric acid (15 ml) and water (85 ml) on boiling. Other rubbers may either produce little change or give an uncharacteristic brown color. Chromic acid oxidation provides a simple test for polyisoprenes. About 0.1 g of sample is gently heated in a test tube in the presence of chromic acid solution (5 ml) prepared by dissolving chromium trioxide (2 g) in water (5 ml) and adding concentrated sulfuric acid (1.5 ml). 1,4-Polyisoprenes (the synthetic equivalent of natural rubber) yield acetic acid, which can be readily identiﬁed by its odor. If the above test is positive, the sample may be further subjected to a modiﬁed Weber color test for polyisoprenes. For this test, an acetone-extracted polymer (0.05 g) is dissolved or suspended in carbon tetrachloride, treated with a little solution of bromine in carbon tetrachloride and heated in a water bath to remove excess bromine. The residue is then warmed with a little phenol. In the presence of polyisoprenes, 402 Plastics Technology Handbook the solid or the solution turns violet or purple and gives a purple solution with chloroform. The test is also positive for natural rubber and butyl rubber. 3.8 Plastics Analysis by Instrumental Methods Plastics analysis is used for many purposes, such as quality control, property prediction, determination of causes of failure, and reverse engineering or deformulation. However, plastics analysis usually poses a difﬁcult challenge to the analyst. This is because plastics may be made of more than one polymer as blends or copolymers and, moreover, may contain a host of additives (to enhance or impart speciﬁc properties), such as inorganic ﬁllers, plasticizers, antioxidants, ﬁre retardants, antistatic agents, cross-linkers, and so on. In fact, a plastic compound may contain more than 10 different ingredients, and some at very low levels, making it unlikely that more than 90%–95% of the plastics formulation can be determined by analysis alone. For a plastics analyst to succeed in this scenario, he should have a good working knowledge of plastics technology. A wide variety of instrumental techniques can be used for plastics analysis. These can be categorized as spectroscopic techniques, chromatographic techniques, and thermal techniques. The spectroscopic techniques include infrared (IR) spectroscopy, ultraviolet (UV) light spectroscopy, nuclear magnetic resonance (NMR) spectroscopy, atomic absorption spectroscopy, x-ray ﬂuorescence spectroscopy, Raman spectroscopy, and energy dispersive analysis. The chromatographic techniques include gas chromatography–mass spectrometry (GC–MS), liquid chromatography–mass spectrometry (LC–MS), gel permeation chromatography (GPC), and thin-layer chromatography (TLC). The thermal techniques include differential scanning calorimetry (DSC), dynamic mechanical thermal analysis (DMTA), and thermogravimetric analysis (TGA). In the present section, however, only two spectroscopic techniques, namely, IR spectroscopy and NMR spectroscopy, which are simple, rapid, convenient, and most widely used for characterization of polymers, will be presented, in fair detail, highlighting their theoretical basis, instrumental aspects, methods of operation, and various applications, with speciﬁc examples, for both qualitative and quantitative analyses of different types of polymers. 3.8.1 IR Spectroscopy The basis of IR spectroscopy [34] rests on the interaction of electromagnetic radiation with mass in the IR region, which ranges from 0.7 to 1000 mm in wavelength, that is, 14,000 to 10 cm−1 in wavenumber. [The ), is the number of waves per centimeter and is wavenumber (cm−1), represented by the symbol “nu bar” (n equal to 104/l, where l is the wavelength in micrometers (mm) with 1 mm = 10−4 cm. Note, however, that is related to frequency, n the wavelength unit is also cited often simply as micron (m). On the other hand, n s−1 , where c is the velocity of light in cm/s.] According to quantum mechanics, atoms and (s−1), by n = cn molecules can hold only certain deﬁnite quantities of energy, or exist in speciﬁc states. If E1 and E2 are is absorbed only when discrete values of energy corresponding to two states, radiation of wavenumber n Equation 3.132 holds, where h is Planck’s constant: = ðE2 − E1 Þ=hc n (3.132) The absorbed energy of the radiation changes the state of the atom or molecule from an initial state (E1) to a ﬁnal state (E2). This is related to an absorption band, which may be deﬁned as a range of wavelengths or frequencies in the electromagnetic spectrum that are characteristic of a particular transition from an initial to a ﬁnal state in a substance. The energy of IR radiation is small to cause transitions between electron energy levels. It can only bring about changes in vibrational and rotational states of a molecule, which are closely related to the molecular structure. 403 Plastics Properties and Testing Bond stretching Bond stretching (symmetric) Bond stretching (asymmetric) A B A B C A B C A B A B A B C C B Bending A B C A C FIGURE 3.91 Two atoms A and B joined by a covalent bond may move back and forth along the bond axis producing bond stretching vibrations, whereas three atoms A, B, and C joined by covalent bonds can produce both stretching and bending vibrations. Bond vibrations are the physical basis of IR spectroscopy. Two atoms joined by a covalent bond may move back and forth along the bond axis, producing a bond stretch, as shown in Figure 3.91, where arrows indicate the directions of atom movement and relative motions of the atoms. Though polymer molecules contain a large number of atoms, their IR spectra are relatively simple. The reason for this lies in the fact that each linear or branched polymer molecule contains end groups joined by a large sequence(s) of repeat units. A regular repetition of a series of identical units in the chain would mean that the actual number of spectroscopically active vibrational modes is relatively small. This concept of group frequencies is very useful for practical interpretation of IR spectra. For a large molecular weight polymer, moreover, the concentration of the end groups is relatively small, and so their absorption bands also are weak. The main contribution to the absorption spectrum then comes usually from vibrations of atoms or groups in the repeating units. This simpliﬁes greatly the analysis of polymers by IR spectroscopy. For simple qualitative work, a double-beam spectrometer, operating between 2.5 and 15 mm, is adequate, since the region between the visible and 2.5 mm is of little use for qualitative identiﬁcation work, while very few characteristic bands are observed beyond 15 mm. 3.8.1.1 Methods of Measurement IR spectra can be obtained by two methods—the transmission (or absorption) method and the reﬂection method. In the ﬁrst method, the fraction of the incident light that is transmitted (or absorbed) is measured, yielding an absorption spectrum, while in the second method, known as attenuated total reﬂection (ATR), a reﬂection spectrum is obtained, which is superﬁcially very similar to an absorption spectrum. In the ATR method [35], the incident radiation beam is passed through a small prism of a material of high refractive index (e.g., silver chloride or thallium bromoiodide) in such a way that the beam suffers total internal reﬂection at one face of the prism. When a sample whose spectrum is to be recorded is pressed into contact with this prism surface (at which the internal reﬂection occurs), the sample attenuates the reﬂection from this surface in such a manner that a spectrum is obtained. This ATR spectrum shows a remarkable similarity to a transmission spectrum, provided there is good contact between the prism surface and the sample. Since a beam penetrates only a few microns into the sample in the ATR method, the thickness of the sample beyond the depth of penetration is of no signiﬁcance and the spectral absorbance is independent of the sample thickness. It is thus found that with samples over about 0.001 in (0.02 mm) thick, the ATR spectrum is independent of the sample thickness and laborious sample 404 Plastics Technology Handbook preparation procedures can be avoided. A particularly important application of the ATR method is thus in the examination of surface coatings and laminates. More common among the two IR methods described above is the transmission method in which the numerical values of percentage transmission, which depend on the sample thickness, are recorded. There are several methods of presenting these spectral data. A common method is to plot percentage transmission or fractional transmission (i.e., percentage transmission divided by 100) on the vertical axis (increasing upward) and to plot wavelength (mm or, simply, m) or wavenumber (cm−1) on the horizontal axis, increasing to the right for wavelength and decreasing to the right for wavenumber. Sometimes, the vertical axis is plotted in reverse with zero transmission (assumed, incorrectly, as 100% absorption) at the top of the ordinate, thus producing an absorption spectrum as a mirror image of the transmission spectrum. The effect is simply to reverse the direction of the vertical scale, but the presentation is not affected. Precalibrated charts are also often used on which the vertical axis is marked “Absorbance,” deﬁned as Absorbance = log10 100 Percentage transmission (3.133) The absorbance scale is nonlinear but since the horizontal scale is not affected, qualitative identiﬁcation of the spectrum is easily done (see Figures 3.91 and 3.92). 3.8.1.2 Instruments The construction principles of conventional IR spectrophotometers (based on dispersive technique) do not differ a great deal from one another. They consist of source of radiation, monochromator, and radiation detector with recording equipment. Common sources of radiation are Globar (i.e., a rod of silicon carbide ﬁred to 1000–1200°C) and the Nernst lamp (the ﬁlament of which is composed of rare earth metal oxides) heated to about 2000°C. The emitted energy of both sources is maximum in the nearIR region (5000–10,000 cm−1) and then decreases rapidly. The beam emanating from the source is split into two paths, that is, the sample beam and the reference beam. The sample beam passes through the sample to be analyzed, while the reference beam passes through the reference cell containing solvent that has been used for making the sample solution. The monochromator disperses both the sample and reference beams according to the wavelength for which prisms and diffraction gratings are used. The intensity of the beam that is passed through the sample and that of the reference beam are compared in the photometric part of the instrument. The recording assembly presents the ratio of the two intensities or the logarithm of the reciprocal of this ratio (i.e., absorbance) as a function of wavelength or wavenumber. Thermocouples, bolometers, or pneumatic Golay cells serve as radiation detectors. 3.8.1.3 Sample Preparation Sample preparation is the ﬁrst important step in IR analysis of polymers. Samples for IR analysis by transmission may be prepared in the form of solution, as ﬁlms and, if insoluble, in pellet or disk form with alkali halide (e.g., potassium bromide), or in nujol mulls. The use of solutions is limited by the large number of absorption bands of most solvents and by the necessity of using at least two solvents if the whole range of spectrum is to be recorded. Pelletization with potassium bromide powder is advantageous for insoluble samples. (The concentration of the sample in KBr should be in the range of 0.2% to 1%.) Instead of potassium bromide, other suitable materials such as thallium bromide, silver chloride, or polyethylene can be used as the basis of sample pellets. Mulls in nujol placed between AgCl, NaCl, KBr, or CsI plates or KBr pellets are also suitable for insoluble samples. (Nujol is a heavy parafﬁn oil with essentially alkane formula CnH2n+2, where n is very large.) The halides are ionic and usually do not have any absorption peaks in the range normally covered in the IR spectrum. However, the presence of absorption bands attributed to CH2 in nujols makes the analysis of vibrations of methylene groups impossible. Polymer samples can also be prepared as very thin platelets by cutting with a microtome, or in the form of ﬁbers if special microscopic equipment is available. 405 Plastics Properties and Testing 1.0 (a) Absorbance 0.8 0.6 0.4 0.2 1.0 (b) Absorbance 0.8 0.6 0.4 0.2 6.0 (c) Absorbance 5.0 4.0 3.0 2.0 1.0 4000 3000 2000 Wavenumber (cm–1) 1000 400 FIGURE 3.92 IR absorption spectra of (a) polyethylene, (b) polypropylene, and (c) polystyrene. (Adapted from NICODOM IR Libraries, http://www.ir.spectra.com.) For the examination of polyethylene and a-oleﬁn resins, such as polypropylene, and generally for thermoplastic resins that do not dissolve rapidly, hot pressing can be a most convenient and quick method of preparing ﬁlms. The polymer is pressed between polished stainless steel plates (using a temperature at which plastic ﬂow occurs readily) in a small hydraulic press (which generates a ram pressure of about 2 kN/cm2) and cut to the same size as platens. To assist in stripping the pressed ﬁlm, the plates are coated with polytetraﬂuoroethylene (PTFE). For ﬁlms with thicknesses greater than about 0.05 mm, feeler gauges may be used as spacers, but for thinner ﬁlms, the sample thickness is adjusted by altering the amount of material used. (Feeler gauges are small lengths of steel leaves or blades of different thicknesses used to check the parallelism and measure the clearance between two parts, typically in thousandths of an inch or hundredths of a millimeter.) [Note: If thin ﬁlms are used, thicknesses differing by an order of magnitude may be required for different polymers or special regions. Thus, for a saturated hydrocarbon polymer, such as polyethylene, 406 Plastics Technology Handbook 0.3 mm may be satisfactory, whereas for materials containing oxygenated groups, e.g., poly(methyl methacrylate), a reduction to 0.03 mm (which represents the lower limit obtainable by hot pressing) or less is necessary.] Casting ﬁlms from solvents is often preferred for latices and soluble resins, as it requires no special apparatus and very thin samples are easily prepared. Moreover, if the resin can be taken into solution, ﬁllers can be removed by ﬁltering or centrifuging before casting the ﬁlm. Films may be cast on a glass plate and subsequently stripped off, or cast directly on a sodium chloride or potassium bromide plate. Aqueous latices may be cast on silver chloride or thallium bromoiodide plates. Solvent may be removed from the ﬁlm by the use of an IR lamp and vacuum oven. Very volatile solvents (e.g., ether, acetone, methylene chloride) should be avoided as they evaporate very rapidly, causing moisture condensation on the sample surface. If nothing is known of the nature of a resin, the following solvents may be tried in the order stated [35]: 1. Ethylene dichloride or 1,2-dichloroethane (dissolves a wide range of thermoplastic resins, including a majority of vinyls and acrylics) 2. Toluene (particularly useful for polyethylene and a-oleﬁn polymers and copolymers) 3. Methyl ethyl ketone (particularly suitable for butadiene copolymers, but more difﬁcult to remove than the aforesaid solvents) 4. Water (dissolves a number of resins rich in –OH, –CO2H, or amine groups) 5. Formic acid (very suitable for polyamides and linear polyurethanes) 6. Dimethyl formamide (suitable for polyacrylonitrile, polyvinyl ﬂuoride, polyvinylidene chloride, and some other resins) For rubbers, and especially for carbon black-ﬁlled compositions, a useful method [36] is a prior extraction with acetone-chloroform mixture followed by reﬂuxing with p-cymene-xylene mixture to dissolve the polymer. Though the alkali halide (e.g., KBr) pellet or disk method is probably the most widely used procedure for preparing samples from simple (non-polymeric) solids, the majority of resins are very difﬁcult to disperse in alkali halides and the spectra obtained by this method are inferior to those measured on solid ﬁlms. Hence, in polymer work, the halide disk method is applied only to those insoluble resins and rubbers that do not respond to the various methods of ﬁlm preparation. In some cases, it is advantageous to pre-grind the polymer before adding the halide powder. It is worth noting, moreover, that some polymers in powder form, especially PTFE and some linear polythenes, will sinter into disks without the addition of alkali halide powder in the usual disk-making procedure. A modiﬁcation of the disk method, which is often successful with rubbers, depends on cooling the rubber below its brittle point (in liquid nitrogen) and grinding it, while in this condition, with KBr, before pelletization. The disk may then be dried by heating in an oven at 140°C for a few minutes. 3.8.1.4 Fourier Transform IR Spectroscopy In the dispersive spectroscopy technique, described above, a monochromatic light beam is shone at the sample, the amount of light absorbed is measured, and the process is repeated for each different wavelength or frequency. The spectrometer is set to start reading at one end and the frequency is swept smoothly across the whole span of the spectrum. The method is thus inherently slow. The inefﬁciency of such a method becomes obvious when one considers taking a spectrum with only one or two sorption peaks in it—with the spectrometer set to sweep from one end to the other, most of the time in this case is spent recording nothing but background noise! A solution to this problem is to adopt a method for measuring all of the aforesaid frequencies simultaneously, rather than individually. This is done in Fourier transform infrared (FT-IR) spectroscopy. FT-IR spectroscopy employs a very simple optical device called an interferometer. The interferometer produces a unique type of signal that has all of the IR frequencies “encoded” into it. In the normal instrumental process, IR radiation is emitted from a glowing black-body source. This beam containing the full spectrum of wavelengths passes through an aperture (which controls the amount of energy) and enters Plastics Properties and Testing 407 the interferometer where the spectral encoding takes place. The resulting interferogram signal then exits the interferometer and enters the sample compartment where it is transmitted through or reﬂected off the surface, depending on the type of analysis (transmission spectroscopy or ATR) being accomplished. Speciﬁc frequencies of energy, which are uniquely characteristic of the sample, are absorbed and the beam then passes to a detector that is capable of measuring the special interferogram signal. The measured signal is sent to a computer, which “decodes” that information to produce a conventionaltype IR spectrum of transmittance (or absorbance) versus wavelength (or wavenumber). Since the signal can be measured very quickly, usually on the order of 1 s or so, the time element per sample is reduced to a matter of a few seconds rather than several minutes of the conventional dispersive method, thus increasing the overall speed, typically by a factor of 10 to 1000. The decoding is accomplished via a wellknown mathematical technique called the Fourier transformation (hence the name “Fourier transform infrared spectroscopy”). The FT-IR spectroscopy is preferred over dispersive or ﬁlter methods of IR spectral analysis for several reasons: (i) it is a precise measurement method that requires no external calibration; (ii) it can increase speed, collecting a scan every second or so; (iii) it can increase sensitivity, for example, by adding 1-s scans together to ratio out random noise; (iv) it has greater optical throughput; and (v) it is mechanically simple with only one moving part. 3.8.1.5 Qualitative Analysis With an increase in the number of atoms in a molecule, the number of normal modes increases rather rapidly and a detailed analysis of the vibrational spectrum becomes impossible. One is therefore content to assign the strongest bands and to identify some of the weaker ones as overtones or combinations. For practical purposes, however, a number of very useful generalizations can be made, as shown below. A particularly important spectral region is 1430–910 cm−1. This region contains many absorptions caused by bending vibrations and also the absorptions caused by several stretching vibrations (see Table 3.11). Since bending vibrations in a molecule are in general more numerous than stretching vibrations, this region of the spectrum is particularly rich in absorption bands. It is seldom possible to assign these bands to particular modes of vibration but the complex of bands as a whole is highly typical of the given molecular structure and is ascribed to skeletal vibrations, which involve all the atoms to much the same extent. Thus, these bands are often referred to as the ﬁngerprint bands, because a molecule or structural moiety may often be recognized merely from the appearance of this part of spectrum. This frequency region, frequently called ﬁngerprint region, is extremely useful in establishing conclusively the identity of two samples. Similar molecules, for instance, may show very similar spectra at frequencies higher than 1430 cm−1, but in the ﬁngerprint region, there will usually be discernible differences. Many commonly occurring functional groups such as −CH3, >C=O, −NH2, and so on give rise to one or more characteristic absorption bands when they are present in a molecule. These bands (group frequencies) are, to a considerable extent, independent of the structure of the molecule as a whole and can be used for analysis. For example, all compounds containing a −CH3 group possess absorption bands in the region of 3000 and 1400 cm−1. Similarly, all compounds with a >C=O group have a strong band in the region of 1700 cm−1. With a few exceptions, these characteristic group frequencies fall in the regions well above or well below the ﬁngerprint region. Table 3.11 contains some of the more important group frequencies along with a qualitative indication of their intensities. A close look at Table 3.11 reveals some logical trends in group frequencies. We see, for instance, that the stretching vibration frequency decreases in the series C–H, C–F, C–Cl, C–Br, and C–I. This trend is attributed to increasing mass of the atom within the group in the same sequence. This also accounts for the fact that the stretching frequency of C=S is less than that of C=O. On the other hand, an increase in the strength of the bond should increase the vibration frequency. This trend is observed in the series C–X, C=X, C≡X, where X is C or N, and also in C–O and C=O. When studying the spectrum of an unknown polymer, it is desirable to have the results of the qualitative elements test available (see Section 3.7.2) on the same polymer. As there is considerable overlapping of the 408 Plastics Technology Handbook TABLE 3.11 Characteristic IR Absorption Frequencies of Organic Functional Groups Type of Vibration Frequency Range (cm−1) C–H (alkane) Stretch 2850–3000 Strong –C–H (alkane) =C–H (alkene) Bending Stretch 1350–1480 3010–3100 Variable Medium Functional Group Intensity of Peak =C–H (alkene) Bending 675–1000 Strong C–H (alkyne) C–H (aromatic) Stretch Stretch 3300 3000–3100 Strong, sharp Medium C=C (alkene) Stretch 1620–1680 Variable –C≡C– (alkyne) C=C (aromatic) Stretch Stretch 2100–2260 1400–1600 Variable Medium-weak, multiple C–F Stretch 1000–1400 Strong C–Cl C–Br Stretch Stretch 600–800 500–600 Strong Strong C–I O–H (alcohol) O–H (alcohol) Stretch 500 Strong Stretch (H-bonded) Stretch (free) 3200–3600 3500–3700 Strong, broad Strong, sharp O–H (acid) Stretch 2500–3300 Strong, very broad C–O (alcohol) C–O (acid) Stretch Stretch 1050–1150 1210–1320 Strong Strong C–O (ester) Stretch 1000–1300 Two bands or more C=O (carbonyl) C=O (acid) Stretch Stretch 1670–1820 1700=1725 Strong Strong C=O (anhydride) Stretch 1800–1830 and 1740–1775 Two bands C=O (ester) C=O (amide) Stretch Stretch 1735–1750 1640–1690 Strong Strong N–H (amine) Stretch 3300–3500 Medium (primary amines have two bands; secondary have one band, often very weak) N–H (amine) Bending 1600 Medium N–H (amide) N–H (amide) Stretch Bending 3100–3500 1550–1640 Two bands C–N (amine) Stretch 1080–1360 Medium-weak CN (nitrile) Stretch 2210–2260 Medium Source: Silverstein, R. M., Bassler, G. C., and Morrill, T. C. 1981. Spectrometric Identiﬁcation of Organic Compounds, 4th ed., John Wiley & Sons, New York. wavelength ranges in which different groups absorb, much time could be saved by immediately rejecting those groups excluded by the elements test. Polymers consisting of carbon and hydrogen, and carbon, hydrogen, and oxygen are usually not separated in the simple elements test. But, in most cases, they can be distinguished without difﬁculty from their characteristic bands in the IR spectrum. Also for polymers that show the presence of additional elements, such as chlorine and sulfur, in the elements test, the characteristic bands arising directly from groups containing these additional elements must be considered along with characteristic bands for hydrocarbons and various oxygen-containing groups. It should be noted that, in some cases, the presence of additional elements may shift the position of these latter groups. Common types of groups occurring in polymers and IR absorption bands associated with them are discussed below. Plastics Properties and Testing 409 Hydrocarbon Groups. A hydrocarbon-type polymer is indicated if the elements test proves to be negative for all elements except carbon and hydrogen. An IR analysis can then be undertaken. All hydrocarbon compounds, composed of only carbon and hydrogen, exhibit an IR band, usually with multiplicity of absorption peaks, near 3000 cm−1 and another between 1470 and 1430 cm−1. Saturated, unsaturated, or aromatic hydrocarbons are distinguished by the absorption bands beyond 1000 cm−1. Unsaturated hydrocarbon groups produce highly characteristic bands between 1000 and 600 cm−1. These bands thus represent a simple method of determining the structure of polythene and of diene polymers. Aromatic structures produce relatively intense absorption bands in the 830 to 600 cm−1 region depending on the type of substitution on the benzene ring. Most aromatic compounds also have one or more sharp bands of weak intensity between 1700 and 1430 cm−1. Figure 3.92, for example, shows the IR absorbance spectra of three common hydrocarbon-type polymers composed of only carbon and hydrogen—polyethylene, polypropylene, and polystyrene. Both polyethylene (Figure 3.92a) and polypropylene (Figure 3.92b) are marked by characteristic absorbance peaks at around 3000, 1500, and between 700 and 750 cm−1. These three peaks result from C–H and C–C absorptions (see Table 3.11) in the IR region. The IR spectrum of polystyrene (mostly atactic and amorphous), however, shows absorbance bands at 3026 and 2849 cm−1, corresponding to aromatic and aliphatic –C–H stretchings (see Table 3.11), respectively. The peaks at 1601 and 1493 cm−1 are assigned to aromatic –C=C– stretchings. The –C–H deformation vibration band of the ﬁve adjacent hydrogens of the benzene ring appears at 758 cm−1, while the ring deformation vibration is observed at 700 cm−1. The peaks at 3000 and 3100 cm−1 are attributed to the –C–H stretching vibrations [37] of ring hydrogens (see Table 3.11). Halide Groups. Once the presence of halogens in a given polymer is conﬁrmed by preliminary elements test, further conﬁrmation can be obtained by conducting an IR spectral run on the polymer and comparing with available spectra. Teﬂon or polytetraﬂuoroethylene (PTFE) with the general formula –[–CF2–CF2–]–, for example, exhibits a strong and distinctive absorption band near 1200 cm−1 owing to the >CF2 stretching vibration. This band is a multiplet and consists of three peaks at 1240, 1215, and 1150 cm−1. Other major bands are located at 641, 554, and 515 cm−1, and these are assigned to the >CF bending modes. In the IR spectrum of poly(vinyl chloride) (PVC), another common polymer with halide groups and having the general formula –[–CH2–CHCl–]–, the peaks found in a range of 2800–3000 cm−1 (Figure 3.93a) correspond to –C–H stretch. The peak at higher wavenumber is assigned to the asymmetric stretching and the lower peak is assigned to the symmetric stretching vibration of –C–H. The peaks around 1400 cm−1 are assigned to –C–H aliphatic bending vibration, while the peak at 1250 cm−1 is attributed to the bending vibration of –C–H near Cl. The –C–C– stretching vibration of the PVC backbone chain occurs in a range of 1000–1100 cm−1. Finally, the peaks in a range of 600–650 cm−1 correspond to the C–Cl bond. Hydroxyl Groups. A careful examination of the hydroxyl stretching region (Table 3.11) can often be valuable in the determination of structure of unknown polymeric compounds. Thus, the presence of a band in the 3700–3150 cm−1 (2.7–3.2 mm) region is a very reliable indication of the presence of hydroxyl groups. [Note, however, the following two exceptions. Since water shows strong absorption in this region, the sample must be fully dry for hydroxyl observation. Halide powders being rarely dry, halide disks for sample preparation are thus best avoided for measurement of hydroxyl groups. Also, N–H groups, if present, can cause interfering absorptions in the hydroxyl region (see Table 3.11).] For illustration, Figure 3.93b shows the IR absorption spectrum of poly(vinyl alcohol) (PVA) with the general formula –[–CH2–CH(OH)–]–. While strong hydroxyl bands of free alcohol, that is, nonbonded –O–H stretching bands, usually occur at 3600–3650 cm−1, PVA exhibits a bonded –O–H stretching band at 3200–3570 cm−1, which would be attributed to the formation of intramolecular and intermolecular hydrogen bonds among PVA chains owing to high hydrophilic forces. The band at 2940 cm−1 is assigned to –C–H stretching, whereas the bands (doublet) at 1330–1440 cm−1 are likely to arise from mixed –C–H and –O–H in-plane bending vibrations [38]. Carbonyl Groups. For compounds containing carbonyl (C=O) groups, the precise range of absorption frequency 1700–1850 cm−1 is often sufﬁcient to determine their presence. [However, the carbonyl group 410 Plastics Technology Handbook 1.0 (a) Absorbance 0.8 0.6 0.4 0.2 1.0 (b) Absorbance 0.8 0.6 0.4 0.2 1.0 (c) Absorbance 0.8 0.6 0.4 0.2 0 4000 3000 2000 Wavenumber (cm–1) 1000 400 FIGURE 3.93 IR absorption spectra of (a) poly(vinyl chloride), (b) poly(vinyl alcohol), and (c) poly(vinyl acetate). (Adapted from NICODOM IR Libraries, http://www.ir.spectra.com.) may bond so strongly to an existing hydroxyl group in its vicinity that the band of the latter in the usual range 3150–3700 cm−1 could become difﬁcult, or impossible, to observe. For example, in carboxylic acids, it is evident only as a broadening of the C–H band at 2870 cm−1.] For illustration, the IR spectrum of poly (vinyl acetate) (PVAc), with the general formula –[–CH2–CH(OCOCH3)–]–, is shown in Figure 3.93c. The most prominent band in this spectrum is seen at 1736 cm−1, which may be assigned to the carbonyl (–C=O) stretching vibration (see Table 3.11). The adjacent region from 1700 to 500 cm−1 is complex. It is composed of stretching –C–O vibration, rocking, wagging, and twisting vibrations of –CH2 groups, out-of-plane bending vibrations of the –C–H chain, and one or more stretching vibrations of the polymer chain [39]. Nitrogen-Containing Groups. If the elements test on an unknown polymeric compound reveals the presence of nitrogen, it then becomes imperative to conduct IR spectral runs to determine the type of 411 Plastics Properties and Testing nitrogen-containing group in the polymer. Nitrogen-containing groups commonly found in synthetic polymers are amines, amides, imide, nitrile, and urethane. In IR spectra, primary amines (–NH2) exhibit two N–H peaks, one near 3350 and one near 3180 cm−1, from asymmetric and symmetric stretching vibrations, respectively. Secondary amines (–NHR), however, give rise to one –N–H stretch peak at 3300. All amides produce a very strong –C=O peak at 1680–1630 cm−1 (usually with the frequency lowered due to hydrogen bonding, if present). In addition, primary amides (–CONH2) and secondary amides (–CONH–) exhibit IR absorptions owing to –N–H bending at 1640–1550 cm−1, with primary amides showing two spikes, as in the case of amines, and secondary amides showing only one spike. Figure 3.94a, for example, shows the IR spectrum of nylon-6, which is a secondary polyamide with the general formula –[–(CH2)5–CONH–]–. The spectrum features, as expected, three strong absorption bands at about 1.0 (a) Absorbance 0.8 0.6 0.4 0.2 1.0 (b) Absorbance 0.8 0.6 0.4 0.2 1.0 (c) Absorbance 0.8 0.6 0.4 0.2 0 4000 3000 2000 Wavenumber (cm–1) 1000 FIGURE 3.94 IR absorption spectra of (a) nylon-6, (b) polyurethane, and (c) polydimethylsiloxane. (Adapted from NICODOM IR Libraries, http://www.ir.spectra.com.) 412 Plastics Technology Handbook 3294 cm−1 (hydrogen bonded >N–H stretch), 1645 cm−1 (C=O), and 1545 cm−1 (CONH). The two moderate peaks at 2932 and 2862 cm−1 are attributed to CH2 asymmetric stretching and CH2 symmetric stretching, respectively. In addition, a number of small peaks appear in the 1400–500 cm−1 region. These are attributed to NH, CO, CONH, CN, methylene sequences, and crystalline forms. It may be mentioned that the major bands of nylon IR spectra are characteristic for polyamides and make identiﬁcation of the generic class a simple task. However, the features distinguishing the subgeneric classes, such as nylon-6 and nylon-6,6, are more subtle. For example, nylon-6 and nylon-6,6 have similar IR spectra and are differentiated by the presence of a weak crystalline band near 935 cm−1 in the nylon-6,6 spectrum. For polyimides, the main bands in the IR spectrum are those attributed to the carbonyl (C=O) groups in the imide [–(–CONRCO–)–] ring, which appear as a doublet at 1780 and 1720 cm−1. The band at 1780 cm−1 is sharp, while that at 1720 cm−1 is broader and stronger. The spectra of polyamide–imides show the characteristic bands of the carbonyl groups of both the amide and imide structure. One can thus see a series of four bands of relatively high intensity in the 1600–1800 cm−1 region, which is a clear indication that the unknown resin sample is a polyamide–imide. The nitrile (–C≡N) group produces an absorption band at 2240–2280 cm−1, a region that is relatively free of other well-known absorption bands and the band can therefore be easily observed. The major characteristic of the IR spectrum of polyacrylonitrile with the general formula –[–CH2–CH(CN)–]– is thus the nitrile band at 2240 cm−1. Polyurethanes containing the urethane linkage [–NH–CO–O–] exhibit the prominent band pair noted earlier for the secondary amide structure, but the pair now occurs at about 1540 and 1700 cm−1. These bands are easily recognized in simple urethanes and in polyether and polyester urethane rubbers. Figure 3.94b, for example, shows a typical polyurethane IR spectrum. The characteristic band at 1730 cm−1 in this spectrum is associated with the C=O group in polyurethane, while other bands are assigned as follows. The absorption band at 3320 cm−1 corresponds to >NH stretching and the sharp peaks at 2860 and 2940 cm−1 are associated with –CH2– stretching, while other modes of –CH2– vibrations give rise to the bands in the region 1500–1300 cm−1. The bands at 1540 cm−1 are attributed to the group of –NH– vibrations. Silicon-Containing Groups. Polymeric organosilicon compounds are commonly referred to as silicones, the most widely used silicone resin being polydimethylsiloxane, with the general formula –[–Si(CH3)2– O–]–. Methyl groups attached to a silicon atom undergo the same C–H stretching and bending vibrations as a CH3 attached to a carbon atom, but the positions of the bands for a Si–CH3 group are different from those for a C–CH3 group, because of electronic effects. The absorption attributed to the umbrella mode (symmetric bend) vibration of the Si–CH3 group produces a very intense band at 1260 ± 5 cm−1, and when a silicon atom has two methyl groups attached to it, denoted as Si(CH3)2, there appears a strong methyl rocking mode band at 800 ± 10 cm−1. The pattern of bands in the spectrum of polydimethylsiloxane is very characteristic—a series of four intense bands between 1200 and 800 cm−1 (see Figure 3.94c). Few other materials give rise to this pattern. The band attributed to the Si–H group occurs at about 2200 cm−1 and is exceedingly prominent. The absorption owing to the Si–O linkage, which forms the backbone of silicone resins, occurs between 1100 and 1000 cm−1, producing a broad, complex, and intense band. Sharp bands arise near 1250 and 1430 cm−1 owing to Si–Me (as noted above) and Si–Phenyl groups, respectively. The group most frequently encountered, however, is the Si–Me group at 1250 cm−1. This band is particularly easy to observe, even in the presence of other materials absorbing in this region of the spectrum. Silicones containing the Si–H group are also easily recognized in mixtures, since few other substances have signiﬁcant absorption in the region where the strong Si–H band appears (2100– 2220 cm−1). The absorption band attributed to the OH group attached to Si is, however, similar to that of the alcoholic OH group. Applications. An IR spectrum may be looked upon as a “ﬁngerprint” of a sample in the form of absorption peaks that correspond to the frequencies of vibrations of the bonds between the atoms making up the material. Because each different material is a unique combination of atoms, no two compounds Plastics Properties and Testing 413 produce exactly the same IR spectrum. Therefore, IR spectroscopy can provide a positive identiﬁcation of every different kind of material, though it can be reasonably expected that unambiguous identiﬁcation of hydrocarbon polymers by IR spectral comparison alone will be difﬁcult because of the similarity of spectra of various possible isomers. However, polymers containing various groups with special elements such as O, S, N, and so on are analyzed relatively easily because they may be analyzed via functional groups. Since the IR spectrum can be looked upon as the molecular ﬁngerprint of a sample, IR spectral analysis can be used in many cases to identify unknown materials and check the quality or consistency of a sample. Moreover, since the size of the peaks in the spectrum is a direct indication of the amount of material present, IR spectroscopy can also be used for quantitative analysis (explained later). Since many functional groups can be easily detected and quantiﬁed by IR spectroscopy, this technique can be conveniently used in many cases to monitor the quality and state of polymeric materials exposed to environmental and other conditions. For example, many polymers easily undergo oxidation, which is indicated by the appearance of an absorption band of the C=O group near 1720 cm−1. Thermal and photo-chemical (UV-induced) oxidations of polyethylene plastics and fabrics may lead to the formation of hydroxyl (of hydroperoxide/alcohol) and carbonyl (of carboxyl and anhydride) groups, which are easily identiﬁed and measured by IR spectroscopy. Since the quality and performance of plastic products depend on the quality of polymer components used in their manufacture, proper identiﬁcation and quality testing are critically important for the plastics industry. IR spectral data can be used for identiﬁcation of polymer samples, qualitative analysis of polymer starting materials, or analysis of in-process samples and product quality control. Comparison of measured spectral data with spectral reference databases provides a rapid and effective identiﬁcation tool for all types of polymer materials, and all sizes and forms including pellets, parts, opaque samples, ﬁbers, powders, wire coatings, and liquids. In appropriate cases, chemical reactions, such as analysis of esteriﬁcation of cellulose by carboxylic acids, can be monitored by IR spectroscopy. The aim of qualitative analysis of polymer mixtures is to determine the presence of individual components and, in the case of copolymers, to determine the presence of individual monomer units. This can be accomplished by considering that the spectrum of a mixture is additively composed of the spectra of the individual components and that all absorption bands of the spectrum should be ascribed to individual components, no band being in surplus and none missing. This is exempliﬁed in Figure 3.95, which presents the IR spectra of butadiene–styrene, butadiene–acrylonitrile, and butadiene–styrene– acrylonitrile copolymers. In the ﬁrst and the third spectra, the characteristic absorption bands of styrene units can be seen at about 700, 760, and 1500 cm−1, while in the second and third spectra, the absorption band at 2250 cm−1 shows the presence of the acrylonitrile unit. The most reliable evidence for the identity of two compounds can be obtained from differential spectrophotometry. In this method, the compound under investigation is inserted into the path of the sample beam and a known compound into the path of the reference beam. If the two compounds are identical, the spectrum will be free of absorption bands. However, if there is a different content of the same group in the two compounds, the spectrum obtained will contain only the bands of this group. Polymer blends are a mixture of chemically different polymers or copolymers with no covalent bonding between them. The IR spectrum will thus be expected to contain absorption bands of all the individual polymers of the blends, the relative intensities of the peaks being dependent on the relative proportions of the constituent polymers. However, if there is a chemical interaction between the polymers, this leads to a considerable difference (shift in peak position) in the blend spectrum. As an illustration, Figure 3.96 shows the FT-IR spectra of two neat resins, polystyrene (PS) and poly(methyl methacrylate) (PMMA), while the FT-IR spectra of their blends in various proportions are shown in Figure 3.97. There are no shifts of the peaks of any group in the blend spectra, which signiﬁes that there is no chemical interaction between the constituent polymers and they remain as a physical mixture. A careful analysis of the IR spectra of the PS/PMMA blends (Figure 3.97) shows that there is a decrease in the transmittance of carbonyl (C=O) and methoxyl (–OCH3) stretchings (at 1732 and 1149 cm−1, respectively) with an increase of PS content, while 414 Plastics Technology Handbook 5000 2000 Wavenumber (cm–1) 1500 1000 700 80 Transmission (%) 20 (a) 0 80 20 (b) 0 80 20 0 (c) 2 5 10 Wavelength (μm) 15 FIGURE 3.95 IR spectra of (a) a butadiene–styrene copolymer; (b) a butadiene–acrylonitrile copolymer, and (c) a butadiene–styrene–acrylonitrile terpolymer. 80 Transmittance (%) 70 60 50 40 30 20 10 0 4500 4000 3500 3000 2500 2000 Wavenumber (cm–1) 1500 1000 500 4000 3500 3000 2500 2000 Wavenumber (cm–1) 1500 1000 500 (a) 80 Transmittance (%) 70 60 50 40 30 20 10 0 4500 (b) FIGURE 3.96 IR spectra of (a) polystyrene and (b) poly(methyl methacrylate). 415 Plastics Properties and Testing 500 Transmittance (%) 400 300 200 100 20:80 40:60 50:50 60:40 80:20 0 4500 FIGURE 3.97 4000 3500 3000 2500 2000 Wavenumber (cm–1) 1500 1000 500 IR spectra of blends of polystyrene and poly(methyl methacrylate) in different proportions. there is an increase in transmittance of these peaks with an increase of PMMA content, clearly indicating the formation of polymer blends. 3.8.1.6 Quantitative Analysis IR spectroscopy is widely used for quantitative analysis in the ﬁeld of synthetic resins. The analysis is based on Lambert–Beer’s law. If an absorption band in the IR spectrum arises from a particular component in the sample, the concentration of the component is related to the absorbance (A) or percentage transmission at the band maximum by this law, written as A = log10 100 = kcl Percentage transmission (3.134) where c is the concentration of the absorbing component, l is the thickness of the sample, and k is a constant. The sample can be prepared by hot pressing the resin between Teﬂon-coated stainless steel plates into sheets of thickness ranging, typically, from 0.2 to 3 mm or solvent casting ﬁlms (typically 0.03 mm thick). The value of k can be determined by measuring the IR absorbance of samples containing a known concentration of the absorbing component. For the determination of absorbance, a base-line calibration is normally used, the base line being drawn tangentially between the minima occurring at each side of the absorption peak. Considering, for example, the absorption band of the acetate group of PVAc at 2.15 mm, shown in Figure 3.98, the absorbance is measured by drawing the straight line background CD above the band and calculating log10(EG/EF). Plotting this absorbance against the thickness of PVAc calibration sample, the graph obtained should be linear and pass through the zero point. From the slope of the linear plot, the constant k in Equation 3.134 can be evaluated using the measured or calculated value of the acetate content of the resin. Using this value of k and the measured absorbance of the band at 2.15 mm, quantitative determination of vinyl acetate in a copolymer, such as vinyl chloride/vinyl acetate copolymer, can be performed. For copolymers containing up to 20% acetate, a sample of about 10 mm thickness gives a band of suitable intensity for measurement. An alternative (and simpler) method for determining the acetate content in vinyl chloride/vinyl acetate copolymer is based on the measurement of the absorbance ratio at 5.8 mm/7.0 mm (i.e., –C=O to –CH2– 416 Plastics Technology Handbook C G Transmission D F B E A Wavelength FIGURE 3.98 Measurement of absorbance of the absorption band at 2.15 mm of the acetate group in poly(vinyl acetate) by the base-line calibration method. ratio). In this method, the sample thickness need not be measured as the acetate content can be determined simply by referring to a calibration curve of 5.8 mm/7.0 mm absorbance ratio versus acetate content. In many cases, some free vinyl acetate (residual monomer) may be present in vinyl acetate polymers or copolymers. Since the band at 2.15 mm is attributed to the acetate group, both free and combined vinyl acetate will absorb at this wavelength. For free vinyl acetate, however, the spectrum can be measured at 1.63 mm (using a relatively thick specimen), where the band is only attributed to the vinyl double bond (–CH=CH2), and hence the monomer content of the sample can be determined. Copolymers of vinyl chloride with acrylates (most commonly, methyl acrylate and ethyl acrylate) produce an ester carbonyl band at 5.8 mm, but the C–O band near 8.5 mm is different in the two cases. Moreover, ethyl acrylate even at low concentrations shows a band at 9.75 mm, which is not present in the methyl acrylate system. IR spectra of the blends of PVC and poly(methyl methacrylate) are similar to those of vinyl chloride/acrylate copolymers. The presence of aliphatic ether in vinyl chloride polymers becomes evident from the appearance of a relatively strong ether band near 9.0 mm. The spectra of vinyl chloride/vinyl isobutyl ether copolymers (containing more than 10% vinyl isobutyl ether), for example, show a prominent, though rather broad, ether band near 9.0 mm, but also a doublet band owing to –CH(CH3)2 at 7.3 mm. The latter is, however, not a conclusive proof for the isobutyl group. IR spectroscopy may provide a simple means of analyzing the composition of polymer blends in many cases. With well-homogenized and unﬁlled blends, the IR spectroscopic measurement can be performed directly on the sample. On the other hand, if the blend is inhomogeneous and/or ﬁlled, an indirect method may be used in which the polymer blend component is separated from the ﬁller by solvent extraction and IR spectroscopy is then applied to the extracted polymer mixture. Considering, as an example, the analysis of polyethylene–polyisobutene (unﬁlled) blend composition by IR spectroscopy using the direct method, a series of standard samples are prepared by blending known amounts of polyethylene and polyisobutene by milling and then hot-pressing the blend into ﬁlms of about 417 Plastics Properties and Testing X´ C Transmission X Y Y´ F B E (2.39 μm) A 10 D 11 12 2.25 Wavelength (μm) 2.50 FIGURE 3.99 Measurement of absorbance of the absorption bands of a polyethylene-polyisobutene blend at 10.53 and 2.4 mm band maxima by the base-line calibration method. 0.3 mm thickness, which is suitable for polyisobutene contents between 5% and 25% (w/w). The IR spectra of all standard and unknown samples are measured over 9.0–12.5 mm and 2.1–2.8 mm wavelength ranges. The absorbances at the band maxima (∼10.5 and ∼2.4 mm) are measured by the base-line method. With straight line backgrounds XX′ and YY′ drawn as shown in Figure 3.99, the absorbances are calculated as log10(AC/AB) and log10(DF/DE). Denoting the latter as A10.5 and A2.4, respectively, the ratio A10.5/A2.4 is plotted against wt% polyisobutene contents of the blend standards to obtain a straight line passing through the origin. This calibration curve can be used to determine the polyisobutene content in an unknown polyethylene–polyisobutene blend from the absorbance ratio A10.5/A2.4 measured with a ﬁlm of the blend. The above method uses a reference band (2.4 mm) as a substitute for ﬁlm thickness. The method proves useful when the ﬁlm thickness is difﬁcult to measure (e.g., for rubbery samples). However, if the samples are sufﬁciently rigid for thickness measurement (such as with a micrometer gauge), the measurement of the reference band absorbance may be omitted and the A10.5/sample thickness ratio is used in a similar type of procedure. If ﬁlms of controlled thickness can be prepared by hot-pressing with, say, 0.05 mm feeler gauges as spacers between stainless steel plates (preferably coated with Teﬂon) and spectra are recorded over the wavelength range 9.5–15 mm, absorbances can then be measured at 10.5 and 13.9 mm band maxima by drawing base lines XX′ and YY′, as shown in Figure 3.100 and applied to Equation 3.134, leading to the following relation: Absorbance at 10:5 mm log10 AC=BC = Absorbance at 13:9 mm log10 DF=EF (3.135) wt% polyisobutene in blend =K wt% polyethylene in blend The value of K can be determined from this relation using the known proportions of polyisobutene and polyethylene of the standard blend. Using the determined value of K and the measured absorbances on the spectrum of an unknown sample, the ratio of the wt% contents of polyisobutene and polyethylene in the blend can be determined. 3.8.2 NMR Spectroscopy NMR spectroscopy [34] is now established as an important technique for characterization and testing of polymers. NMR spectra can be observed from a number of atomic nuclei, but for the organic chemist, the spectra of the 1H nucleus, that is, the 1H NMR spectra or proton resonance spectra, are of the greatest practical importance, the reason being that most polymers are organic compounds containing hydrogen and a great deal can be learned about their structure if the relative positions and chemical environment of the hydrogen atoms in the molecule can be established. This information can be derived from proton 418 Plastics Technology Handbook Y X A D Transmission X´ E B 13.9 μm 10.5 μm F C 10 Y´ 11 12 13 Wavelength (μm) 14 FIGURE 3.100 Measurement of absorbance of the absorption bands of a polyethylene–polyisobutene blend at 10.5 and 13.9 mm by the base-line calibration method. resonance spectra. Though, in principle, complementary information would be obtained from carbon NMR spectra, for the majority of organic compounds, the naturally abundant isotope 12C is inactive, while the 13C nucleus, which is magnetically active, has low natural abundance and gives only weak resonance signals. The NMR spectroscopy of organic compounds is thus conﬁned mainly to proton resonance spectra. Unless otherwise mentioned, NMR spectroscopy/spectra in this book will always mean proton NMR (or 1H NMR) spectroscopy/spectra. 3.8.2.1 General Principles Certain atomic nuclei like H, D, 13C, F, and so on possess an intrinsic mechanical spin. Since a charged particle spinning about its axis is equivalent to a circular electric current, which, in turn, gives rise to a magnetic ﬁeld, a spinning nucleus behaves as a tiny bar magnet whose axis is coincident with the axis of the spin, and its potential energy (Un) in a magnetic ﬁeld of strength H is Un = −Hm cos q (3.136) where m is the magnetic moment of the nucleus (i.e., the strength of the nuclear dipole) and q is the angle between the magnetic moment vector and the direction of the magnetic ﬁeld. According to quantum mechanics, the angle q is not a continuous variable but can have only certain discrete values; in other words, in an applied magnetic ﬁeld, the spin angular momentum vector for a nucleus cannot point in any arbitrary direction, but can have only a discrete set of orientations. This is the result of a phenomenon, known as space quantization. The angular momentum vector can point only such that its components along the direction of the magnetic ﬁeld are given by mI(h/2p), where the quantum number mI can have any of the values I, I − 1, …, −(I − 1), −I, with I representing the spin of the nucleus. Thus, for I = 1, the possible values of mI are 1, 0, and −1. and the nucleus can have three spin orientations. For proton, however, I = 1/2 and mI can be only +1/2 and −1/2, and so proton can have only two spin orientations. Each orientation represents a spin state and transition of a nucleus from one spin state to an adjacent one may occur by absorption or emission of an appropriate quantum of energy. From a classical point of view, the behavior of a spinning proton (pictured as a tiny bar magnet rotating about its axis) is analogous to that of a gyroscope spinning in frictionless bearings. It is a known fact that 419 Plastics Properties and Testing when a force (torque) is applied to a spinning gyroscope, its axis does not tilt but merely precesses about the direction of the force. Similarly, a spinning proton, behaving as a magnetic gyroscope, will precess about the direction of the applied ﬁeld (see Figure 3.101), keeping the angle q constant. The expression for the precessional frequency (w) of such a spinning proton has the same form as the frequency of the precessional motion of an orbiting electron (known as Larmor frequency), namely, w= Magnetic moment H radian=s Angular moment mH cycles=s = 2ppn where pn is the spin angular momentum of the nucleus given by h pﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ pn = I(I + 1) 2p (3.137) (3.138) So long as the angle q between its spinning axis and the ﬁeld direction is constant, the potential energy of the proton in the ﬁeld, given by Equation 3.136, will also be constant and no energy will be absorbed from the ﬁeld. In order to cause a change of the angle q, a second magnetic ﬁeld is to be applied perpendicular to the main ﬁeld. The secondary magnetic ﬁeld must not be a stationary one, however, as otherwise its effect in one half-cycle of the precessional motion will be cancelled by the effect in the other half-cycle. To produce a net effect, the secondary ﬁeld must also rotate about the direction of the main ﬁeld with a frequency equal to that of the precessing proton. The secondary magnetic ﬁeld can then interact with the precessing proton and energy can be exchanged; if the frequencies differ, there will be no interaction. Thus, when the frequency of the rotating secondary ﬁeld and the frequency of the precessing nucleus are equal, they are said to be in resonance, since in this condition, transition from one nuclear spin state to another can readily occur, giving rise to absorption or emission of energy. A rotating magnetic ﬁeld can be produced in a simple way by sending the output current of a radiofrequency crystal oscillator through a helical coil (solenoid) of wire. In NMR spectroscopy, the sample under investigation is taken in a small glass tube placed between the pole faces of a dc electromagnet (Figure 3.102). The coil that transmits the radio-frequency ﬁeld is placed with its axis perpendicular to the direction of the main ﬁeld produced by the electromagnet. The coil is made in two halves to allow N θ H S FIGURE 3.101 A spinning proton, behaving as a magnetic gyroscope, precesses about the direction of the applied magnetic ﬁeld H, with a constant angle q. 420 Plastics Technology Handbook Sweep current Test tube Poles of 15,000 G magnet Receiver coil Radio receiver Transmitter coil Radio frequency transmitter (typical frequency = 60 Mc/s) Recorder FIGURE 3.102 Schematic representation of an NMR spectrometer. the insertion of the sample holder. The electric current passing through the coil produces in it a magnetic ﬁeld directed along its axis, and this ﬁeld reverses its direction with the same frequency as the current. It is a general property of vectors that the resultant of two identical vectors rotating with the same frequency in opposite directions is equivalent to a vector having constant direction and periodic change to magnitude, which, in fact, represents an alternating motion. The alternating magnetic ﬁeld along the axis of the coil, described above, is thus equivalent to two magnetic ﬁelds rotating with the same frequency but in opposite directions. Of these two rotating magnetic ﬁelds, the one whose direction of rotation is the same as the direction of the precessional motion of the nucleus will act as the secondary magnetic ﬁeld. The other ﬁeld rotating in the opposite direction can be ignored, since its average effect is zero. To obtain a condition of resonance (absorption or emission of energy), the magnetic ﬁeld at the proton and the frequency of the alternating current supplied to the coil from the oscillator must be such that Equation 3.137 is satisﬁed. The nuclei can then absorb energy from the secondary ﬁeld. Coils located within the pole gap or wound about the poles of the magnet allow a sweep to be made through the applied magnetic ﬁeld to bring about this condition of resonance. When resonance absorption of energy takes place, it can be thought of as producing nuclei in the excited state, which will then tend to return to the lower level in order to approach the Boltzmann distribution ratio. The radiation emitted in this process is picked up by the receiver coil, which is a separate radio-frequency coil consisting of a few turns of wire wound tightly around the sample tube. The receiver coil is perpendicular to both the magnetic ﬁeld and the radio-frequency transmitter coil in order to minimize pickup from these ﬁelds. It is the sample that provides this coupling between the receiver and the transmitter. The signal from the receiver coil can be displayed on an oscilloscope or a recorder chart. Experimentally, the resonance condition may be obtained in two alternative ways. We might vary either the ﬁeld strength of the electromagnet or the frequency of the oscillator, keeping the other ﬁxed. Suppose we apply a ﬁxed magnetic ﬁeld and the Larmor frequency produced by it is, say, 60 Mc/s; if the frequency of the oscillator is then varied over a range including 60 Mc/s, resonance absorption will occur exactly at that frequency. On the other hand, we can ﬁx the oscillator frequency at 60 Mc/s and vary the applied ﬁeld over a range until absorption occurs. The latter arrangement is simple and widely used in practice. Most NMR spectrometers in use today employ a ﬁxed oscillator frequency of either 60 Mc/s, 100 Mc/s or 220 Mc/s. 421 Plastics Properties and Testing 3.8.2.2 Chemical Shift A very important characteristic of the NMR technique is that it can distinguish protons in different molecular environments. If the resonance frequencies of all protons in a molecule were the same, as given by Equation 3.137, then the NMR spectrum would show only one peak for the compound, and as such would be of little use to the organic chemist. However, we must consider the fact that the ﬁeld strength represented in Equation 3.137 is the ﬁeld strength experienced by the protons in the sample and is not the same as the strength of the applied magnetic ﬁeld. Protons whether in hydrogen atoms or molecules are surrounded by an electromagnetic charge cloud having approximately spherical symmetry. A magnetic ﬁeld induces electronic circulations in the charge cloud in a plane perpendicular to the applied ﬁeld and in such a direction as to produce a ﬁeld opposing the applied ﬁeld, as shown in Figure 3.103. The induced ﬁeld is directly proportional to the applied ﬁeld H and so can be represented by sH, where s is a constant. The effective magnetic ﬁeld experienced by the proton is therefore Heff = ðH − s H Þ = H ð1 − s Þ (3.139) We can thus say that proton is shielded from the external ﬁeld by diamagnetic electron circulation and s represents the shielding constant. The extent of shielding of a proton depends on the electron density around it in a molecule. A molecule may contain protons in different chemical environments. The average electron concentrations in these environments are different and so are the shielding effects on the protons. Consider, for example, the C–H and O–H bonds. Since oxygen is more electronegative than carbon, the electron density around the CH proton (i.e., proton in C–H bonds) should be considerably higher than that around the OH proton. We should thus expect that the shielding constant of the CH proton is greater than that of the OH proton. It then follows from Equation 3.139 that, for a given applied ﬁeld, the effective ﬁeld at the OH proton is greater than that of the CH proton. Consequently, as the applied magnetic ﬁeld is increased, the OH proton will come into resonance before the CH proton. The separation between the resonances (absorption peaks) for the same nucleus in different chemical environments (causing different degrees of shielding) is known as the chemical shift. It is easy to see that the magnitude of the chemical shift will be proportional to the strength of the applied ﬁeld. From the chemical shifts, we should be able to know how many different types of protons there are in a molecule. It has, however, no effect on the signal or peak intensity. The intensity of absorption at a given ﬁeld strength will be proportional to the number of protons in a given environment in the molecule and a spectrum will thus tell us how many of each type of protons are present. These are the two important facets 70° S N H (a) (b) FIGURE 3.103 (a) Diamagnetic circulation of an electron about a nucleus produces a ﬁeld opposing the applied ﬁeld, H. (b) Field of point magnetic dipole. 422 Plastics Technology Handbook of NMR spectroscopy that make it a qualitative and quantitative analytical technique. As an illustration, we consider below the NMR spectrum of ethanol obtained under low resolution (Figure 3.104a). The spectrum shows three absorption peaks of intensity (area) ratio 1:2:3. The smallest peak is assigned to the single proton in the OH group, the next peak is assigned to the two protons in the CH2 group, and the largest peak is assigned to the three protons in the CH3 group. That the absorption peak for the OH proton occurs at the lowest value of the applied ﬁeld accords with the fact that the OH bond is polar with the bonding electrons being, on the average, closer to the oxygen atom, and as a result, the OH proton is relatively bare and unshielded from the applied ﬁeld. On the other hand, the C–H bond being almost nonpolar, the bonding electrons in the CH3 group are nearly equally distributed between the carbon atom and the three protons. The CH3 protons are thus well shielded and a larger applied ﬁeld is required to bring them into resonance. The CH2 peak occurs in between OH and CH3 peaks, because the electron withdrawal by the oxygen of the OH group also has some effect on the electron distribution in the adjacent CH2 group and reduces the degree of shielding for the CH2 protons. It is interesting to note that dimethyl ether (CH3OCH3), which is an isomer of ethanol, has six equivalent protons, and accordingly, the spectrum shows a single absorption peak. Chemical shifts are very small compared to the strength of the applied ﬁeld; their magnitude being a few milligauss at a ﬁeld strength of about 10,000 gauss. Moreover, since electronic shielding (sH) is directly proportional to the strength of the applied ﬁeld, the chemical shift value also varies with the ﬁeld strength. Therefore, in NMR analysis, chemical shifts are usually expressed on a relative basis, in a form independent of ﬁeld strength. For a proton in a given environment, the chemical shift is deﬁned by d= Hr − Hs 106 ppm Hr (3.140) where Hs and Hr are the applied ﬁeld strengths at which resonance occurs in a given substance and a reference substance, respectively. The reference substance is dissolved in the same solution as the sample (internal reference) so that both experience the same magnetic ﬁeld. The chemical shift parameter d is dimensionless; the factor 106 is included to express it as parts per million (ppm). The substance now almost universally selected as reference for proton resonances is tetramethyl silane, Si(CH3)4, or TMS in short. Its chief advantages are as follows: (i) it gives a single sharp peak (since all 12 protons in it are equivalent); (ii) its resonance peak occurs at an exceptionally high ﬁeld, which is on the higher side of almost all other proton resonances in organic molecules, and (iii) it is chemically inert, OH (a) H (b) H H H C C H H H FIGURE 3.104 Schematic representation of NMR spectrum (energy absorption versus applied ﬁeld strength) of ethanol (a) under low resolution and (b) under high resolution. 423 Plastics Properties and Testing magnetically isotropic, and low boiling (boiling point, 27°C), so that it can be readily recovered from most samples after use. Resonance positions are indicated on the d-scale or the t-scale. On the d-scale (Equation 3.140), d is zero for TMS and it increases downﬁeld. On the t-scale, t is given a value of 10 at the resonance peak of TMS and it decreases downﬁeld, being related to d by t = 10 − d. The d-scale has the disadvantage that a large numerical value of d implies a low-ﬁeld resonance and hence a small shielding of the nucleus from the applied ﬁeld. In this respect, the t-scale is more convenient, a larger value of t implying a greater shielding of the nucleus. 3.8.2.3 Shielding Mechanisms As illustrated in Figure 3.103, the magnetic ﬁeld generated by the induced circulation opposes the applied magnetic ﬁeld (H) and is called diamagnetic shielding. For protons, three main kinds of diamagnetic shielding can be recognized, which arise from three kinds of induced electron circulation, namely, local diamagnetic shielding, neighboring diamagnetic shielding by anisotropic groups, and interatomic diamagnetic circulation in aromatic rings. The origin of local diamagnetic shielding has been explained in Section 3.8.2.2. It should be noted that the charge distribution surrounding protons has approximately spherical symmetry irrespective of the presence of magnetic ﬁeld or chemical bonding. Thus, even if a molecule, of which the proton is a part, changes its orientation with respect to the applied ﬁeld direction owing to incessant rotation and vibration, mainly diamagnetic currents will ﬂow around the proton because of the almost axial symmetry of the charge cloud with respect to the ﬁeld direction. The degree of shielding of a proton is, however, dependent on the electron density around it—the higher the electron density, the higher the shielding and hence the higher the ﬁeld at which the proton absorbs (as reﬂected in a lower d and a higher t). The circulation of electrons about neighboring atoms can be effective at a proton only if the generated secondary magnetic ﬁelds are anisotropic. As shown in Figure 3.105, the diamagnetic anisotropy of the triple bond in acetylene serves to decrease the external magnetic ﬁeld (i.e., increase the shielding) of the acetylenic protons. In contrast, the diamagnetic anisotropy of the carbonyl group in an aldehyde causes Secondary magnetic ﬁeld generated by circulating electrons H C C Induced circulation of π electrons H C C H Restricted diamagnetic circulation H Applied ﬁeld H Applied ﬁeld H (a) (b) FIGURE 3.105 The origin of diamagnetic shielding of acetylenic protons attributed to the anisotropy of acetylene. Axis of molecule is (a) parallel and (b) perpendicular to the applied ﬁeld. 424 Plastics Technology Handbook deshielding, that is, decreases shielding at the aldehydic proton (see Figure 3.106), while protons attached to a single-bonded carbon actually experience slight deshielding (Figure 3.107). The third type of diamagnetic shielding is associated with aromatic and pseudo-aromatic rings, having cyclically delocalized p electrons, which are readily induced into circulation in the plane of the ring by an applied magnetic ﬁeld. This generates a secondary magnetic ﬁeld that causes pronounced shielding at the center of the ring but deshielding outside in the plane of the ring where the protons lie (Figure 3.108). This accounts for the relatively low value of t for aromatic protons, as shown in Table 3.12. Substitution of benzene shifts the t value for the absorption of the aromatic protons (Table 3.12). Electron-withdrawing groups (e.g., –NO2, –CO2H) lower the t value, while electron-donating groups (–OH, –NH2) raise the t value relative to benzene (2.73t), because of the local diamagnetic shielding effect. Because of the p-electron circulation (Figure 3.108), aromatic rings provide a strong source of longrange shielding and deshielding. Long-range shielding applies to shielding that operates through space rather than directly through chemical bonds; the latter is short-range shielding. Thus, of the three main kinds of diamagnetic shielding, the ﬁrst is short-range and the last two are long-range. Secondary magnetic ﬁeld generated by circulating electrons R C O H Induced circulation of π electrons Applied ﬁeld H FIGURE 3.106 Deshielding of an aldehydic proton owing to diamagnetic anisotropic effects. Induced circulation of electrons C C Secondary magnetic ﬁeld generated by circulating electrons Applied ﬁeld H FIGURE 3.107 Diamagnetic shielding by the carbon–carbon single bond. 425 Plastics Properties and Testing Secondary ﬁeld generated by electron circulation Induced circulation of π electrons H H Applied ﬁeld H FIGURE 3.108 The magnetic ﬁeld generated by the induced circulation of electrons in benzene. TABLE 3.12 Typical Chemical Shifts for Various Types of Protons Type of Proton D t Type of Proton 0.9 1.3 9.1 8.7 RNH2 C6H5NH2 1.5 3.4 Protons on saturated carbons RCH3 RCH2R d Other protons R3CH 1.5 8.5 CH3CN CH3COCH3 1.97 2.09 8.03 7.91 RCHO RCOOH 9.7 11.0–12.0 ROH Highly variable CH3CHO 2.15 7.85 C6H5CH3 CH3OR 2.54 3.4 7.66 6.6 RCH2OR 3.7 6.3 RCH2Cl Protons on unsaturated carbons 3.7 6.3 RC≡CH 2.4 7.6 R2C=CH2 4.5–5.0 5.0–5.5 R2C=CHR Protons on aromatic rings 5.0–5.5 4.5–5.0 C6H6 7.27 2.73 C6H5NO2a 8.22 7.48 1.78 2.52 7.60 2.40 6.52 7.07 3.48 2.93 6.65 3.35 C6H5NH2a a The three values given refer to ortho-, meta-, and para-protons, respectively. 426 Plastics Technology Handbook 3.8.2.4 Spin–Spin Coupling The NMR spectrum of a substance is usually much more complex than would be expected solely on the basis of chemical shift effect. Thus, the resonance absorptions are often split into a number of components and may appear as doublets, triplets, quartets, or more complex patterns. The phenomenon, observed only with spectrometers of high resolution, is attributed to the effect on a nucleus (responsible for absorption peak) of other magnetic nuclei in the same molecule. This effect is referred to as spin–spin coupling. It may be heteronuclear, for example, between protons and adjacent ﬂuorine nuclei, but the most important effect is homonuclear coupling between protons that are very close together in the same molecule but are in different chemical environments. As an illustration, we consider the high-resolution spectrum of a common sample of ethanol represented in Figure 3.104b. By comparing it with Figure 3.104a, we see that the methyl (CH3) absorption is now split into a triplet with areas of the three components in the approximate ratios of 1:2:1 and methylene (CH2) absorption is split into a quartet, having relative areas of approximately 1:3:3:1. The separation between the components of the triplet is the same and equal to the separation between the components of the quartet. This separation, quoted in hertz (Hz, cycles per second), is known as the coupling constant J between the nuclei that interact. If the difference (in terms of frequency) between the absorption resonances of the interacting protons is large compared with J, then a simple rule of peak multiplicity is The number of components into which a resonance absorption peak is split = n + 1, (3.141) where n = number of identical nuclei involved in the interaction. In accordance with this rule, the methyl proton resonance peak is split into three components because two identical nuclei (i.e., the two methylene protons) are adjacent to it (i.e., on the next carbon atom). Similarly, the methylene proton resonance peak is split into four by interaction with the three identical methyl protons. [Note: The proton of the adjacent OH group does not partake in spin–spin coupling; otherwise, the methylene peak would split further. This is attributed to a rapid chemical exchange (proton transfer) of the hydroxyl proton among different ethyl alcohol molecules over a period of time, with the result that the methylene protons experience simply an average, nonsplitting ﬁeld from the hydroxyl proton. The hydroxyl proton also experiences an averaged effect of the spin orientations of the methylene protons and its absorption peak therefore shows only a singlet.] In simple cases of interacting nuclei, the component peaks of a multiplet are symmetrical about a midpoint and their relative areas are numerically proportional to the coefﬁcients of the binomial expansion (1 + r)n, where n is the number of protons on adjacent atoms. To illustrate, the following multiplicities will be predicted for the following compounds—n–propyl iodide (CH3CH2CH2I): a three-proton triplet (relative areas 1:2:1), a two-proton sextet (1:5:10:10:5:1), and a two-proton triplet (1:2:1); isopropyl iodide [(CH3)2.CHI]: a six-proton doublet (1:1) and a one-proton septet (1:6:15:20:15:6:1); methyl ethyl ether (CH3OCH2CH3): a three-proton singlet, a two-proton quartet (1:3:3:1), and a three-proton triplet (1:2:1). Observing the aforesaid patterns of relative intensities (i.e., relative areas) of the component peaks, doublets, triplets, quartets, and so on may sometimes be recognized in NMR spectra, even though they are partially overlapped by other bands. When plotted on a d- or t-scale, a spin–spin coupling pattern is compressed in the higher ﬁeld strength spectrum. Thus, spectra recorded at, say, 60 MHz frequency and 100 MHz frequency will usually show differences. In principle, such differences may be used to differentiate spin–spin coupling from chemical shift. 3.8.2.5 Applications in Polymer Analysis The NMR spectrum is presented as a plot of d or t shift on the horizontal axis against resonance energy as a vertical axis, usually on precalibrated paper. Field strength conventionally increases to the right. Unless otherwise indicated, it is presumed that the resonance of TMS is coincident with the zero of the d (ppm) scale, which is on the right edge of the recorder chart. Most spectrometers have provision for integration Plastics Properties and Testing 427 of areas under resonances. The integral is presented, using an arbitrary vertical scale, on the same chart as the resonance spectrum and in alignment with the latter on the horizontal axis. To obtain sharp and well-deﬁned NMR spectra, it is necessary to ensure that chemically equivalent nuclei in each molecule in the sample experience effectively the same magnetic environment. For this reason, samples are used in the liquid state and preferably in solution, as molecules then having higher mobility can assume all possible orientations with respect to the magnetic ﬁeld. Each nucleus experiencing the same average magnetic environment ensures that the resonances are sharp. This condition can be achieved with monomeric substances in solution, but not with polymers. In practice, however, if rapid segmental motion of the polymer chain can be achieved, comparatively sharp resonances with substantially reduced width may be observed. Polymers are therefore examined in solution and often at 100°C to 150°C. Even under these conditions, resonances obtained for polymers are usually broader and the ﬁne detail characteristic of monomeric substances is rarely achieved. The NMR spectroscopy has been used for the analysis of polymers. Qualitative analysis and identiﬁcation of polymers, quantitative analysis such as polymer end-group analysis and molecular weight determination, copolymer composition and sequence distribution analysis, polymer branching, structural isomerism, and polymer tacticity measurement may be mentioned as some of the applications. These are often done in combination with IR spectroscopic studies and elemental analysis as they may provide helpful leads and conﬁrmation. Some examples of such applications are given below. Polymer Molecular Weights. NMR analysis offers an easy method for determination of molecular weight of relatively low-molecular-weight polymers by end-group analysis. The method can be applied if the polymer has identiﬁable end-group protons that are distinguishable from protons of repeating monomer units by NMR and if accurate integration of resonance absorptions of different types of protons is provided by the instrument. Considering, for example, poly(ethylene glycol) diacrylate represented by the formula H2 C¼CH — CO—ð—O—CH2 —CH2 —Þn —O—CO—CH¼CH2 : The end groups have formula weights (FW) 55 and 71, while the repeat unit has FW 44. Adding the area integrals of all the end-group proton resonances and dividing by the total number of end-group protons (i.e., 6) yields the value (say X) of the integral per end-group proton. Thereafter, dividing the sum of the integrals of all the repeat unit protons by the number of protons of one repeat unit (i.e., 4) gives the value (say Y) of the integral per proton of one repeat unit. Therefore, the number of repeat units in the polymer molecule is n = Y/X and polymer molecular weight = (55 + 71) + 44n. Branching in Polyethylene. Branching in polyethylene occurs in two forms—short-chain branching, which results from copolymerization of ethylene with another oleﬁn (e.g., butene or octene), and long chain branching, where the length of the branches is on the same scale as that of the backbone. To detect and quantify these different types of branching, 13C NMR, instead of 1H NMR, is used based on the differences of the chemical shifts of the backbone chain carbon atoms attached or adjacent to the branch. The chemical shift depends on the length of the branch for branches up to 6 carbons in length, but is independent of branch length for all branches six carbons long or longer. On 13C NMR spectra of branched polyethylene, carbons can be classiﬁed as methylene carbon (C atom bonded to two other C atoms) on backbone and long-chain branches, methine carbon (C atom bonded to three other C atoms), aC (C atom immediately adjacent to a methine carbon), bC (C atom immediately adjacent to an aC), g C (C atom immediately adjacent to a bC), 2C (second C atom from end) on short branches, and chain-end methyl C. Tacticity in Polymers. NMR spectroscopy is very useful for observation of stereoisomerism in polymers. Considering, for example, poly(methyl methacrylate) (PMMA) (Figure 3.109a), the two methylene protons are not magnetically equivalent in isotactic PMMA and so have different (though close) chemical shift values, forming an AB set, and there are two such sets, thus giving an appearance of four lines (Figure 3.109b). In syndiotactic PMMA (Figure 3.109b), the two methylene protons are magnetically equivalent and produce a single resonance in the center of the four preceding ones, while in atactic PMMA, the two methylene protons exhibit a combination of the above two resonance patterns. The three protons of the backbone methyl group are, however, sensitive to triad arrangements (Figure 3.109c). The isotactic, 428 Plastics Technology Handbook (a) Isotactic CH3 CH2 C n C O O CH3 Isotactic Syndiotactic CH3 C O-Me SiMe4 CH2 Syndiotactic Atactic Atactic 4.0 ppm = –COOMe 0.0 = –Me = –CH2 (c) (b) FIGURE 3.109 Observation of tacticity in poly(methyl methacrylate) by 1H NMR spectroscopy. (a) Repeating unit of PMMA. (b) Proton NMR spectra of isotactic, syndiotactic, and atactic PMMA. (c) Arrangements of substituents in isotactic, syndiotactic, and atactic triads. syndiotactic, and atactic triads are resolved at different chemical shifts and a line appears at different locations for the three arrangements, the atactic line being between the iso and syndio lines (Figure 3.109b). Figure 3.110 compares the proton NMR spectra of a predominantly syndiotactic PMMA and a predominantly isotactic PMMA recorded by an NMR spectrometer employing ﬁxed oscillator frequency of 60 Mc/s. (a) (b) 7.0 8.0 9.0 10.0 τ FIGURE 3.110 1H NMR spectra of (a) predominantly syndiotactic PMMA and (b) predominantly isotactic PMMA with a 60-MHz NMR spectrometer. (Adapted from Bovey, F. A. 1972. High Resolution NMR of Macromolecules, Academic Press, New York.) 429 Plastics Properties and Testing Copolymer Composition and Sequence Distribution. NMR spectroscopy can offer an easy method for the determination of composition of copolymers. The method is based on identifying the resonance absorption of a suitable proton in the comonomer units and recording their resonance absorption integrals. Considering, for example, a typical proton NMR spectrum of a copolymer of methyl methacrylate (MMA) and hexyl methacrylate (HMA), the resonance absorptions of O–CH3 and O–CH2 protons are identiﬁed at d = 3.6 and 3.9 ppm, respectively (Figure 3.111). The content of MMA in the copolymer is then calculated CH3 CH3 CH2 C O C CH2 C x O CH3 δ = 3.6 ppm (a) O y O CH2 (CH2)4 C CH3 δ = 3.9 ppm (b) 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0 ppm FIGURE 3.111 1H NMR spectrum of a copolymer of methyl methacrylate and hexyl methacrylate. (Adapted from Bovey., F. A. 1972. High Resolution NMR of Macromolecules, Academic Press, New York.) (a) (b) 4 (c) 2 1 6 3 5 6 7 7 8 9 τ FIGURE 3.112 1H NMR spectra of (a) a poly(vinylidene chloride) homopolymer, (b) a polyisobutylene homopolymer, and (c) a copolymer of vinylidene chloride (VDC) and isobutylene (IB). Tetrad sequences in the copolymer associated with the resonance peaks are—1: VDC-VDC-VDC-VDC, 2: VDC-VDC-VDC-IB, 3: IB-VDC-VDC-IB, 4: VDC-VDC-IB-VDC, 5: IB-VDC-IB-VDC, 6: VDC-VDC-IB-IB, 7: IB-VDC-IB-IB. (Adapted from Bovey, F. A. 1972. High Resolution NMR of Macromolecules, Academic Press, New York.) 430 Plastics Technology Handbook from the respective area integrals (A), using the following equation: %MMA = A3:6 ppm =3 100 A3:6 ppm =3 + A3:9 ppm =2 Proton NMR spectroscopy has been used for observing sequence distribution of monomer units in copolymers. As a typical example, Figure 3.112a and b shows simple resonance peaks in poly(vinylidene chloride) and polyisobutylene homopolymers, respectively. In comparison, the spectrum of a copolymer of vinylidene chloride (VDC) and isobutylene (IB), shown in Figure 3.112c, exhibits a multiplicity of peaks, which are attributed to various tetrad sequences as noted. References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. Ogorkiewicz, R. M. 1970. Engineering Properties of Thermoplastics, John Wiley, New York. Williams, J. G. 1973. Stress Analysis of Polymers, Longmans, London. Turner, S. 1973. Mechanical Testing of Plastics, CRC Press, Cleveland, OH. Crawford, R. J. 1981. Plastics Engineering, Pergamon, London. Ferry, J. D. 1970. Viscoelastic Properties of Polymers. 2nd Ed., John Wiley, New York. Tobolsky, A. V. 1960. Properties and Structure of Polymers, John Wiley, New York. Bartenev, G. M. and Zuyev, Y. S. 1968. Strength and Failure of Viscoelastic Materials, Pergamon, London. McLoughlin, J. R. and Tobolsky, A. V. 1952. J. Colloid Sci., 7, 555. Williams, M. L., Landel, R. F., and Ferry, D. 1955. J. Am. Chem. Soc., 77, 3701. Lenk, R. S. 1978. Polymer Rheology, Applied Science Publishers, London. Graessley, W. W. 1967. J. Chem. Phys., 47, 1942. Bueche, F. and Harding, W. S. 1958. J. Polym. Sci., 32, 177. Herzog, R. and Weissenberg, K. 1928. Colloid Z., 46, 227. Nason, H. K. 1945. J. Appl. Phys., 16, 338. Tordella, J. P. 1969. Rheology—Theory and Applications, Vol. 5, F. R. Eirich, ed., p. 355. Academic Press, San Diego, Chap. 2. vanWazer, J. R., Lyons, J.W., Kim, K. Y., and Colwell, R. E. 1963. Viscosity and Flow Measurements, Interscience Publishers, New York. Bagley, E. B. 1961. Trans. Soc. Rheol., 5, 355. Rabinowitsch, B. 1929. Z. Physik. Chem., 145A, 1. Vincent, P. I. 1970. Impact Tests and Service Performance of Thermoplastics, Plastics Institute, London. Fakirov, S. 2007. J. Mater Sci., 40, 1131. Braden, M. and Gent, A. N. 1960. J. Appl. Polym. Sci., 3, 90. Gaube, E. 1959. Kunststoffe, 49, 446. Holloway, L. 1977. Glass Reinforced Plastics in Construction, Surrey University Press, Surrey. Richardson, M. O. W. 1977. Polymer Engineering Composites, Applied Science Publishers, London. Johnston, A. F. 1979. Engineering Design Properties of GRP, British Plastics Federation, England. Brophy, J. J. and Buttrey, J. W. 1962. Organic Semiconductors, Macmillan, New York. Saunders, D. W. 1965. Optical properties of high polymers. In Physics of Plastics, P. D. Ritchie, ed., Van Nostrand Reinhold, New York. Hornak, L. 1991. Polymers for Lightwave and Integrated Optics, Marcel Dekker, New York. Daum, W., Krauser, J., Zamzow, P. E., and Ziemann, O. 2002. Polymer Optical Fibers for Data Communication, Springer, Berlin, Heidelberg, New York. Gordon, M. 1965. Thermal properties of high polymers. In Physics of Plastics, P. D. Ritchie, ed., Van Nostrand Reinhold, New York. Glanville, A. B. 1971. The Plastics Engineer’s Data Book, Industrial Press, New York. Plastics Properties and Testing 431 32. Krause, A. and Lange, A. 1969. Introduction to Chemical Analysis of Plastics, Iliffe Books Ltd, London. 33. Ghosh, P. 1990. Polymer Science and Technology of Plastics and Rubber, Tata McGraw-Hill, New Delhi, India. 34. Chanda, M. 2000. Atomic Structure and Chemical Bond, Including Molecular Spectroscopy, Tata McGraw-Hill, New Delhi, India. 35. Haslam, J., Willis, H., and Squirrell, D. 1972. Identiﬁcation and Analysis of Plastics, Butterworth, London. 36. Barnes, R. B., Williams, V. Z., Davis, A. R., and Geisecke, P. 1944. Ind. Eng. Chem. Analyt. Ed., 16, 9. 37. Kaniappan, K. and Latha, S. 2011. Int. J. Chem. Tech. Res. 3 (2), 708. 38. Krimm, S., Liang, C. Y., and Sutherland, G. B. B. M. 1956. J. Polym. Sci. Part A Polym. Chem., 22, 101, 227. 39. Hossain, U. H., Seidl, T., and Ensinger, W. 2014. Polym. Chem. (Roy. Soc. Chem.), 5, 1001. http://taylorandfrancis.com 4 Industrial Polymers 4.1 Introduction The ﬁrst completely synthetic plastic, phenol-formaldehyde, was introduced by L. H. Baekeland in 1909, nearly four decades after J. W. Hyatt had developed a semisynthetic plastic—cellulose nitrate. Both Hyatt and Baekeland invented their plastics by trial and error. Thus the step from the idea of macromolecules to the reality of producing them at will was still not made. It had to wait till the pioneering work of Hermann Staudinger, who, in 1924, proposed linear molecular structures for polystyrene and natural rubber. His work brought recognition to the fact that the macromolecules really are linear polymers. After this it did not take long for other materials to arrive. In 1927 poly(vinyl chloride) (PVC) and cellulose acetate were developed, and 1929 saw the introduction of urea-formaldehyde (UF) resins. The production of nylon-6,6 (ﬁrst synthesized by W. H. Carothers in 1935) was started by DuPont in 1938, and the production of nylon-6 (perlon) by I. G. Farben began in 1938, using the caprolactam route to nylon developed by P. Schlock. The latter was the ﬁrst example of ring-opening polymerization. The years prior to World War II saw the rapid commercial development of many important plastics, such as acrylics and poly(vinyl acetate) in 1936, polystyrene in 1938, melamine–formaldehyde (formica) in 1939, and polyethylene and polyester in 1941. The amazing scope of wartime applications accelerated the development and growth of polymers to meet the diverse needs of special materials in different ﬁelds of activity. The development of new polymeric materials proceeded at an even faster pace after the war. Epoxies were developed in 1947, and acrylonitrile–butadiene–styrene (ABS) terpolymer in 1948. The polyurethanes, introduced in Germany in 1937, saw rapid development in the United States as the technology became available after the war. The discovery of Ziegler–Natta catalysts in the 1950s brought about the development of linear polyethylene and stereoregular polypropylene. These years also saw the emergence of acetal, polyethylene terephthalate, polycarbonate, and a host of new copolymers. The next two decades saw the commercial development of a number of highly temperature-resistant materials, which included poly(phenylene oxide) (PPO), polysulfones, polyimides, polyamide-imides, and polybenzimidazoles. Numerous plastics and ﬁbers are produced from synthetic polymers: containers from polypropylene, coating materials from PVC, packaging ﬁlm from polyethylene, experimental apparatus from Teﬂon, organic glasses from poly(methyl methacrylate), stockings from nylon ﬁber—there are simply too many to mention them all. The reason why plastics materials are popular is that they may offer such advantages as transparency, self-lubrication, lightweight, ﬂexibility, economy in fabricating, and decorating. Properties of plastics can be modiﬁed through the use of ﬁllers, reinforcing agents, and chemical additives. Plastics have thus found many engineering applications, such as mechanical units under stress, low-friction components, heat- and chemical-resistant units, electrical parts, high-light-transmission applications, housing, building construction functions, and many others. Although it is true that in these applications plastics have been used in a proper manner according to our needs, other ways of utilizing both natural and synthetic polymers may still remain. To investigate these further possibilities, active 433 434 Plastics Technology Handbook research has been initiated in a ﬁeld called specialty polymers. This ﬁeld relates to synthesis of new polymers with high additional value and speciﬁc functions. Many of the synthetic plastic materials have found established uses in a number of important areas of engineering involving mechanical, electrical, telecommunication, aerospace, chemical, biochemical, and biomedical applications. There is, however, no single satisfactory deﬁnition of engineering plastics. According to one deﬁnition, engineering plastics are those which possess physical properties enabling them to perform for prolonged use in structural applications, over a wide temperature range, under mechanical stress and in difﬁcult chemical and physical environments. In the most general sense, however, all polymers are engineering materials, in that they offer speciﬁc properties which we judge quantitatively in the design of end-use applications. For the purpose of this discussion, we will classify polymers into three broad groups: addition polymers, condensation polymers, and special polymers. By convention, polymers whose main chains consist entirely of C–C bond are addition polymers, whereas those in which hetero atoms (e.g., O, N, S, Si) are present in the polymer backbone are considered to be condensation polymers. Grouped as special polymers are those products which have special properties, such as temperature and ﬁre resistance, photosensitivity, electrical conductivity, and piezoelectric properties, or which possess speciﬁc reactivities to serve as functional polymers. Further classiﬁcation of polymers in the groups of addition polymers and condensation polymers has been on monomer composition, because this provides an orderly approach, whereas classiﬁcation based on polymer uses, such as plastics, elastomers, ﬁbers, coatings, etc. would result in too much overlap. For example, polyamides are used not only as synthetic ﬁbers but also as thermoplastics molding compounds and polypropylene, which is used as a thermoplastic molding compound has also found uses as a ﬁberforming material. All vinyl polymers are addition polymers. To differentiate them, the homopolymers have been classiﬁed by the substituents attached to one carbon atom of the double bond. If the substituent is hydrogen, alkyl or aryl, the homopolymers are listed under polyoleﬁns. Oleﬁn homopolymers with other substituents are described under polyvinyl compounds, except where the substituent is a nitrile, a carboxylic acid, or a carboxylic acid ester or amide. The monomers in the latter cases being derivatives of acrylic acid, the derived polymers are listed under acrylics. Under oleﬁn copolymers are listed products which are produced by copolymerization of two or more monomers. Condensation polymers are classiﬁed as polyesters, polyamides, polyurethanes, and ether polymers, based on the internal functional group being ester (–COO–), amide (–CONH–), urethane (–OCONH–), or ether (–O–). Another group of condensation polymers derived by condensation reactions with formaldehyde is described under formaldehyde resins. Polymers with special properties have been classiﬁed into three groups: heat-resistant polymers, silicones and other inorganic polymer, and functional polymers. Discussions in all cases are centered on important properties and main applications of polymers. 4.2 Part I: Addition Polymers Addition polymers are produced in largest tonnages among industrial polymers. The most important monomers are ethylene, propylene, and butadiene. They are based on low-cost petrochemicals or natural gas and are produced by cracking or reﬁning of crude oil. Polyethylene, polypropylene, poly(vinyl chloride), and polystyrene are the four major addition polymers and are by far the least-expensive industrial polymers on the market. In addition to these four products, a wide variety of other addition polymers are commercially available. For addition polymers four types of polymerization processes are known: free-radical-initiated chain polymerization, anionic polymerization, cationic polymerization, and coordination polymerization (with Ziegler–Natta catalysts). By far the most extensively used process is the free-radical-initiated chain polymerization. However, the more recent development of stereo regular polymers using certain 435 Industrial Polymers organometallic coordination compounds called Ziegler–Natta catalysts, which has added a new dimension to polymerization processes, is expected to play a more important role in coming years. The production of linear low-density polyethylene (LLDPE) is a good example. Ionic polymerization is used to a lesser extent. Thus, anionic polymerization is used mainly in the copolymerization of oleﬁns, such as the production of styrene–butadiene elastomers, and cationic polymerization is used exclusively in the production of butyl rubber. Different processes are used in industry for the manufacture of polymers by free-radical chain polymerization. Among them homogeneous bulk polymerization is economically the most attractive and yields products of higher purity and clarity. But it has problems associated with the heat of polymerization, increases in viscosity, and removal of unreacted monomer. This method is nevertheless used for the manufacture of PVC, polystyrene, and poly(methyl methacrylate). More common processes are homogeneous solution polymerization and heterogeneous suspension polymerization. Solution polymerization is used for the manufacture of polyethylene, polypropylene, and polystyrene, but by far the most widely used process for polystyrene and PVC is suspension polymerization. In the latter process (also known as bead, pearl, or granular polymerization because of the form in which the ﬁnal products may be obtained), the monomer is dispersed as droplets (0.01–0.05 cm in diameter) in water by mechanical agitation. Various types of stabilizers, which include water-soluble organic polymers, electrolytes, and water-insoluble inorganic compounds, are added to prevent agglomeration of the monomer droplets. Each monomer droplet in the suspension constitutes a small bulk polymerization system and is transformed ﬁnally into a solid bead. Heat of polymerization is quickly dissipated by continuously stirring the suspension medium, which makes temperature control relatively easy. 4.2.1 Polyoleﬁns 4.2.1.1 Polyethylene CH2 Monomer Ethylene Polymerization LDPE: free-radical-initiated chain polymerization HDPE: Ziegler–Natta or metal-oxide catalyzed chain polymerization CH2 n Major Uses LDPE: ﬁlm and sheet (55%), housewares and toys (16%), wire and cable coating (5%) HDPE: bottles (40%), housewares, containers, toys (35%), pipe and ﬁttings (10%), ﬁlm and sheet (5%) Polyethylene is the most widely used thermoplastic material and is composed of ethylene. The two main types are LDPE and high-density polyethylene (HDPE) [1]. 4.2.1.1.1 Manufacturing Processes LDPE is manufactured by polymerization of ethylene under high pressures (15,000–50,000 psi, i.e., 103– 345 MPa) and elevated temperatures (200–350°C) in the presence of oxygen (0.03–0.1%) as free-radical initiator. Ethylene is a supercritical ﬂuid with density 0.4–0.5 g/cm3 under these conditions. Polyethylene remains dissolved in ethylene at high pressures and temperatures but separates in the lower ranges. Branch polyethylene is produced due to chain transfer to polymer. The type and extent of branching depends on the local reaction temperature and concentrations of monomer and polymer. The molecular weight distributions and the frequencies of long and short branches on polymer chains depend strongly on reactor geometry and operation. The branched products (LDPE) are less crystalline and rigid than higher density species (HDPE) made by low pressure coordination polymerization. Linear polyethylenes are produced in solution, slurry, and increasingly, gas-phase low-pressure processes. The Phillips process developed during the mid 1950s used supported chromium trioxide catalysts 436 Plastics Technology Handbook in a continuous slurry process (or particle-form process) carried out in loop reactors. Earlier, Standard Oil of Indiana patented a process using a supported molybdenum oxide catalyst. The polyethylenes made by both these processes are HDPE with densities of 0.950–0.965 g/cm3 and they are linear with very few side-chain branches and have a high degree of crystallinity. During the late 1970s, Union Carbide developed a low-pressure polymerization process (Unipol process) capable of producing polyethylene in the gas phase that required no solvents. The process employed a chromium based catalyst. In this process (Figure 4.1) ethylene gas and solid catalysts are fed continuously to a ﬂuidized bed reactor. The ﬂuidized material is polyethylene powder which is produced as a result of polymerization of the ethylene on the catalyst. The ethylene, which is recycled, supplies monomer for the reaction, ﬂuidizes the solid, and serves as a heat-removal medium. The reaction is exothermic and is normally run at temperatures 25–50°C below the softening temperatures of the polyethylene powder in the bed. This operation requires very good heat transfer to avoid hot spots and means that the gas distribution and ﬂuidization must be uniform. The keys to the process are active catalysts. These are special organochromium compounds on particular supports. The catalysts yield up to about 106 kg of polymer per kilogram of metallic chromium. Branching is controlled by the use of comonomers like propylene or 1-butene, and hydrogen is used as a chain transfer agent. The catalyst is so efﬁcient that its concentration in the ﬁnal product is negligible. The absence of a solvent and a catalyst removal step makes the process less expensive. The products marketed as linear LLDPE can be considered as linear polyethylenes having a signiﬁcant number of branches (pendant alkyl groups). The linearity imparts strength, the branches impart toughness. During the late 1970s, the Dow Chemical Co. also began producing polyethylene using a proprietary solution process based on Ziegler–Natta-type catalysts. Resins are made at low pressures and with lower densities in a system derived essentially from high-density resin technology (Ziegler–Natta). Higher boiling comonomers (1-hexene and 1-octene) are used to produce LLDPE having superior mechanical properties and ﬁlm-drawing tendencies. A difference of the Dow solution products from gas-phase Gas recycle Cyclone Gas analyzer Catalyst feeder v Disengagement zone Reactor Timer Distribution plate v v Filter Product Gas feed Cooler Compressor FIGURE 4.1 Union Carbide gas phase process for the production of polyethylene. 437 Industrial Polymers TABLE 4.1 Types of Polyethylene Material Chain Structure Density (g/cm3) Crystallinity (%) Process LDPE Branched 0.912–0.94 50 High pressure LLDPE Linear/less branched 0.92–0.94 50 Low pressure HDPE Linear 0.958 90 Low pressure products is that they are produced in standard pellet form with any needed additives incorporated into the pellet. Polyethylene is partially amorphous and partially crystalline. Linearity of polymer chains affords more efﬁcient packing of molecules and hence a higher degree of crystallinity. On the other hand, side-chain branching reduces the degree of crystallinity. Increasing crystallinity increases density, stiffness, hardness, tensile strength, heat and chemical resistance, creep resistance, barrier properties, and opacity, but it reduces stress-crack resistance, permeability, and impact strength. Table 4.1 shows a comparison of three types of polyethylene. Polyethylene has excellent chemical resistance and is not attacked by acids, bases, or salts. (It is, however, attacked by strong oxidizing agents.) The other characteristics of polyethylene which have led to its widespread use are low cost, easy process ability, excellent electrical insulation properties, toughness and ﬂexibility even at low temperatures, freedom from odor and toxicity, reasonable clarity of thin ﬁlms, and sufﬁciently low permeability to water vapor for many packaging, building, and agricultural applications. Major markets for LDPE are in packaging and sheeting, whereas HDPE is used mainly in blow-molded products (milk bottles, household and cosmetic bottles, fuel tanks), and in pipe, wire, and cable applications. Ultra-high-molecular-weight polyethylene materials (see later), which are the toughest of plastics, are doing an unusual job in the textile machinery ﬁeld. 4.2.1.1.2 Chlorinated Polyethylene Low chlorination of polyethylene, causing random substitution, reduces chain order and thereby also the crystallinity. The low chlorine products (22–26% chlorine) of polyethylene are softer, more rubber-like, and more compatible and soluble than the original polyethylene. However, much of the market of such materials has been taken up by chlorosulfonated polyethylene (Hypalon, DuPont), produced by chlorination of polyethylene in the presence of sulfur dioxide, which introduces chlorosulfonyl groups in the chain. Chlorosulfonated LDPE containing about 27% chlorine and 1.5% sulfur has the highest elongation. Chlorosulfonated polyethylene rubbers, designated as CSM rubbers, have very good heat, ozone, and weathering resistance together with a good resistance to oils and a wide spectrum of chemicals. The bulk of the output is used for fabric coating, ﬁlm sheeting, and pit liner systems in the construction industry, and as sheathing for nuclear power cables, for offshore oil rig cables, and in diesel electric locomotives. 4.2.1.1.3 Cross-Linked Polyethylene Cross-linking polyethylene enhances its heat resistance (in terms of resistance to melt ﬂow) since the network persists even about the crystalline melting point of the uncross-linked material. Cross-linked polyethylene thus ﬁnds application in the cable industry as a dielectric and as a sheathing material. Three main approaches used for cross-linking polyethylene are (1) radiation cross-linking, (2) peroxide crosslinking, and (3) vinyl silane cross-linking. Radiation cross-linking is most suitable for thin sections. The technique, however, requires expensive equipment and protective measure against radiation. Equipment requirements for peroxide curing are simpler, but the method requires close control. The peroxide molecules break up at elevated temperatures, producing free radicals which then abstract hydrogen from the polymer chain to produce a polymer free 438 Plastics Technology Handbook radical. Two such radicals can combine and thus cross-link the two chains. It is important, however, that the peroxide be sufﬁciently stable thermally so that premature cross-linking does not take place during compounding and shaping operations. Dicumyl peroxide is often used for LDPE but more stable peroxides are necessary for HDPE. For cross-linking polyethylene in cable coverings, high curing temperatures, using high-pressure steam in a long curing tube set into the extrusion line, are normally employed. Copolymers of ethylene with a small amount of vinyl acetate are often preferred for peroxide crosslinking because the latter promotes the cross-linking process. Large amounts of carbon black may be incorporated into polyethylene that is to be cross-linked. The carbon black is believed to take part in the cross-linking process, and the mechanical properties of the resulting product are superior to those of the unﬁlled material. In the vinyl silane cross-linking process (Sioplas process) developed by Dow, an easily hydrolysable trialkoxy vinyl silane, CH2═CHSi(OR)3, is grafted onto the polyethylene chain, the site activation having been achieved with the aid of a small amount of peroxide. The material is then extruded onto the wire. When exposed to hot water or low-pressure steam, the alkoxy groups hydrolyze and then condense to form a siloxane cross-link: ( Si O Si ) The cross-linking reaction is facilitated by the use of a cross-linking catalyst, which is typically an organotin compound. There are several variations of the silane cross-linking process. In one process, compounding, grafting, and extrusion onto wire are carried out in the same extruder. Cross-linked LDPE foam (see “Polyoleﬁn Foams” in Chapter 2) has been produced by using either chemical cross-linking or radiation cross-linking. These materials have been used in the automotive industry for carpeting, boot mats, sound deadening, and pipe insulation, and as ﬂotation media for oilcarrying and dredging hose. 4.2.1.1.4 Linear Low-Density Polyethylene Chemically, LLDPE can be described as linear polyethylene copolymers with alpha-oleﬁn comonomers in the ethylene chain. They are produced primarily at low pressures and temperatures by the copolymerization of ethylene with various alpha-oleﬁns such as butene, hexene, octene, etc., in the presence of suitable catalysts. Either gas-phase ﬂuidized-bed reactors or liquid-phase solution-process reactors are used. (In contrast, LDPE is produced at very high pressures and temperatures either in autoclaves or tubular reactors.) Polymer properties such as molecular weight, molecular-weight distribution (MWD), crystallinity, and density are controlled through catalyst selection and control of reactor conditions Among the LLDPE processes, the gas-phase process has shown the greatest ﬂexibility to produce resins over the full commercial range. The molecular structure of LLDPE differs signiﬁcantly from that of LDPE: LDPE has a highly branched structure, but LLDPE has the linear molecular structure of HDPE, though it has less crystallinity and density than the latter (see Table 4.1). The stress-crack resistance of LLDPE is considerably higher than that of LDPE with the same melt index and density. Similar comparisons can be made with regard to puncture resistance, tensile strength, tensile elongation, and low- and high-temperature toughness. Thus LLDPE allows the processor to make a stronger product at the same gauge or an equivalent product at a reduced gauge. LLDPE is now replacing conventional LDPE in many architectures because of the combination of favorable production economics and product performance characteristics. For many architectures (blow molding, injection molding, rotational molding, etc.) existing equipment for processing LDPE can be used Industrial Polymers 439 to process LLDPE. LLDPE ﬁlm can be treated, printed, and sealed by using the same equipment used for LDPE. Heat-sealing may, however, require slightly higher temperatures. LLDPE ﬁlms provide superior puncture resistance, high tensile strength, high impact strength, and outstanding low-temperature properties. The resins can be drawn down to thicknesses below 0.5 mil without bubble breaks. Slot-cast ﬁlms combine high clarity and gloss with toughness. LLDPE ﬁlms are being increasingly used in food packaging for such markets as ice bags and retail merchandise bags, and as industrial liners and garment bags. Good ﬂex properties and environmental stress-crack resistance combined with good low-temperature impact strength and low warp age make LLDPE suitable for injection-molded parts for housewares, closures, and lids. Extruded pipe and tubing made from LLDPE exhibit good stress-crack resistance and good bursting strength. Blow-molded LLDPE parts such as toys, bottles, and drum liners provide high strength, ﬂex life, and stress-crack resistance. Light-weight parts and faster blow-molding cycle times can be achieved. The combination of good high- and low-temperature properties, toughness, environmental stress-crack resistance, and good dielectric properties suit LLDPE for wire and cable insulation and jacketing applications. A new class of linear polyethylene copolymers with densities ranging between 0.890 and 0.915 g/cm3, known as very low-density polyethylene (VLDPE), was introduced commercially in late 1984 by Union Carbide. These resins are produced by copolymerization of ethylene and alpha-oleﬁns in the presence of a catalyst. VLDPE provides ﬂexibility previously available only in lower-strength materials, such as ethylene– vinyl acetate (EVA) copolymer (see later) and plasticized PVC, together with the toughness and broader operating temperature range of LLDPE. Its unique combination of properties makes VLDPE suited for a wide range of applications. Generally, it is expected that VLDPE will be widely used as an impact modiﬁer. Tests suggest that it is suited as a blending resin for polypropylene and in HDPE ﬁlms for improved tear strength. On its own, VLDPE should ﬁnd use in applications requiring impact strength, puncture resistance, and dart drop resistance combined with ﬂexibility. The drawdown characteristics of VLDPE allow for very thin ﬁlms to be formed without pinholing. Soft ﬂexible ﬁlms for disposable gloves, furniture ﬁlms, and high-performance stretch and shrink ﬁlm are potential markets. 4.2.1.1.5 High-Molecular-Weight High-Density Polyethylene High-molecular-weight high-density polyethylene (HMW-HDPE) is deﬁned as a linear homopolymers or w ) in the range of approximately 200,000–500,000. copolymer with a weight-average molecular weight (M HMW-HDPE resins are manufactured using predominantly two basic catalyst systems: Ziegler-type catalysts and chromium oxide-based catalysts. These catalysts produce linear polymers which can be either homopolymers when higher-density products are required or copolymers with lower density. Typical comonomers used in the latter type of products are butene, hexane, and octenes. HMW-HDPE resins have high viscosity because of their high molecular weight. This presents problems in processing and, consequently, these resins are normally produced with broad MWD. The combination of high molecular weight and high density imparts the HMW-HDPE good stiffness characteristics together with above-average abrasion resistance and chemical resistance. Because of the relatively high melting temperature, it is imperative that HMW-HDPE resins be specially stabilized with antioxidant and processing stabilizers. HMW-HDPE products are normally manufactured by the extrusion process; injection molding is seldom used. The principal applications of HMW-HDPE are in ﬁlm, pressure pipe, large blow-molded articles, and extruded sheet. HMW-HDPE ﬁlm now ﬁnds application in T-shirt grocery sacks (with 0.6–0.9-mil thick sacks capable of carrying 30 lb of produce), trash bags, industrial liners, and specialty roll stock. Sheets 20– 100 mils thick and 18–20 ft wide are available that can be welded in situ for pond and tank liners. 440 Plastics Technology Handbook HMW-HDPE piping is used extensively in gas distribution, water collection and supply, irrigation pipe, industrial efﬂuent discharge, and cable conduit. The availability of pipe materials with signiﬁcantly higher hydrostatic design stress (800 psi compared with 630 psi of the original HDPE resins) has given added impetus to their use. Large-diameter HMW-HDPE piping has found increasing use in sewer relining. Large-blow-molded articles, such as 55-gal shipping containers, are produced. Equipment is now available to blow mold very large containers, such as 200- and 500-gal capacity industrial trash receptacles and 250-gal vessels to transport hazardous chemicals. 4.2.1.1.6 Ultrahigh-Molecular-Weight Polyethylene Ultrahigh-molecular-weight polyethylene (UHMWPE) is deﬁned by ATM as “polyethylene with molecular weight over three million (weight average).” The resin is made by a special Ziegler-type polymerization. Being chemically similar to HDPE, UHMWPE shows the typical polyethylene characteristics of chemical inertness, lubricity, and electrical resistance, while its very long substantially linear chains provide greater impact strength, abrasion resistance, toughness, and freedom from stress cracking. However, this very high molecular weight also makes it difﬁcult to process the polymer by standard molding and extrusion techniques. Compression molding of sheets and ram extrusion of proﬁles are the normal manufacturing techniques. Forms produced by compression molding r specialty extrusion can be made into ﬁnal form by machining, sintering, or forging Standard wood-working techniques are employed for machining; sharp tools, low pressures, and good cooling are used. Forging can be accomplished by pressing a perform and billet and then forging to then ﬁnal shape. Parts are also formed from compression-molded or skived sheets by heating them above 300°F (∼150°C) and stamping them in typical metal-stamping equipment. Such UHMPWE items have better abrasion resistance than do steel or polyurethanes. They also have high impact strength even at very low temperatures, high resistance to cyclic fatigue and stress cracking, low coefﬁcient friction, good corrosion and chemical resistance, good resistance to nuclear radiation, and resistance to boiling water. Fillers such as graphite, talc, glass beads or ﬁbers, mica, and powdered metals can be incorporated to improve stiffness or to reduce deformation and deﬂection under load. Resistance to abrasion and deformation can be increased by peroxide cross-linking, described earlier. UHMWPE was ﬁrst used in the textile machinery ﬁeld picker blocks and throw sticks, for example. Wear strips, timing wheels, and gears made of the UHMW polymer are used in material handling, assembly, and packaging lines. Chemical resistance and lubricity of the polymer are important in its applications in chemical, food, beverage, mining, mineral processing, and paper industries. All sorts of self-unloading containers use UHMWPE liners to reduce wear, prevent sticking, and speed up the unloading cycles. The polymer provides slippery surfaces that facilitate unloading even when the product is wet or frozen. The polymer ﬁnds applications in transportation, recreation, lumbering, and general manufacturing. Metal equipment parts in some cases are coated or replaced with UHMWPE parts to reduce wear and prevent corrosion. Sewage plants have used this polymer to replace cast-iron wear shoes and rails, bearings, and sprockets. There is even an effort to use UHMW polymer chain to replace metal chain, which is corroded by such environments. Porous UHMW polymer is made by sintering to produce articles of varied porosity. It has found growing use for controlled-porosity battery separators. Patents have been issued on the production of ultrahigh strength, very lightweight ﬁbers from UHMW polymer by gel spinning. 4.2.1.2 Polypropylene CH2 CH CH3 n 441 Industrial Polymers Monomer Propylene Polymerization Ziegler–Natta catalyzed chain polymerization Major Uses Fiber products (30%), housewares and toys (15%), automotive parts (15%), appliance parts (5%) Since its conception in the late 1950s, the propylene polymerization is being revolutionized in terms of both manufacturing hardware and, more importantly, catalyst technology [2–4]. New catalyst technologies, in conjunction with state-of-the-art polymerization, have established polypropylene as the kingpin in the ﬁeld of polyoleﬁns. A brief survey of manufacturing processes and catalyst technologies are presented below. 4.2.1.2.1 Manufacturing Processes Commercial production of crystalline polypropylene (PP) was ﬁrst put on stream in late 1959 by Hercules in the United States, by Montecatini in Italy, and by Farbenwerke Hoechst AG in Germany. The workhorse process for commercial production of PP has been slurry polymerizations in liquid hydrocarbon diluent, for example, hexane or heptane. These are carried out either in stirred batch or continuous reactors. High purity (>99.5%) propylene is fed to the reactor containing diluent, as a suspension of solid (Ziegler–Natta) catalyst particles is metered in. The reaction is carried out at 50–80°C and 5–20 atmospheric pressure. The crystalline polymer produced is insoluble and forms a ﬁnely divided granular solid enveloping the solid catalyst particles. Monomer addition is continued until the slurry reaches 20–40% solids. Residence time varies from minutes to several hours, depending on the catalyst concentration and activity, as well as the speciﬁc reaction conditions. Molecular weight is controlled preferentially by the addition of hydrogen as the chain transfer agent. The reactor slurry of PP is discharged to a stripping unit where the unreacted monomer ﬂashes out for recycling. The catalyst is then deactivated and solubilized by the addition of alcohol. The bulk of the diluent, solubilized catalysts, and atactic polypropylene is solution are removed at this point by centrifuging. The crystalline polymer is puriﬁed by steam distillation and/or by water washing with surface active agents, followed by ﬁltration and centrifuging and then drying. The dried polymer can be stored, transported, or premixed with stabilizers to be used with or without pelletization. The efﬁciency of the Ziegler–Natta catalysts is of the order of 1500 g polymer formed per gram of transition metal. Residual catalyst has adverse effects on the corrosiveness, color, and light stability of the polymer, and extraction processes must be used to remove it from the product. However, by utilizing state-of-the-art, high-mileage (supported) catalyst systems, polymer yields are obtained which are several orders of magnitude higher than those obtained with ﬁrst generation, Ziegler catalysts. Typically the highmileage catalysts produce about 300,000 g of polymer per gram of transition metal. Since there are only about 3 ppm of residual metal in the polymer, catalyst removal is unnecessary. The expensive catalyst removal (deashing) steps required for the products made by earlier Ziegler catalyst systems are thus eliminated. Propylene is readily polymerized in bulk; that is, in the liquid monomer itself. Arco, El Paso, Phillips, and Shell are practitioners of bulk processing in stirred or loop reactor systems. In either case, liquid propylene (and ethylene, if random copolymer is desired) is continuously metered to the polymerization reactor along with a high-activity/high-stereospeciﬁcity catalyst system. Polymerization temperatures are normally in the range of 45–80°C with pressures sufﬁcient to maintain propylene in the liquid phase (250– 500 psi, that is, 1.7–3.5 MPa). Hydrogen is used for molecular weight control. The polymer slurry (approximately 30–50% solids in liquid propylene) is continuously discharged from loop reactors through a series of sequence valves into a zone maintained essentially at atmospheric pressure an containing terminating agents. Technologies also exist for production of propylene/ethylene block copolymers via bulk polymerization employing stirred or loop reactors [5]. 442 Plastics Technology Handbook Additives Steam Modern vapor-phase polymerization is represented in one form by the stirred gas-phase process originally developed by BASF and licensed by Norchem in the United States [6]. BASF process reactors contain a spiral or double-helical agitator to stir the polymer bed. Cooling of the bed is maintained by continuous injection of fresh, high-purity propylene in a liquid or partly liqueﬁed state into the reaction zone. The unreacted propylene is removed from the top of the reactor during polymerization, condensed, and reinjected with fresh propylene. Evaporation of the unreacted propylene absorbs the heat of polymerization and also brings about intense mixing of the solid polymer particles with the gas phase. Energy costs of the process are economically attractive [7]. The diluent-free BASF process provides sufﬁciently high yield of polymer per unit of catalyst so that deashing is not required. Although products made in this way contain relatively high levels of titanium and aluminum residues, a unique ﬁnishing step during extrusion palletizing reduces active chlorides to an innocuous level [6]. With dramatic improvements in Ziegler–Natta catalyst technology, the Spheripol process, ﬁrst developed by Montedison and Mitsui Petrochemical with simpliﬁed bulk (liquid propylene) process technology operating with loop reactors, is capable of directly producing a relatively large round bead with suitable density to eliminate the need for pelletizing for many applications [5]. Subsequently, Montedison and Hercules, Inc., which assumed responsibility for all polypropylene operations and technology of the parent companies. The Himont spheripol loop reactor process is initiated by injecting specially prepared supported catalyst and cocatalyst into liquid propylene circulated in a relatively simple high L/D ratio loop reactor, followed by monomer removal (Figure 4.2). The homopolymers so produced can be circulated through ethylene and ethylene/propylene gas phase reactors for insertion of copolymer fractions before ﬁnal monomer stripping. The Unipol low-pressure gas-phase ﬂuidized-bed process, which was introduced by Union Carbide in 1977 for LLDPE, has also been adapted to the production of PP homopolymers and block copolymers using Shell Chemicals high-activity (Ziegler-type) catalyst technology. The Spheripol and Unipol processes are capable of producing polymer in crumb bead or granular forms, with the potential for direct marketing without any pelletizing ﬁnishing operation. Regardless of the polymerization process used, the PP homo- and co-polymer must be stabilized to some degree to prevent oxidative degradation. The general practice is to incorporate a small quantity of stabilizer in the polymer prior to the ﬁrst exposure to elevated temperatures of a drying operation or longterm storage. Inert-gas (nitrogen) blanketing is also used in some storage/transfer systems. Additional stabilizers, up to 1%, are added to the polymer during pelletizing. Most commercial PP compositions Propylene Catalyst and FIGURE 4.2 CW A simpliﬁed ﬂow diagram of the Himont spherical loop reactor process. Nitrogen CW Steam Liquid phase loop reactors chemicals Spherical polyproylene to storage Industrial Polymers 443 contain mixtures of hindered phenols and hydroperoxide decomposers or various phosphates (see Chapter 1). 4.2.1.2.2 Catalyst Technology The origin of sophisticated catalyst used today are to be found in the early work of Karl Ziegler (1950) and Guilio Natta (1954). In the last ﬁve decades, several distinct “generations” of catalyst technologies have emerged. The earliest commercial catalysts ( ﬁrst generation) were essentially titanium trichloride, simply prepared by reducing TiCl4 with alkylaluminums to yield brown (b) TiCl3, which was subsequently heated to convert it to the stereospeciﬁc purple (g) form. In the 1970s, improved or second-generation catalysts were developed. The essence of the improvement was that catalyst poisons AlCl3 or AlEtCl2, which are cocrystallized with or absorbed onto the TiCl3 catalyst, were removed by using dialkyl ethers (especially di-n-butyl ether and di-isoamyl ether). The 1980s heralded the widespread commercial implementation of supported catalysts. These thirdgeneration catalysts comprise of TiCl4 on a specially prepared MgCl2 support. Commercially available MgCl2 is converted to “active MgCl2” by treating with “activating agents,” which are electron donors (Lewis bases) such as ethyl benzoate, diisobutylphthalate, and phenyl triethoxy silane. These are also used in conjunction with the cocatalyst (trialkylaluminum) as a “selectivity control agent.” While the ﬁrst-generation catalysts were suitable for slurry process, in which polymerization occurs in a parafﬁnic solvent, third-generation supported catalysts with dramatically higher activity (typically 1500 kg PP/g Ti compared to 15 kg/g Ti for ﬁrst generation TiCl3 catalyst) and stereo-speciﬁcity not only allowed the full exploitation of the advantages of a solventless polymerization, but also made substantial simpliﬁcation of slurry process possible through elimination of atactic removal and catalyst de-ashing. These third-generation catalysts can be used in both the bulk and simpliﬁed slurry processes. They are, however, unsuitable for a gas-phase process. The major advance offered by the later ( fourth generation) supported catalysts is their controlled morphology, which rendered them suitable for all commercial polymerization processes. These are the only catalysts suitable for full exploitation of the advantages of the solvent-less polymerization including the gas-phase process. These catalysts come in a variety of regular shapes, such as spherical, cubical, or cylindrical, as single particles or clusters of several particles, and are characterized by sufﬁciently narrow particle size distributions with the minimum of ﬁnes or coarse particles. During polymerization, replication of the catalyst shape occurs, and hence spherical PP beads are obtained directly from the reactor, if the catalyst is spherical. By producing catalysts with a dense, spherical, and uniform particle shape it is thus possible to generate PP particles that, for many applications, can be shipped and used without a granulation or extrusion step to generate the “nibs” desired by most customers. This concept is already part of the Himont Spheripol process mentioned earlier. While attractive in terms of economy, since the energy consuming extrusion step is obviated in the above process, there are various disadvantages such as the difﬁculty of homogeneously administrating additives/stabilizers, the loss of the options for molecular weight modiﬁcations via extruder operation, and the fact that the process yields particles of smaller size than what is preferred by the industry simply because such large-size particles cannot be kept in suspension during polymerization. Montecatini has developed spherical morphology MgCl2 supported catalysts for ethylene and propylene polymerization. The name of this process is “spherilene.” The major advantage in such system is the preparation of marketable powders directly from polymerization plants and the catalyst is suitable for slurry, bulk, and gas-phase processes. In the so-called reactor granule technology developed by Montecatini, the growing particle itself serves as the reactor within which the polymerization takes place. Thus alloys and blends previously possible by mixing and melt extrusion of polyoleﬁns can be made with this technology directly in the reactor from the individual monomers, dramatically decreasing the producing cost by eliminating the need for compounding equipment. 444 Plastics Technology Handbook A variety of specialty polyoleﬁns and polyoleﬁn alloys can now be made directly in the reactor taking advantage of the new technology. Examples are the catalloy materials from Himont, which are polyoleﬁn alloys made by synthesis and not by the conventional route of compounding. Hivalloy is a polypropylene/ polystyrene alloy made by synthesis and combines the properties of both crystalline and amorphous engineering polymers. Such materials could challenge the established positions of several thermoplastic elastomers. Despite the domination of supported catalysts in polyoleﬁn production, industrial research work has shifted towards new generation, single-site, homogeneous, Ziegler–Natta catalyst systems, the main compound of which is the Group IVB transition metallocenes (titanocene, zirconocenes, and hafnocenes) [8]. The molecular structure of the two famous Brintzinger catalysts, ethylenebis (indenyl) zirconium dichloride Et(ind)2ZrCl2 and ethylenebis (tetrahydroindenyl) zirconium dichloride Et(H4Ind)2ZrCl2 are depicted in Figure 4.3. Methylalumoxane (MAO) is the most important co-catalyst that activates the group IVB metallocenes in homogeneous Ziegler–Natta polymerization. Before the discovery of the MAO cocatalyst, the homogeneous Ziegler–Natta catalyst Cp2TiCl2 (Cp = cyclopentadiene) was activated with alkyl aluminum chloride, which led to poor catalyst activity. The use of MAO cocatalyst raised the catalyst activity by several orders of magnitude. Possible structures of MAO are shown in Figure 4.4. Using the chiral zirconocene catalyst in combination with the MAO cocatalyst, polypropylene can be obtained in high purity and high yield [9] even at relatively high temperatures (room temperature and above). The zirconium catalysts are three orders of magnitude more active than their titanium counterparts. Unlike the heterogeneous Ziegler–Natta catalysts in which usually several types of active catalyst sites are present, metallocene catalysts are single-site in nature, that is, they have a single active catalyst site on the catalyst structure and thus they make one type of polymer, ensuring a high degree of purity. Cl Zr Cl Cl Et (ind)2 ZrCl2 FIGURE 4.3 Zr Cl Et (H4ind)2 ZrCl2 Structures of the Brintzinger catalysts. CH3 Linear Al O CH3 l Al CH3 Cyclic CH3 O Al CH3 n CH3 Al O CH3 l Al CH3 O Al n n = 4–20 O FIGURE 4.4 Possible structures of methylalumoxane. (After Kaminsky, W., Sinn, H., and Woldt, R. 1983. Macromol. Chem. Rapid Commun., 4, 417.) Industrial Polymers 445 The very high selectivity of the zirconocene single-site catalysts and their high activity, which approaches that of the MgCl2-supported catalysts described in the previous section, makes them a serious contender for future processes. It can be expected [4] that the relative ease of tailoring of homogeneous catalysts compared to complicated heterogeneous systems will enable these catalysts to be further improved and exploited in terms of activity and selectivity (isotacticity, molecular weight, and distribution). It is worthwhile to mention that homogeneous Ziegler–Natta catalysts other than metallocene-based catalysts have also been developing rapidly in recent years The catalyst precursors of these systems are nonmetallocene organometallic compounds, such as monocyclopentadienyl derivatives. A representative of monocyclopentadienyl catalysts is the constrained geometry (CG) catalyst. This new type of homogeneous catalyst was developed by Dow Plastics [10]. The catalyst system is based on group IVB transition metals such as Ti, covalently bonded to a cyclopentadienyl group bridged with a heteroatom such as nitrogen. The components are linked in such a way that a constrained cyclic structure is formed with Ti at the center. The bond angle between the monocyclopentadienyl group, Ti center, and heteroatom is less than 115°. The catalyst is activated by strong Lewis acid systems to a highly efﬁcient cationic form. The CG catalysts produce highly processable polyoleﬁns with a unique combination of narrow MWD and long chain branches. Ethylene–octene copolymers produced with CG catalysts have useful properties across a range of densities and melting indexes. These novel copolymer families are called polyoleﬁn plastomers (POP) and polyoleﬁn elastomers (POE). POPs possess plastic and elastic properties while POEs containing greater than 20 wt% octene comonomer units have higher elasticity. The CG catalyst technology is not limited to the typical selection of C2–C8 a-oleﬁns, but can include higher a-oleﬁns. The open structure of the CG catalyst signiﬁcantly increases the ﬂexibility to insert higher a-oleﬁn comonomers into the polymer structure. This technology also allows addition of vinylended polymer chains to produce long chain branching. 4.2.1.2.3 Properties “Polypropylene” is not one or even 100 products. Rather it is a multidimensional range of products with properties and characteristics interdependent on the type of polymer (homopolymers, random, or block copolymer), molecular weight and molecular weight distribution, morphology and crystalline structure, additives, ﬁllers and reinforcing ﬁllers, and fabrication techniques. Commercial homopolymers are usually about 90–95% isotactic, the other structures being atactic and syndiotactic (a rough measure of isotacticity is provided by the “isotactic index”—the percentage of polymer insoluble in heptane): the greater the degree of isotacticity the greater the crystallinity and hence the greater the softening point, stiffness, tensile strength, modulus, and hardness. Although very similar to HDPE, PP has a lower density (0.90 g/cm3) and a higher softening point, which enables it to withstand boiling water and many steam sterilizing operations. It has a higher brittle point and appears to be free from environmental stress-cracking problems, except with concentrated sulfuric acid, chromic acid, and aqua regia. However, because of the presence of tertiary carbon atoms occurring alternately on the chain backbone, PP is more susceptible to UV radiation and oxidation at elevated temperatures. Whereas PE cross-links on oxidation, PP undergoes degradation to form lowermolecular-weight products. Substantial improvement can be made by the inclusion of antioxidants, and such additives are used in all commercial PP compounds. The electrical properties of PP are very similar to those of HDPE. Because of its reasonable cost and good combination of the foregoing properties, PP has found many applications, ranging from ﬁbers and ﬁlament to ﬁlms and extrusion coatings. A signiﬁcant portion of the PP produced is used in moldings, which include luggage, stacking chairs, hospital sterilizable equipment, toilet cisterns, washing machine parts, and various auto parts, such as accelerator pedals, battery cases, dome lights, kick panels, and door frames. 446 Plastics Technology Handbook Although commercial PP is a highly crystalline polymer, PP moldings are less opaque when unpigmented than are corresponding HDPE moldings, because the differences between amorphous and crystal densities are less with PP (0.85 and 0.94 g/cm3, respectively) than with polyethylene (0.84 and 1.01 g/cm3, respectively). A particularly useful property of PP is the excellent resistance of thin sections to continued ﬂexing. This has led to the production of one-piece moldings for boxes, cases, and accelerator pedals in which the hinge is an integral part of the molding. Monoaxially oriented polypropylene ﬁlm tapes have been widely used for carpet backing and for woven sacks (replacing those made from jute). Combining strength and lightness, oriented PP straps have gained rapid and widespread acceptance for packaging. Nonoriented PP ﬁlm, which is glass clear, is used mainly for textile packaging. However, biaxially oriented PP ﬁlm is more important because of its greater clarity, impact strength, and barrier properties. Coated grades of this material are used for packaging potato crisps, for wrapping bread and biscuits, and for capacitor dielectrics. In these applications PP has largely replaced regenerated cellulose. (The high degree of clarity of biaxially oriented PP is caused by layering of the crystalline structures. Layering reduces the variations in refractive index across the thickness of the ﬁlm, which thus reduces the amount of light scattering.) Polypropylene, produced by an oriented extrusion process, has been uniquely successful as a ﬁber. Its excellent wear, inertness to water, and microorganisms, and its comparatively low cost have made it extensively used in functional applications, such as carpet backing, upholstery fabrics, and interior trim for automobiles. Random ethylene–propylene copolymers, another important variety of polypropylene, are noted for high clarity, a lower and broader melting range than homopolymers grades, reduced ﬂexural modulus, and higher melt strengths. They are produced by the random addition of ethylene to a polypropylene chain as it grows. The melt-ﬂow rate of random copolymers ranges from 1 g/10 min for a blow-molding grade to 35 g/10 min for an injection-molding grade. The density is about 0.90 g/cm3 and the notched Izod impact strength of the materials ranges from under 1 to more than 5 ft.-lb/in. The blow-molded bottles capitalize on the good clarity provided by the random copolymer. The high gloss and very broad heat sealing range of this resin is useful in such cast-ﬁlm applications as trading cards and document protectors. Polypropylene copolymers with a melt-ﬂow rate of 35 g/10 min or above ﬁnd applications in thin wall parts, usually for injection molded food packaging such as delicatessen containers or yogurt cups. Such containers have walls with a length-to-thickness ratio as high as 400:1; yet they retain the properties of top-load strength, impact resistance, and recyclability that are typical of polypropylene. Block copolymers, preferably with ethylene, are classed as having medium, high, or extra-high impact resistance with particular respect to subzero temperatures. Block copolymers consist of a crystalline PP matrix containing segments of EPR-type elastomer and/or crystalline PE for energy impact absorption in the rubber phase [5]. The level of the ethylene comonomer as well as the size of these segments has an important bearing on the physical properties of the ﬁnal block copolymer. 4.2.1.2.4 Use Pattern Few materials are as compatible with as many processing techniques or are used in as many commercial applications as polypropylene. It is found in everything from ﬂexible and rigid packaging to ﬁbers and large molded parts for automotive and consumer products. Largely conforming to this diversity of applications is the fact that the material can be processed by most methods, including extrusion, extrusion coating, blown and cast ﬁlm, blow molding, injection molding, and thermoforming. Polypropylene ﬁbers and ﬁlaments form the largest market area, which is comprised of several segments with carpeting applications being the largest; these include primary and secondary woven and nonwoven uses, carpet backing face yarns, indoor/outdoor constructions, automotive interior mats and trunk linings, and synthetic turf. 447 Industrial Polymers The good wickability of PP is utilized in such nonwoven applications as disposable diaper. Other applications include clothing inner liners, drapes and gowns, sleeping bags, wall coverings, wiping cloths, and tea bags. Furniture and automotive upholstery fabrics are produced from both continuous monoﬁlament and staple ﬁbers. The second largest PP market is ﬁlm, both oriented (OPP) and cast. Large users of OPP ﬁlms are packaging for snack foods, bakery products, dry foods, candy, gum, cheese, tobacco products, and electrical capacitors. Cast (unoriented) ﬁlm is used for packaging textile soft goods, cheese, snack foods, and bakery products. Largest users of injection molded PP are in transportation, particularly automotive and truck batter cases. PP copolymers have secured about 90% of this market as a result of a drive by automotive manufacturers to reduce weight and cost. In addition to being lightweight, PP also provides outstanding resistance to creep and fatigue, high temperature rigidity, impact strength, and resistance to corrosion. The next largest molded product market for PP is packaging, especially closures and containers. Childresistant, tamperproof, linerless features are important design factors as also inherent chemical resistance, stress-crack resistance, and high productivity at low cost. Housewares utilize random copolymers for refrigerator and shelf-storage containers and lids. Medium-impact copolymers are used for hot/cold thermos containers, lunch boxes, coolers, and picnic ware. Medical applications of PP such as disposable syringes, hospital trays, and labware are contingent on sterilizability, either autoclaving or radiation. Disposable syringes that are sterilized by radiation require special formulations to prevent discoloration (yellowing) or brittleness as a consequence of degradation and cross-linking. PP ﬁnds highly successful uses in both major and small appliances. Washing machines, dish washers, tub liners, agitators, bleach and detergent dispensing units, valve and control assemblies, drain tubes, pump housings, door liners, coffee makers, hair dryers, vacuum cleaners, can openers, knife sharpeners, room humidiﬁers and dehumidiﬁers, ﬂoor and ceiling fans, and window air-conditioner units are some examples. 4.2.1.3 Polyallomer ( CH2 CH )m ( CH2 CH2 ) n CH3 A proprietary polymerization process, developed in the mid 1960s by staff researchers of Eastman Chemical Products, produces copolymers of 1-oleﬁns that give a degree of crystallinity normally obtained only with homopolymers. The term polyallomer was coined to identify the polymers manufactured by this process and to distinguish them from conventional copolymers. The polyallomer materials available today are based on block copolymers of propylene and ethylene. Polyallomers combine the most desirable properties of both crystalline polypropylene and high-density polyethylene (HDPE) and can offer impact strengths three or four times that of polypropylene. Resistance to heat distortion is better than that of HDPE but not quite as good as that of polypropylene. Polyallomer has better abrasion resistance than polypropylene and comparable hinge-forming characteristics. Polyallomer lightweight cases can be molded entirely in one piece. Back, front, hinges, handles, and snap clasps can be molded in at the same time in a wide range of colors. Polyallomer is thus used in such injection molded items as ﬁshing tackle boxes, typewriter cases, gas-mask cases, and bowlingball bags. Shoe toes molded of polyallomer resist cracking and denting under repeated hammer blows. They withstand temperatures from −40°C to 150°C and can withstand up to 300 pounds of force (1335 N). Polyallomers can be processed easily on conventional molding and extruding equipment. Polypropylene color concentrates can be used to color polyallomer, since these two polymers are compatible. 448 Plastics Technology Handbook 4.2.1.4 Poly(Vinyl Chloride) [ CH2 CH [ n CI Monomer Vinyl chloride Polymerization Free-radical-initiated chain polymerization Major Uses Pipe and ﬁttings (35%), ﬁlm and sheet (15%), ﬂooring materials (10%), wire and cable insulation (5%), automotive parts (5%), adhesives and coatings (5%) PVC is produced by polymerization of vinyl chloride by free-radical mechanisms, mainly in suspension and emulsion, but bulk and solution processes are also employed to some extent [11–15]. (The control of vinyl chloride monomer escaping into the atmosphere in the PVC production plant has become important because cases of angiosarcoma, a rare type of liver cancer, were found among workers exposed to the monomer. This led to setting of stringent standards by governments and modiﬁcation of manufacturing processes by the producers to comply with the standards.) At processing temperatures used in practice (150–200°C), sufﬁcient degradation may take place to render the product useless. Evidence points to the fact that dehydrochlorination occurs at an early stage in the degradation process and produces polyene structures: CH2 CH Cl CH2 CH HCl CH CH CH CH Cl It is believed that the liberated hydrogen chloride can accelerate further decomposition and that oxygen also has an effect on the reaction. However, incorporation of certain materials known as stabilizers retards or moderates the degradation reaction so that useful processed materials can be obtained. Many stabilizers are also useful in improving the resistance of PVC to weathering, particularly against degradation by UV radiation. 4.2.1.4.1 Characterization of Commercial Resins w = Commercial PVC polymers are largely amorphous with molecular weights in the range M 100, 000–200, 000 and Mn = 45, 000–64, 000, although values may be as low as 40,000 and as high as w . In practice, the ISO viscosity number is often used to characterize the molecular weight of 480,000 for M a PVC polymer. Table 4.2 compares typical correlations between number and weight average molecular weights with ISO numbers. Most general purpose polymer for use in plasticized PVC compounds have ISO numbers of about 125. Because of processing problems the polymer used for unplasticized PVC compounds have lower molecules weights, typical ISO numbers being 105 for pipe, 85–95 for rigid sheet, and as low as 70 for injection molding compounds [14]. With commercial polymers the major differences are in the characteristics of the particle, i.e., its shape, size, size distribution, and porosity. Such differences considerably affect the processing behavior of a polymer. Considerable effort has been expended to develop suitable process to control porosity, surface area, and diffusivity of PVC particles and this has led to great improvements over the years in the processability of PVC. If PVC polymer particles are mixed, at room temperature, with plasticizers, the immediate product may take one of two forms. If the plasticizer quantity is insufﬁcient to ﬁll all the gaps between the particles, 449 Industrial Polymers TABLE 4.2 Molecular Weight Characterization of PVC Average Molecular Weight Weight Number ISO/R174-1961(E): Viscosity Number 54,000 70,000 26,000 36,000 57 70 1,00,000 45,500 87 1,40,000 2,00,000 55,000 64,000 105 125 2,60,000 73,000 145 3,40,000 82,000 165 Source: Matthews, G. A. R. 1972. Vinyl and Allied Polymers, Vol. 1. Vinyl Chloride and Vinyl Acetate Polymers, Iliffe, London, UK. a mush will be produced. If all the voids are ﬁlled then the particles will become suspended in the excess plasticizer and a paste will be formed. The viscosity of a PVC paste (see “Plastisol Casting” in Chapter 2) made from a ﬁxed polymer– plasticizer ratio depends to a great extent on the particle size and size distribution. To obtain a low viscosity paste the amount of plasticizer required to ﬁll the voids between particles should be low so that more plasticizer is available to act as a lubricant for the particles, facilitating their general mobility in suspension. Thus in general PVC pastes in which the polymer has a wide particle-size distribution (but within limits set by problems of signiﬁcant plasticizer absorption even at room temperature by very small particles and settling caused by large particles) so that particles pack efﬁciently and leave less voids (see Figure 4.5a) will have lower viscosity than those of constant particle size (Figure 4.5b). The use of “ﬁller” polymers in increasing quantities in PVC paste technology is an extension of this principle. These ﬁller polymers are made by suspension (granular, dispersion) polymerization and by themselves the particles are too large to make stable pastes. However, in the presence of much smaller paste polymer particles they remain in stable suspension. As shown in Figure 4.6, the replacement in space of a mixture of paste-polymer particles and plasticizer by a large granular polymer particle releases plasticizer which then acts as a lubricant, i.e., a viscosity depressant. PVC pastes exhibit complex rheological behavior with the viscosities showing dependence on the shear rate and on the time of shear. A paste viscosity may increase with shear rate (dilatancy) or decrease (shear thinning or pseudoplasticity). Some pastes may show dilatant tendencies over one range of shear rates but be shear thinning over another range. The viscosities may also decrease with time of stirring (thixotropy) or increase with it (rheopexy) [14]. It has been observed that spherical particles with distribution of size giving a high degree of packing are closest to Newtonian liquids in their behavior. Spherical particles of homogeneous size, however, give (a) (b) FIGURE 4.5 (a) PVC paste polymer particles with homogeneous particle size—less efﬁcient packing. (b) PVC paste polymer particles with distribution of size—efﬁcient packing. (After Brydson, J. A. 1982. Plastics Materials. Butterworth Scientiﬁc, London, UK.) 450 Plastics Technology Handbook (a) (b) FIGURE 4.6 (a) PVC paste polymer suspended in plasticizer. (b) PVC paste containing ﬁller polymer. Less plasticizer is required to ﬁll voids in unit volume. (After Brydson, J. A. 1982. Plastics Materials. Butterworth Scientiﬁc, London, UK.) shear thinning pastes. This may be due to the fact that these particles tend to aggregate at rest while shearing causes disaggregation and hence easier movement of particles. Very coarse and lumpy uneven granules do not slide past each other in pastes and tend to become more entangled as shear rate increases. Such pastes commonly show dilatant behavior. 4.2.1.4.2 Compounding Ingredients PVC is a colorless rigid material with limited heat stability and with a tendency to adhere to metallic surfaces when heated. For these and other reasons it is necessary to compound the polymer with other ingredients to make useful products. It is possible in this way to make a wide range of products including rigid piping and soft elastic cellular materials. A PVC compound may contain, besides the polymer, the following ingredients: stabilizers, plasticizers, extenders, lubricants, ﬁllers, pigments, and polymeric processing aids. Other ingredients also used occasionally include impact modiﬁers, ﬁre retardants, optical bleaches, and blowing agents. 4.2.1.4.3 Stabilizers The most important class of stabilizers are the lead compounds which form lead chloride on reaction with the hydrogen chloride evolved during decomposition. Basic lead carbonate (white lead), which has a low weight cost, is more commonly used. A disadvantage of lead carbonate is that it may decompose with the evolution of carbon dioxide at higher processing temperatures and lead to a porous product. For this reason, tribasic lead sulfate, which gives PVC products with better electrical insulation properties than lead carbonate, is often used despite its somewhat higher weight cost. Other lead stabilizers are of much more speciﬁc applications. For example, dibasic lead phthalate, which is an excellent heat stabilizer, is used in heat-resistant insulation compounds (e.g., in 150°C wire), in high-ﬁdelity gramophone records, in PVC coatings for steel, and in expanded PVC formulation. The use of lead compounds as stabilizers has been subjected to regulation because of its toxicity. Generally, lead stabilizers are not allowed in food-packaging PVC materials, but in most countries they are allowed in PVC pipes for conveying drinking water, with reduction in the level of use of such stabilizers. Today the compounds of cadmium, barium, calcium, and zinc have gained prominence as PVC stabilizers. A modern stabilizing system may contain a large number of components. A typical cadmium– barium packaged stabilizer may have the following composition: cadmium–barium phenate 2–3 parts, epoxidized oils 3–5 parts, trisnonyl phenyl phosphite 1 part, stearic acid 0.5–1 part, and zinc octoate 0.5 part by weight. For ﬂooring compositions, calcium–barium, magnesium–barium, and copper–barium compounds are sometimes used in conjunction with pentaerythritol (which has the function of reducing color by chelating iron present in asbestos). Another group of stabilizers are the organotin compounds. Development of materials with low toxicity, excellent stabilizing performance, and improving relative price situation has led to considerable 451 Industrial Polymers growth in the organotin market during the last decade. Though the level of toxicity of butyltins is not sufﬁciently low for application in contact with foodstuffs, many of the octylins, such as dioctyltin dilaurate and dioctyltin octylthiog–lycolate, meet stringent requirements for use in contact with foodstuffs. Further additions to the class of organotins include the estertins characterized by low toxicity, odor, and volatility, and the methyltins having higher efﬁciency per unit weight compared with the more common organotins. Organotin compounds that are salts of alkyltin oxides with carboxylic acids (e.g., dioctyltin dilaurate) are usually called organotin carboxylates. Organotin compounds with at least one tin-sulfur bond (e.g., dioctyltin octylthioglycolates) are generally called organotin mercaptides. The latter are considered to be the most efﬁcient and most universal heat stabilizers. The important products which are on the market have the following structures: R1 S R2 R2 Sn R1 S R2 Sn S R2 S R2 S R2 Where R1 is H3C–, n-C4H9–, n-C8H17–, n-C12H25 or alkyl–O–CO–CH2–CH2–, and R2 is –CH2–CO– O–alkyl, –CH2–CH2–CO–O–alkyl, –CH2–CH2–O–CO–alkyl or –alkyl. One of the most important properties—not only of the sulfur-containing tin stabilizers but also of the whole group of organotin stabilizers—is absolute crystal clarity, which can be achieved by means of a proper formulation. Clarity is required for bottles, containers, all kinds of packaging ﬁlms, corrugated sheets, light panels, and also for hose, proﬁles, swinging doors, and transparent top coats of ﬂoor or wall coverings made from plasticized PVC. 4.2.1.4.4 Plasticizers In addition to resin and stabilizers, a PVC compounds may contain ingredients such as plasticizers, extenders, lubricants, ﬁllers, pigments, polymeric processing aids, and impact modiﬁers. Plasticizers (see also Chapter 1) are essentially nonvolatile solvents for PVC. At the processing temperature of about 150°C, molecular mixing occurs in a short period of time to give products of greater ﬂexibility. Phthalates prepared from alcohols with about eight carbon atoms are by far the most important class and constitute more than 70% of plasticizers used. For economic reasons, diisooctyl phthalate (DIOP), di-2-ethylhexyl phthalate (DEHP or DOP), and the phthalate ester of the C7–C9 oxo-alcohol, often known as dialphanyl phthalate (DAP) because of the ICI trade name “Alphanol-79” for the C7-C9 alcohols, are used. DIOP has somewhat less odor, whereas DAP has the greatest heat stability. Dibutyl phthalate and diisobutyl phthalate are also efﬁcient plasticizers and continue to be used in PVC (except in thin sheets) despite their high volatility and water extractability. Phosphate plasticizers such as tritolyl phosphate and trixylyl phosphate are generally used where good ﬂame resistance is required, such as in insulation and mine belting. These materials, however, are toxic and give products with poor low-temperature resistance, i.e., with a high cold ﬂex temperature (typically, −5°C). For applications where it is important to have a compound with good low-temperature resistance, aliphatic ester plasticizers are of great value. Dibutyl sebacate, dioctyl sebacate, and, more commonly, cheaper esters of similar effect derived from mixed acids produced by the petrochemical industry are used. These plasticizers give PVC products with a cold ﬂex temperature of −42°C. Esters based on allyl alcohol, such as diallyl phthalate and various polyunsaturated acrylates, have proved useful in improving adhesion of PVC to metal. They may be considered as polymerizable plasticizers. In PVC pastes they can be made to cross-link by the action of peroxides or perbenzoates when the paste is spread on to metal, giving a cured coating with a high degree of adhesion [14]. The high adhesion of these rather complex compounds has led to their development as metal-to-metal adhesives used, for example, in car manufacture. Metal coatings may also be provided from plasticized powders containing polymerizable plasticizers by means of ﬂuidized bed or electrostatic spraying techniques. 452 Plastics Technology Handbook 4.2.1.4.5 Extenders In the formulation of PVC compounds it is not uncommon to replace some of the plasticizer with an extender, a material that is not in itself a plasticizer because of its very low compatibility but that can be used in conjunction with a true plasticizer. Commercial extenders are cheaper than plasticizers and can often be used to replace up to one-third of the plasticizer without seriously affecting the properties of the compound. Three commonly employed types of extenders are chlorinated parafﬁn waxes, chlorinated liquid parafﬁnic fractions, and oil extracts [14]. 4.2.1.4.6 Lubricants In plasticized PVC it is common practice to incorporate a lubricant whose main function is to prevent sticking of the compound to processing equipment [14]. The material used should have limited compatibility such that it will sweat out during processing to form a ﬁlm between the bulk f the compound and the metal surfaces of the processing equipment. The additives used for such a purpose are known as external lubricants. In the United States normal lead stearate is commonly used. This material melts during processing and lubricates like wax. Also used is dibasic lead stearate, which does not melt but lubricates like graphite and improves ﬂow properties. In Britain, stearic acid is mostly used with transparent products, calcium stearate with nontransparent products. An unplasticized PVC formulation usually contains at least one other lubricant, which is mainly intended to improve the ﬂow of the melt, i.e., to reduce the apparent melt viscosity. Such materials are known as internal lubricants. Unlike external lubricants they are reasonably compatible with the polymer and are more like plasticizers in their behavior at processing temperatures, whereas at room temperature this effect is negligible. Among materials usually classiﬁed as internal lubricants are montan wax derivatives, glyceryl monostearate, and long-chain esters such as cetyl palmitate. 4.2.1.4.7 Fillers Fillers are commonly employed in opaque PVC compounds to reduce cost and to improve electrical insulation properties, to improve heat deformation resistance of cables, to increase the hardness of a ﬂooring compound, and to reduce tackiness of highly plasticized compounds. Various calcium carbonates (such as whiting, ground limestone, precipitated calcium carbonate) are used for general-purpose work, china clay is commonly employed for electrical insulation, and asbestos for ﬂooring applications. Also employed occasionally are the silicas and silicates, talc, light magnesium carbonate, and barytes (barium sulfate). 4.2.1.4.8 Pigments Many pigments are now available commercially for use with PVC. Pigment selection should be based on the pigments ability to withstand process conditions, its effect on stabilizer and lubricant, and its effect on end-use properties, such as electrical insulation. 4.2.1.4.9 Impact Modiﬁers and Processing Aids Unplasticized PVC present some processing difﬁculties due to its high melt viscosity; in addition, the ﬁnished product is too brittle for some applications. To overcome these problems and to produce toughening, certain polymeric additives are usually added to the PVC. These materials, known as impact modiﬁers, are generally semicompatible and often some what rubbery in nature [14]. Among the most important impact modiﬁers in use today are butadiene–acrylonitrile copolymers (nitrile rubber), acrylonitrile–butadiene–styrene (ABS) graft terpolymers, methacrylate–butadiene–styrene (MBS) terpolymers, chlorinated polyethylene, and some polyacrylates. ABS materials are widely used as impact modiﬁers, but they cause opacity and have only moderate aging characteristics. Many grades also show severe stress whitening, A phenomenon advantageously 453 Industrial Polymers employed in labeling tapes, such as Dymotape. MBS modiﬁers have been used such as where tough PVC materials of high clarity are desired (e.g., bottles and ﬁlm). Chlorinated polyethylene has been widely used as an impact modiﬁer where good aging properties are required. A number of polymeric additives are also added to PVC as processing aids. They are more compatible with PVC and are included mainly to ensure more uniform ﬂow and thus improve the surface ﬁnish. In chemical constitution they are similar to impact modiﬁers and include ABS, MBS, acrylate–methacrylate copolymers, and chlorinated polyethylene. Typical formulations of several PVC compounds for different applications are given in Appendix A5. 4.2.1.4.10 Properties and Applications PVC is one of the most versatile of plastics and its usage ranges widely from building construction to toys and footwear. PVC compounds are made in a wide range of formulations, which makes it difﬁcult to make generalizations about their properties. Mechanical properties are considerably affected by the type and amount of plasticizer. Table 4.3 illustrates differences in some properties of three distinct types of compound. To a lesser extent, ﬁllers also affect the physical properties. Unplasticized PVC (UPVC) is a rigid material, whereas the plasticized material is tough, ﬂexible, and even rubbery at high plasticizer loadings. Relatively high plasticizer loadings are necessary to achieve many signiﬁcant improvement in impact strength. Thus incorporation of less than 20% plasticizer does not give compounds with impact strength higher than that of unplasticized grades. Lightly plasticized grades are therefore used when the ease of processing is more important than achieving good impact strength. PVC is resistant to most aqueous solutions, including those of alkalis and dilute mineral acids. The polymer also has a good resistance to hydrocarbons. The only effective solvents appear to be those which are capable of some form of interaction with the polymer. These include cyclohexanone and tetrahydrofuran. At ordinary temperatures, PVC compounds are reasonably good electrical insulators over a wide range of frequencies, but above the glass transition temperature their value as an insulator is limited to lowfrequency applications. The volume resistivity decreases as the amount of plasticizer increases. PVC has the advantage over other thermoplastic polyoleﬁns of built-in ﬁre retardancy because of its 57% chlorine content. Copolymers of vinyl chloride with vinyl acetate have lower softening points, easier processing, and better vacuum-forming characteristics than the homopolymers. They are soluble in ketones, esters, and certain chlorinated hydrocarbons, and have generally inferior long-term heat stability. About 90% of the PVC produced is used in the form of homopolymers, the other 10% as copolymers and terpolymers. The largest application of homopolymers PVC compounds, particularly unplasticized grades, is for rigid pipes and ﬁttings, most commonly as suspension homopolymers of high bulk density compounded as powder blends. TABLE 4.3 Properties of Three Types of PVC Compounds Property Speciﬁc gravity Tensile strength lbf/in.2 Unplasticized PVC PVC+DIOP (50 Parts Per 100 Resin) Vinyl Chloride–Vinyl Acetate Copolymer (Sheet) 1.4 1.31 1.35 8500 2700 7000 MPa Elongation at break (%) 58 5 19 300 48 5 Vicat softening (°C) 80 Flexible at room temperature 70 Source: Brydson, J. A. 1982. Plastics Materials, Butterworth Scientiﬁc, London, UK. 454 Plastics Technology Handbook In addition to the afore said properties, UPVC has an excellent resistance to weathering. Moreover, when the cost of installation is taken into account, the material frequently turns out to be cheaper. UPVC is therefore becoming used increasingly in place of traditional materials. Important uses include translucent roof sheathing with good ﬂame-retarding properties, window frames, and piping that neither corrodes nor rots. As a pipe material PVC is widely used in soil pipes and for drainage and above-ground applications. Piping with diameters of up to 60 cm is not uncommon. UPVC is now being increasingly used as a wood replacement due to its more favorable economics, taking into account both initial cost and installation. Speciﬁc applications include bench-type seating at sports stadia, window ﬁttings, wall-cladding, and fencing. UPVC bottles have better clarity, oil resistance, and barrier properties than those made from polyethylene. Compared with glass they are also lighter, less brittle, and possess greater design ﬂexibility. These products have thus made extensive penetration into the packaging market for fruit juices and beverages, as well as bathroom toiletry. Sacks made entirely of PVC enable fertilizers and other products to be stored outdoors. The largest applications of plasticized PVC are wire and cable insulation and as ﬁlm and sheet. PVC is of great value as an insulator for direct-current and low-frequency alternating-current carriers. It has almost completely replaced rubber in wire insulation. PVC is widely used in cable sheathing where polyethylene is employed as the insulator. Other major outlets of plasticized PVC include ﬂoor coverings, leathercloth, tubes and proﬁles, injection moldings, laminates, and paste processes. When a thin layer of plasticized PVC is laminated to a metal sheet, the bond may be strong enough that the laminate can be punched, cut, or shaped without parting the two layers. A pattern may be printed or embossed on the plastic before such fabrication. Typewriter cases and appliance cabinets have been produced with such materials. PVC leathercloth has been widely used for many years in upholstery and trim in car applications, house furnishings, and personal apparel. The large-scale replacement of leather by PVC initiated in the 1950s and 1960s was primarily due to the greater abrasion resistance, ﬂex resistance, and washability of PVC. Ladies handbags are frequently made from PVC leathercloth. House furnishing applications include kitchen upholstery, printed sheets, and bathroom curtains. Washable wallpapers are obtained by treating paper with PVC compounds. Special grades of PVC are used in metal-ﬁnishing applications, for example, in stacking chairs. Calendered plasticized PVC sheet is used in making plastic rainwear and baby pants by the high-frequency welding technique. The application of PVC in mine belting is still important in terms of the actual tonnage of material consumption. All-PVC shoes are useful as beachwear and standard footwear. PVC has also proved to be an excellent abrasion-resistant material for shoe soles. PVC adhesives, generally containing a polymerizable plasticizer, are useful in many industries. The two main applications of vinyl chloride–vinyl acetate copolymers are phonograph records and vinyl ﬂoor tiles. The copolymers contain an average of about 13% of vinyl acetate. They may be processed at lower temperatures than those used for the homopolymers. Phonograph records contain only a stabilizer, lubricant, pigment, and, possibly, an antistatic agent; there are no ﬁllers. Preformed resin biscuits are normally molded in compression presses at about 130–140°C. The press is a ﬂash mold that resembles a wafﬂe iron. The faces of the mold may be nickel negatives of an original disc recording that have been made by electrodeposition. Floor tiles contain about 30–40 parts plasticizer per 100 parts copolymer and about 400 parts ﬁller (usually a mixture of asbestos and chalk). Processing involves mixing in an internal mixer at about 130°C, followed by calendering at 110–120°C. 4.2.1.4.11 Pastes A PVC paste is obtained when the voids between the polymer particles in a powder are completely ﬁlled with plasticizer so that the particles are suspended in it. To ensure a stable paste, there is an upper limit 455 Industrial Polymers and a lower limit to the order of particle size. PVC paste polymers have an average particle size of about 0.2–1.5 mm. The distribution of particle sizes also has signiﬁcant inﬂuences on the ﬂow and ﬂuxing characteristics of the paste. The main types of PVC pastes are plastisols, organosols, plastisols incorporating ﬁller polymers (including the rigisols), plastigels, hot-melt compounds, and compounds for producing cellular products. Typical formulations of the ﬁrst three types are shown in Table 4.4. The processing methods were described in Chapter 2. Plastisols are of considerable importance commercially. They are converted into tough, rubbery products by heating at about 160°C (gelation). Organosols are characterized by the presence of a volatile organic diluent whose sole function is to reduce the paste viscosity. The diluent is removed after application and before gelling the paste. Another method of reducing paste viscosity is to use a ﬁller polymer to replace a part of the PVC paste polymer. The ﬁller polymer particles are too large to make stable pastes by themselves, but in the presence of pastepolymer particles they remain in stable suspension. Being very much larger than paste-polymer particles and having a low plasticizer absorption, the take up large volumes in the paste and make more plasticizer available for particle lubrication, thus reducing paste viscosity. The use of ﬁller polymers has increased considerably in recent years. Pastes prepared using ﬁller polymers and only small quantities of plasticizer (approximately 20 parts per 100 parts of polymer) are termed rigisols. The incorporation of such materials as aluminum stearate, fumed silicas, or certain bentonites gives a paste that shows pronounced Bingham Body behavior (i.e., it only ﬂows on application of shearing stress above a certain value). Such putty-like materials (called pastigels), which are usually thixotropic may be hand-shaped and subsequently gelled (see ‘Plastisol Casting’ in Chapter 2). Plastigels are often compared with hot-melt PVC compounds. These later materials are prepared by ﬂuxing polymer with large quantities of plasticizers and extenders. They melt at elevated temperatures and become very ﬂuid, so they may be poured. These compounds are extensively used for casting and prototype work. Sigma-blade trough mixers are most commonly used for mixing PVC pastes. It is common practice to mix the dry ingredients initially with part of the plasticizer so that the shearing stresses are high enough to break down the aggregates. The remainder of the plasticizer is then added to dilute the product. The mix is preferably deaerated to remove air bubbles before ﬁnal processing. A large proportion of PVC paste is used in the manufacture of leathercloth by a spreading technique. A layer of paste is smeared on the cloth by drawing the latter between a roller or endless belt and a doctor blade against which there is a rolling bank of paste. The paste is gelled by passing through a heated tunnel or under infrared heaters. Embossing operations may be carried out by using patterned rollers when the gelled paste is still hot. The leathercloth is then cooled and wound up. Where it is desired that the paste should enter the interstices of the cloth, a shear-thinning (pseudo-plastic) paste is employed. Conversely, TABLE 4.4 Typical Formulationsa of Three Types of PVC Pastes Ingredient Plastisol Organosol Plastigel PVC paste polymer 100 100 100 Plasticizer (e.g., DOP) 80 30 80 Filler (e.g., china clay) Stabilizer (e.g., white lead) 10 4 10 4 10 4 Naphtha – 50 – Aluminum stearate – – 4 Source: Brydson, J. A. 1982. Plastics Materials. Butterworth Scientiﬁc, London, UK. a Parts by weight. 456 Plastics Technology Handbook where strike-through should be minimized, a dilatant paste (viscosity increases with shear rate) is employed. Numerous methods exist for producing cellular products (see Chapter 2) from PVC pastes. Closed-cell products can be made if a blowing agent such as azodiisobutyronitrile is incorporated into the paste. The paste is then heated in a mold to cause the blowing agent to decompose and the compound to gel. Since the mold is full, expansion does not take place at this stage. The unexpanded block is removed after thoroughly cooling the mold and is heated in an oven at about 100°C to produce uniform expansion. One method of producing a ﬂexible, substantially open-cell product is to blend the paste with carbon dioxide (either as dry ice or under pressure). The mixture is heated to volatilize the carbon dioxide to produce a foam, which is then gelled at a higher temperature. 4.2.1.4.12 Chlorinated PVC Closely related to PVC, but with distinct properties of its own, is chlorinated poly(vinyl chloride) (CPVC), a polymer produced by postchlorination of PVC. The effect of adding more chlorine to the PVC molecule is to raise the Tg of the base resin to 115–135°C (239–275°F) range and the heat deﬂection temperature under load to around 115°C (239°F). CPVC also has higher tensile strength, higher modulus, and greater resistance to combustion and smoke generation. The compounding process for CPVC is similar to that used in PVC compounding, but is more complex. Processing of CPVC is done by the traditional thermoplastic operations of extrusion, calendaring, and injection molding. However, because of the high temperature of CPVC polymer melt (205– 230°C), extrusion of the resin requires chrome-plated or stainless steel dies. Injection molding of CPVC requires low-compression screws with good exit depth and molds should be stainless steel or chrome- or nickel-plated. Traditional applications of CPVC compounds are hot and cold water distribution piping, ﬁttings, and valves that can handle industrial liquids and chemicals. The increasing popularity of CPVC in its application in hot and cold water pipes in residential units stems from its continuous-use rating of 80°C (176°F) and 100 psi, its approval for potable water by the National Sanitation Foundation (U.S.A.), and its low heat loss along with lack of sweating and scale buildup. The high-heat capability, low combustion ratings, and resistance to grease- and oil-induced cracking have made CPVC a strong contender for applications in automotive interiors. With its combination of excellent properties of PVC and the added ability to perform at elevated temperatures, CPVC is beginning to penetrate markets formerly dominated by metals or the more expensive engineering polymers. 4.2.1.5 Poly(Vinylidene Chloride) Cl CH2 C n Cl Monomer Vinylidene chloride Polymerization Major Uses Free-radical-initiated chain polymerization Film and sheeting for food packaging The polymer may be prepared readily by free-radical mechanisms in bulk, emulsion, and suspension; the latter technique is usually preferred on an industrial scale. Copolymers of vinylidene chloride with vinyl chloride, acrylates, and acrylonitrile are also produced. Since the poly(vinylidene chloride) molecule has an extremely regular structure (and the question of tacticity does not arise), the polymer is capable of crystallization. Because of the resultant close packaging and the presence of heavy chloride atoms the polymer has a high speciﬁc gravity (1.875) and a low 457 Industrial Polymers permeability to vapors and gases. The chlorine present gives a self-extinguishing polymer. Vinylidene chloride–vinyl chloride copolymers are also self-extinguishing and possess very good resistance to a wide range of chemicals, including acids and alkalies. Because of a high degree of crystallization, even in the copolymers, high strengths are attained even though the products have relatively low molecular weights (∼20,000–50,000). Both poly(vinylidene chloride) and copolymers containing vinylidene chloride are used to produce ﬂexible ﬁlms and coatings. Flexible ﬁlms are used extensively for food packaging because of their superior barrier resistance to water and oxygen. The coating resins are used for cellophane, polyethylene, paper, fabric, and container liner applications. Dow’s trade name for a copolymer of vinylidene chloride (87%) and vinyl chloride (13%) is Saran. Biaxially stretched Saran ﬁlm is a useful, though expensive, packaging material possessing exceptional clarity, brilliance, toughness, and impermeability to water and gases. Vinylidene chloride–vinyl chloride copolymers are used in the manufacture of ﬁlaments. The ﬁlaments have high toughness, ﬂexibility, durability, and chemical resistance. They ﬁnd use in car upholstery, deckchair fabrics, decorative radio grilles, doll hair, ﬁlter presses, and other applications. A ﬂame-resisting ﬁber said to be a 50:50 vinylidene chloride–acrylonitrile copolymer is marketed by Courtaulds with the name Teklan. 4.2.1.6 Polytetraﬂuoroethylene and Other Fluoropolymers F F C C F F n Monomer Tetraﬂuoroethylene, hexaﬂuoropropylene, chlorotriﬂuoroethylene, vinyl ﬂuoride, vinylidene ﬂuoride, perﬂuoroalkyl vinyl ether Polymerization Major Uses Free-radical-initiated chain polymerization Coatings for chemical process equipment, cable insulation, electrical components, nonsticking surfaces for cookware Polytetraﬂuoroethylene (PTFE) was discovered in 1947. Today PTFE probably accounts for at least 85% of the ﬂuorinated polymers and, in spite of its high cost, has a great diversity of applications [16–18]. It is produced by the free-radical chain polymerization of tetraﬂuoroethylene. With a linear molecular structure of repeating –CF2–CF2–units, PTFE is a highly crystalline polymer with a melting point of 327°C. Density is 2.13–2.19 g/cm3. Commercially PTFE is made by two major processes—one leading to the so-called granular polymer, the second to a dispersion of polymers of much ﬁner particle size and lower molecular weight. Since the carbon–ﬂuorine bond is very stable and since the only other bond present in PTFE is the stable C–C bond, the polymer has a high stability, even when heated above its melting point. Its upper-use temperature is given as 260°C. It is reported to give ductile rather than brittle failures at temperatures just above absolute zero, signifying a useful temperature range of more than 500°C. In many instances PTFE has been used satisfactorily as a totally enclosed gasket for considerable periods of time at temperatures well above the recommended upper-use temperature. Because of its high crystallinity (>90%) and incapability of speciﬁc interaction, PTFE has exceptional chemical resistance and is insoluble in all organic solvents. (PTFE dissolves in certain ﬂuorinated liquids such as perﬂuorinated kerosenes at temperatures approaching the melting point of the polymer.) At room 458 Plastics Technology Handbook temperature it is attacked only by alkali metals and, in some cases, by ﬂuorine. Treatment with a solution of sodium metal in liquid ammonia sufﬁciently alters the surface of a PTFE sample to enable it to be cemented to other materials by using epoxide resin adhesives. PTFE has good weathering resistance. The polymer is not wetted by water and has negligible water absorption (0.005%). The permeability of gases is low—the rate of transmission of water vapor is approximately half that of poly(ethylene terephthalate) and low-density polyethylene. PTFE is however degraded by high-energy radiation. For example, the tensile strength of a given sample may be halved by exposure to a dosage of 70 Mrad. PTFE is a tough, ﬂexible material of moderate tensile strength (2500–3800 psi, i.e., 17–21 MPa) at 23°C. Temperature has a considerable effect on its properties. It remains ductile in compression at temperatures as low as 4 K (−269°C). The creep resistance is low in comparison to other engineering plastics. Thus, even at 20°C unﬁlled PTFE has a measurable creep with compression loads as low as 300 psi (2.1 MPa). The coefﬁcient of friction is unusually low and is lower than almost any other material. The values reported in the literature are usually in the range 0.02–0.10 for polymer to polymer and 0.09–0.12 for polymer to metal. The polymer has a high oxygen index and will not support combustion. PTFE has outstanding insulation properties over a wide range of temperatures and frequencies. The volume resistivity exceeds 1020 ohm-m. The power factor (<0.003 at 60 Hz and <0.0003 at 106 Hz) is negligible in the temperature range −60°C to +250°C. PTFEs low dielectric constant (2.1) is unaffected by frequency. The dielectric strength of the polymer is 16–20 kV/mm (short time on 2-mm thick sheet). 4.2.1.6.1 Processing PTFE is commonly available in three forms [14] (a) granular polymers with average particle size of 300– 600 mm; (b) dispersion polymers (obtained by coagulation of a dispersion) consisting of agglomerates with an average diameter of 450 mm, which are made up of primary particles 0.1 mm in diameter; and (c) dispersion (lattices) containing about 60% polymer in the form of particles with an average diameter of about 0.16 mm. ASTM Standard Speciﬁcation D1457 covers the forms sold as powders and deﬁnes seven type of PTFE molding and extrusion materials ranging from general-purpose granular resin (Type I) to presintered resin (Type VII). ASTM D4441 describes eight types of aqueous dispersion that differ, primarily, in the solids content and the amount of surfactant. PTFE cannot be processed by the usual thermoplastics processing techniques because of its exceptionally high melt viscosity (∼1010–1011 P at 350°C). Granular polymers are processed by press and sinter methods used in powder metallurgy. In principle, these methods involve performing the sieved powder by compressing in a mold at 1.0–3.5 tonf/in.2 (16–54 MPa) usually at room temperature or at 100°C, followed by sintering at a temperature above the melting point (typically at about 370°C), and then cooling. Free-sintering of the perform in an oven at about 380°C is also satisfactory. The sintering period depends on the thickness of the sample. For example, a 0.5-in. (1.25-cm)-thick sample will need sintering for 3.5 h. Granular polymers may also be extruded, though at very low rates (2.5–16 cm/min), by screw and ram extruders. The extrudates are reasonably free of voids. Billets in sizes varying from less than a kilogram up 700 kg (among the largest moldings made of any plastic material) are made by perform sintering. The powder is placed in a mold at or slightly above room temperature and compressed at pressures from 2000–5000 psi (14–34 MPa). After being removed from the mold, the perform is sintered by heating it unconﬁned in an oven at temperatures in the range of 360– 380°C for times ranging from a few hours to several days. The time-temperature schedule depends on the size and shape of the billet. The billets are often in the form of cylinder, which are then mounted on a mandrel. Sheeting is prepared by skiving, much like plywood is cut from large logs. The sheeting is cut in the thickness range of 0.025 mm (0.001 in.) to 2.5 mm (0.1 in.). With sheet molding, the procedures are very similar to those used as described above for billet molding except for the shape and size of the molding. This process is used for sheeting above 2.5 mm (0.1 in.) up to Industrial Polymers 459 large blocks. The latter are used for a block method skiving operation similar to that used traditionally for cellulose nitrate. With ram extrusion, the PTFE powder is fed into a cavity at one end of a heated tube. A reciprocating ram compacts the powder and forces in into the tube. While it is being transported down the length of the tube, the PTFE is melted and coalesced. Continuous lengths of sintered rod or shapes comes out of the other end of the extruder. PTFE moldings and extrudates may be machined without difﬁculty. Continuous ﬁlm may be obtained by peeling a pressure-sintered ring and welding it to a similar ﬁlm by heat sealing under pressure at about 350°C. A PTFE dispersion polymer leads to products with improved tensile strength and ﬂex life. Preforms are made by mixing the polymer with 15–25% of a lubricant and extruded. This step is followed by lubricant removal and sintering. In a typical process a mixture of PTFE dispersion polymer (83 parts) and petroleum ether (17 parts) with a 100–120°C boiling range is compacted into a perform billet which is then extruded by a vertical ram extruder. The extrudate is heated in an oven at about 105°C to remove the lubricant and is then sintered at about 380°C. Because of the need to remove the lubricant, only thin sections can be produced by this process. Thin-walled tubes with excellent fatigue resistance can be produced, or wire can be coated with very thin coatings of PTFE. Tapes also may be made by a similar process. However, in this case the lubricant used in a nonvolatile oil. The perform is extruded in the shape of a rod, which is then passed between a pair of calendared rolls at about 60–80°C. The unsintered tape ﬁnds an important application in pipe-thread sealing. If sintered tape is required, the calendered product is ﬁrst degreased by passing through boiling trichloroethylene and then sintered by passing through a salt bath. The tape made in this way is superior to that obtained by machining granular polymer molding [14]. A convenient way to apply PTFE dispersions to surfaces is by use of aqueous dispersion coatings. For this a dispersion as provided by the manufacturer may be used. There are also many coating formulations that can be purchased for use in such applications as cookware and bakeware. Coating is done as dip coating and, with thickened dispersions, as roller coating. The dispersion should have a concentration of 45–50% PTFE and 6–9% of a wetting agent, based on the amount of PTFE. Usually a nonionic agent is used, such as Triton X-100 from Rohm and Haas or similar materials from other suppliers. The amount of PTFE deposited on each coat must be restricted so as to prevent mud cracking when the coating is dried. For drying the coating, infrared lamps or forcedconvention ovens at 85–95°C are usually used. Multiple coats are used to obtain thicker ﬁlms. Drying is followed by baking and sintering to remove the wetting agent and then coalesce the PTFE. Temperatures of 260–315°C are used for the baking and 360–400°C for the sintering. The time required for each of these steps varies from several seconds to a few minutes, depending on the shape of the surface and thickness of the coating. To obtain good homogeneity, multiple dips are used, with baking and sintering after each. Casting is used to make thin ﬁlms for use in such applications as heart–lung machines, special electronic equipment and various specialty applications. Often it is done by allowing the dispersion to ﬂow onto a support surface, usually a polished stainless-steel belt, that caries the ﬁlm through the successive steps of coating, drying, baking, sintering, recoating, and ﬁnally, stripping from the belt. Glass-coated PTFE laminates may be produced by piling up layers of glass cloth impregnated with PTFE dispersions and pressing at about 330°C. Asbestos–PTFE laminates may be produced in a similar way. The dispersions can also be used for producing ﬁlled PTFE molding material. The process typically involves stirring ﬁllers into the dispersion, coagulating with acetone, drying at 280–290°C, and disintegrating the resulting cake of material. 4.2.1.6.2 Applications The exceptional properties of PTFE make it highly useful. It is selected for a wide range of applications that affect every person. The applications fall in the ﬁve areas that require one or more of its chemical, 460 Plastics Technology Handbook mechanical, electrical, thermal, and surface properties. In essentially every instance, a successful application employs at least two of the outstanding properties of this polymer. However, because of its high volume cost, PTFE is not generally used to produce large objects. In many cases it is possible to coat a metal object with a layer of PTFE to meet the particular requirement. Nonstick home cookware is perhaps the best-known example. PTFE is used for lining chutes and coating other metal objects where low coefﬁcients of friction, chemical inertness, nonadhesive characteristics are required. The same properties make PTFE useful for coverings on rollers in food processing equipment, xerographic copiers and saw blades, coatings on snow shovels, and many other similar applications. The Alaskan oil pipeline, for example, rests on PTFE-coated steel plates. Most new bridges and tunnels use similar supports. Because of its exceptional chemical resistance over a wide temperature range, PTFE is used in a variety of seals, gaskets, packings, valve and pump parts, and laboratory equipment. Its excellent electrical insulation properties and heat resistance lead to its use in high-temperature wire and cable insulation, molded electrical components, insulated transformers, hermetic seals for condensers, laminates for printed circuitry, and many other electrical applications. Reinforced PTFE applications include bushings and seals in compressor hydraulic applications, automotive applications, and pipe liners. A variety of moldings are used in aircraft and missiles and also in other applications where use at elevated temperatures is required. An important application for aqueous dispersions of PTFE is the architectural fabric market. This product consists of ﬁberglass fabric coated with special forms of the aqueous dispersion. The resulting material is used as roofs in a wide variety of buildings, especially where a large area must be covered with minimum support. Notable examples of such use are the Pontiac “Silver Dome,” the airport terminal in Jeddah, Saudi Arabia (where 105 acres are enclosed), and many college and university stadiums or union buildings. Copolymers of tetraﬂuoroethylene were developed in attempts to provide materials with the general properties of PTFE and the melt process-ability of the more conventional thermoplastics. Two such copolymers are tetraﬂuoroethylene–hexaﬂuoropropylene (TFE–HFP) copolymers (Teﬂon FEP resins by DuPont; FEP stands for ﬂuorinated ethylene propylene) with a melting point of 290°C and tetraﬂuoroethylene–ethylene (ETFE) copolymers (Tefzel by DuPont) with a melting point of 270°C. These products are melt processable. A number of other ﬂuorine containing melt processable polymers have been introduced. Polychlorotriﬂuoroethylene (PCTFE) was the ﬁrst ﬂuorinated polymer to be produced on an experimental scale and was used in the United States early in World War II. It was also used in the handling of corrosive materials, such as uranium hexaﬂuoride, during the development of the atomic bomb. PCTFE is a crystalline polymer with a melting point of 218°C and density of 2.13 g/cm3 The polymer is inert to most reactive chemicals at room temperature. However, above 100°C a few solvents dissolve the polymer, and a few, especially chlorinated types, swell it. The polymer is melt processable, but processing is difﬁcult because of its high melt viscosity and its tendency to degrade, resulting in deterioration of its properties. PCTFE is marketed by Hoechst as Hostaﬂon C2 and in the United States by Minnesota Mining and Manufacturing (3M) as Kel-F and by Allied Chemical as Halon. The ﬁlm is sold by Allied Corp. as Aclar. PCTFE is used in chemical processing equipment and cryogenic and electrical applications. Major applications include wafer boats, gaskets, O-rings, seals, and electrical components. PCTFE has outstanding barrier properties to gases, the PCTFE ﬁlm has the lowest water-vapor transmission of any transparent plastic ﬁlm. It is used in pharmaceutical packaging and other applications for its vapor barrier properties, including electroluminescent lamps. Other melt-processable ﬂuoroplastics include ethylene–chlorotriﬂuoroethylene (ECTFE) copolymer (melting point 240°C), polyvinylidene ﬂuoride (PVDF) (melting point 170°C), and polyvinyl ﬂuoride (PVF), which is commercially available only as ﬁlm. 461 Industrial Polymers PVF is tough and ﬂexible, has good abrasion and staining resistance, and has outstanding weathering resistance. It maintains useful properties over a temperature range of −70°C to 110°C. PVF can be laminated to plywood, hardboard, vinyl, reinforced polyesters, metal foils, and galvanized steel. These laminates are used in aircraft interior panels, lighting panels, wall coverings, and a variety of building applications. PVF is also used as glazing in solar energy collectors. PVF ﬁlm is marketed by DuPont as Tedlar. PVDF is correctly named poly(1,1-diﬂuoroethylene) and represented by (–CF2CH2–)n. It is a hard, tough thermoplastic ﬂuoropolymer. PVDF is prepared by free-radical initiated polymerization, either in suspension or (usually) in emulsion systems. The basic raw material for PVDF is vinylidene ﬂuoride (CH2═CF2), a preferred synthesis of which is dehydrochlorination of chlorodiﬂuoroethane. PVDF has the lowest melting point of any of the commercial ﬂuoropolymers. As a result, its upper use temperature is limited to about 150°C, compared to values of 200°C for FEP and 260°C for PTFE. At temperatures in its useful range, however, PVDF maintains its stiffness and toughness very well. PVDF exhibits the excellent resistance to harsh environments, characteristic of ﬂuoropolymers. It is widely used in the chemical processing industry, in piping systems, vales, tanks (both molded and lined), and other areas where its combination of excellent mechanical properties and superb resistance to most chemicals make it an ideal material for ﬂuid handling equipment. Increasingly important is the use of PVDF as the base resin for long-life, exterior coatings on aluminum, steel, masonry, wood, and plastics. The high dielectric loss and high dielectric constant of PVDF (8–9) both restrict its use in some electrical applications and provide superior performance in others. PVDF has very unusual piezoelectric and pyroelectric properties which are opening up many new applications with a very high value in use (see Chapter 5). PVDFs cost, about the lowest of the melt-processible ﬂuoropolymers, is an important advantage. Essentially all the common procedures available for thermoplastic polymers can be used with PVDF. Pennwalt Corp. is the leading producer and markets a full line of PVDF resins under the trade name of Kynar. Perﬂuoroalkoxy (PFA) resins represent another class of commercially available class of meltprocessable ﬂuoroplastics. Their general chemical structure is CF2 CF2 CF CF2 CF2 O Rf where Rf = –CnF2n+1. PFA resin has somewhat better mechanical properties than FEP above 150°C and can be used up to 260°C. In chemical resistance it is about equal to PTFE. PFA resin is sold by DuPont under the Teﬂon trademark. 4.2.1.7 Polyisobutylene CH3 C CH2 CH3 n Monomer Polymerization Major Uses Isobutylene Cationic-initiated chain polymerization Lubricating oils, sealants 462 Plastics Technology Handbook High-molecular-weight polyisobutylene (PIB) is produced by cationic chain polymerization in methyl chloride solution at −70°C using aluminum chloride as the catalyst. Such polymers are currently available from Esso (Vistanex) and BASF (Oppanol). PIB ﬁnds variety of uses. It is used as a motor oil additive to improve viscosity characteristics, as a blending agent for polyethylene to improve its impact strength and environmental stress-cracking resistance, as a base for chewing gum, as a tackiﬁer for greases; it is also used in caulking compounds and tank linings. Because of its high cold ﬂow, it has little use as a rubber in itself, but copolymers containing about 2% isoprene to in troduce unsaturation for cross-linking are widely used (butyl rubber; see later). 4.2.1.8 Polystyrene CH2 CH n Monomer Styrene Polymerization Major Uses Free-radical-initiated chain polymerization Packaging and containers (35%), housewares, toys and recreational equipment (25%), appliance parts (10%), disposable food containers (10%) Polystyrene is made by bulk or suspension polymerization of styrene. Polystyrene is very low cost and is extensively used where price alone dictates. Its major characteristics [19,20] include rigidity, transparency, high refractive index, no taste, odor, or toxicity, good electrical insulation characteristics, low water absorption, and ease of coloring and processing. Polystyrene has excellent organic acid, alkali, salts, and lower alcohol resistance. It is, however, attacked by hydrocarbons, esters, ketones, and essential oils. A more serious limitation of polystyrene in many applications is its brittleness. This limitation led to the development of rubber modiﬁed polystyrenes (containing usually 5–15% rubber), the so-called high impact polystyrenes (HIPS). The most commonly used are styrene–butadiene rubber and cis-1,4polybutadiene. The method of mixing the polystyrene and rubber has a profound effect on the properties of the product. Thus, much better results are obtained if the material is prepared by polymerization of styrene in the presence of t he rubber rather than by simply blending the two polymers. The product of the former method contains not only polystyrene and straight rubber but also a graft copolymer in which polystyrene side chains are attached to the rubber. Compared to straight or general-purpose polystyrenes, high-impact polystyrene materials have much greater toughness and impact strength, but clarity, softening point, and tensile strength are not as good. Expanded or foamed polystyrene (see Chapter 2), which has become very important as a thermal insulating material, has a low density, has a low weight cost, is less brittle, and can be made ﬁre retarding. End uses for all types of polystyrene are packaging, toys, and recreational products, housewares, bottles, lenses, novelties, electronic appliances, capacitor dielectrics, low-cost insulators, musical instrument reeds, light-duty industrial components, furniture, refrigeration, and building and construction uses (insulation). Packaging is by far the largest outlet: bottle caps, small jars and other injection-molded containers, blow-molded containers, toughened polystyrene liners (vacuum formed) for boxed goods, and oriented polystyrene ﬁlm for foodstuffs are some of its uses. The second important outlet is refrigeration equipment, including door liners and inner liners (made from toughened polystyrene sheet), molding for refrigerator furnishings, such as ﬂip lids and trays, and expanded polystyrene for thermal insulation. Expanded polystyrene products have widely increased the market for polystyrene resin (see the section on polystyrene foams in Chapter 2). With as light a weight as 2 lb/ft3 (0.032 g/cm3), the thermal 463 Industrial Polymers conductivity of expanded polystyrene is very low, and its cushioning value is high. It is an ideal insulation and packaging material. Common applications include ice buckets, water coolers, wall panels, and general thermal insulation applications. Packaging uses of expanded polystyrene range from thermoformed egg boxes and individually designed shipping packages for delicate equipment, such as cameras and electronic equipment, to individual beads (which may be about 1 cm in diameter and up to 5 cm long) for use as a loose ﬁll material in packages. 4.2.1.9 Polybutadiene (Butadiene Rubber) CH2 CH CH CH2 n Monomer Polymerization Major Uses Butadiene Ziegler–Natta-catalyzed chain polymerization Tires and tire products (90%) Polybutadiene is made by solution polymerization of butadiene using Ziegler–Natta catalysts. Slight changes in catalyst composition produce drastic changes in the stereoregularity of the polymer. For example, polymers containing 97–98% of trans-1,4 structure can be produced by using Et3Al/VCl3 catalyst, those with 93–94% cis-1,4 structure by using Et2AlCl/CoCl2, and those with 90% 1,2-polybutadiene by using Et3Al/Ti(OBu)4. The stereochemical composition of polybutadiene is important if the product is to be used as a base polymer for further grafting. For example, a polybutadiene with 60% trans-1,4, 20% cis-1,4, and 20% 1,2 conﬁguration is used in the manufacture of ABS resin. Polybutadiene rubbers generally have a higher resilience than natural rubbers at room temperature, which is important in rubber applications. On the other hand, these rubbers have poor tear resistance, poor tack, and poor tensile strength. For this reason polybutadiene rubbers are usually used in conjunction with other materials for optimum combination of properties. For example, they are blended with natural rubber in the manufacture of truck tires and with styrene–butadiene rubber (SBR) in the manufacture of automobile tires. Polybutadiene is also produced in low volume as specialty products. These include low-molecularweight, liquid 1,2-polybutadienes (60–80%, 1,2 content) used as potting compounds for transformers and submersible electric motors and pumps, liquid trans-1,4-polybutadienes used in protective coatings inside metal cans, and hydroxy-terminated polybutadiene liquid resins for use as a binder and in polyurethane and epoxy resin formulations. 4.2.1.10 Polyisoprene CH2 C CH3 CH CH2 n Monomer Polymerization Major Uses Isoprene Ziegler–Natta-catalyzed chain polymerization Car tires (55%), mechanical goods, sporting goods, footwear, sealants, and caulking compounds Polyisoprene is produced by solution polymerization using Ziegler–Natta catalysts. The cis-1,4polyisoprene is a synthetic equivalent of natural rubber. However, the synthetic polyisoprenes have cis contents of only about 92–96%; consequently, these rubbers differ from natural rubber in several ways. The raw synthetic polyisoprene is softer than raw natural rubber (due to a reduced tendency for a stressinduced crystallization because of the lower cis content) and is therefore more difﬁcult to mill. On the 464 Plastics Technology Handbook other hand, the unvulcanized synthetic material ﬂows more readily; this feature makes it easier to injection mold. The synthetic product is somewhat more expensive than natural rubber. Polyisoprene rubbers are used in the construction of automobile tire carcasses and inner liners and truck and bus tire treads. Other important applications are in mechanical goods, sporting goods, footwear, sealants, and caulking compounds. 4.2.1.11 Polychloroprene CH2 C CH CH2 Cl Monomer n Polymerization Chloroprene (2-chlorobuta1,3-diene) Major Uses Free-radical-initiated chain polymerization (mostlyemulsion polymerization) Conveyor belts, hose, seals and gaskets, wire and cable sheathing The polychloroprenes were ﬁrst marketed by DuPont in 1931. Today these materials are among the leading special-purpose or non-tire rubbers and are well known under such commercial names as Neoprene (DuPont), Baypren (Bayer), and Butachlor (Distagul). A comparison of polychloroprene and natural rubber or polyisoprene molecular structures shows close similarities. However, while the methyl groups activates the double bond in the polyisoprene molecule, the chlorine atom has the opposite effect in polychloroprene. Thus polychloroprene is less prone to oxygen and ozone attack than natural rubber is. At the same time accelerated sulfur vulcanization is also not a feasible proposition, and alternative vulcanization or curing systems are necessary. Vulcanization of polychloroprene rubbers is achieved with a combination of zinc and magnesium oxide and added accelerators and antioxidants. The vulcanizates are broadly similar to those of natural rubber in physical strength and elasticity. However, the polychloroprene vulcanizates show much better heat resistance and have a high order of oil and solvent resistance (though less resistant than those of nitrile rubber). Aliphatic solvents have little effect, although aromatic and chlorinated solvents cause some swelling. Because of chlorine chloroprene rubber is generally self-extinguishing. Because of their greater overall durability, chloroprene rubbers are used chieﬂy where a combination of deteriorating effects exists. Products commonly made of chloroprene rubber include conveyor belts, V-belts, diaphragms, hoses, seals, gaskets, and weather strips. Some important construction uses are highway joint seals, pipe gaskets, and bridge mounts and expansion joints. Latexes are used in gloves, balloons, foams, adhesives, and corrosion-resistant coatings. 4.2.2 Oleﬁn Copolymers 4.2.2.1 Styrene–Butadiene Rubber CH2 CH CH CH2 CH2 CH n Monomer Styrene, butadiene Polymerization Major Uses Free-radical-initiated chain polymerization (mostly emulsion polymerization) Tires and tread (65%), mechanical goods (15%), latex (10%), automotive mechanical good (5%) 465 Industrial Polymers In tonnage terms SBR is the worlds most important rubber [21,22]. Its market dominance is primarily due to three factors: low cost, good abrasion resistance, and a higher level of product uniformity than can be achieved with natural rubber. Reinforcement of SBR with carbon black leads to vulcanizates which resemble those of natural rubber, and the two products are interchangeable in most applications. As with natural rubber, accelerated sulfur systems consisting of sulfur and an activator comprising a metal oxide (usually zinc oxide) and a fatty acid (commonly stearic acid) are used. A conventional curing system for SBR consists of 2.0 parts sulfur, 5.0 parts zinc oxide, 2.0 parts stearic acid, and 1.0 part N-t-butylbenzothiazole-2-sulfenide (TBBS) per 100 parts polymers. The most important application of SBR is in car tires and tire products, but there is also widespread use of the rubber in mechanical and industrial goods. SBR latexes, which are emulsions of styrene–butadiene copolymers (containing about 23–25% styrene), are used for the manufacture of foam rubber backing for carpets and for adhesive and molded foam applications. 4.2.2.2 Nitrile Rubber CH2 CH2 CH CH CH CH2 CN n Monomer Acrylonitrile, butadiene Polymerization Major Uses Free-radical-initiated chain polymerization (mostly emulsion polymerization) Gasoline hose, seals, gaskets, printing tools, adhesive, footwear Nitrile rubber (acrylonitrile–butadiene copolymer) is a unique elastomer. The acrylonitrile content of the commercial elastomers ranges from 25% to 50% with 34% being a typical value. This nonhydrocarbon monomer imparts to the copolymer very good hydrocarbon oil and gasoline resistance. The oil resistance increases with increasing amounts of acrylonitrile in the copolymer. Nitrile rubber is also noted for its high strength and excellent resistance to abrasion, water, alcohols, and heat. Its drawbacks are poor dielectric properties and poor resistance to ozone. Because of the diene component, nitrile rubbers can be vulcanized with sulfur. A conventional curing system consists of 2.5 parts sulfur, 5.0 parts zinc oxide, 2.0 parts stearic acid, and 0.6 parts N-tbutylbenzothiazole-2-sulfenamide (TBBS) per 100 parts polymer. Nitrile rubbers (vulcanized) are used almost invariably because of their resistance to hydrocarbon oil and gasoline. They are, however, swollen by aromatic hydrocarbons and polar solvents such as chlorinated hydrocarbons, esters, and ketones. 4.2.2.3 Ethylene–Propylene Elastomer CH2 CH2 CH2 CH CH3 n Monomer Ethylene, propylene Polymerization Major Uses Ziegler–Natta-catalyzed chain polymerization Automotive parts, radiator and heater hoses, seals 466 Plastics Technology Handbook Two types of ethylene–propylene elastomers are currently being produced; ethylene–propylene binary copolymers (EPM rubbers) and ethylene–propylene-diene ternary copolymers (EPDM rubbers). Because of their saturated structure, EPM rubbers cannot be vulcanized by using accelerated sulfur systems, and the less convenient vulcanization with free-radical generators (peroxide) is required. In contrast, EPDM rubbers are produced by polymerizing ethylene and propylene with a small amount (3–8%) of a diene monomer, which provides a cross-link site for accelerated vulcanization with sulfur. A typical vulcanization system for EPDM rubber consists of 1.5 parts sulfur, 5.0 parts zinc oxide, 1.0 part stearic acid, 1.5 parts 2-mercaptobenzothiazole (MBT), and 0.5 part tetramethylthiuram disulﬁde (TMTD) per 100 parts polymer. The EPDM rubbers, though hydrocarbon, differ signiﬁcantly from the diene hydrocarbon rubbers considered earlier in that the level of unsaturation in the former is much lower, giving rubbers much better heat, oxygen, and ozone resistances. Dienes commonly used in EPDM rubbers include dicyclopentadiene, ethylidene norbornene, and hexa-1,4-diene (Table 4.5). Therefore the double bonds in the polymer are either on a side chain or as part of a ring in the main chain. Hence, should the double bond become broken, the main chain will remain substantially intact, which also accounts for the greater stability of the product. The use of EPDM rubbers for the manufacture of automobile and truck tires has not been successful, mainly because of poor tire cord adhesion and poor compatibility with most other rubbers. However, EPDM rubbers have become widely accepted as a moderately heat-resisting material with good weathering, oxygen, and ozone resistance. They ﬁnd extensive use in nontire automobile applications, including body and chassis parts, car bumpers, radiator and heater hoses, weatherstrips, seals, and mats. Other applications include wire and cable insulation, appliance parts, hoses, gaskets and seals, and coated fabrics. These rubbers are now also being blended on a large scale with polyoleﬁn plastics, particularly polypropylene, to produce an array of materials ranging from tough plastics at one end to the thermoplastic polyoleﬁn rubbers (see later) at the other. TABLE 4.5 Principal Diene Monomers Used in EPDM Manufacture Monomer Predominant Structure Present in Terpolymer Dicyclopentadiene CH CH CH2 CH2 4-Ethylidenenorborn-2-ene CH CH CH2 CH3 CH CH2 CH Hexa-1,4-diene CH2═CH–CH2–CH═CH–CH3 CH2 CH3 CH CH2 CH CH CH3 467 Industrial Polymers 4.2.2.4 Butyl Rubber CH3 CH2 C CH2 CH CH3 C CH2 CH3 n Monomer Isobutylene, isoprene Polymerization Major Uses Cationic chain polymerization of isobutylene with 0.5– 2.5 mol% of isoprene Tire inner tubes and inner liners of tubeless tires (70%), inﬂatable sporting goods Although polyisobutylene described earlier is a nonrubbery polymer exhibiting high cold ﬂow, the copolymer containing about 2% isoprene can be vulcanized with a powerful accelerated sulfur system to give rubbery polymers. Being almost saturated, they are broadly similar to the EPDM rubbers in many properties. The most outstanding property of butyl rubber is its very low air permeability, which has led to its extensive use in tire inner tubes and liners. A major disadvantage is its lack of compatibility with SBR, polybutadiene, and natural rubber. An ozoneresistant copolymer of isobutylene and cyclopentadiene has also been marketed. 4.2.2.5 Thermoplastic Elastomers Monomer Butadiene, isoprene, styrene Polymerization Major Uses Anionic block polymerization Footwear, automotive parts, hot-melt adhesives Conventional rubbers are vulcanized, that is, cross-linked by primary valence bonding. For this reason vulcanized rubbers cannot dissolve or melt unless the network structure is irreversibly destroyed. These products cannot therefore be reprocessed like thermoplastics. Hence, if a polymer could be developed which showed rubbery properties at normal service temperatures but could be reprocessed like thermoplastics, it would be of great interest. During the past few decades several groups of materials have been developed that could be considered as being in this category. Designated as thermoplastic elastomers, they include (1) styrene–diene–styrene triblock copolymers; (2) thermoplastic polyester elastomers and thermoplastic polyurethane elastomers; and (3) thermoplastic polyoleﬁn rubbers (polyoleﬁn blends). 4.2.2.5.1 Styrene–Diene–Styrene Triblock Elastomers The styrene–diene–styrene triblocks consist of a block of diene units joined at each end to a block of styrene units and are made by sequential anionic polymerization of styrene and a diene. In this way two important triblock copolymers have been produced—the styrene–butadiene–styrene (SBS) and styrene– isoprene–styrene (SIS) materials, developed by Shell. The commercial thermoplastic rubbers, Clarifex TR (Shell), are produced by joining styrene–butadiene or styrene–isoprene diblocks at the active ends by using a difunctional coupling agent. Similarly, copolymer molecules in the shape of a T, X, or star have been produced (e.g., Solprene by Phillips) by using coupling agents of higher functionality. 468 Plastics Technology Handbook Polybutadiene chain segments Polystyrene domains FIGURE 4.7 Schematic representation of polystyrene domain structure in styrene–butadiene–styrene triblock copolymers. (After Kaminsky, W. and Steiger, R. 1988. Polyhedra, 7(22), 2375.) The outstanding behavior of these rubbers arises from the natural tendency of two polymer species to separate. However, this separation is restrained in these polymers since the blocks are covalently linked to each other. In a typical commercial SBS triblock copolymer with about 30% styrene content, the styrene blocks congregate into rigid, glassy domains which act effectively to link the butadiene segments into a network (Figure 4.7) analogous to that of cross-linked rubber. As the SBS elastomers is heated above the glass transition temperature (Tg) of the polystyrene, the glass domains disappear and the polymer begins to ﬂow like a thermoplastic. However, when the molten material is cooled (below Tg), the domains reharden and the material once again exhibits properties similar to those of a cross-linked rubber. Below Tg of polystyrene the glassy domains also fulﬁll another useful role by acting like a reinforcing particulate ﬁller. It is also an apparent consequence of this role that SBS polymers behave like carbonblack-reinforced elastomers with respect to tensile strength. The styrene–diene–styrene triblock copolymers are not used extensively in traditional rubber applications because they show a high level of creep. The block copolymers can, however, be blended with many conventional thermoplastics such as polystyrene, polyethylene, and polypropylene, to obtain improved properties. A major area of use is in footwear, where blends of SBS and polystyrene have been used with remarkable success for crepe soles. Other important uses are adhesives and coatings. A wide variety of resins, plasticizers, ﬁllers, and other ingredients commonly used in adhesives and coatings can be used with styrene–diene–styrene triblock copolymers. With these ingredients properties such as tack, stiffness, softening temperatures, and cohesive strength can be varied over a wide range. With aliphatic resin additives the block copolymers are used for permanently tacky pressure-sensitive adhesives, and in conjunction with aromatic resins they are used for contact adhesives. The copolymers can be compounded into these adhesives by solution or hot-melt techniques. The block copolymers are also used in a wide variety of sealants, including construction, industrial, and consumer-grade products. They are unique in that they can be formulated to produce a clear, water white product. Other applications include bookbinding and product assembly and chemical milling coatings. 4.2.2.5.2 Thermoplastic Polyester Elastomers Because of the relatively low Tg of the short polystyrene blocks, the styrene–diene–styrene triblock elastomers have very limited heat resistance. One way to overcome this problem is to use a block Industrial Polymers 469 copolymer in which one of the blocks is capable of crystallization and has a melting temperature well above room temperature. This approach coupled with polyester technology has led to the development of thermoplastic polyester elastomers (Hytrel by DuPont and Arnitel by Akzo). A typical such polymer consists of relatively long sequences of tetra-methylene terephthalate (which segregate into rigid domains of high melting point) and softer segments of polyether (see Section 4.8 for more details). Being polar polymers, these rubbers have good oil and gasoline resistance. They have a wider service temperature range than many general-purpose rubbers, and they also exhibit a high resilience, good ﬂex fatigue resistance, and mechanical abuse resistance. These rubbers have therefore become widely accepted in such applications as seals, belting, water hose, etc. 4.2.2.5.3 Thermoplastic Polyurethane Elastomers Closely related to the polyether–ester thermoplastic elastomers are thermoplastic polyurethane elastomers, which consist of polyurethane or urethane terminated polyurea hard blocks, with Tg above normal ambient temperature, separated by soft blocks of polyol, which in the mass are rubbery in nature (see Section 4.11 for more details). The main uses of thermoplastic rubbers (e.g., Estane by Goodrich) are for seals, bushes, convoluted bellows, and bearings. One particular form of thermoplastic polyurethane elastomer is the elastic ﬁber known as Spandex. Several commercial materials of this type have been introduced, which include Lycra (DuPont), Dorlastan (Bayer) Spanzelle (Courtaulds), and Vyrene (U.S. Rubber). Spandex ﬁbers have higher modulus, tensile strength, and resistance to oxidation, and are able to produce ﬁner deniers than natural rubber. They have enabled lighter-weight garments to be produced. Staple ﬁber blends of Spandex ﬁber with non-elastic ﬁbers have also been introduced. 4.2.2.5.4 Thermoplastic Polyoleﬁn Elastomers Blends of EPDM rubbers with polypropylene in suitable ratios have been marketed as thermoplastic polyoleﬁn rubbers. Their recoverable high elasticity is believed to be due to short propylene blocks in the EPDM rubber co-crystallizing with segments of the polypropylene molecules so that these crystalline domains act like cross-linking agents. Having good weathering properties, negligible toxicity hazards, and easy processability, these rubbers have received rapid acceptance for use in a large variety of nontire automotive applications such as bumper covers, headlight frames, radiator grilles, door gaskets, and other auto parts. They have also found use in cable insulation. FIGURE 4.8 Schematic representation of domain structure in ionic elastomers. 470 Plastics Technology Handbook 4.2.2.5.5 Ionic Elastomers Ionic elastomers have been obtained using sulfonated EPDM. In one case an EPDM terpolymer consisting of 55% ethylene units, 40% propylene units, and 5% ethylidene norbornene units is sulfonated to introduce about 1 mol sulfonate groups (appended to some of the unsaturated groups of the EPDM). The sulfonic acid group is then neutralied with zinc acetate to form the zinc salt. The ionized sulfonic groups create inoic cross-links in the intermolecular structure (Figure 4.8), giving properties normally associated with a cross-linked elastomer. However, being a thermoplastic material, it can be processed in conventional molding machines. This rubber, however, has a very high melt viscosity, which must be reduced by using a polar ﬂow promoter, such as zinc stearate, at levels of 9.5–19%. 4.2.2.6 Fluoroelastomers Monomers Vinylidene ﬂuoride (CH2═CF2), chlorotriﬂuoroethylene (CF2═CFCl), tetraﬂuoroethylene (CF2═CF2), hexaﬂuoropropylene (CF3CF═CF2), perﬂuoromethyl vinyl ether (CF2CFOCF3) Polymerization Major Uses Free-radical-initiated chain polymerization Aerospace industry (20%), industrial equipment, wire and cable jacketing, and other insulation applications The ﬂuoroelastomers are a general family of ﬂuorinated oleﬁn copolymers. To be rubbery, the copolymer must have a ﬂexible backbone and be sufﬁciently irregular in structure to be noncrystalline. A number of important ﬂuororubbers are based on vinylidene ﬂuoride (CH2═CF2). Several common products are listed in Table 4.6. The most important of the above products are the copolymers of vinylidene ﬂuoride and hexaﬂuoropropylene (VF2–HFP), as typiﬁed by the DuPont product Viton A. The terpolymer of these two monmers together with tetraﬂuoroethylene (VF2–HFP–TFE) is also of importance (e.g., DuPont product Viton B). This terpolymer is the best among oil-resistant rubbers in its resistance to heat aging, TABLE 4.6 Commercial Fluoroelastomers Composition Trade Name (Manufacturer) Vinylidene ﬂuoride–hexaﬂuoropropylene copolymer Viton A (DuPont) Vinylidene ﬂuoride–hexaﬂuoropropylene-tetraﬂuoroethylene terpolymer Viton B (DuPont) Vinylidene ﬂuoride–chlorotriﬂuoroethylene copolymer Kel-F 3700 (MMM) Remarks 60–85% VF. Largest tonnage production among ﬂuororubbers Fluorel (MMM) Superior resistance to heat, chemical, and solvent Daiel G-501 (Daikin Kogyo) Superior resistance to oxidizing acids Kel-F 5500 (MMM) Vinylidene ﬂuoride-1-hydropentaﬂuoropropylene– tetraﬂuoroethylene terpolymer Tecnoﬂon T (Montecatini) Superior resistance to oil, chemical, and solvent Tetraﬂuoroethylene–perﬂuoro(methyl vinyl ether)+cure site monomer terpolymer Kalrez (DuPont) Excellent air oxidation resistance to 315°C Tetraﬂuoroethylene–propylene cure site monomera terpolymer Aﬂas (Asahi Glass) Cross-linked by peroxides. Resistant to inorganic acids and bases. Cheaper alternative to Kalrez a Suggested as triallyl cyanurate. 471 Industrial Polymers TABLE 4.7 Commercial Elastomer Products Type Properties Major Uses Natural rubber Excellent properties of vulcanizates under conditions not demanding high levels of heat, oil, and chemical resistance Tires, bushings, couplings, seals, footwear and belting; second place in global tonnage Styrene–butadiene (SBR) Reinforcement with carbon black leads to vulcanizates which resemble those of natural rubber; more effectively stabilized by antioxidants than natural rubber Tires, tire products, footwear, wire and cable covering, adhesives; highest global tonnage Polybutadiene Higher resilience than similar natural rubber compounds, good low-temperature behavior and adhesion to metals, but poor tear resistance, poor tack, and poor tensile strength Blends with natural rubber and SBR; manufacture of high-impact polystyrene Polyisoprene Similar to natural rubber, but excellent ﬂow characteristics during molding Tires, belting, footwear, ﬂooring Butyl Outstanding air-retention property, but low Tire inner tubes, inner liners, seals, coated resiliency and poor resistance to oils and fuels fabrics Ethylene–propylene Outstanding resistance to oxygen and ozone, poor fatigue resistance, poor tire-cord adhesion Excellent resistance to oils and solvents, poor low-temperature ﬂexibility and poor resistance to weathering Nitrile Nontire automotive parts, radiator and heater hoses, wire and cable insulation Hoses, seals, gaskets, footwear Chloroprene High order of oil and solvent resistance (but less than nitrile rubber), good resistance to most chemicals, oxygen and ozone, good heat resistance, high strength but difﬁcult to process Mechanical automotive goods, conveyor belts, disphragms, hose, seals and gaskets Silicone Outstanding electrical and high-temperature properties, retention of elasticity at low temperature, poor tear and abrasion resistance, relatively high price Polyurethane High tensile strength, tear, abrasion and oil resistance, relatively high price Gaskets and sealing rings for jet engines, ducting, sealing strips, vibration dampers and insulation equipment in aircraft, cable insulation in naval craft, potting and encepsulation Oil seals, shoe soles and heels, forklift truck tires, disaphragms, ﬁber coatings resistant to dry cleaning, variety of mechanical goods Chlorosulfonated polyethylene Very good heat, ozone, and weathering resistance, good resistance to oil and a wide range of chemicals, high elasticity, good abrasion resistance Wire and cable coating, chemical plant hose, fabric coating, ﬁlm sheeting, footwear, pond liners Polysulﬁde Excellent oil, solvent, and water resistance, high impermeability to gases, low strength, unpleasant odor (particularly during processing) Adhesive, sealants, binders, hose Epichlorohydrin Low air permeability, low resilience, excellent ozone resistance, good heat resistance, ﬂame resistance, and weathering resistance Outstanding heat resistance, superior oil, chemical, and solvent resistance, highestpriced elastomer Seals, gaskets, wire and cable coating Fluoroelastomers Aerospace applications, high quality seals, and gaskets 472 Plastics Technology Handbook although its actual strengths are lower than for some other rubbers. The copolymers of vinylidene ﬂuoride and chlorotriﬂuoroethylene (VF2–CTFE) are notable for their superior resistance to oxidizing acids such as fuming nitric acid. Fluoroelastomers with no C–H groups will be expected to exhibit a higher thermal stability. DuPont thus developed a terpolymer of tetraﬂuoroethylene, perﬂuoro(methyl vinyl ether) and, in small amounts, a cure site monomer of undisclosed composition. This product, marketed as Kalrez, has excellent airoxidation resistance up to 315°C and exhibits extremely low swelling in a wide range of solvents, which is unmatched by any other commercial ﬂuroelastomer. Table 4.7 lists a number of commercial elastomers with their main properties and applications. 4.2.2.7 Styrene–Acrylonitrile Copolymer CH2 CH CH2 CH CN n Monomer Acrylonitrile, styrene Polymerization Major Uses Free-radical-initiated chain polymerization Components of domestic appliances, electrical equipment and car equipment, picnic ware, housewares Because of the polar nature of the acrylonitrile molecule, styrene–acrylonitrile (SAN) copolymers have better resistance to hydrocarbons, oils, and greases than polystyrene. These copolymers have a higher softening point, a much better resistance to stress cracking and crazing, and a higher impact strength than the homopolymer polystyrene, yet they retain the transparency of the latter. The toughness and chemical resistance of the copolymer increases with the acrylonitrile content but so do the difﬁculty in molding and the yellowness of the resin. Commercially available SAN copolymers have 20–30% acrylonitrile content. They are produced by emulsion, suspension, or continuous polymerization. Due to their rigidity, transparency, and thermal stability, SAN resins have found applications for dials, knobs, and covers for domestic appliances, electrical equipment, car equipment, dishwasher-safe housewares, such as refrigerator meat and vegetable drawers, blender bowls, vacuum cleaner parts, humidiﬁer parts, plus other industrial and domestic applications with requirements more stringent than can be met by polystyrene. SAN resins are also reinforced with glass to make dashboard components and battery cases. Over 35% of the total SAN production is used in the manufacture of ABS blends. 4.2.2.8 Acrylonitrile–Butadiene–Styrene Terpolymer CH2 CH CH2 CH CH CH2 CH2 CH CN n Monomers Acrylonitrile, butadiene styrene Polymerization Major Uses Free-radical-initiated chain polymerization Pipe and ﬁttings (30%), automotive and appliance (15%), telephones and business machine housings Industrial Polymers 473 A range of materials popularly referred to as ABS polymers ﬁrst became available in the early 1950s. They are formed basically from three different monomers: acrylonitrile, butadiene, and styrene, Acrylonitrile contributes chemical resistance, heat resistance, and high strength; butadiene contributes toughness, impact strength, and low-temperature property retention; styrene contributes rigidity, surface appearance (gloss), and processability. Not only may the ratios of the monomers be varied, but the way in which they can be assembled into the ﬁnal polymer can also be the subject of considerable variations. The range of possible ABS-type polymers is therefore very large [23]. The two most important ways of producing ABS polymers are (1) blends of styrene–acrylonitrile copolymers with butadiene–acrylonitrile rubber, and (2) interpolymers of polybutadiene with styrene and acrylonitrile, which is now the most important type. A typical blend would consist of 70 parts styrene– acrylonitrile (70:30) copolymer and 40 parts butadiene–acrylonitrile (65:35) rubber. Interpolymers are produced by copolymerizing styrene and acrylonitrile in the presence of polybutadiene rubber (latex) by using batch or continuous emulsion polymerization. The resultant materials are a mixture of polybutadiene, SAN copolymer, and polybutadiene grafted with styrene and acrylonitrile. The mixture is made up of three phases: a continuous matrix of SAN, a dispersed phase of polybutadiene, and a boundary layer of SAN graft. ABS polymers are processable by all techniques commonly used with thermoplastics. They are slightly hygroscopic and should be dried 2–4 h at 180–200°F (82–93°C) just prior to processing. A dehumidifying circulating air-hopper dryer is recommended. ABS can be hot stamped, painted, printed, vacuum metallized, electroplates, and embossed. Common fabrication techniques are applicable, including sawing, drilling, punching, riveting, bonding, and incorporating metal inserts and threaded and non-threaded fasteners. The machining characteristics of ABS are similar to those of nonferrous metals. ABS materials are superior to the ordinary styrene products and are commonly described as tough, hard, and rigid. This combination is unusual for thermoplastics. Moreover, the molded specimens generally have a very good surface ﬁnish, and this property is particularly marked with the interpolymer type ABS polymers. Light weight and the ability to economically achieve a one-step ﬁnished appearance part have contributed to large-volume applications of ABS. Adequate chemical resistance is present in the ABS materials for ordinary applications. They are affected little by water, alkalis, weak acids, and inorganic salts. Alcohol and hydrocarbon may affect the surfaces. ABS has poor resistance to outdoor UV light; signiﬁcant changes in appearance and mechanical properties will result after exposure. Protective coatings can be applied to improve resistance to UV light. ABS materials are employed in thousand of applications, such as house-hold appliances, business machine and camera housings, telephone handsets, electrical hand tools (such as drill housings), handles, knobs, cams, bearings, wheels, gears, pump impellers, automotive trim and hardware, bathtubs, refrigerator liners, pipe and ﬁttings, shower heads, and sporting goods. Business machines, consumer electronics, and telecommunications applications represent the fastest-growing areas for ABS. Painted, electroplated, and vacuum-metallized parts are used throughout the automotive, business machine, and electronics markets. Multilayered laminates with an ABS outer layer can be produced by coextrusion. In this process two or three different polymers may be combined into a multilayered ﬁlm or sheet. Adhesion is enhanced by cooling the extruded laminate directly from the melt rather than in a separate operation after the components of the sheet have been formed and cooled separately. In one process ﬂows from individual extruders are combined in a ﬂow block and then conveyed to a single manifold die. All the polymer streams should have approximately the same viscosity so that laminar ﬂow can be maintained. Multilayered ﬁlms and sheets have the advantage that a chemically resistant sheet can be combined with a good barrier to oxygen and water diffusion, or a decorative glossy sheet can be placed over a tough, strong material. One commercial example is an ABS-high-impact-polystyrene sheet, which can be thermoformed (see Chapter 2) to make the inside door and food compartment of a refrigerator. Another example is a four-layered sheet comprising ABS, polyethylene, polystyrene, and rubber-modiﬁed polystyrene for butter and margarine packages. However, not all combinations adhere equally well, so there are limits to the design of such structures. 474 Plastics Technology Handbook A variety of special ABS grades have been developed. These include high-temperature-resistant grades have been developed. (for automotive instrument panels, power tool housings), ﬁre-retardant grades (for appliance housings, business machines, television cabinets), electroplating grades (for automotive grilles and exterior decorative trim), high-gloss, low-gloss, and matte-ﬁnish grades (for molding and extrusion applications), clear ABS grades (using methyl methacrylate as the fourth monomer), and structural foam grades (for molded parts with high strength-to-weight ratio). The structural foam grades are available for general purpose and ﬂame-retardant applications. The cellular structure can be produced by injecting nitrogen gas into the melt just prior to entering the mold or by using chemical blowing agents in the resin (see also the section on foaming processes in Chapter 2). 4.2.2.9 Ethylene–Methacrylic Acid Copolymers (Ionomers) CH3 CH2 CH2 CH2 C COO n Monomers Ethylene, methacrylic acid Polymerization Major Uses Free-radical-initiated chain polymerization Packaging ﬁlm, golf ball covers, automotive parts, footwear 4.2.2.9.1 Ionomers Ionomer is a generic name for polymers containing interchain ionic bonding [24]. Introduced in 1964 by DuPont, they have the characteristics of both thermoplastics and thermosetting materials and are derived by copolymerizing ethylene with a small amount (1–10% in the basic patent) of an unsaturated acid, such as methacrylic acid, using the high-pressure process. The carboxyl groups in the copolymer are then neutralized by monovalent and divalent cations, resulting in some form of ionic cross-links (see Figure 4.8) which are stable at normal ambient temperatures but which reversibly break down on heating. These materials thus process the advantage of cross-linking, such as enhanced toughness and stiffness, at ambient temperatures, but they behave as linear polymers at elevated temperatures, so they may be processed and even reprocessed without undue difﬁculty. Copolymerization used in making ionomers has had the effect of depressing crystallinity, although not completely eliminating it, so the materials are also transparent. Ionomers also have excellent oil and grease resistance, excellent resistance to stress cracking, and a higher water vapor permeability than does polyethylene. The principal uses of ionomers are for ﬁlm lamination and coextrusion for composite food packaging. The ionomer resin provides an outer layer with good sealability and signiﬁcantly greater puncture resistance than an LDPE ﬁlm. Sporting goods utilize the high-impact toughness of ionomers. Most major golf ball manufacturers use covers of durable ionomer. Such covers are virtually cut proof in normal use and retain a greater resiliency over a wider temperature range; they are superior to synthetic transpolyisoprene in these respects. Automotive uses (bumper pads and bumper guards) are based on impact toughness and paintability. In footwear applications, resilience and ﬂex toughness of ionomers are advantages in box toes, counters, and shoe soles. Ski boot and ice skate manufacturers produce light weight outer shells of ionomers. Sheet and foamed sheet products include carpet mats, furniture tops, ski lift seat pads, boat bumpers, and wrestling mats. 475 Industrial Polymers Ionomers should be differentiated from polyelectrolytes and ion-exchange resins, which also contain ionic groups. Polyelectrolytes show ionic dissociation in water and are used, among other things, as thickening agents. Common examples are sodium polyacrylate, ammonium poly-methacrylate (both anionic polyacrylate) and poly(N-butyl-4-vinyl–pyridinium bromide), a cationic polyelectrolyte. Ionexchange resins used in water softening, in chromatography, and for various industrial purposes, are cross-linked polymers containing ionic groups. Polyelectrolytes and ion-exchange resins are, in general, intractable materials and not processable on conventional plastics machinery. In ionomers, however, the amount of ionic bonding is limited to yield useful and tractable plastics. Using this principle, manufacturers can produce rubbers which undergo ionic cross-linking to give the effect of vulcanization as they cool on emergence from an extruder or in the mold of an injection-molding machine (see the section on thermoplastic polyoleﬁn rubbers). 4.2.3 Acrylics Acrylic polymers may be considered structurally as derivatives of acrylic acid and its homologues. The family of acrylics includes a range of commercial polymers based on acrylic acid, methacrylic acid, esters of acrylic acid and of methacrylic acid, acrylonitrile, acrylamide, and copolymers of these compounds. By far the best known applications of acrylics are acrylic ﬁbers and acrylonitrile copolymers such as NBR, SAN, and ABS. 4.2.3.1 Polyacrylonitrile CH2 CH CN n Monomer Acrylonitrile Polymerization Major Uses Free-radical-initiated chain polymerization Fibers in apparel (70%) and house furnishings (30%) Polyacrylonitrile and closely related copolymers have found wide use as ﬁbers [25]. The development of acrylic ﬁbers started in the early 1930s in Germany. In the United States they were ﬁrst produced commercially about 1950 by DuPont (Orlon) and Monsanto (Acrilan). In polyacrylonitrile appreciable electrostatic forces occur between the dipoles of adjacent nitrile groups on the same polymer molecule. This restricts the bond rotation and leads to a stiff, rodlike structure of the polymer chain. As a result, polyacrylonitrile has a very high crystalline melting point (317°C) and is soluble in only a few solvents, such as dimethylformamide and dimethylacetamide, and in concentrated aqueous solutions of inorganic salts, such as calcium thiocyanate, sodium perchlorate, and zinc chloride. Polyacrylonitrile cannot be melt processed because its decomposition temperature is close to the melting point. Fibers are therefore spun from solution by either wet or dry spinning (see Chapter 2). Fibers prepared from straight polyacrylonitrile are difﬁcult to dye. To improve dyeability, manufacturers invariably add to monomer feed minor amounts of one or two comonomers, such as methyl acrylate, methyl methacrylate, vinyl acetate, and 2-vinyl–pyridine. Small amounts of ionic monomers (sodium styrene sulfonate) are often included for better dyeability. Modacrylic ﬁbers are composed of 35–85% acrylonitrile and contain comonomers, such as vinyl chloride, to improve ﬁre retardancy. Acrylic ﬁbers are more durable than cotton, and they are the best alternative to wool for sweaters. A major portion of the acrylic ﬁbers produced are used in apparel (primarily hosiery). Other uses include pile fabrics (for simulated fur), craft yarns, blankets, draperies, carpets, and rugs. 476 Plastics Technology Handbook 4.2.3.2 Polyacrylates CH2 CH COOR n Monomer Acrylic acid esters Polymerization Major Uses Free-radical-initiated chain polymerization Fiber modiﬁcation, coatings, adhesives, paints The properties of acrylic ester polymers depend largely on the type of alcohol from which the acrylic acid ester is prepared [26]. Solubility in oils and hydrocarbons increases as the length of the side chain increases. The lowest member of the series, poly(methyl acrylate), has poor low-temperature properties and is water sensitive. It is therefore restricted to such applications as textile sizes and leather ﬁnishes. Poly(ethyl acrylate) is used in ﬁber modiﬁcations and in coatings; and poly(butyl acrylate) and poly (2-ethylhexyl acrylate) are used in the formulation of paints and adhesives. The original acrylate rubbers ﬁrst introduced in 1948 by B. F. Goodrich and marketed as Hycar 4021 were a copolymer of ethyl acrylate with about 5% of 2-chloroethyl vinyl ether (CH2═CH–O–CH2– CH2Cl) acting as a cure site monomer. Such polymers are vulcanized through the chlorine atoms by amines (such as triethylenetetramine), which are, however, not easy to handle. Therefore, 2-chloroethyl vinyl ether has been replaced with other cure site monomers, such as vinyl and allyl chloroacetates; the increased reactivity of the chlorine in these monomers permits vulcanization with ammonium benzoate (which decomposes on heating to produce ammonia, the actual cross-linking agent) rather than amines. Acrylate rubbers have good oil resistance. In heat resistance they are superior to most rubbers, exceptions being the ﬂuororubbers, the ﬂuororubbers, the silicones, and the ﬂuorosilicones. It is these properties which account for the major use of acrylate rubbers, i.e., in oil seals for automobiles. They are, however, inferior in low-temperature properties. A few acrylate rubbers (such as Hycar 2121X38) are based on butyl acrylate. These materials are generally copolymers of butyl acrylate and acrylonitrile (∼10%) and may be vulcanized with amines. They have improved low-temperature ﬂexibility compared to ethyl acrylate copolymers but swell more in aromatic oils. Ethylene copolymers with acrylates represent a signiﬁcant segment of the ethylene copolymer market, as many LDPE producers use copolymerization as a strategy to obtain products more resistant to displacement by HDPE and LLDPE. Ethylene copolymers with methyl methacrylate and ethyl, butyl, and methyl acrylates are similar to EVA copolymers in properties (discussed later) but have improved thermal stability during extrusion and increased low-temperature ﬂexibility. The commercial products in this category generally contain 15–30% of the acrylate or methacrylate comonomer. Applications include medical packaging, disposable gloves, hoses, tubing, gaskets, cable insulation, and squeeze toys. Use of ethylene–ethyl acrylate copolymers for making vacuum cleaner hoses demonstrates the increased ﬂexibility and long ﬂex life that is possible with such materials. Copolymers in which the acrylate monomer is the major component are useful as ethylene–acrylate elastomers (trade name: Vamac). These are terpolymers containing a small amount of an alkenoic acid to introduce sites (C═C) for subsequent cross-linking via reaction with primary diamines [see Equation 34 in Chapter 1]. These elastomers have excellent oil resistance and stability over a wide temperature range (−50°C to 200°C), being superior to chloroprene and nitrile rubbers. Although not superior to silicone and ﬂuoroelastomers, they are less costly; uses include automotive (hydraulic systems seals, hoses) and wire and cable insulation. 477 Industrial Polymers 4.2.3.3 Polymethacrylates CH3 CH2 C COOR n A large number of alkyl methacrylates, which may be considered as esters of poly(methacrylic acid), have been prepared [26]. By far the most important of these polymers is poly(methyl methacrylate), which is an established major plastics material. As with other linear polymers, the mechanical and thermal properties of polymethacrylates are largely determined by the intermolecular attraction, spatial symmetry, and chain stiffness. As the size of the ester alkyl group increases in a series of poly(n-alkyl methacrylate)s, the polymer molecules become spaced further apart and the intermolecular attraction is reduced. Thus, as the length of the side chain increases, the softening point decreases, and the polymers become rubbery at progressively lower temperatures (Figure 4.9). However, when the number of carbon atoms in the side chain exceeds 12, the polymers become less rubbery, and the softening point, brittle point, and other properties related to the glass transition temperature rise with an increase in chain length (Table 4.8 and Table 4.9). As with the polyoleﬁns, this effect is due to side-chain crystallization. Poly(alkyl methacrylate)s in which the alkyl group is branched have higher softening points (see Table 4.8) and are harder than their unbranched isomers. This effect is not simply due to the better packing possible with the branched isomers. The lumpy branched structures impede rotation about the carbon– carbon bond on the main chain, thus contributing to stiffness of the molecule and consequently a higher transition temperature. Similarly, since the a-methyl group in polymethacrylates reduces chain ﬂexibility, the lower polymethacrylates have higher softening points than the corresponding polyacrylates do. Poly(methyl methacrylate (PMMA) is by far the predominant polymethacrylate used in rigid applications because it has crystal clear transparency, excellent weatherability (better than most other plastics), 100 80 Brittle point (°C) 60 Poly (n-alkyl methacrylate)s 40 20 0 –20 –40 Poly (n-alkyl acrylate)s –60 2 4 6 8 10 12 14 Carbon atoms in the alkyl group 16 FIGURE 4.9 Brittle points of poly(n-alkyl acrylate)s and poly(n-alkyl methacrylate)s. (After Rehberg, C. E. and Fisher, C. H. 1948. Ind. Eng. Chem., 40, 1431.) 478 Plastics Technology Handbook TABLE 4.8 Vicat Softening Points of Polymethacrylates Derived from Monomers of Type CH2═C(CH3)COOR R– Softening Point (°C) CH3– 119 CH3–CH2– 81 CH3–CH2–CH2– CH3–CH2–CH2–CH2– 55 30 CH3–CH2–CH2–CH2–CH2– —a (CH3)2CH– (CH3)2CH–CH2– 88 67 (CH3)3C– 104 (CH3)2CH–CH2–CH2 (CH3)3C–CH2– 46 115 (CH3)3C CH 119 CH3 a Too rubbery for testing. TABLE 4.9 Glass Transition Temperatures of Polymsethacrylates Ester Group Tg (°C) Methyl 105 Ethyl n-Butyl 65 20 n-Decyl −70 n-Hexadecyl −9 and a useful combination of stiffness, density, and moderate toughness. The glass transition temperature of the polymer is 105°C (221°F), and the heat deﬂection temperatures range from 75 to 100°C (167– 212°F). The mechanical properties of PMMA can be further improved by orientation of heat-cast sheets. PMMA is widely used for signs, glazing, lighting, ﬁxtures, sanitary wares, solar panels, and automotive tail and stoplight lenses. The low index of refraction (1.49) and high degree of uniformity make PMMA an excellent lens material for optical applications. Methyl methacrylate has been copolymerized with a wide variety of other monomers, such as acrylates, acrylonitrile, styrene, and butadiene. Copolymerization with styrene gives a material with improved melt-ﬂow characteristics. Copolymerization with either butadiene or acrylonitrile, or blending PMMA with SBR, improves impact resistance. Butadiene–methyl methacrylate copolymer has been used in paper and board ﬁnishes. Higher n-alkyl methacrylate polymers have commercial applications. Poly(n-butyl-), poly(n-octyl-) and poly(n-nonyl methacrylate)s are used as leather ﬁnishes; poly(lauryl methacrylate) is used to depress the pour point and improve the viscosity-temperature property of lubricating oils. Mention may also be made here of the 2-hydroxyethyl ester of methacrylic acid, which is the monomer used for soft contact lenses. Copolymerization with ethylene glycol dimethacrylate produces a hydrophilic network polymer (a hydrogel). Hydrogel polymers are brittle and glassy when dry but become soft and plastic on swelling in water. Terpolymers based on methyl methacrylate, butadiene, and styrene (MBS) are being increasingly used as tough transparent plastics and as additives for PVC. 479 Industrial Polymers 4.2.3.4 Polyacrylamide CH2 CH CONH2 n Monomer Polymerization Major Uses Acrylamide Free-radical-initiated chain polymerization Flocculant, adhesives, paper treatment, water treatment, coatings Polyacrylamide exhibits strong hydrogen bonding and water solubility. Most of the interest in this polymer is associated with this property. Polymerization of acrylamide monomer is usually conducted in an aqueous solution, using free-radical initiators and transfer agents. Copolymerization with other water-soluble monomers is also carried out in a similar manner. Cationic polyacrylamides are obtained by copolymerizing with ionic monomers such as dimethylaminoethyl methacrylate, dialkyldimethylammonium chloride, and vinylbenzyltrimethylammonium chloride. These impart a positive charge to the molecule. Anionic character can be imparted by copolymerizing with monomers such as acrylic acid, methacrylic acid, 2-acrylamido-2-methyl-propanesulfonic acid, and sodium styrene sulfonate. Partial hydrolysis of polyacrylamide, which converts some of the amide groups to carboxylate ion, also results in anionic polyacrylamides. Polyacrylamides have several properties which lead to a multitude of uses in diverse industries. Table 4.10 lists the main functions of the polymers and their uses in various industries. Polyacrylamides are used as primary ﬂocculants or coagulant aids in water clariﬁcation and mining application. They are effective for clariﬁcation of raw river water. The capacity of water clariﬁers can be increased when the polymer is used as a secondary coagulant in conjunction with lime and ferric chloride. Polyacrylamides, and especially cationic polyacrylamides, are used for conditioning municipal and industries sludges for dewatering by porous and empty sand beds, vacuum ﬁlters, centrifuges, and other mechanical devices. Certain anionic polyacrylamides are approved by the U.S. Environmental Protection Agency for clariﬁcation of potable water. Polymer treatment also allows ﬁlters to operate at higher hydraulic rates. The function of clariﬁcation is not explained by a simple mechanism. The long-chain linear polymer apparently functions to encompass a number of individual ﬁne particles of the dispersed material in water, attaching itself to the particles at various sites by chemical bonds, electrostatic attraction, or other TABLE 4.10 Applications of Polyacrylamide Function Flocculation Adhesion Rheology control Application Industry Water clariﬁcation General Waste removal Solids recovery Sewage Mining Retention aid Paper Drainage aid Dry strength Paper Paper Wallboard cementing Construction Waterﬂooding Viscous drag reduction Petroleum Petroleum Fire ﬁghting Irrigation pumping 480 Plastics Technology Handbook attractive forces. Relatively stable aggregates are thus produced, which be removed by ﬁltration, settling, or other convenient means. Polyacrylamides are useful in the paper industry as processing aids, in compounding and formulating, and as ﬁller-retention aids. Polyacrylamides and copolymers of acrylamide and acrylic acid are used to increase the dry strength of paper. Polyacrylamides are used as ﬂooding aids in secondary oil recovery from the producing oil well. Water, being of low viscosity, tends to ﬁnger ahead of the more viscous oil. However, addition of as little as 0.05% polyacrylamide to the waterﬂood reduces oil bypass and give signiﬁcantly higher oil to water ratios at the producing wellhead. Greatly increased yields of oil result from adding polymer to waterﬂooding. Solutions containing polyacrylamide are very slippery and can be used for water-based lubrication. Small amounts of polymer, when added to an aqueous solution, can signiﬁcantly reduce the friction in pipes, thereby increasing the throughput or reducing the power consumption. Other applications include additives in coatings and adhesives and binders for pigments. Lightly cross-linked polyacrylamide is used to make superabsorbents of water. Astarch-g-polyacrylamide/ clay superabsorbent composite has been synthesized [27] by graft copolymerization reaction of acrylamide, potato starch, and kaolinite micropowder (<1 mm) followed by hydrolysis with sodium hydroxide. Such a superabsorbent of compositin: 20% kaolinite, 20% potato starch, 60% acrylamide, 2% initiator (ceric amonium nitrate), and 0.04% cross-linker (N,N-methylenebisacrylamide) is found to absorb 2250 g H2O/g at room temperature at swelling equilibrium. 4.2.3.5 Poly(Acrylic Acid) and Poly(Methacrylic Acid) CH3 CH2 CH and n Acrylic acid and methacrylic acid C COOH COOH Monomers CH2 n Polymerization Major Uses Free-radical-initiated chain polymerization Sodium and ammonium-salts as polyelectrolytes, thickening agents Poly(acrylic acid) and poly(methacrylic acid) may be prepared by direct polymerization of the appropriate monomer, namely, acrylic acid or methacrylic acid, by conventional free-radical techniques, with potassium persulfate used as the initiator and water as the solvent (in which the polymers are soluble); or if a solid polymer is required, a solvent such as benzene, in which the polymer is insoluble, can be used, with benzoyl peroxide as a suitable initiator. The multitude of applications of poly(acrylic acid) and poly(methacrylic acid) is reminiscent of the fable of the man who blew on his hands to warm them and blew on his porridge to cool it. They are used as adhesives and release agents, as ﬂocculants and dispersants, as thickeners and ﬂuidizers, as reaction inhibitors and promoters, as permanent coatings and removable coatings, etc. Such uses are the direct result of their varied physical properties and their reactivity. Many of the applications thus depend on the ability of these polymers to form complexes and to bond to substrates. Monovalent metal and ammonium salts of these polymers are generally soluble in water. These materials behave as anionic polyelectrolytes and are used for a variety of purposes, such as thickening agents, particularly for rubber latex. Superabsorbent polymers are loosely cross-linked networks of partially neutralized acrylic acid polymers capable of absorbing large amounts of water and retaining the absorbed water under pressure. Although superabsorbent polymers have been abundantly used in the disposable diaper industry for the 481 Industrial Polymers past 30 years, their applications are still being expanded to many ﬁelds, including agriculture and horticulture, sealing composites, artiﬁcial snow, drilling ﬂuid additives, drug delivery systems, and so on. The swelling of a hydrophilic polymer is dependent on the rubbery elasticity, ionic osmotic pressure, and afﬁnity of the polymer toward water. Although superabsorbent polymers have the greatest absorbency in water, the addition of an inorganic salt or organic solvent will reduce the absorbency. Clay and mineral ﬁllers have been used for reducing production costs and improving the comprehensive water absorbing properties of superabsorbent materials. For example, a poly(acrylic acid)/mica superabsorbent has been synthesized with water absorbency higher than 1100 g H2O/g. In a typical method of preparation, acrylic acid monomer is neutralized at ambient temperature with an amount of aqueous sodium hydroxide solution to achieve 65% neutralization (optimum). Dry ultraﬁne (<0.2 mm) mica powder (10 wt%) is added, followed by cross-linker N,N-methylene-bisacrylamide (0.10 wt%) and radical initiator, potassium persulfate. The mixture is heated to 60–70°C in a water bath for 4 h. The product is washed, dried under vacuum at 50°C, and screened. 4.2.3.6 Acrylic Adhesives Acrylic adhesives are essentially acrylic monomers which achieve excellent bonding upon polymerization. Typical examples are cyanoacrylates and ethylene glycol dimethacrylates. Cyanoacrylates [28] are obtained by depolymerization of a condensation polymer derived from a malonic acid derivative and formaldehyde. CN CH2 CN + CH2O H2O COOR C CN Δ CH2 COOR C CH2 COOR n Cyanoacrylates are marketed as contact adhesives. Often popularly known as superglue, they have found numerous applications. In dry air and in the presence of polymerization inhibitors, methyl- and ethyl-2-cyanoacrylates have a storage life of many months. As with many acrylic monomers, air can inhibit or severely retard polymerization of cyanoacrylates. These monomers are, however, prone to anionic polymerization, and even a very weak base such as water can bring about rapid polymerization. CN CN C COOR CH2 + H2O C CH2 COOR n In practice, a trace of moisture occurring on a substrate is adequate to cause polymerization of the cyanoacrylate monomer to provide strong bonding within a few seconds of closing the joint and excluding air. Cyanoacrylate adhesives are particularly valuable because of their speed of action, which obviates the need for complex jibs and ﬁxtures. The amount of monomer applied should be minimal to obtain a strong joint. Larger amounts only reduce the strength. Notable uses of cyanoacrylates include surgical glue and dental sealants; morticians use them to seal eyes and lips. Dimethacrylates, such as tetramethylene glycol dimethacrylate, are used as anaerobic adhesives. Air inhibition of polymerization of acrylic monomers is used to advantage in this application because the monomers are supplied along with a curing system (comprising a peroxide and an amine) as part of a one-part pack. When this adhesive is placed between mild steel surfaces, air inhibition is prevented since the air is excluded, and polymerization can take place. Though the metal on the surface acts as a polymerization promoter, it may be necessary to use a primer such as cobalt naphthenate to expedite the 482 Plastics Technology Handbook polymerization. The anaerobic adhesives are widely used for sealing nuts and bolts and for miscellaneous engineering purposes. Dimethacrylates form highly cross-linked and, therefore, brittle polymers. To overcome brittleness, manufacturers often blend dimethacrylates with polyurethanes or other polymers such as low-molecularweight vinyl-terminated butadiene–acrylonitrile copolymers and chlorosulfonated polyethylene. The modiﬁed dimethacrylate systems provide tough adhesives with excellent properties. These can be formulated as two-component adhesives, the catalyst component being added just prior to use or applied separately to the surface to be bonded. One-component systems also have been formulated which can be conveniently cured by ultraviolet radiation. 4.2.4 Vinyl Polymers If the R substituent in an oleﬁn monomer (CH2═CHR) is either hydrogen, alkyl, aryl, or halogen, the corresponding polymer in the present discussion is grouped under polyoleﬁns. If the R is a cyanide group or a carboxylic group or its ester or amide, the substance is an acrylic polymer. Vinyl polymers include those polyoleﬁns in which the R substituent in the oleﬁn monomer is bonded to the unsaturated carbon through an oxygen atom (vinyl esters, vinyl ethers) or a nitrogen atom (vinyl pyrrolidone, vinyl carbazole). Vinyl polymers constitute an important segment of the plastics industry. Depending on the speciﬁc physical and chemical properties, these polymers ﬁnd use in adhesives, in treatments for paper and textiles, and in special applications. The commonly used commercial vinyl polymers are described next. 4.2.4.1 Poly(Vinyl Acetate) CH2 CH OCCH3 O n Monomer Vinyl acetate Polymerization Major Uses Free-radical-initiated chain Polymerization (mainly emulsion polymerization) Emulsion paints, adhesives, sizing Since poly(vinyl acetate) is usually used in an emulsion form, it is manufactured primarily by freeradical-initiated emulsion polymerization. The polymer is too soft and shows excessive cold ﬂow, which precludes its use in molded plastics. The reason is that the glass transition temperature of 28°C is either slightly above or (at various times) below the ambient temperatures. Vinyl acetate polymers are extensively used in emulsion paints, as adhesives for textiles, paper, and wood, as a sizing material, and as a permanent starch. A number of commercial grades are available which differ in molecular weight and in the nature of comonomers (e.g., acrylate, maleate, fumarate) which are often used. Two vinyl acetate copolymers of particular interest to the plastics industry are EVA and vinyl chloride–vinyl acetate copolymers. EVA copolymers represent the largest-volume segment of ethylene copolymer market and are the products of low-density polyethylene (LDPE) technology. Commercial preparation of EVA copolymer is based on the same process as LDPE with the addition of controlled comonomer stream into the reactor. EVA copolymers are thermoplastic materials consisting of an ethylene chain incorporating 5–20 mol% vinyl acetate (VA), in general. The VA produces a copolymer with lower crystallinity than conventional ethylene homopolymer. 483 Industrial Polymers These lower crystallinity resins have lower melting points and heat seal temperatures, along with reduced stiffness, tensile strength, and hardness. EVAs have greater clarity, low-temperature ﬂexibility, stress-crack resistance, and impact strength than LDPE. EVA resins are more permeable to oxygen, water vapor, and carbon dioxide. Chemical resistance is similar to that of LDPE, with somewhat better resistance to oil and grease for EVA resins of higher VA content. The VA groups contribute to improved adhesion in extrusions or hot-melt adhesive formulations. The outdoor stability of EVA resins is superior to that of LDPE by virtue of their greater ﬂexibility. Addition of UV stabilizers can extend the outdoor life of clear compounds to three to ﬁve years, depending on the degree of exposure. Outdoor life expectancy is also enhanced by the addition of carbon black. In addition to specialty applications involving ﬁlm and adhesives production, EVA are used in a variety of molding, compounds, and extrusion applications. Some typical end uses include ﬂexible hose and tubing, footwear components, toys and athletic goods, wire and cable compounding, extruded gaskets, molded automotive parts (such as energy-absorbing bumper components), cap and closure seals, and color concentrates. In footwear applications, EVA resins are used in canvas box toes and ﬂocked or fabric-laminated contours. Foamed and cross-linked EVA is used in athletic or leisure shoe midsoles and in sandals. With increasing VA content, EVA resin properties range from those of LDPE to those of highly plasticized PVC. High VA resins are soft and ﬂexible, with excellent toughness and good stress-crack resistance. With EVA resins these properties are permanent and do not dissipate with time because of the loss of the liquid plasticizer. EVAs also have exceptional low temperature toughness and ﬂexibility. The EVA copolymers are slightly less ﬂexible than normal rubber compounds but have the advantage of simpler processing since no vulcanization is necessary. The materials have thus been largely used in injection molding in place of plasticized PVC or vulcanized rubber. Typical applications include turntable mats, based pads for small items of ofﬁce equipment, buttons, car door protection strips, and for other parts where a soft product of good appearance is required. A substantial use of EVA copolymers is as wax additives and additives for hot-melt coatings and adhesives. Cellular cross-linked EVA copolymers are used in shoe parts. EVA copolymers with only a small vinyl acetate content (∼3 mol%) are best considered as a modiﬁcation of low-density polyethylene. These copolymers have less crystallinity and greater ﬂexibility, softness, and, in case of ﬁlm, surface gloss. EVA can be processed by all standard plastics processing techniques, including injection and blow molding, thermoforming, and extrusion into sheet and shapes. They accommodate high loadings of ﬁllers, pigments, and carbon blacks. They are compatible with other thermoplastics, and thus are frequently used for impact modiﬁcation and improvement of stress-crack resistance. This combination of properties makes EVA highly adaptable vehicles for color concentrates. EVA resins can be formulated with blowing agents and cross-linking to produce low density foams via compression molding. Hydrolysis of EVA copolymers yields ethylene–vinyl alcohol copolymers (EVOH). EVOH has exceptional gas barrier properties as well as oil and organic solvent resistance. The poor moisture resistance of EVOH is overcome by coating, coextrusion, and lamination with other substrates. Applications include containers for food (ketchup, jelly, mayonnaise) as well as chemicals and solvents. 4.2.4.2 Poly(Vinyl Alcohol) CH2 CH OH n Manufacture Alcoholysis of poly(vinyl acetate) Major Uses Paper sizing, textile sizing, cosmetics 484 Plastics Technology Handbook Poly(vinyl alcohol) (PVA) is produced by alcoholysis of poly(vinyl acetate), because vinyl alcohol monomer does not exist in the free state [29]. (The term hydrolysis is often used incorrectly to describe this process.) Either acid or base catalysts may be employed for alcoholysis. Alkalien catalysts such as sodium hydroxide or sodium methoxide give more rapid alcoholysis. The degree of alcoholysis, and hence the residual acetate content, is controlled by varying the catalyst concentration. CH2 CH OCCH3 + CH3OH CH2 CH OH + CH3COCH3 O O The presence of hydroxyl groups attached to the main chain renders the polymer hydrophilic. PVA therefore dissolves in water to a greater or lesser extent according to the degree of hydrolysis. Polymers with a degree of hydrolysis in the range of 87–90% readily dissolve in cold water. Solubility decreases with an increase in the degree of hydrolysis, and fully hydrolyzed polymers are water soluble only at higher temperatures (>85°C). This apparently anomalous behavior is due to the higher degree of crystallinity and the greater extent of hydrogen bonding in the completely hydrolyzed polymers. Commercial PVA is available in a number grades which differ in molecular weight and degree of hydrolysis. The polymer ﬁnds a variety of uses. It functions as a nonionic surface active agent and is used in suspension polymerization as a protective colloid. It also serves as a binder and thickener and is widely used in adhesives, paper coatings, paper sizing, textile sizing, ceramics, and cosmetics. Completely hydrolyzed grades of PVA ﬁnd use in quick-setting, water-resistant adhesives. Combinations of fully hydrolyzed PVA and starch are used as a quick-setting adhesive for paper converting. Borated PVA, commonly called “tackiﬁed,” are combined with clay and used in adhesive applications requiring a high degree of wet tack. They are used extensively to glue two or more plies of paper together to form a variety of shapes such as tubes, cans, and cores. Since PVA ﬁlm has little tendency to adhere to other plastics, it can be used to prevent sticking to mold. Films cast from aqueous solution of PVA are used as release agents in the manufacture of reinforced plastics. Partially hydrolyzed grades have been developed for making tubular blown ﬁlm (similar to that with polyethylene) for packages for bleaches, bath salts, insecticides, and disinfectants. Use of water-soluble PVA ﬁlm for packaging preweighed quantities of such materials permits their addition to aqueous systems without breaking the package or removing the contents, thereby saving time and reducing material losses. Film made from PVA may be used for hospital laundry bags that are added directly to the washing machine. A process has been developed in Japan for producing ﬁbers from poly(vinyl alcohol). The polymer is wet spun from a warm aqueous solution into a concentrated aqueous solution of sodium sulfate containing sulfuric acid and formaldehyde, which insolubilizes the alcohol by formation of formal groups (see below). These ﬁbers are generally known as vinal of vinylon ﬁbers. 4.2.4.3 Poly(Vinyl Acetals) Poly(vinyl acetals) are produced by treating poly(vinyl alcohol) with aldehydes. (They may also be made directly from poly(vinyl acetate) without separating the alcohol.) Since the reaction with aldehyde involves a pair of neighboring hydroxyl groups on the polymer chain and the reaction occurs at random, some hydroxyl groups become isolated and remain unreacted. A poly(vinyl acetal) molecule will thus contain acetal groups and residual hydroxyl groups. In addition, there will be residual acetate groups due to incomplete hydrolysis of poly(vinyl acetate) to the poly(vinyl alcohol) used in the acetalization reaction. The relative proportions of these three types of groups may have a signiﬁcant effect on speciﬁc properties of the polymer. 485 Industrial Polymers CH2 CH CH2 OH + RCH CH CH2 CH2 O O OH CH CH O C R H When the aldehyde in this reaction is formaldehyde, the product is poly(vinyl formal). This polymer is, however, made directly from poly(vinyl acetate) and formaldehyde without separating the alcohol. The product with low hydroxyl (5–6%) and acetate (9.5–13%) content (the balance being formal) is used in wire enamel and in structural adhesives (e.g., Redux). In both applications the polymer is used in conjunction with phenolic resins and is heat cured. When the aldehyde in the acetalization reaction is butyraldehyde, i.e., R=CH7CH2CH2–, the product is poly(vinyl butyral). Sulfuric acid is the catalyst in this reaction. Poly(vinyl butyral) is characterized by high adhesion to glass, toughness, light stability, clarity, and moisture insensitivity. It is therefore extensively used as an adhesive interlayer between glass plates in the manufacture of laminate safety glass and bulletproof composition. 4.2.4.4 Poly(Vinyl Cinnamate) ( CH2 CH ( ( CH2 OH CH ( O C=O CH CH Poly(vinyl cinnamate) is conveniently made by the Schotten–Baumann reaction using poly(vinyl alcohol) in sodium or potassium hydroxide solution and cinnamoyl chloride in methyl ethyl ketone. The product is, in effect, a copolymer of vinyl alcohol and vinyl cinnamate, as shown. The polymer has the ability to cross-link on exposure to light, which has led to its important applications in photography, lithography, and related ﬁelds as a photoresist (see also Chapter 5). 4.2.4.5 Poly(Vinyl Ethers) CH2 CH OR n Commercial uses have developed for several poly(vinyl ethers) in which R is methyl, ethyl, and isobutyl. The vinyl alkyl ether monomers are produced from acetylene and the corresponding alcohols, and the polymerization is usually conducted by cationic initiation using Friedel–Craft-type catalysts. Poly(vinyl methyl ether) is a water-soluble viscous liquid which has found application in the adhesive and rubber industries. One particular applications has been as a heat sensitizer in the manufacture of rubber-latex dipped goods. Ethyl and butyl derivatives have found uses as adhesives. Pressure-sensitive adhesive tapes made from poly(vinyl ethyl either) incorporating antioxidants are said to have twice the shelf life of similar tapes 486 Plastics Technology Handbook made from natural rubber. Copolymers of vinyl isobutyl ether with vinyl chloride, vinyl acetate and ethyl acrylate are also produced. 4.2.4.6 Poly(Vinyl Pyrrolidone) CH2 CH N n O Poly(vinyl pyrrolidone) is produced by free-radical-initiated chain polymerization of N-vinyl pyrrolidone. Polymerization is usually carried out in aqueous solution to produce a solution containing 30% polymer. The material is marketed in this form or spray dried to give a ﬁne powder. Poly(vinyl pyrrolidone) is a water-soluble polymer. Its main value is due to its ability to form loose addition compounds with many substances. It is thus used in cosmetics. The polymer has found several applications in textile treatment because of its afﬁnity for dyestuffs. In an emergency it is used as a blood plasma substitute. Also, about 7% polymer added to whole blood allows it to be frozen, stored at liquid nitrogen temperatures for years, and thawed out without destroying blood cells. 4.2.4.7 Poly(Vinyl Carbazole) CH2 CH N n Poly(vinyl carbazole) is produced by polymerization of vinyl carbazole using free-radial initiation or Ziegler–Natta catalysis. Poly(vinyl carbazole) has a high softening point, excellent electrical insulating properties, and good photoconductivity, which has led to its application in xerography. 4.3 Part II: Condensation Polymers According to the original classiﬁcation of Carothers, condensation polymers are formed from bi- or polyfunctional monomers by reactions which involve elimination of some smaller molecule. A condensation polymer, according to t his deﬁnition, is one in which the repeating unit lacks certain atoms which were present in the monomers(s) from which the polymer was formed. With the development of polymer science and synthesis of newer polymers, this deﬁnition of condensation polymer was found to be inadequate. For example, in polyurethanes, which are classiﬁed as condensation polymers, the repeat unit has the same net composition as the two monomers-that is, a diol and a diisocyanate, which react without the elimination of any small molecule. Similarly the polymers produced by the ring-opening polymerization of cyclic monomers, such as cyclic ethers and amides, are generally classiﬁed as condensation polymers based on the presence of functional groups, such as the ether and amide linkages, in the polymer chains, even though the polymerization occurs without elimination of any small molecule. To overcome such problems, an alternative deﬁnition has been introduced. According to this deﬁnition, polymers whose main chains consist entirely of C–C bonds are classiﬁed as addition polymers, whereas those in which heteratoms (O, N, S, Si) are present in the polymer backbone are considered to be condensation polymers. A polymer which satisﬁes both the original deﬁnition (of Carothers) and the 487 Industrial Polymers alternative deﬁnition or either of them, is classiﬁed as a condensation polymer. Phenol–formaldehyde condensation polymers, for example, satisfy the ﬁrst deﬁnition but not the second. Condensation polymers described in Part II are classiﬁed as polyesters, polyamides, formaldehyde resins, polyurethanes, and ether polymers. 4.3.1 Polyesters Polyesters were historically the ﬁrst synthetic condensation polymers studied by Carothers in his pioneering work in the early 1930s. Commercial polyesters [30] were manufactured by polycondensation reactions, the methods commonly used being melt polymerization of diacid and diol, ester interchange of diester and diol, and interfacial polymerization (Schotten–Baumann reaction) of diacid chloride and diol. In a polycondensation reaction a by-product is generated which has to be removed as the reaction progresses. Thermoplastic saturated polyesters are widely used in synthetic ﬁbers and also in ﬁlms and molding applications. The production of polyester ﬁbers accounts for nearly 30% of the total amount of synthetic ﬁbers. Unsaturated polyesters are mainly used in glass-ﬁber reinforced plastic products. 4.3.1.1 Poly(Ethylene Terephthalate) C C O CH2 O O Monomers Dimethyl terephthalate or terephthalic acid, ethylene glycol CH2 O n Polymerization Major Uses Bulk polycondensation Apparel (61%), home furnishings (18%), tire cord (10%) Whinﬁeld and Dixon, in the UK, developed polyethylene terephthalate ﬁbers (Dacron, Terylene). This ﬁrst Dacron polyester plant went into operation in 1953. Ester interchange (also known as ester exchange or alcoholysis) was once the preferred method for making polyethylene terephthalate (PET) because dimethyl terephthalate can be readily puriﬁed to the high quality necessary for the production of the polymer. The process is carried out in two steps. Dimethyl terephthalate (DMT) is reacted with excess ethylene glycol (mole ratio 1:2.1–2.2) at 150°C and 100 kPa (1 atm = 101 kPa) The output of the process is bis(hydroxyethyl) terephthalate (BHET) The pre-polymerization step (250–280°C, 2–3 kPa) follows in which BHET is polymerized to a degree of poymerization (DP) of up to 30. The next step is the polycondensation process where the DP is further increased to 100 by heating under vacuum, the process conditions being 280–290°C and 50–100 Pa. Up to this stage, PET is suitable for applications that do not require high molecular weight or high intrinsic viscosity [h], such as ﬁbers and sheets Solid-state polymerization is used to further increase the DP to 150. The operating conditions are 200–240°C at 100 kPa and 5–25 h Bottle-grade PET that has an [h] of 0.73–0.81 dl g−1 is normally produced by solid-state polymerization at 210°C for around 15–20 h [31]. In recent years methods have been developed to produce terephthalic acid with satisfactory purity, and direct polycondensation reaction with ethylene glycol is now the preferred route to this polymer. Virgin PET is produced at different speciﬁcations because different applications require different properties [32]. Examples of intrinsic viscosity [h] for different applications are recording tape 0.60, carbonated drink bottles 0.73–0.81, and industrial tire cord 0.85 dl g−1. PET granules can be processed in many ways depending on application and ﬁnal product requirements. 488 Plastics Technology Handbook PET is widely used in synthetic ﬁbers designed to simulate wool cotton, or rayon, depending on the processing conditions. They have good wash-and-wear properties and resistance to wrinkling. In the production of ﬁber the molten polymer is extruded through spinnerets and rapidly cooled in air. The ﬁlaments thus formed are, however, largely amorphous and weak. They are therefore drawn at a temperature (80°C) above Tg and ﬁnally heated at 190°C under tension, whereby maximum molecular orientation, crystallinity, and dimensional stability are achieved. The melting point of highly crystalline PET is 271°C. Crystalline PET has good resistance to water and dilute mineral acids but is degraded by concentrated nitric and sulfuric acids. It is soluble at normal temperature sonly in proton donors which are capable of interaction with the ester group, such as chlorinated and ﬂuorinated acetic acids, phenols, and anhydrous hydroﬂuoric acid. PET is also used in ﬁlm form (Melinex, Mylar) and as a molding material. The manufacture of PET ﬁlm closely resembles the manufacture of ﬁber. The ﬁlm is produced by quenching extruded sheet to the amorphous state and then reheating and stretching the sheet approximately threefold in the axial and transverse directions at 80–100°C. To stabilize the biaxially oriented ﬁlm, it is annealed under restraint at 180–210°C. This operation increases the crystallinity of PET ﬁlm and reduces its tendency to shrink on heating. The strength of PET in its oriented form is outstanding. The principal uses of biaxially oriented PET ﬁlm are in capacitors, in slot liners for motors, and for magnetic tape. Although a polar polymer, its electrical insulation properties at room temperature are good (even at high frequencies) because at room temperature, which is well below Tg (69°C), dipole orientation is severely restricted. The high strength and dimensional stability of the polyester ﬁlm have also led to its use for x-ray and photographic ﬁlm and to a number of graphic art and drafting applications. The ﬁlm is also used in food packaging, including boil-in-bag food pouches. Metallized polyester ﬁlms have many uses as a decorative material. Because of its rather high glass transition temperature, only a limited amount of crystallization can occur during cooling after injection molding of PET. The idea of molding PET was thus for many years not a technical proposition. Toward the end of the 1970s DuPont introduced Rynite, which is a PET nucleated with an ionomer, containing a plasticizer and only available in glass-ﬁber-ﬁlled form (at 30, 45, and 55% ﬁll levels.) The material is very rigid, exceeding that of polysulfone, is less water sensitive than an unﬁlled polymer, and has a high heat-deﬂection temperature (227°C at 264 psi). In the late 1970s the beneﬁts of biaxial stretching PET were extended from ﬁlm to bottle manufacture. Producing carbonated beverages PET bottles by blow molding has gained prominence (particularly in the United States) because PET has low permeability to carbon dioxide. The process has been extended, particularly in Europe, to produce bottles for other purposes, such as fruit juice concentrates and sauces, wide-necked jars for coffee, and other materials. Because of its excellent thermal stability, PET is also used for microwave and conventional ovens. Virgin PET manufacturers have tended in recent years to produce PET copolymer, such as isophthalic acid modiﬁed PET, rather than homopolymer PET. PET bottles are normally made from copolymer PET because of its lower crystallinity, improved ductility, better process ability, and better clarity. Some of the most improtant PET copolymers [33] are shown in Figure 4.10. 4.3.1.2 Poly(Butylene Terephthalate) C C O ( CH2 ( 4 O O O n 489 Industrial Polymers O C O C C O O O (CH2)2 O (CH2)2 C O O Poly[(ethylene terephthalate)-co-(ethylene 2,6-naphthalate)][PET/PEN] O C C O O O (CH2)2 O C C O O (CH2)2 O Poly[(ethylene terephthalate)-co-(ethylene isophathalate)][PET/PEI] N O C C O O O (CH2)2 O C O N O C O (CH2)2 O Poly[(ethylene terephthalate)-co-(ethylene 2,5-bis(4-carboxyphenyl)1,3,4,-oxadiazole)[PET/PEOD] FIGURE 4.10 Some of the most important PET copolymers. (After Awaja, F., and Pavel, D. 2005. Eur. Polymer J., 41, 1453. With permission.) Monomers Dimethyl terephthalate or terephthalic acid, butanediol Polymerization Major Uses Bulk polycondensation Machine parts, electrical applications, small appliances Poly(butylenes terephthalate), often abbreviated to PBT or PBTP, is manufactured by condensation polymerization of dimethyl terephthalate and butane-1,4-diol in the presence of tetrabutyl titanate. The polymer is also known as poly(tetramethylene terephthalate), PTMT in short. Some trade names for this engineering thermoplastic are Tenite PTMT (Eastman Kodak), Valox (General Electric), Celanex (Celanese) in America and Arnite PBTP (Akzo), Ultradur (BASF), Pocan (Bayer), and Crastin (CibaGeigy) in Europe. Because of the longer sequence of methylene groups in the repeating unit poly(butylene terephthalate) chains are both more ﬂexible and less polar than poly(ethylene terephthalate). This leads to lower values for melting point (about 224°C) and glass transition temperature (22–43°C). The low glass transition temperature facilitates rapid crystallization when cooling in the mold, and this allows short injectionmolding cycles and high injection speeds. PBT ﬁnds use as an engineering material due to its dimensional stability, particularly in water, and its resistance to hydrocarbon oils without showing stress cracking. PBT also ahs high mechanical strength and excellent electrical properties but a relatively low heat-deﬂection temperature 130°F (54°C) at 264 psi (1.8 MPa). The low water absorption of PBT—less than 0.1% after 24-h immersion—is outstanding. Both dimensional stability and electrical properties are retained under conditions of high humidity. The lubricity of the resin results in outstanding wear resistance. As with PET, there is particular interest in glass-ﬁlled grades of PBT. The glass has a profound effect on such properties as tensile strength, ﬂexural modulus, and impact strength, as can be seen from the values of these properties for unﬁlled and 30% glass-ﬁlled PBT: 8200 vs. 17,000 psi (56 vs. 117 MPa), 340,000 vs. 1.1 – 1.2 × 106 psi (2350 vs. 7580–8270 MPa) and 0.8–1.0 vs. 1.3–1.6 ft.-lbf/in.2 (Izod), respectively. Reinforcing with glass ﬁber also results in an increase in heat-deﬂection temperature to over 400°F (204°C) at 264 psi (1.8 MPa). 490 Plastics Technology Handbook Typical applications of PBT include pump housings, impellers, bearing bushings, gear wheels, automotive exterior and under-the-hood parts, and electrical parts such as connectors and fuse cases. 4.3.1.3 Poly(Dihydroxymethylcyclohexyl Terephthalate) C C O O O CH2 CH2 O n In 1958, Eastman Kodak introduced a more hydrophobic polyester ﬁber under the trade name Kodel. The raw material for this polyester is dimethyl terephthalate. Reduction leads to 1,4-cyclohexylene glycol, which is used with dimethyl terephthalate in the polycondensation (ester exchange) reaction. CH3OC O HOCH2 COCH3 H2 HOCH2 CH2OH O CH2OH + CH3OC COCH3 O O CH2 –CH3OH O CH2 O C C O O n Eastman Kodak also introduced in 1972 a copolyester based on 1,4-cyclohexylene glycol and a mixture of terephthalic and isophthalic acids. The product is sold as Kodar PETG. Being irregular in structure, the polymer is amorphous and gives products of brilliant clarity. In spite of the presence of the heterocyclic ring, the deﬂection temperature under load is as low as that of the poly(butylenes terephthalate)s, and the polymer can be thermoformed at draw ratios as high as 4:1 without blustering or embrittlement. Because of its good melt strength and low molding shrinkage, the material performs well in extrusion blow molding and in injection molding. The primary use for the copolymer is extrusion into ﬁlm and sheeting for packaging. Ethylene glycol-modiﬁed polyesters of the Kodel type are used in blow-molding applications to produce bottles for packaging liquid detergents, shampoos, and similar products. One such product is Kodar PETG 6703 in which one acid (terephthalic acid) is reacted with a mixture of glycols (ethylene glycol and 1,4-cyclohexylene glycol). A related glass-reinforced grade (Ektar PCTG) has also been offered. The principle of formation of segmented or block copolymers (see Section 4.4.5) has also been applied to polyesters, with the “hard” segment formed from butanediol and terephthalic acid, and the “soft” segment provided by a hydroxyl-terminated polyether [polytetramethylene ether glycol (PTMEG)] with molecular weight 600–3000. In a typical preparation, dimethyl terephthalate is transesteriﬁed with a mixture of PTMEG and a 50% excess of butane-1,4-diol in the presence of an ester exchange catalyst. The stoichiometry is such that relatively long sequences of tetramethylene terephthalate (TMT) are produced which, unlike the polyether segments, are crystalline and have a high melting point. Since the sequences of TMT segregate into rigid domains, they are referred to as “hard ”segments, and the softer polyether terephthalate (PE/T) segments are referred to as “soft” segments (Figure 4.11). DuPont markets this polyester elastomer under the trade name Hytrel. These elastomers are available in a range of stiffnesses. The harder grades have up to 84% TMT segments and a melting point of 214°C, and the softest grades contain as little as 33% TMT units and have a melting point of 163°C. Processing of these thermoplastic rubbers is quite straightforward. The high crystallization rates of the hard segments facilitate injection molding, while the low viscosity at low shear rates facilitates a low shear process, such as rotational molding. 491 Industrial Polymers HO (CH2)4 O CH3OC H n ~14 O COCH3 HO (CH2)4 OH O –CH3OH (CH2)4 O n ~14 C O PE/T soft segment FIGURE 4.11 C O O (CH2)4 O C C O O O TMT hard segment Formation of polyester-type segmented or block copolymer. These materials are superior to conventional rubbers in a number of properties (see the section on thermoplastic rubbers). Consequently, in spite of their relatively high price they have become widely accepted as engineering rubbers in many applications. 4.3.1.4 Unsaturated Polyesters Monomers Phthalic anhydride, maleic anhydride, fumaric acid, isophthalic acid, ethylene glycol, propylene glycol, diethylene glycol, styrene Polymerization Major Uses Bulk polycondensation followed by free-radical-initiated chain polymerization Construction, automotive applications, marine applications Unsaturated polyester laminating resins [14,34] are viscous materials of a low degree of polymerization (i.e., oligomers) with molecular weights of 1500–3000. They are produced by condensing a glycol with both an unsaturated dicarboxylic acid (maleic acid) and a saturated carboxylic acid (phthalic or isophthalic acid). The viscous polyesters are dissolved in styrene monomer (30–50% concentration) to reduce the viscosity. Addition of glass ﬁbers and curing with peroxide initiators produces a cross-linked polymer (solid) consisting of the original polyester oligomers, which are now interconnected with polystyrene chains (Figure 4.12b). The unsaturated acid residues in the initial polyester oligomer provide a site for cross-linking in this curing step, while the saturated acid reduces the brittleness of the ﬁnal crosslinked product by reducing the frequency of cross-links. In practice, the peroxide curing system is blended into the resin before applying the resin to the reinforcement, which is usually glass ﬁber (E type), as perform, cloth, mat, or rovings, but sisal or more conventional fabrics may also be sued. The curing system may be so varied (in both composition and quality) that curing times may range from a few minutes to several hours, and the cure may be arranged to proceed either at ambient or elevated temperatures. The two most important peroxy materials used for room temperature curing of polyester resins are methyl ethyl ketone peroxide (MEKP) and cyclohexanone peroxide. These are used in conjunction with a cobalt compound such as a naphthenate, octoate, or other organic-solvent-soluble soap. The peroxides are referred to as catalysts (though, strictly speaking, these are polymerization initiators) and the cobalt compound is referred to as an accelerator. In room-temperature curing it is obviously necessary to add the resin to the reinforcement as soon as possible after the curing system has been blended and before gelation can occur. Benzoyl peroxide is most commonly used for elevated-temperature curing. It is generally supplied as a paste (∼50%) in a liquid such as dimethyl phthalate to reduce explosion hazards and to facilitate mixing. Since the cross-linking of polyester–styrene system occurs by a free-radical chain-reaction mechanism across the double bonds in the polyesters with styrene providing the cross-links, the curing reaction 492 Plastics Technology Handbook CH2=CH CH2=CH CH2=CH CH2=CH CH2=CH CH2=CH Accelerator Styrene monomer CH2=CH Initiator (catalyst) (a) (CH2–CH)n Unsaturated polyester (CH2–CH)n (CH2–CH)n (CH2–CH)n (CH2–CH)n (CH2–CH)n Cross-links (CH2–CH)n Polyester (CH2–CH)n (b) FIGURE 4.12 Curing of unsaturated polyesters. (a) Species in polyester resin ready for laminating. (b) Structures present in cured polyester resin. Cross-linking takes place via an addition copolymerization reaction. The value of n∼2–3 on average in general-purpose resins. (After Brydson, J. A. 1982. Plastics Materials. Butterworth Scientiﬁc, London, UK.) does not give rise to volatile by-products (unlike phenolic and amino resins) and it is thus possible to cure without applying pressure. This fact as well as that room temperature cures are also possible makes unsaturated polyesters most useful in the manufacture of large structures such as boats and car bodies. Unsaturated polyesters ﬁnd applications mainly in two ways: polyester–glass-ﬁber laminates and polyester molding compositions (discussed later). 493 Industrial Polymers 4.3.1.4.1 Raw Materials and Resin Preparation General purpose resins generally employ either maleic acid (usually as the anhydride) or its trans-isomer fumaric acid as the unsaturated acid: H HC H C C HOOC O COOH Maleic acid CH C C H COOH C O C HOOC O Maleic anhydride H Fumaric acid Maleic anhydride or fumaric acid confers the fundamental unsaturation to the polyester which provides the reactivity with coreactant monomers such as styrene. Maleic anhydride is a crystalline solid melting at 52.6°C (the acid melts at 130°C), while fumaric acid is a solid melting at 284°C. The latter is sometimes preferred to maleic anhydride because it is less corrosive, tends to give lighter colored products, higher impact strength, and slightly greater heat resistance. Phthalic anhydride (melting point 131°C) is most commonly used to play the role of the saturated acid because it provides an inﬂexible link and maintains the rigidity in the cured resin. It is also preferred because its low price enables cheaper resins to be made. Use of isophthalic acid (melting point 347°C) in place of phthalic anhydride yields resins having higher heat distortion temperature and ﬂexural moduli, better craze resistance, and often better water and alkali resistance. These resins are also useful in the preparation of resilient gel coats. Where a ﬂexible resin is required, adipic acid may be used since, unlike the phthalic acids which give a rigid link, adipic acid gives highly ﬂexible link and hence ﬂexibility in the cured resin. Flexible resins are of value in gel coats. COOH O O O Phthalic anhydride HOOC(CH2)4COOH COOH Isophthalic acid Adipic acid Polyester laminating resins are manufactured by heating the component acids and glycols for several hours at 150–200°C in steel reactor vessels with moderate agitation to prevent local overheating and overhead condenser systems to collect the aqueous by-products, initially under ordinary pressure and, in the last phase, under vacuum. To prevent discoloration and premature gelation caused by oxygen, the reactor is continuously purged with nitrogen or carbon dioxide. Reaction systems are fabricated from 304 or 316 stainless steels. Copper and brass valves and ﬁttings are avoided because dissolved copper salts can affect the curing characteristics of the ﬁnal resins. A typical charge for a general purpose resin would be propylene glycol 170 parts, maleic anhydride 132 parts, and phthalic anhydride 100 parts, corresponding to molar ratios of 1.1: 0.67: 0.33. (The slight excess of glycol is primarily to allow for evaporation losses.) The glycols and dibasic acid can be substituted by other components of similar functionality, e.g., propylene glycol by ethylene and diethylene glycols, maleic anhydride by fumaric acid, and phthalic anhydride by isophthalic acid. A fusion-type reator process, also known as fusion-melt process, is commonly used for condensing liquid glycols with dibasic acids of low melting temperature, including phthalic anhydride and aliphatic dibasic acids. The components are added in liquid form to facilitate loading, that is, glycol followed by molten anhydrides, thus saving the heating time needed when solid anhydrides are used. Heat is supplied by hot oil circulating through internal coils. 494 Plastics Technology Handbook The polycondensation reaction occurs according to third-order kinetics with the progressive development of higher molecular weight polymers aproaching an asymptotic limit. This requires extended reaction periods, e.g., 15 h at 190°C for phthalic resins to attain a satisfactory molecular weight. Steps must be taken to free the viscous melt of water, as the latter retards the reaction. At 190°C, some glycol is vaporized and lost with the water of condensation. The reactors may be equipped with fractionating condenser systems to prevent this glycol loss. While the reaction rate is enhanced by acid catalysts, the latter also promote the formation of volatile ethers that are lost as byproducts. A 5% excess glycol added to the initial charge compensates for glycol losses that may occur during the course of the reaction. Addition of glycol during the ﬁnal stages is, however, not recommended. The long reaction time in conventional fusion melt reactors, usually at temperatures above 180°C, causes the maleate ester to isomerize to the corresponding fumarate. This isomerization is of fundamental importance because the fumarate polymers display reactivities almost 20 times more than those of the maleate reaction products in subsequent polymerization reactions with styrene. The rate and extent of isomerization can be promoted at lower temperatures by a cycloaliphatic amine catalyst, e.g., morpholine. The polyesteriﬁcation reaction is followed by measuring the acid number (deﬁned as the number of milligrams of potassium hydroxide equivalent to the acidity present in one gram of resin) of small samples periodically removed from the reactor. Where there are equimolar proportions of glycol and acid in the initial charge, the number average molecular weight is given by 5600/acid number. When the acid number value between 25 and 50 is reached, the heaters are switched off, the reactor is cooled to 150°C, and the contents transferred under vigorous agitation to a blend tank containing suitably inhibited styrene monomer at 30°C. The ﬁnal temperature of the blend reaches about 80°C. At this temperature, even an inhibited styrenated resin polymerizes in several hours unless the blend is rapidly cooled to ambient temperature. A mixture of inhibitors is commonly employed for the styrene diluent in order to obtain a balance of properties in respect of color, storage stability, and gelation rate of catalyzed resin. Thus, a typical composition of the diluent based on the above polyester formulation would be: styrene 172 parts, benzyltrimethylammonium chloride 0.44 part, hydroquinone 0.06 part, and quinone 0.006 part. After cooling to the ambient temperature, the resin is transferred into drums for storage and shipping. Isophthalic acid is widely used as a substitute for phthalic anhydride since, as mentioned earlier, it improves certain properties of the cross-linked polymer. However, isophthalic acid is insoluble in the initial melt charge of maleic anhydride and propylene glycol and it also reacts more slowly with glycols than maleic anhydride. Thus while soluble components in the melt, namely, propylene glycol and maleic anhydride, react to form propylene glycol maleate polymers at lower temperatures (<190°C), isophthalic acid remains inert and requires heating subsequently to temperatures of 240°C to dissolve and react over prolonged periods, thus giving rise to undesirable discoloration of the resin. Since maleic anhydride has a preference to form esters with the more reactive primary hydroxyl group on the propylene glycol molecule, producing propylene glycol maleate esters with a preponderance of terminal secondary hydroxyl groups, subsequent condensation with isophthalic acid proceeds much more slowly than if the available hydroxyl functionality was primary, thus contributing to the slowness of transesteriﬁcation reactions. Furthermore, sufﬁcient unreacted isophthalic acid may be present in the ﬁnal polymer, thus leading to precipitation upon blending with styrene and hence to a hazy resin product. To reduce reaction time and eliminate color problems of one-step processes, as explained above, a twostep process is used to produce isophthalic resins. In the ﬁrst stage, only the isophthalic acid is reacted with the glycol at a relatively high temperature, which may be elevated rapidly to over 220°C without concern for discoloration. Since ethylene glycol and propylene glycol boil at lower temperatures under ordinary pressure, the reactors must be provided with fractionating condensers or operated under pressure to prevent glycol loss. At 220°C, a clear melt is obtained from isophthalic acid and propylene glycol in about 8 h. However, esteriﬁcation catalysts, such as tetrabutyl titanate, stannous oxalate, and dibutyl tin oxide can be used to accelerate the reaction. 495 Industrial Polymers At the end of the ﬁrst stage, the melt is cooled to 150°C, maleic anhydride is added, and the temperature is raised to 180°C to re-start the process of condensation. The progress of the reaction is monitored by the measurement of carboxylic functionality and viscosity. As the reaction proceeds rapidly, the temperature only needs to be elevated to 210°C to drive the reaction. The stability of isophthalic acid esters at higher temperatures allows the development of polymers with higher molecular weight. (In contrast, highmolecular weight phthalic resins cannot be produced by the fusion melt process, since phthalic anhydride sublimes above 200°C.) 4.3.1.4.2 Polyester–Glass–Fiber Laminates (GRP, FRP) Methods of producing FRP laminates with polyesters have been described in Chapter 2. The major process today is the hand layup technique in which the resin is brushed or rolled into the glass mat (or cloth) by hand (see Figure 2.44). Since unsaturated polyesters are susceptible to polymerization inhibition by air, surfaces of the hand layup laminates may remain under-cured, soft, and, in some cases, tacky if freely exposed to air during the curing. A common way of avoiding this difﬁculty is to blend a small amount of parafﬁn wax (or other incompatible material) in with the resin. This blooms out on the surface and forms a protective layer over the resin during cure. For mass production purposes, matched metal molding techniques involving higher temperatures and pressures are employed (see Figure 2.45). A number of intermediate techniques also exist involving vacuum bag, pressure bag, pultrusion, and ﬁlament winding (see Figure 2.46 through Figure 2.48). Glass ﬁbers are the preferred form of reinforcement for polyester resins. Glass ﬁbers are available in a number of forms, such as glass cloth, chopped strands, mats, or rovings (see “Glass Fibers” in Chapter 2). Some typical properties of polyester–glass laminates with different forms of glass reinforcements are given in Table 4.11. it may be seen that laminates can have very high tensile strengths. Being relatively cheaper, polyesters are preferred to epoxide and furan resins for general-purpose laminates. Polyesters thus account for no less than 95% of the low-pressure laminates produced. The largest single outlet is in sheeting for rooﬁng and building insulation. For the greatest transparency of the laminate the refractive indices of glass-cured resins and binder should be identical. The second major outlet is in land transport. Polyester–glass laminates are used in the building of sports car bodies, translucent rooﬁng panel in lorries, and in public transport vehicles. In such applications the ability to construction large polyester–glass moldings without complicated equipment is used to advantage. Polyester resins in conjunction with glass cloth or mat are widely used in the manufacture of TABLE 4.11 Typical Properties of Polyester–Glass Laminates Property Speciﬁc gravity Mat Laminate (Hand Layup) Mat Laminate (Press Formed) Fine Square Woven Cloth Laminate Rod from Rovings 1.4–1.5 1.5–1.8 2.0 2.19 8–17 55–117 18–25 124–173 30–45 210–310 150 1030 10–20 69–138 20–27 138–190 40–55 267–380 155 1100 5 3440 6 4150 10–20 6890–1380 66 45,500 Tensile strength 103 lbf/in.2 MPa Flexural strength 103 lbf/in.2 MPa Flexural modulus 105 1bf/in.2 MPa Dielectric constant (106 Hz) Power factor (106 Hz) Water absorption (%) 3.2–4.5 3.2–4.5 3.6–4.2 – 0.02–0.08 0.2–0.8 0.02–0.08 0.2–0.8 0.02–0.05 0.2–0.8 – – Source: From Brydson, J. A. 1982. Plastics Materials. Butterworth Scientiﬁc, London, UK. 496 Plastics Technology Handbook boat hulls up to 153 ft. (∼46 m) in length. Such hulls are competitive in price and are easier to maintain and to repair [14]. The high strength-to-weight ratio, microwave transparency, and corrosion resistance of the laminates have led to their use in air transport applications as in aircraft radomes, ducting, spinners, and other parts. Land, sea, and air transport applications account for nearly half the polyester resin produced. Other applications include such diverse items as chemical storage vessels, chemical plant components, swimming pools, stacking chairs, trays, and sports equipment. 4.3.1.4.3 Polyester Molding Compositions Four types of polyester molding compounds may be recognized [14]: (1) dough-molding compound (DMC), (2) sheet-molding compound (SMC), (3) alkyd-molding compositions, sometimes referred to as polyester alkyds, (4) diallyl phthalate (DAP) and diallyl isophthalate (DAIP) compounds. Dough-molding compounds of puttylike consistency are prepared by blending resins, catalyst, powdered mineral ﬁller, reinforcing ﬁber (chopped strand), pigment, and lubricant in a dough mixer, usually of the Z-blade type. Formulations for three typical DMC grades are given in Table 4.12. The tendency of thick sections of DMC structural parts to crack has been overcome by using lowproﬁle polyester resins (or low-shrink resins). These are prepared by making a blend of a thermoplastic (e.g., acrylic) polymer–styrene system with a polyester–styrene system. Moldings of this blend cure at elevated temperatures exhibit negligible shrinkage and minimal warpage and have very smooth surface, to which paint may be applied with very little pretreatment. A wide spectrum of properties may be obtained by varying the ratios of thermoplastics, polyester, and styrene in the blend. Among the thermoplastics quoted in the literature for such blending are poly (methyl methacrylate), polystyrene, PVC, and polyethylene. High-gloss DMCs using low-shrink resins have found uses in kitchen appliances such as toaster end plates, steam iron bases, and casings for electric heaters. Since the manufacture of DMC involves intensive shear that causes extensive damage to ﬁbers, DMC moldings have less strength than GRP laminates. This problem is largely avoided with the sheet-molding compounds (SMC). In the SMC process, unsaturated polyester resin, curing systems, ﬁller thickening agents, and lubricant are blended together and coated onto two polyethylene ﬁlms. Chopped-glass rovings are supplied between the resin layers, which are then sandwiched together and compacted as shown in Figure 4.13. Thickening occurs by the reaction of free carboxyl end groups with magnesium oxide. This converts the soft, sticky mass to a handleable sheet, which takes usually a day or two. A typical formulation consists of 30% chopped-glass ﬁber, 30% ground limestone and resin. For molding, the sheet may be easily cut to the appropriate weight and shape and placed between the halves of the heated mold. The main applications of SMCs are in car parts, baths, and doors. The polyester “alkyd” molding compositions are also based on a polyester resin similar to those used for laminating. (The term alkyd is derived from alcohol and acid.) They are prepared by blending the resin TABLE 4.12 Typical Formulations of DMC Grades Ingredients Polyester resin E glass (1/4-in. length) E glass (1/2-in. length) Sisal Calcium carbonate Low-Cost General-Purpose High-Grade Mechanical High-Grade Electrical 100 20 100 – 100 90 – 85 – 40 240 – 150 – – Benzoyl peroxide 1 1 1 Pigment Calcium stearate 2 2 2 2 2 2 Source: From Brydson, J. A. 1982. Plastics Materials. Butterworth Scientiﬁc, London, UK. 497 Industrial Polymers Let-oﬀ for polyethylene ﬁlm Resin spreader Roll-up for SMC Rovings cutter Resin spreader Compaction rolls Let-oﬀ for polyethylene ﬁlm FIGURE 4.13 Schematic outline of machine used for sheet-molding compounds. with cellulose pulp, mineral ﬁller, pigments, lubricants, and peroxide curing systems on hot mills to the desired ﬂow properties. The mix is then removed, cooled, crushed, and ground. Diallyl phthalate (DAP) is a diester of phthalic acid and allyl alcohol and contains two double bonds. O C O CH2 CH CH2 C O CH2 CH CH2 O On heating with a peroxide, DAP therefore polymerizes and eventually cross-links, forming an insoluble network polymer. However, it is possible to heat the DAP monomer under carefully controlled conditions, to give a soluble and stable partial polymer in the form of a white powder. The powder may then be blended with peroxide catalysts, ﬁllers, and other ingredients to form a molding powder in the same manner as polyester alkyds. Similar products can be obtained from diallyl isophthalate (DAIP). Both DAP and DAIP alkyd moldings are superior to the phenolics in their tracking resistance and in their availability in a wide range of colors; however, they tend to show a higher shrinkage on cure. The DAIP materials are more expensive than DAP but have better heat resistance. They are supposed to be able to withstand temperatures as high as 220°C for long periods. The polyester alkyd resins are cheaper than the DAP alkyd resins but are mechanically weaker and do not maintain their electrical properties as well under severe humid conditions. Some pertinent properties of the polyester molding composition as compared in Table 4.13 along with those of a GP phenolic composition. The alkyd molding compositions are used almost entirely in electrical applications where the cheaper phenolic and amino resins are not suitable. 4.3.1.5 Aromatic Polyesters Monomers p-Hydroxy benzoic acid, bisphenol A, diphenyl isophthalate Polymerization Major Uses Bulk polycondensation High-temperature engineering thermoplastics, plasma coatings, abradable seals 498 Plastics Technology Handbook TABLE 4.13 Properties of Thermosetting Polyester Moldings Property Molding temperature (°C) Phenolic (GP) DMC (GP) Polyester Alkyd DAP Alkyd DAIP Alkyd 150–170 140–165 140–165 150–165 150–165 Cure time (cup ﬂow test) (sec) 60–70 25–40 20–30 60–90 60–90 Shrinkage (cm/cm) Speciﬁc gravity 0.007 1.3 0.004 2.0 0.009 1.7 0.009 1.6 0.006 1.8 Impact strength (ft.-lb) 0.12–0.2 2.0–4.0 0.13–0.18 0.12–0.18 0.09–0.13 Volume resistivity (ohm-m) Dielectric constant (106 Hz) 1012–1014 4.5–5.5 1016 5.6–6.0 1016 4.5–5.0 1016 3.5–5.0 1016 4.0–6.0 Dielectric strength (90°C) (kV/cm) Power factor (106 Hz) Water absorption (mg) 39–97 78–117 94–135 117–156 117–156 0.03–0.05 45–65 0.01–0.03 15–30 0.02–0.04 40–70 0.02–0.04 5–15 0.04–0.06 10–20 Source: Brydson, J. A. 1982. Plastics Materials. Butterworth Scientiﬁc, London, UK. In the 1960s the Carborundum Company introduced the homopolymer of p-hydroxybenzoic acid under the trade name Ekonol [35]. It is used in plasma coating. This wholly aromatic homopolyester is produced in practice by the self-ester exchange of the phenyl ester of p-hydroxybenzoic acid. HO C O C6H5 –C6H5OH O C O O n The homopolyester (mol. wt. 8000–12,000) is insoluble in dilute acids and bases and all solvents up to their boiling points. It melts at about 500°C and is difﬁcult to fabricate. It can be shaped only by hammering (like a metal), by impact molding and by pressure sintering (420°C at 35 MPa). The difﬁculty in fabrication has severely limited the wider application of these polymers. The homopolyester is available as a ﬁnely divided powder in several grades, based on particle size. The average particle size ranges from 35 to 80 mm. The material can be blended with various powdered metals, such as bronze, aluminum, and nickel-chrome, and is used in ﬂame-spray compounds. Plasma-sprayed coatings are thermally stable, self-lubricating, and were and corrosion resistant. Applications include abradable seals for jet aircraft engine parts. The polymer can also be blended up to 25% with PTFE. Such blends have good temperature and wear resistance and are self-lubricating. Applications include seals, bearings, and rotors. Copolymeric aromatic polyesters, though possessing a somewhat lower level of heat resistance are easier to fabricate than are the wholly aromatic polymers; they also possess many properties that make them of interest as high-temperature materials. These materials, called polyarylates, are copolyester of terephthalic acid, and bisphenol A in the ratio of 1:1:2. CH3 O C CH3 CH3 O C O C O O C CH3 O C C O O The use of two isomeric acids leads to an irregular chain which inhibits crystallization. This allows the polymer to be processed at much lower temperatures than would be possible with a crystalline homopolymer. Nevertheless the high aromatic content of these polyesters ensures a high Tg (∼90°C). The polymer is self-extinguishing with a limiting oxygen index of 34 and a self-ignition temperature of 545°C. The heat-deﬂection temperature under load (1.8 MPa) is about 175°C. Industrial Polymers 499 Among other distinctive properties of the polyarylate are its good optical properties (luminous light transmission 84–88% with 1–2% haze, refractive index 1.61), high impact strength between that polycarbonate and polysulfone, exceptionally high level of recovery after deformation (important in applications such as clips and snap fasteners), good toughness at both elevated and low temperatures with very little notch sensitivity, and high abrasion resistance which is superior to that of polycarbonates. Polyarylates weatherability and ﬂammability (high oxygen index, low ﬂame spread) are inherent and are achieved without additives. The weatherability properties therefore do not deteriorate signiﬁcantly with time. (Tests show that over 5000 h of accelerated weathering results in virtually no change in performance with respect to luminous light transmittance, haze, gloss, yellowness, and impact.) Having no ﬂame-retardant additives, the combustion products of polyarylate are only carbon dioxide, carbon monoxide, and water, with no formation of toxic gas. Several companies have marketed polyarylates under the trade names: U-polymer (Unitika of Japan), Arylef (Solvay of Belgium), Ardel (Union Carbide), and Arylon (DuPont). These are noncrystallizing copolymers of mixed phthalic acids with a bisphenol and have repeat units of the type shown above. They are melt processable with Tg and heat distortion temperatures in the range of 150–200°C and have similar mechanical properties to polycarbonate and polyethersulfones (see later). The polymers are useful for electrical and mechanical components that require good heat resistance and for lighting ﬁxtures and consumer goods that operate at elevated temperatures, such as microwave ovens and hair dryers. Potential uses of the somewhat cheaper arylon type polyesters include exterior car parts, such as body panels and bumpers. Typical properties of aromatic polyesters mentioned above are shown in Table 4.14. A different approach to obtaining polymers with good melt processability coupled with high softening point has led to another type of aromatic copolyester, the so-called liquid crystalline polymers (LCPs) (see Chapter 5). In these polymers, marketed under the trade names Vectra (Celanese) and Xydar (Dartco Manufacturing), the retention of liquid crystalline order in the melt gives lower melt viscosities than would otherwise be achieved. Heat distortion temperatures are also in the high range of 180–240°C (Table 4.14). LCPs have thus heralded a new era of readily molded engineering and electrical parts for high temperature use. 4.3.1.6 Wholly Aromatic Copolyester A high-performance, wholly aromatic copolyester suitble for injection molding was commercial introduced in late 1984 by Dartco Manufacturing under the trade name Xydar. Xydar injection-molding resins are based on terephthalic acid, p,p′-dihydroxybiphenyl, and p-hydroxybenzoic acid. Polymers of this class contain long relatively rigid chains which are thought to undergo parallel ordering in the melt, resulting in low melt viscosity and good injection-molding characteristics, although at relatively high melt temperatures—750°F to 806°F (400–430°F) (400–430°C). The melt solidiﬁes to form tightly packed ﬁbrous chains in the molded parts, which give rise to exceptional physical properties. The tensile modulus of the molded unﬁlled resin is 2.4 × 106 psi (16,500 MPa) at room temperature and 1.2 × 106 psi (8300 MPa) at 575°F (300°C) tensile strength is about 20,000 psi (138 MPa), compressive strength is 6000 psi (41 MPa), and elongation is approximately 5%. Mechanical properties are claimed to improve at subzero temperatures. The wholly aromatic copolyester is reported to have outstanding thermal oxidative stability, with a decomposition temperature in air of 1040 F (560°C) and 1053°F (567°C) in a nitrogen atmosphere. The resin is inherently ﬂame retardant and does not sustain combustion. Its oxygen index is 42, and smoke generation is extremely low. The resin is extremely inert, resists attack by virtually all chemicals, including acids, solvents, boiling water, and hydrocarbons. It is attacked by concentrated, boiling caustic but is unaffected by 30 days of immersion in 10% sodium hydroxide solution at 127°F (53°C). It withstands a high level of UV radiation and is transparent to microwaves. 500 Plastics Technology Handbook TABLE 4.14 Properties of Unﬁlled Aromatic Polyesters Property Ekonol Tensile strength 103 lbf/in.2 MPa 11 10 10 20–35 70 (yield) 70 140–240 – 50 25 1.6–7 – 3.0 2.9 14–58 – 2.1 2.0 10–40 105 lbf/in.2 GPa Flexural modulus GPa Heat distortion temperature (°C) Vectra (Range for Various Grades) Arylon 74 (ﬂexural) Elongation at break (%) Tensile modulus 105 lbf/in.2 Arylef or Ardel 10 2.9 3.0 14–51 7.1 >550 2.0 175 2.1 155 10–35 180–240 – – 2.8–4.7 150–250 5.4 288 1–10 53–530 – 34 26 35–50 Impact strength, notched Izod ft.-lbf/in. J/m Limiting oxygen index (%) The wholly aromatic copolyester for injection molding is available in ﬁlled and unﬁlled grades. It can be molded into thin-wall components at high speeds. The high melt ﬂow also enables it to be molded into heavy-wall parts. No mold release is required because of the inherent lubricity and nonstick properties. No post-curing is necessary because the material is completely thermoplastic in nature. The material is expected to have many applications because of its moldability and its resistance to high temperatures, ﬁre, and chemicals. 4.3.1.7 Polycarbonates CH3 O C O C O CH3 Monomers Bisphenol A, phosgene n Polymerization Major Uses Interfacial polycondensation, solution polycondensation, transesteriﬁcation Glazing (37%), electrical and electronics (15%, appliances (15%), compact discs The major processes for polycarbonate manufacture include (1) transesteriﬁcation of bisphenol A with diphenyl carbonate [36,37]: CH3 n HO C OH + n O C O ~300°C Low pressure O CH3 CH3 2n OH + O C CH3 O C O n 501 Industrial Polymers (2) solution phosgenation in the presence of an acid acceptor such as pyridine: CH3 n HO C OH + n CI C CI CH2CI2 soln. Pyridine O CH3 CH3 O C CH3 O C O + 2nHCI n and (3) interfacial phosgenation in which the basic reaction is the same as in solution phosgenation, but it occurs at the interface of an aqueous phase and an organic phase. Here the acid acceptor is an aqueous phase and an organic phase. Here the acid acceptor is aqueous sodium hydroxide, which dissolves the bisphenol A and a monohydric phenol used for molecular-weight control (without which very highmolecular-weight polymers of little commercial value will be obtained), and the organic phase is a solvent for phosgene and the polymer formed. A mixture of methylene chloride and chlorobenzene is a suitable solvent. The interfacial polycondensation method is the most important process at present for the production of polycarbonate. Interestingly, polycarbonate represents the ﬁrst commercial application of interfacial polycondensation. Fire-retardant grades of polycarbonates are produced by using tetrabromobisphenol A as comonomer. Polycarbonate resin is easily processed by all thermoplastic-molding methods. Although it is most often injection molded or extruded into ﬂat sheets, other options include blow molding, proﬁle extrusion, and structural foam molding. Polycarbonate sheet can be readily thermoformed. The resin should be dried to less than 0.02% moisture before processing to prevent hydrolytic degradation at the high temperatures necessary for processing. The chemical resistance of polyester materials is generally limited due to the comparative ease of hydrolysis of the ester groups, but the bisphenol A polycarbonates are somewhat more resistant. This resistance may be attributed to the shielding of the carbonate group by the hydrophobic benzene rings on either side. The resin thus shows resistance to dilute mineral acids; however, it has poor resistance to alkali and to aromatic and chlorinated hydrocarbons. Polycarbonates have an unusual combination of high impact strength (12–16 ft.-lbf per inch notch for 1/2-in. × 1/8-in. bar), heat-distortion temperature (132°C), transparency, very good electrical insulation characteristics, virtually self-extinguishing nature, and physiological inertness. As an illustration of the toughness of polycarbonate resins, it is claimed that an 1/8-in.-thick molded disc will stop a 22 caliber bullet, causing denting but not cracking. In creep resistance, polycarbonates are markedly superior to acetal and polyamide thermoplastics. Because of a small dipole polarization effect, the dielectric constant of polycarbonates (e.g., 3.0 at 103 Hz) is somewhat higher than that for PTFE and the polyoleﬁns (2.1–2.5 at 103 Hz). The dielectric constant is also almost unaffected by frequency changes up to 106 Hz and temperature changes over the normal range of operations. (Note that for satisfactory performance electrical insulating materials should have a low dielectric constant for low dissipation factor but high dielectric strength. For dielectrics used in capacitors, however, a high dielectric constant is desirable.) At low frequencies (60 Hz) and in the ordinary temperature range (20–100°C), the power factor of polycarbonates (∼0.0009) is remarkably low for a polar polymer. It increases, however, at higher frequencies, reaching a value of 0.010 at 106 Hz. The polycarbonates have a high volume resistivity (2.1 × 1020 ohm-cm at 23°C) and a high dielectric strength (400 kV/in., 1/8-in. sample). Because of the low water absorption, these properties are affected little by humidity. Polycarbonates, however, do have a poor resistance to tracking. 502 Plastics Technology Handbook Although the electrical properties of polycarbonates are not as impressive as those observed with polyethylene, they are adequate for many purposes. These properties, coupled with the high impact strength, heat and ﬂame resistance, transparency, and toughness have led to the extensive use of these resins in electronics and electrical engineering, which remains the largest single ﬁeld of their application. Polycarbonate is the only material that can provide such a combination of properties, at least at a reasonable cost. Known for many years, epoxy oligomers made from tetrabromobisphenol A are still used as the ﬂame retardant in polycarbonates because they minimally affect the heat distortion temperature and even show a positive effect on impact strength. About 6–9 wt% of the epoxy oligomer is required for achieving V-0 rating and a thermotropic liquid crystal polyester helps to improve melt ﬂow, so that thin-walled parts can be molded [38]. Antimony trioxide is not normally used in combination with halogen-containing additives in PC, because it causes loss of clarity. At General Electric, it was found that very low additions (<1 wt%) of alkali or alkaline earth metal salts of certain arylsulfonates provide self-extinguishing performance to PC. Potassium diphenylsulfone sulfonate, sodium trichlorobenzene sulfonate, and potassium perﬂuorobutane sulfonate are effective in PC at one-tenth of a percent level and these salts are used on a commercial scale. The salts are mostly active in the condensed phase where they strongly destabilize PC upon heating, thus promoting fast decomposition and melt ﬂow which removes heat. Phosphate esters are rarely used in plain PC because of partial loss of clarity, tendency to stresscracking, and somewhat reduced hydrolytic stability. However, aromatic phosphates are currently the products of choice for ﬂame-retarding PC-based blends [39]. Triphenyl phosphate and mixed tri (t-butylphenyl phenyl) phosphate are reasonably effective in PC/ABS blends and are used commercially, though they have the disadvantage of relatively high volatility. However, bridged aromatic diphenyl phosphates, especially resorcinol bis(diphenyl phosphate) and bisphenol A bis(diphenyl phosphate), have found much broader application than monophosphates because of good thermal stability, high efﬁciency, and low volatility. Nano-scale inorganic materials have been shown to improve ﬁre-retardant performance of aromatic phosphates and provide enhanced thermal dimensional stability for PC/ABS blends. Polycarbonate covers for time switches, batteries, and relays utilize the god electrical insulation characteristics in conjunction with transparency, toughness, and ﬂame resistance of the polymer. Its combination of properties also accounts for its wide use in making coil formers. Many other electrical and electronic applications include moldings for computers, calculating machines and magnetic disc pack housing, contact strips, switch plates, and starter enclosures for ﬂuorescent lamps. Polycarbonate ﬁlms of high molecular weight are used in the manufacture of capacitors. Traditional applications of polycarbonate in the medical market, such as ﬁlter housings, tubing connectors, and surgical staplers, have relied on the materials unique combination of strength, purity, transparency, and ability to stand all sterilization methods (steam, ethylene oxide gas, and gamma radiation). Polycarbonate-based blends blends and copolymers have further extended the materials usefulness to medical applications. Recent years have seen a continuing growth of the market for polycarbonate glazing and light transmission units. Applications here include lenses and protective domes as well as glazing. The toughness and transparency of polycarbonates have led to many successful glazing applications of the polymer, such as bus shelters, telephone kiosks, gymnasium windows, lamp housings for street lighting, trafﬁc lights, and automobiles, strip-lighting covers at ground level, safety goggles, riot-squad helmets, armor, and machine guards. The limited scratch and weathering resistance of the polycarbonates is a serious drawback in these applications, and much effort is being directed at overcoming these problems. One approach is to coat the polycarbonate sheet with a glasslike composition by using a suitable priming material (e.g., Margard, marketed by the General Electric Company) to ensure good adhesion between coating and the base plastic. Polycarbonates modiﬁed with ABS (acrylonitrile and styrene grafted onto polybutadiene) and MBS (methyl methacrylate and styrene grafted onto polybutadiene) resins have been available for many years. Usually used to the extent of 2–9%, the styrene-based terpolymers are claimed to reduce the notch sensitivity of the polycarbonate and to improve its resistance to environmental stress cracking while retaining 503 Industrial Polymers from some grades the high impact strength of the unmodiﬁed polycarbonate. These materials ﬁnd use in the electrical industry, in the automotive industry (instrument panels and glove compartment ﬂaps), and for household appliances (coffee machine housings, hair drier housings, and steam handles). Elastomer modiﬁed polycarbonates have been used for automobile front ends and bumpers (e.g., 1982 Ford Sierra). Polycarbonate is used for making compact audio discs, which are based on digital recording and playback technology and can store millions of bits of information in the form of minute pits in an area only a few inches in diameter. It is the presence or absence of the pits that is read by the laser. Each “track” comprising a spiral of these pits is laid in polycarbonate which is backed with reﬂective aluminum and coated with a protective acrylic layer. The processability of the polycarbonate, or any other material used as the substrate, is crucial in the manufacture of all optical discs. Bayers polycarbonate grade Makrolon CD-2000 has been specially developed to ﬁt such requirements. 4.3.2 Polyamides The early development of polyamides started with the work of W.H. Carothers and his colleagues, who, in 1935, ﬁrst synthesized nylon-6,6—a polyamide of hexamethylene diamine and adipic acid—after extensive and classical researches into condensation polymerization. Commercial production of nylon-6,6 and its conversion into ﬁbers was started by the DuPont Company in 1939. In a parallel development in Germany, Schlack developed polyamides by ring-opening polymerization of cyclic lactams, and nylon-6 derived from caprolactam was introduced in 1939. Today nylon-6,6 and nylon-6 account for nearly all of the polyamides produced for ﬁber applications. Nylon-6,6 and nylon-6 are also used for plastics applications. Besides these two polyamides, very many other aliphatic polyamides, have been prepared in the laboratory, and a few of them (nylon-11, nylon-12, and nylon-6,10 in particular) have attracted specialized interest as plastics materials. However, only about 10% of the nylons produced are used for plastics production. Virtually all of the rest goes for the production of ﬁbers where the market is shared, roughly equally, between nylon-6 and nylon-6,6. (Nylon is the trade name for the polyamides from unsubstituted, non-branched aliphatic monomers. A polyamide made from either an amino acid or a lactam is called nylon-x, where x is the number of carbon atoms in the repeating unit. A nylon made from a diamine and a dibasic acid is designated by two numbers, in which the ﬁrst represents the number of carbons in the diamine chain and the second the number of carbons in the dibasic acid.) For a variety of technical reasons the development of aromatic polyamides was much slower in comparison. Commercially introduced in 1961, the aromatic polyamides have expanded the maximum temperature well above 200°C. High-tenacity, high-modulus polyamide ﬁbers (aramid ﬁbers) have provided new levels of properties ideally suited for tire reinforcement. More recently there has been considerable interest in some new aromatic glassy polymers, in thermoplastic polyamide elastomers, and in a variety of other novel materials. 4.3.2.1 Aliphatic Polyamides C NH(CH2)6NH O Nylon-6,6 NH(CH2)5 C O n Nylon-6 (CH2)4 C O n 504 Plastics Technology Handbook Monomers Adipic acid, hexamethylenediamine, caprolactam Polymerization Major Uses Bulk polycondensation Home furnishings, apparel, tire cord Aliphatic polyamides are produced commercially by condensation of diamines with dibasic acids, by self-condensation of an amino acid, or by self-condensation of an amino acid, or by ring-opening polymerization of a lactam [14,40,41]. To obtain polymers of high molecular weight, there should be stoichiometric equivalence of amine and acid groups of the monomers. For amino acids and lactams the stoichiometric balance is ensured by the use of pure monomers; for diamines and dibasic acids this is readily obtained by the preliminary formation of a 1:1 ammonium salt, often referred to as a nylon salt. Small quantities of monofunctional compounds are often used to control the molecular weight. The nylon-6,6 salt (melting point 190–191°C) is prepared by reacting hexamethylenediamine and adipic acid in boiling methanol, so that the comparatively insoluble salt precipitates out. A 60% aqueous slurry of the salt together with a trace of acetic acid to limit the molecular weight to the desired level (9000–15,000) is heated under a nitrogen blanket at about 220°C in a closed autoclave under a pressure of about 20 atmospheres (atm). The polymerization proceeds to approximately 80–90% without removal of by-product water. The autoclave temperature is then raised to 270–300°C, and the steam is continuously driven off to drive the polymerization to completion. n nH2N(CH2)6NH2 + nHO2C(CH2)4CO2H NH (CH2)6 NH C O (CH2)4 C – O2C(CH2)4CO2– +H N(CH ) NH + 3 26 3 OH (2n + I)H2O O n The later stages of polymerization reaction constitute a melt polycondensation, since the reaction temperature is above the melting point of the polyamide. The molten polymer is extruded by nitrogen pressure on to a water-cooled casting wheel to form a ribbon which is subsequently disintegrated. In a continuous process for the production of nylon-6,6 similar reaction conditions are used, but the reaction mixture moves slowly through various zones of a reactor. Nylon-6,10 is prepared from the salt (melting point 170°C) of hexamethylenediamine and sebacic acid by a similar technique. Nylon-6,9 uses azelaic acid. Decane-1,10-dicarboxylic acid is used for nylon-6,12. In a typical batch process for the production of nylon-6 by ring-opening polymerization, a mixture of caprolactam, water (5–10% by weight), which acts as a catalyst, and a molecular-weight regulator [e.g., acetic acid (∼0.1%)] is heated in a reactor under a nitrogen blanket at 250°C for above 12 h, a pressure of about 15 atm being maintained by venting off steam. The product consists of high-molecular-weight polymer (about 90%) and low-molecular-weight material (about 10%), which is mainly monomer. To obtain the best physical properties, the low-molecular-weight materials may be removed by leaching and/ or by vacuum distillation. In the continuous process, similar reaction conditions are used. In one process a mixture of molten caprolactam, water, and acetic acid is fed continuously to a reactor operating at about 260°C. The residence time is 18–20 h. A simpler technique for the preparation of nylon-6 is the polymerization casting of caprolactam in situ in the mold. In this process rapid formation of polymer is achieved by anionic polymerization, initiated by strong bases such as metal amides, metal hydrides, and alkali metals. However, the anionic polymerization of lactams by strong bases alone is relatively slow because it is associated with an induction period due to a Industrial Polymers 505 slow step in the initiation sequence leading to an N-acyl lactam which participates in the propagation reaction. The induction period may, however, be eliminated by adding along with the strong base a preformed N-acyl lactam or related compound at the start of the reaction. A typical system for polymerization casting of caprolactam thus uses as a catalyst 0.1–1 mol% N-acetyl caprolactam and 0.15–0.50 mol% of the sodium salt of caprolactam. The reaction temperature is initially about 150°C, but during polymerization it rises to about 200°C. The technique is especially applicable to the production of large, complex shapes that could not be made by the more conventional plastics processing techniques. Important advantages of the process are the low heats and low pressures involved. Although the polymerization process is exothermic, the relatively low heat of polymerization of caprolactam, coupled with its low melting point, makes the process easy to control and simpliﬁes the heat transfer problem generally associated with the production of massive parts. Moldings of cast nylon-6 up to 1 tn are claimed to have been produced by these techniques. Nylon parts made by polymerization casting of caprolactam exhibit higher molecular weights and a highly crystalline structure and are, therefore, slightly harder and stiffer than conventionally molded nylon-6. Applications for cast nylon-6 include hug gears (e.g., a 150-kg nylon gear for driving a large steel drum drier) and bearings, gasoline and fuel tanks, buckets, building shutters, and various components for paper production machinery and mining and construction equipment. Later development has centered on adding reinforcing materials to the monomer before polymerization to produce parts with higher heat distortion temperature, impact strength and tensile strength. Nylon-12 is produced by the ring-opening polymerization of laurolactam (dodecyl lactam) such as by heating the lactam at about 300°C in the presence of aqueous phosphoric acid. Unlike the polymerization of caprolactam, the polymerization of dodecyl lactam does not involve an equilibrium reaction. Hence, an almost quantitative yield of nylon-12 polymer is obtained by the reaction, and the removal of lowmolecular-weight material is unnecessary. Nylon-11 is produced by the condensation polymerization of w-aminoundecanoic acid at 200–220°C with continuous removal of water. The latter stages of the reaction are conducted under reduced pressure to drive the polymerization to completion. Nylon copolymers can be obtained by heating a blend of two or more different nylons above the melting point so that amide interchange occurs. Initially, block copolymers are formed, but prolonged reaction leads to random copolymers. For example, a blend of nylon-6,6 and nylon-6,10 heated for 2 h gives a random copolymer (nylon-6,6–nylon-6,10) which is identical with a copolymer prepared directly from the mixed monomers. Other copolymers of this type are available commercially. 4.3.2.1.1 Properties Aliphatic polyamides are linear polymers containing polar –CONH– groups spaced at regular intervals by aliphatic chain segments. The principal structural difference between the various types of nylon is in the length of aliphatic chain segments separating the adjacent amide groups. The polar amid groups give rise to high interchain attraction in the crystalline zones, and the aliphatic segments impart a measure of chain ﬂexibility in the amorphous zones. This combination of properties yields polymers which are tough above their glass transition temperatures. The high intermolecular attraction also accounts for high melting points of nylons, which are usually more than 200°C. The melting point, however, decreases (which facilitates processing) as the length of the aliphatic segment in the chain increases, as indicated in Table 4.15. Because of the high cohesive energy and their crystalline state, the nylons are resistant to most solvents. They have exceptionally good resistance to hydrocarbons and are affected little by esters, alkyl halides, and glycols. There are only a few solvents for the nylons, of which the most common are formic acid, glacial acetic acid, phenols, and cresols. Alcohols generally have some swelling action and may dissolve some copolymers (e.g., nylon-6,6, nylon-6,10, nylon-6). Nylons have very good resistance to alkalis at room 506 Plastics Technology Handbook TABLE 4.15 Melt Temperatures of Aliphatic Polyamides Polyamide Tm (°C) Nylon-6,6 265 Nylon-6,8 240 Nylon-6,10 Nylon-6,12 225 212 Nylon-6 230 Nylon-7 Nylon-11 223 188 Nylon-12 180 temperature. Mineral acids attack nylons, built the rate of attack depends on the nature and concentration of acids and the type of nylon. Nitric acid is generally active at all concentrations. Because of the presence of amide groups, the nylons absorb water. Figure 4.14 shows how the equilibrium water absorption of different nylons varies with humidity at room temperature, and Figure 4.15 shows how the rate of moisture absorption of nylon-6,6 is affected by the environmental conditions. Since dimensional changes may occur as a result of water absorption this effect should be considered when 12 Nylon-6 Moisture content of equlibrium (%) 10 Nylon-6/6,10/6,6 8 Nylon-6,6 6 4 Nylon-6,10 2 Nylon-11 0 50 100 Relative humidity (%) FIGURE 4.14 Effect of relative humidity on the equilibrium moisture absorption of the nylons. Moisture absorption (%) 10 Immersion in water (Room temperature) 8 Immersion in boiling water 6 4 2 0 1 10 Time in days Standing in air (100%RH) Standing in air (65%RH) 100 1,000 FIGURE 4.15 Effect of environmental conditions on rate of moisture absorption of nylon-6,6 (1/8-in.- thick specimens). 507 Industrial Polymers dimensional accuracy is required in a speciﬁc application. Manufacturers commonly supply data on the dimensional changes of their products with ambient humidity. The various types of nylon have generally similar physical properties, being characterized by high toughness, impact strength, and ﬂexibility (Table 4.16). Mechanical properties of nylons are affected signiﬁcantly by the amount of crystallization in the test piece, ambient temperature (Figure 4.16), and humidity (Figure 4.17), and it is necessary to control these factors carefully in the determination of comparative properties. Moisture has a profound plasticizing inﬂuence on the modulus. For example, the Youngs modulus values for nylon-6,6 and nylon-6 decreases by about 40% with the absorption of 2% moisture. Nylons have extremely good abrasion resistance. This property can be further enhanced by addition of external lubricants and by providing a highly crystalline hard surface to the bearings. The surface crystallinity can be developed by the use of hot injection molds and by annealing in a nonoxidizing ﬂuid at an elevated temperature (e.g., 150–200°C for nylon-6,6). The coefﬁcient of friction of nylon-6,6 is lower than mild steel but is higher than the acetal resins. The fractional heat buildup, which determines the upper working limits for bearing applications, is related to the coefﬁcient of friction under working conditions. The upper working limits measured by the maximum LS value (the product of load L in psi on the projected bearing area and the peripheral speed S in ft./min) TABLE 4.16 Comparative Propertiesa of Typical Commercial Grades of Nylon Property 6,6 6 6,10 11 12 6,6/6,10/6 (40:30:30) Speciﬁc gravity 1.14 1.13 1.09 1.04 1.02 1.09 Tensile stress at yield 103 lbf/in.2 11.5 11.0 8.5 5.5 6.6 – 80 76 55 38 45 – 80–100 100–200 100–150 300 200 300 MPa Elongation at break (%) Tension modulus 105 lbf/in.2 4.3 4 3 2 2 2 30 28 21 14 14 14 1.0–1.5 1.5–3.0 1.6–2.0 1.8 1.9 – R118 R112 R111 R108 R107 R83 75 60 55 55 51 30 10−5 cm/cm/°C 10 9.5 15 15 12 – Volume resistivity ohm-m (dry) >1017 >1017 >1017 – – – – 1016 – – 1015 102 MPa Impact strength ft.-lbf/1/2-in. notch Rockwell hardness Heat distortion temperature (264 lbf/in.2) (°C) Coefﬁcient of linear expansion ohm-m (50% RH) Dielectric constant (103 Hz dry) 10 15 3.6–6.0 3.6–6.0 3.6–6.0 – – – Power factor (103 Hz dry) Dielectric strength 0.04 0.02–0.06 0.02 – – – (kV/cm) (25°C, 50% RH) >100 >100 >100 – – – Source: Brydson, J. A. 1982. Plastics Materials. Butterworth Scientiﬁc, London, UK. a ASTM tests for mechanical and thermal properties. 508 Plastics Technology Handbook Nylon-6,6 30 4.0 25 3.0 Nylon-6 20 15 2.0 10 1.0 5 0 10 (a) 20 30 40 Temperature (°C) 50 Young’s modulus (102 MPa) Young’s modulus (105 lbt/in2) 5.0 60 Impact strength (ft-lbf/1/2-in.notch) 10 5 0 (b) 20 40 60 80 Temperature (°C) 100 120 FIGURE 4.16 Effect of temperature on (a) Young’s modulus of nylon-6,6 and nylon 6 and (b) impact strength of nylon-6,6. are 500–1000 for continuous operation of unlubricated nylon-6,6. For intermittent operation initially oiled nylon bearings can be used at LS values of 8000. Higher LS values can be employed with continuously lubricated bearings. The electrical insulation properties of the nylons are reasonably good at room temperature, under conditions of low humidity, and at low frequencies. Because of the presence of polar amide groups, they are not good insulators for high-frequency work, and since they absorb water, the electrical insulation properties deteriorate as the humidity increases (see Figure 4.18). The properties of nylons are considerably affected by the amount of crystallization and by the size of morphological structures, such as spherulites, which in turn, are generally inﬂuenced by the processing conditions. Thus, a molding of nylon-6, slowly cooled and subsequently annealed, may be 50–60% crystalline, whereas a rapidly cooled thin-walled molding may be only 10% crystalline. Slowly cooled melts may form bigger spherulites, but rapidly cooled surface layers may be quite different from that of the more slowly cooled centers. The use of nucleating agents (e.g., about 0.1% of a ﬁne silica) can give smaller spherulites and thus a more uniform structure in an injection molding. Such a product may have greater tensile strength, hardness, and abrasion resistance at the cost of some reduction in impact strength and elongation at break: the higher the degree of crystallinity the less the water absorption, and hence the less will be the effect of humidity on the properties of the polymer. Nylon molding materials are available in a number of grades which many differ in molecular weight and/or in the nature of additives which may be present. The various types of additives used in nylon can be grouped as heat stabilizers, light stabilizers, lubricants, plasticizers, pigments, nucleating agents, ﬂame retarders, and reinforcing ﬁllers. 509 Industrial Polymers 30 Nylon-6,6 4 3 20 2 10 Nylon-6 1 0 (a) 1 Moisture content (%) 2 1 2 Moisture content (%) 3 Young’s modulus (102 MPa) Young’s modulus (105 lbf / in2) 5 Lzod impact strength (ft-bf/ 1/2-in.notch) 5 4 3 2 1 0 (b) FIGURE 4.17 Effect of moisture content on (a) Young’s modulus of nylon-6,6 and nylon-6 and (b) impact strength of nylon-6,6. Volume resistivity (ohm cm) 1015 1014 1013 1012 1011 1010 109 FIGURE 4.18 0 1 2 3 4 5 Water content (%) 6 7 Effect of moisture content on the volume resistivity of nylon-6,6. Heat stabilizers include copper salts, phosphoric acid esters, mercaptobenzothiazole, mercaptobenzimidazole, and phenyl-b-naphthyl-amine. Among light stabilizers are carbon black and various phenolic materials. Self-lubricating grades of nylon which are of value in some gear and bearing applications incorporate lubricants such as molybdenum disulﬁde (0.2%) and graphite (1%). Plasticizers may be added to nylon to lower the melting point and to improve toughness and ﬂexibility particularly at low temperatures. A plasticizer used commercially is a blend of o- and p-toluene ethyl sulfonamide. Substances used as nucleating agents include silica and phosphorus compounds. Nucleating agents are used to control the size of morphological structures of the molding. 510 Plastics Technology Handbook There have been substantial efforts to improve the ﬂame resistance of nylons. Various halogen compounds (synergized by zinc oxide or zinc borate) and phosphorus compounds have been used (see the section on Flame Retardation in Chapter 1). They are, however, dark in color. Glass-reinforced nylons have become available in recent years. Two main types of glass ﬁllers used are glass ﬁbers and glass beads. From 20% to 40% glass in used. Compared to unﬁlled nylons, glass-ﬁber reinforcement leads to a substantial increase in tensile strength (160 vs. 80 MPa), ﬂexural modulus (8000 vs. 3000 MPa), hardness, creep resistance (at least three times as great), and heat-distortion temperature under load (245 vs. 75°C under 264 psi), and to a signiﬁcant reduction in coefﬁcient of expansion (2.8 × 10−5 vs. 9.9 × 10−5 cm/cm-°C). The glass-ﬁber-ﬁlled types can be obtained in two ways. One route involves passing continuous lengths of glass ﬁber (as rovings) through a polymer melt or solution to produce glass-reinforced nylon strand that is chopped into pellets. Another route involves blending a mixture of resin and glass ﬁbers about 1/4 in. (0.6 cm) long in an extruder. Usually E-grade glass with a diameter of about 0.001 cm treated with a coupling agent, such as a silane, to improve the resin-glass bond is used. Nylons ﬁlled with 4.0% glass spheres have a compressive strength about eightfold higher than unﬁlled grades, besides showing good improvement in tensile strength, modulus, and heat-distortion temperature. Having low melt viscosity, glass-bead-ﬁlled nylons are easier to process than the glass-ﬁber-ﬁlled varieties. They are also more isotropic in their mechanical properties and show minimum warpage. Glass ﬁllers, both ﬁbers and beads, tend to improve self-extinguishing characteristics of nylons. 4.3.2.1.2 Applications The most important application of nylons is as ﬁbers, which account for nearly 90% of the world production of all nylons. Virtually all of the rest is used for plastic applications. Because of their high cost, they have not become general-purpose materials, such as polyethylene and polystyrene, which are available at about one-third the price of nylons. Nylons have nevertheless found steadily increasing application as plastics materials for specialty purposes where the combination of toughness, rigidity, abrasion resistance, reasonable heat resistance, and gasoline resistance is important. The largest plastics applications of nylons have been in mechanical engineering [14]—nylon-6, nylon6,6, nylon-6,10, nylon-11, and nylon-12 being mainly used. These applications include gears, cams, bushes, bearings, and valve seats. Zippers made of nylon last longer than traditional ones of fabric or metal. Nylon moving parts have the advantage that they may often be operated without lubrication, and they may often be molded in one piece. Among the aforesaid nylons, nylon-11 and nylon-12 have the lowest water absorption and are easy to process, but there is some loss in mechanical properties. For the best mechanical properties, the nylon-6,6 would be considered, but this material is also the most difﬁcult to process and has high water absorption. Nylon-6 is easier to process but has even higher water absorption (see Figure 4.14). Other applications include sterilizable nylon moldings in medicine and pharmacy, nylon hair combs, and nylon ﬁlm for packaging foodstuffs (a typical example being milk pouches made of coextruded multilayered ﬁlms of LDPE/LLDP/nylon-6, with nylon-6 as the barrier layer) and pharmaceutical products. The value of nylon in these latter applications is due to its low odor transmission and the boilin-the-bag feature. Nylons have reasonable heat resistance. Spatula blades and spoons of nylon-6,6 withstand highest cooking temperatures. Besides ﬁlm, other extruded applications of nylons are as monoﬁlaments, which have found applications in surgical sutures, brush tufting, wigs, sports equipment, braiding, outdoor upholstery, and angling. Production of moldings by polymerization casting of caprolactam and the ability to produce large objects in this way have widened the use of nylon plastics in engineering and other applications. The process gives comparatively stress-free moldings having a reasonably consistent morphological structure with a 45–50% crystallinity, which is higher than melt-processed materials, and thus leads to higher tensile strength, modulus, hardness, and resistance to creep. Products made by polymerization casting 511 Industrial Polymers include main drive gears for use in the textile and papermaking industries, conveyor buckets used in the mining industry, liners for coal-washing equipment, and propellers for small marine craft. Glass-ﬁlled nylons form the most important group of glass-ﬁlled varieties of thermoplastics. Glassreinforced nylon plastics have high rigidity, excellent creep resistance, low coefﬁcient of friction, high heat-deﬂection temperature, good low-frequency electrical insulation properties, and they are nonmagnetic in nature. Therefore they have replaced metals in many applications. Nylons reinforced with glass ﬁbers are thus widely used in domestic appliances, in housings and casing, in car components, including radiator parts, and in the telecommunication ﬁeld for relay coil formers and tag blocks. Glass-bad-ﬁlled nylons have found use in bobbins. Carbon-ﬁber reinforcement has been used with nylon-6 and nylon-6–nylon-12 mixtures. These materials have found use in the aerospace ﬁeld and in tennis rackets. A signiﬁcant development is the appearance of supertough nylon plastics, which are blends in nylon6,6 with other resins, such as an ionomer resin used in the initial grades or a modiﬁed ethylene– propylene–diene terpolymer rubber (EPDM rubber) used in later grades. Nylon has been blended with PPO (Vydyne, Noryl GTX), PC, HDPE (Selar), PP, SAN, ABS (Triax, Elemid), PBT (Bexloy), and polyarylates (Bexlar). These blends have lower water absorption than nylon6,6. Nylon-SAN, which has a high impact strength of 16 ft.-lb/in. of notch (854 J/m), has received a UL (Underwriters Laboratory) rating of 104°C. The nylon-arylate blend is transparent and has a heatdistortion temperature under load (264 lbf/in.2) value of 154°C. Interest has been aroused by the appearance of novel elastomeric polyamides. The products introduced by Huls under the designation XR3808 and X4006 may be considered as the polyether-amide analogue of the polyether-ester thermoplastic elastomers introduced in the 1970s by DuPont as Hytrel (see Figure 4.10). The polyether-amide is a block copolymer prepared by the condensation of polytetramethylene ether glycol (i.e., polytetrahydrofuran) with laurin lactam and decane-1,10-dicarboxylic acid. The elastomeric polyamide XR3808 is reported to have a speciﬁc gravity of 1.02, yield stress of 24 MPa, a modulus of elasticity of 300 MPa, and an elongation at break of 360%. 4.3.2.2 Aromatic Polyamides Aromatic polyamide ﬁbers, better known as aramid ﬁbers, have been deﬁned as “a long chain synthetic polyamide in which at least 85% of the amide linkages are attached directly to two aromatic rings [42].” The ﬁrst signiﬁcant material of this type was introduced in 1961 by DuPont as Nomex. It is poly (m-phenyleneisophthalamide), prepared from m-phenylenediamine and isophthaloyl chloride by interfacial polycondensation. H2N NH2 CIC + O CCI HN –HCI O NHC C O O n The ﬁber may be spun from a solution of the polymer in dimethylformamide containing lithium chloride. In 1973, DuPont commenced production of another aromatic polyamide ﬁber, a poly(p-phenylene terephthalamide) marketed as Kevlar. It is produced by the reaction of p-phenylenediamine with terephthaloyl chloride in a mixture of hexametylphosphoramide and N-methyl pyrrolidone (2:1) at −10 C. H2N NH2 + CIC O HN –HCI CCI O NHC C O O n 512 Plastics Technology Handbook Kevlar ﬁbers are as strong as steel but have one-ﬁfth the weight. Kevlar is thus ideally suited as tire cord materials and for ballistic vests. The ﬁbers have a high Tg (>300°C) and can be heated without decomposition to temperatures exceeding 500°C. The dimensional stability of Kevlar is outstanding: It shows essentially no creep or shrinkage as high as 200°C. In view of the high melting temperatures of the aromatic polyamides and their poor solubility in conventional solvents, special techniques are required to produce the ﬁbers. For example, Kevlar is wet spun from a solution in concentrated sulfuric acid. Similar ﬁber-forming materials have been made available by Monsanto. Thus the product marketed as PABH-T X-500 is made by reacting p-aminobenzhydrazide with terephthaloyl chloride. H2N CNHNH2 + CIC CCI –HCI O O O HN CNHNHC C O O O n Polymers have also been prepared from cyclic amines such as piperazine and bis(p-aminocyclohexyl) methane. The latter amine is condensed with decanedioic acid to produce the silklike ﬁber Qiana (DuPont). H2N NH2 + HOC(CH2)10 COH CH2 O O HN CH2 NHC(CH2)10C O O n Qiana ﬁbers have a high glass transition temperature (135°C, as compared to 90°C for nylon-6,6), which assures that the polymer will remain in the glassy state during fabric laundering and resist wrinkles and creases. Synthetic ﬁbers range in properties from low-modulus, high-elongation ﬁbers like Lycra (see Section 4.11) to high-modulus high-tenacity ﬁbers such as Kevlar. A breakthrough in ﬁber strength and stiffness has been achieved with Kevlar. Another high-performance ﬁber in commercial application is graphite. The use of these new ﬁbers has resulted in the development of superior composite materials, generally referred to as ﬁber-reinforced plastics or FRPs (see Chapter 2 and Chapter 3), which have shown promise as metal-replacement materials by virtue of their low density, high speciﬁc strength (strength/density), and high speciﬁc modulus (modulus/density). Today a host of these FRP products are commercially available as tennis rackets, golf clubs shaft, skis, ship masts, and ﬁshing rods, which are ﬁlament wound with graphite and Kevlar ﬁbers. Signiﬁcant quantities of graphite composites and graphite/Kevlar hybrid composites are used in boeing 757 and 767 planes, which make possible dramatic weight saving. Boron, alumina, and silicon carbide ﬁbers are also high-performance ﬁbers but they are too expensive for large-scale commercial applications. Partially aromatic, melt processable, polyamides are produced as random copolymers, which do not crystallize and are therefore transparent, but are still capable of high-temperature use because of their high Tg values. Several commercial polymers of this type that have glass-like clarity, high softening point, and oil and solvent resistance have been developed. For example, Trogamid T (Dynamit Nobel) contains repeat units of CH3 C C O O NH CH2 CH2 CH CH3 CH2 C CH2 CH3 NH 513 Industrial Polymers TABLE 4.17 Properties of Polyamides Property Nylon-6,6 Ixef (High-Impact Grade) Trogamid T Nomex (Fiber) 9.4 25.7 8.7 97.2 435 65 177 60 670 3000 4.6 19.4 4.4 25.5 194 3.2 13.4 3.0 17.6 134 100 2.7 132 22 2.6 4.8 14.8 – – – 3.3 10.2 – – – Tensile strength 103 lbf/in.2 MPa Tensile modulus 105 lbf/in.2 GPa Elongation at break (%) Flexural modulus 105 lbf/in.2 GPa Impact strength, notched Izod Kevlar ft.-lb/in. 1.3 3.0 – – – J/m 69 159 – – – and of 2,4,4-trimethyl isomer, and has a Tg of about 150°C. Grilamid TR (Emser) with a Tg of about 160°C is a copolymer with units of H3C C C O O CH3 NH CH2 NH and of C ( CH2 ( 10 NH O A crystalline, partially aromatic polyamide, poly-m-xylylene-adipamide, (also known as MXD-6) with repeat units of NH CH2 CH2 (CH2)4 C NH O C O is available as a heat resistant engineering plastic (e.g., Ixef by Solvay) generally similar in properties to nylon-6,6, having a Tm of 243°C but with reduced water absorption, greater stiffness, and a Tg of about 90° C. Although its heat distortion temperature is only 96°C, with 30% glass ﬁlling this is increased to about 270°C. Typical properties of some of these polyamides, along with those of nylon-6,6 for comparison, are shown in Table 4.17. Copolymers containing amide and imide units, the polyamideimides, are described in the following section on polyimides. 4.3.2.3 Polyimides R N O O C C C C O O N R n 514 Plastics Technology Handbook The polyimides have the characteristic functional group CO N CO and are thus closely related to amides [43]. The branched nature of the imide functional group enables production of polymers having predominantly ring structures in the backbone and hence high softening points. Many of the structures exhibit such a high level of thermal stability that they have become important for application at much higher service temperatures than had been hitherto achieved with polymer. The use of tetracarboxylic acid anhydride instead of the dicarboxylic acids used in the manufacture of polyamides yields polyimides. The general method of preparation of the original polyimides by the polymerization of pyromellitic dianhydride and aromatic diamine is shown in Figure 4.18a. A number of diamines have been investigated, and it has been found that certain aromatic amines, which include m-phenylendediamine, benzidine, and di-(4-aminophenyl)ether, give polymers with a high degree of oxidative and thermal stability. The aromatic amine di(4-aminophenyl)ether is employed in the manufacture of polyimide ﬁlm, designated as Kapton (DuPont). Other commercial materials of this type introduced by DuPont in the early 1960s included a coating resin (Pyre ML) and a machinable block form (Vespel). In spite of their high price these materials have found established uses because of their exceptional heat resistance and good retention of properties at high temperatures. Since the polyimides are insoluble and infusible, they are manufactured in two stages. The ﬁrst stage involves an amidation reaction carried out in a polar solvent (such as dimethylformamide and dimethylacetamide) to produce an intermediate poly(amic acid) which is still soluble and fusible. The poly (amic acid) is shaped into the desired physical form of the ﬁnal product (e.g., ﬁlm, ﬁber, coating, laminate) and then the second stage of the reaction is carried out. In the second stage the poly(amic acid) is cyclized in the solid state to the polyimide by heating at moderately high temperatures above 150°C. A different approach, avoiding the intermediate poly(amic acid) step, was pioneered by Upjohn. The Upjohn process involves the self-condensation of the isocyanate of trimellitic acid, and the reaction by-product is carbon dioxide (Figure 4.19b). Polypyromellitimides (Figure 4.19a) have many outstanding properties: ﬂame resistance, excellent electrical properties, outstanding abrasion resistance, exceptional heat resistance, and excellent resistance to oxidative degradation, most chemicals (except strong bases), and high-energy radiation. After 1000 h of exposure to air at 300°C the polymers retained 90% of their tensile strength, and after 1500 h exposure to a radiation of about 10 rad at 175°C, they retained form stability, although they became brittle. The ﬁrst commercial applications of polypyromellitimides were as wire enamels, as insulating varnishes, as coating for glass cloth (Pyre ML, DuPont), and as ﬁlm (Kapton, DuPont). A fabricated solid grade was marketed as Vespel (DuPont). Laminates were produced by impregnation of glass and carbon ﬁber, with the polyimide precursor followed by pressing and curing at about 200°C and further curing at temperatures of up to 350°C. Such laminates could be used continuously at temperatures up to 250°C and intermittently up to 400°C. The laminates have thus found important application in the aircraft industry, particularly in connection with supersonic aircraft. At the present time the applications of polyimides include compressor seals in jet engines, sleeves, bearings, pressure discs, sliding and guide rolls, and friction elements in data processing equipment, valve shafts in shutoff valves, and parts in soldering and welding equipment. Polyimides have also found a number of specialist applications. Polyimide foams (Skybond by Monsanto) have been used for sound deadening of jet engines. Polyimides ﬁbers have been produced by Upjohn and by Rhone-Poulenc (Kermel). A particular drawback of the polyimides is that they have limited resistance to hydrolysis and may crack in water or steam at temperatures above 100°C. Consequently, polyimides have encountered 515 Industrial Polymers O O C C C C O O O O H2N O O C C N C O O NH2 –H2O N C R R O O N C O (b) FIGURE 4.19 acid. C NH HO C C C NH R OH O C C O O O Poly (amic acid) (a) O O O C –CO2 C N N C O C N O (a) Synthesis of polyimides by polycondensation. (b) Self-condensation of isocyanate of trimellitic competition from polyetheretherketones (PEEK), which are not only superior in this regard but are also easier to mold. 4.3.2.3.1 Modiﬁed Polyimides The application potential of polyimides is quite limited because, being infusible, they cannot be molded by conventional thermoplastics techniques [14]. In trying to overcome this limitation, scientists, in the early 1970s, developed commercially modiﬁed polyimides, which are more tractable materials than polyimides but still possessing signiﬁcant heat resistance. The important groups of such modiﬁed polyimides are the polyamideimides (e.g., Torlon by Amoco Chemicals), the polybismaleinimides (e.g., Kinel by RhonePoulenc), the polyester–imides (e.g., Icdal Ti40 by Dynamit Nobel), and the polyether-imides (e.g., Ultem by General Electric). If trimellitic anhydride is used instead of pyromellitic dianhydride in the reaction shown in Figure 4.19a, then polyamide-imide is formed (see Figure 4.20a). Other possible routes to this type of product involve the reaction of trimellitic anhydride with diisocyanates (Figure 4.20b) or diurethanes (Figure 4.20c). Closely related is the Upjohn process for polyimide by self-condensation of the isocyanate of trimellitic acid, as illustrated in Figure 4.19b, although the product in this case is a true polyimide rather than a polyamide-imide. Polyamide-imides may also be produced by reacting together pyromellitic dianhydride, a diamine, and a diacid chloride. Alternatively, it may be produced in a two-stage process in which a diacid chloride is reacted with an excess of diamine to produce a low-molecular-weight polyamide with amine end groups which may then be chain extended by reaction with pyromellitic dianhydride to produce imide linkages. The Torlon materials produced by Amoco Chemicals are polyamide-imides of the type shown in Figure 4.20a. Torlon has high strength, stiffness, and creep resistance, shows good performance at moderately high temperatures, and has excellent resistance to radiation. The polymers are unaffected by all types of hydrocarbons (including chlorinated and ﬂuorinated products), aldehydes, ketones, ethers, esters, and dilute acids, but resistance to alkalis is poor. 516 Plastics Technology Handbook O O C C O H2N R NH2 –H2O N R NH HO C C C C O O O O n (a) O O C O OCN R C –CO2 NCO N R NH HO C C C C O O O O n (b) O C O HO C C O O R'C C NH R NH C O O OR' -CO2 -R'OH O C N R NH (c) C C O O n FIGURE 4.20 Synthesis of polyamide-imides from trimellitic anhydride and (a) diamine, (b) diisocyanate, and (c) diurethane. Torlon has been marketed both as a compression-molding grade and as an injection-molding grade. The compression-molding grade, Torlon 2000, can accept high proportions of ﬁller without seriously affecting many of its properties. For compression molding, the molding compound is preheated at 280°C before it is molded at 340°C at pressures of 4350 psi (30 MPa); the mold is cooled at 260°C before removal. For injection molding, the melt at temperatures of about 355°C is injected into a mold kept at about 230°C. To obtain high-quality moldings, prolonged annealing cycles are recommended. Uses of polyamide-imides include pumps, valves, refrigeration plant accessories, and electronic components. The polymers have low coefﬁcient of friction, e.g., 0.2 (to steel), which is further reduced to as little as 0.02–0.08 by blending with graphite and Teﬂon. In solution form in N-methyl-2-pyrrolidone, Torlon has been used as a wire enamel, as a decorative ﬁnish for kitchen equipment, and as an adhesive and laminating resin in spacecraft. The polyimides and polyamide-imides are produced by condensation reactions which give off volatile low-molecular-weight by-products. The polybismaleinimides may however be produced by rearrangement polymerization with no formation of by-products. The starting materials in this case are the bismaleimides, which are synthesized by the reaction of maleic anhydride with diamines (Figure 4.21). The bismaleimides can be reacted with a variety of bifunctional compounds to form polymers by rearrangement reactions. These include amines, mercaptans, and aldoximes (Figure 4.22). If the reaction is carried out with a deﬁciency of the bifunctional compound, the polymer will have terminal double bonds to serve as a cure site for the formation of a cross-linked polymer via a double bond polymerization mechanism during molding. The cross-linking in this case occurs without the formation of any volatile by-products. The Kinel materials produced by Rhone-Poulenc are polybismaleinimides of the type shown in Figure 4.22. These materials having chain-end double bonds, as explained previously, can be processed like 517 Industrial Polymers O O O C NH R NH C C HC H2N R NH2 O CH CH HC C C OH O HO C O O O O C NH R NH C CH –H2O HC CH C OH HO C O N R N O O O FIGURE 4.21 O HC O Bismaleimide Synthesis of bismaleimides by the reaction of maleic anhydride with diamines. O N R N (a) O O O O O N R N HN HS O O O O N R N (c) O O O O O O n S R' N R N R´ SH S (b) NH R' N R N H2N R´ NH2 O O O n O O O N R N HON CH R´ CH NOH NH C R' C NH O O O O n FIGURE 4.22 Formation of polymers by reaction of bismaleimides with (a) amines, (b) mercaptans, and (c) aldoximes. conventional thermosetting plastics. The properties of the cured polymers are broadly similar to the polyimides and polyamide-imides. Molding temperatures are usually from 200°C to 260°C. Post-curing at 250°C for about 8 h is necessary to obtain the optimum mechanical properties. Polybismaleinimides are used for making laminates with glass- and carbon-ﬁber fabrics, for making printed circuit boards, and for ﬁlament winding. Filled grades of polybismaleinimides are available with a variety of ﬁllers such as asbestos, glass ﬁber, carbon ﬁber, graphite, Teﬂon, and molybdenum sulﬁde. They ﬁnd application in aircraft, spacecraft, and rocket and weapons technology. Speciﬁc uses include fabrication of rings, gear wheels, friction bearings, cam discs, and brake equipment. The polyester–imides constitute a class of modiﬁed polyimide. These are typiﬁed by the structure shown in Figure 4.23. Polyether-imides form yet another class of modiﬁed polyimide. These are highperformance amorphous thermoplastics based on regular repeating ether and imide linkages. The aromatic imide units provide stiffness, while the ether linkages allow for good melt-ﬂow characteristics and processability. One of several synthetic routes to polyetherimides of a general structure involves a cyclization reaction of form the imide rings and a displacement reaction to prepare the ether linkages and form the polymer 518 Plastics Technology Handbook O O C C CH3 N N O C C O CH3 O R C O n The ﬁrst step of this synthesis is to form a bis-imide monomer formed by the reaction of nitrophthalic anhydride and a diamine (see Figure 4.21). The second step of polyetherimide synthesis involves the formation of a bisphenol dianion by treatment of a diphenol with two equivalents of base, followed by removal of water. The polymerization step involves displacement of the nitrogroups of the bis-imide by the bisphenol dianion to form the ether linkages of the polymer. A large number of polyetherimides can be prepared by the synthetic route. Polyetherimides are suitable for applications that require high temperature stability, high mechanical strength, inherent ﬂame resistance with extremely low smoke evolution, outstanding electrical properties over a wide frequency and temperature range, chemical resistance to aliphatic hydrocarbons, acids and dilute bases, UV stability, and ready processability on conventional equipment. Ultem, introduced by General Electric in 1982, is a polyether-imide. It was designed to complete with heat- and ﬂame-resistory, high-performance engineering polymers, polysulfones, and polyphenylene sulﬁde. Some typical properties of Ultem 1000 are speciﬁc gravity 1.27, tensile yield strength 105 MPa, ﬂexural modules 3300 MPa, hardness Rockwell M109, Vicat softening point 219°C, heat-distortion temperature (1.82 MPa) 200°C, and limiting oxygen index 47. Speciﬁc applications include circuit breaker housings and microwave oven stirrer shafts. Typical properties of unﬁlled polyimides are compared in Table 4.18. O O C C N FIGURE 4.23 O R´ O O O C C N C C O O R² Typical structure of polyester–imides. TABLE 4.18 Properties of Unﬁlled Polyimides Property Vespel (ICI) Torlon (Amoco) Kinel (Rhone-Poulenc) Ultem (General Electric) 90 186 ∼40 100 67 105 – – 58 52 ∼25 – 25°C 3.5 4.6 3.8 3.3 150°C 260°C 2.7 2.3 3.6 3.0 – 2.8 2.5 – Heat distortion temperature (°C) 357 282 – 200 Limiting oxygen index (%) 35 42 – 47 Tensile strength (MPa) 25°C 150°C 260°C Flexural modulus (GPa) 519 Industrial Polymers 4.3.3 Formaldehyde Resins The phenol-formaldehyde and urea-formaldehyde resins are the most widely used thermoset polymers. The phenolic resins were the ﬁrst truly synthetic polymers to be produced commercially. Both phenolic and urea resins are used in the highly cross-linked ﬁnal form (C-stage), which is obtained by a stepwise polymerization process. Lower-molecular-weight prepolymers are used as precursors (A-stage resins), and the ﬁnal form and shape are generated under heat and pressure. In this process water is generated in the form of steam because of the high processing temperatures. Fillers are usually added to reduced resin content and to improve physical properties. The preferred form of processing is compression molding. The phenolics and urea resins are high-volume thermosets which owe their existence to the relatively low cost of the starting materials and their superior thermal and chemical resistance. Today these resins are widely used in molding applications, in surface coatings and adhesives, as laminating resins, casting resins, binders and impregnants, and in numerous other applications. However, as with all products based on formaldehyde, there is concern about the toxicity of these resins during processing and about the residual traces of formaldehyde in the ﬁnished product. 4.3.3.1 Phenol–Formaldehyde Resins Monomers Phenol, formaldehyde Polymerization Major Uses Base- or acid-catalyzed stepwise polycondensation Plywood adhesives (34%), glass-ﬁber insulation (19%), molding compound (8%) Since the cross-linked polymer of phenol-formaldehyde reaction is insoluble and infusible, it is necessary for commercial applications to produce ﬁrst a tractable and fusible low-molecular-weight prepolymer which may, when desired, be transformed into the cross-linked polymer [14,44,45]. The initial phenol-formaldehyde products (prepolymers) may be of two types: resols and novolacs. 4.3.3.1.1 Resols Resols are produced by reacting a phenol with a molar excess of formaldehyde (commonly about 1:1.5–2) by using a basic catalyst (ammonia or sodium hydroxide). This procedure corresponds to Baekeland’s original technique. Typically, reaction is carried out batchwise in a resin kettle equipped with stirrer and jacketed for heating and cooling. The resin kettle is also ﬁtted with a condenser such that either reﬂux or distillation may take place as required. A mixture of phenol, formalin, and ammonia (1–3% on the weight of phenol) is heated under reﬂux at about 100°C for 0.25–1 h, and then the water formed is removed by distillation, usually under reduced pressure to prevent heat hardening of the resin. Two classes of resins are generally distinguished. Resols prepared with ammonia as catalysts are spiritsoluble resins having good electrical insulation properties. Water-soluble resols are prepared with caustic soda as catalyst. In aqueous solutions (with a solids content of about 70%) these are used mainly for mechanical grade paper and cloth laminates and in decorative laminates. The reaction of phenol and formaldehyde in alkaline conditions results in the formation of o- and p-methylol phenols. These are more reactive towards formaldehyde than the original phenol and undergo rapid substitution with the formation of di- and trimethylol derivatives. The methylol phenols obtained are relatively stable in an alkaline medium but can undergo self-condensation to form dinuclear and polynuclear phenols (of low molecular weight) in which the phenolic nuclei are bridged by methylene groups. Thus in the base-catalyzed condensation of phenol and formaldehyde, there is a tendency for polynuclear phenols, as well as mono-, di-, and trimethylol phenols to be formed. Liquid resols have an average of less than two phenolic nuclei per molecule, and a solid resol may have only three or four. Because of the presence of methylol groups, the resol has some degree of water tolerance. However, for the same reason, the shelf life of resols is limited. 520 Plastics Technology Handbook OH OH CH2OCH2 OH OH CH2OH OH CH2 –H2O O CH2 FIGURE 4.24 Other products Curing mechanisms for resols. Resols are generally neutralized or made slightly acidic before cure (cross-linking) is carried out. Network polymers are then obtained simply by heating, which results in cross-linking via the uncondensed methylol groups or by more complex mechanisms (see Figure 4.24). Above 160°C it is believed that quinone methide groups, as depicted on the bottom of Figure 4.24, are formed by condensation of the ether linkages with the phenolic hydroxyl groups. These quinone methide structures can be cross-linked by cycloaddition and can undergo other chemical reactions. It is likely that this formation of quinone methide and other related structures is responsible for the dark color of phenolic compression moldings made at higher temperatures. Note that cast phenol-formaldehyde resins, which are cured at much lower temperatures, are water white in color. If they are heated to about 180°C, they darken considerably. 4.3.3.1.2 Novolac The resols we have described are sometimes referred to as one-stage resins, since cross-linked products may be made from the initial reaction mixture only by adjusting the pH. The resol process is also known as the one-stage process. On the other hand, the novolacs are sometimes referred to as two-stage resins because, in this case, it is necessary to add, as we will show, some agent to enable formation of cross-linked products. Novolac resins are normally prepared by the reaction of a molar excess of phenol with formaldehyde (commonly about 1.25:1) under acidic conditions. The reaction is commonly carried out batchwise in a resin kettle of the type used for resol manufacture. Typically, a mixture of phenol, formalin, and acid is heated under reﬂux at about 100°C. The acid is usually either hydrochloric acid (0.1–0.3% on the weight of phenol) or oxalic acid (0.5–2%). Under acidic conditions the formation of methylol phenols is rather slow, and the condensation reaction thus takes approximately 2–4 h. When the resin reaches the requisite degree of condensation, it become hydrophobic, and the mixture appears turbid. Water is then distilled off until a cooled sample of the residual resin shows a melting point of 65–75°C. The resin is then discharged and cooled to give a hard, brittle solid (novolac). Unlike resols, the distillation of water for novolac is normally carried out without using a vacuum. Therefore the temperature of the resin increases as the water is removed and the reaction proceeds, the temperature reaching as high as 160°C at the end. At these temperatures the resin is less viscous and more easily stirred. The mechanism of phenol-formaldehyde reaction under acidic conditions is different from that under basic conditions described previously. In the presence of acid the products o- and p-methylol phenols, which are formed initially, react rapidly with free phenol to form dihydroxy diphenyl methanes (Figure 4.25). The latter undergo slow reaction with formaldehyde and phenolic species, forming polynuclear phenols by further methylolation and methylol link formation. Reactions of this type continue until all the 521 Industrial Polymers OH OH OH CH2OH CH2O CH2OH OH HO CH2 CH2 HO FIGURE 4.25 CH2 OH OH CH2 OH OH Higher oligomers (Novolac) Formation of novolac in an acid-catalyzed reaction of phenol and formaldehyde. formaldehyde has been used up. The ﬁnal product thus consists of a complex mixture of polynuclear phenols linked by o- and p-methylene groups. The average molecular weight of the ﬁnal product (novolac) is governed by the initial molar ratio of phenol and formaldehyde. A typical value of average molecular weight is 600, which corresponds to about six phenolic nuclei per chain. The number of nuclei in individual chains is usually 2–13. A signiﬁcant feature of novolacs is that they represent completed reactions and as such have no ability to continue increasing in average molecular weight. Thus there is no danger of gelation (cross-linking) during novolac production. Resols, however, contain reactive methylol groups and so are capable of crosslinking on heating. To convert novolacs into network polymers, the addition of a cross-linking agent (hardener) is necessary. Hexamethylenetetramine (also known as hexa or hexamine) is invariably used as the hardener. The mechanism of the curing process is complex. Because of the exothermic reaction on curing and the accompanying shrinkage, it is necessary to incorporate inert ﬁllers to reduce resin content. Fillers also serve to reduce cost ad may give additional beneﬁts, such as improving the shock resistance. Commonly used ﬁllers are wood ﬂour, cotton ﬂock, textile shreds, mica, and asbestos. Wood ﬂour, a ﬁne sawdust preferably from soft woods, is the most commonly used ﬁller. Good adhesion occurs between the resin and the wood ﬂour, and some chemical bonding may also occur. Wood ﬂour reduces exotherm and shrinkage, improves the impact strength of the moldings, and is cheap. For better impact strength cotton fabric or chopped fabric may be incorporated. Asbestos may be used for improved heat and chemical resistance, and iron-free mica powder may be used for superior electrical insulation resistance characteristics. Other ingredients which may be incorporated into a phenolic molding powder include accelerators (e.g., lime or magnesium oxide) to promote the curing reaction, lubricants (e.g., stearic acid and metal stearates) to prevent sticking to molds, plasticizers (e.g., naphthalene, furfural, and dibutyl phthalate) to improve ﬂow properties during cure, and pigments or dyes (e.g., nigrosine) to color the product. Some typical formulations of phenolic molding powders are given in Table 4.19. Since the phenolic resins cure with evolution of volatiles, compression molding is performed using molding pressures of 1–2 ton/in.2 (15–30 MPa) at 155–170°C. Phenolic molding compositions may be preheated by high frequency or other methods. Preheating reduces cure time, shrinkage, and required molding pressures. It also enhances the ease of ﬂow, with consequent reduction of mold wear and danger of damage to inserts. Molding shrinkage of general-purpose grades is about 0.005–0.08 in./in. Highly loaded mineral-ﬁlled grades exhibit lower shrinkage. Phenol–formaldehyde molding compositions are traditionally processed on compression- and transfermolding machines with a very small amount being extruded. However, the injection-molding process as modiﬁed for thermosetting plastics is being increasingly used and it is today the most common method used 522 Plastics Technology Handbook TABLE 4.19 Typical Formulationsa of Phenolic Molding Grades Ingredient Novolac resin Hexa General-Purpose Grade Medium ShockResisting Grade High ShockResisting Grade Electrical Grade 100 12.5 100 12.5 100 17 100 14 Magnesium oxide 3 2 2 2 Magnesium stearate Nigrosine dye 2 4 2 3 3.3 3 2 3 100 – – – Cotton ﬂock Textile shreds – – 110 – – 150 – – Asbestos – – – 40 Mica – – – 120 Wood Flour Source: Brydson, J. A. 1982. Plastics Materials. Butterworth Scientiﬁc, London, UK. a Parts by weight. to process phenolic compounds. The shorter cycle times and low waste factor available with screw injection molding, which contribute to the lowest unit costs for extended runs, have induced phenolic molding compounders to develop products for this molding process. 4.3.3.1.3 Properties and Applications Since the polymer in phenolic moldings is highly cross-linked and interlocked, the moldings are hard, infusible, and insoluble. The chemical resistance of the moldings depends on the type of resin and ﬁller used. General-purpose PF grades are readily attacked by aqueous sodium hydroxide, but cresol- and xylenol-based resins are more resistant. Phenolic moldings are resistant to acids except formic acid, 50% sulfuric acid, and oxidizing acids. The resins are ordinarily stable up to 200°C. The mechanical properties of phenolic moldings are strongly dependent on the type of ﬁller used (Table 4.20). Being polar, the electrical insulation properties of phenolics are not outstanding but are generally adequate. A disadvantage of phenolics as compared to aminoplasts and alkyds is their poor tracking resistance under high humidity, but this problem is not serious, as will be evident from the wide use of phenolics for electrical insulation applications. Perhaps the most well-known applications of PF molding compositions are in domestic plugs and switches. However, in these applications PF has now been largely replaced by urea-formaldehyde plastics because of their better antitracking property and wider range of color possibility. (Because of the dark color of the phenolic resins molded above 160°C, the range of pigments available is limited to relatively darker colors—blacks, browns, deep blues, greens, reds, and oranges.) Nevertheless, phenolics continue to be used as insulators in many applications because their properties have proved quite adequate. There are also many applications of phenolics where high electrical insulation properties are not as important, and their heat resistance, adequate shock resistance, and low cost are important features: for example, knobs, handles, telephones, and instrument cases. In some of these applications phenolics have been replaced by ureaformaldehyde, melamine–formaldehyde, alkyd, or newer thermoplastics because of the need for brighter colors and tougher products. In general, phenolics have better heat and moisture resistance than urea-formaldehyde moldings. Heatresistant phenolics are used in handles and knobs of cookware, welding tongs, electric iron parts, and in the automobile industry for fuse box covers, distributor heads, and other applications where good electrical insulation together with good heat resistance is required. 523 Industrial Polymers TABLE 4.20 Properties of Phenolic Moldingsa Property Speciﬁc gravity Shrinkage (cm/cm) General-Purpose Grade Medium ShockResisting Grade High ShockResisting Grade Electrical Grade 1.35 0.006 1.37 0.005 40 0.002 1.85 0.002 8000 55 7000 48 6500 45 8500 58 0.16 0.22 0.29 0.39 0.8–1.4 1.08–1.9 0.14 0.18 6.0–10.0 4.5–5.5 5.5–5.7 – 6.0–10.0 – 4.0–6.0 4.3–5.4 150–300 58–116 200–275 78–106 150–250 58–97 275–350 106–135 0.1–0.4 0.03–0.05 0.1–0.35 – 0.1–0.5 – 0.03–0.05 0.01–0.02 1012–1014 1012–1014 1011–1013 1013–1016 45–65 30–50 50–100 2–6 Tensile strength lbf/in.2 MPa Impact strength ft.-lbf J Dielectric constant at 800 Hz at 106 Hz Dielectric strength (20°C) V/mil kV/cm Power factor at 800 Hz at 106 Hz Volume resistivity (ohm-m) Water absorption (24 h, 23°C) mg Source: Brydson, J. A. 1982. Plastics Materials. Butterworth Scientiﬁc, London, UK. a Testing according to BS 2872. Bottle caps and closures continue to be made in large quantities from phenolics. The development of machines for injection molding of thermosetting plastics and availability of fast-curing grades of phenolics have stimulated the use of PF for many small applications in spite of competition from other plastics. Among the large range of laminated plastics available today, the phenolics were the ﬁrst to achieve commercial signiﬁcance, and they are still of considerable importance. In these applications one-stage resins (resols) are used, sine they have sufﬁcient methylol groups to enable curing without the need of a curing agent. Caustic soda is commonly used as the catalyst for the manufacture of resols for mechanical and decorative laminates. However, it is not used in electrical laminates because it adversely affects the electrical insulation properties. For electrical-grade resols ammonia is the usual catalyst, and the resins are usually dissolved in industrial methylated spirits. The use of cresylic acid (m-cresol content 50–55%) in place of phenol yields laminating resins of better electrical properties. In the manufacture of laminates for electrical insulation, paper (which is the best dielectric) is normally used as the base reinforcement. Phenolic paper laminates are extensively used for high-voltage insulation applications. Besides their good insulation properties, phenolic laminates also possess good strength, high rigidity, and machinability. Sheet, tubular, and molded laminates are employed. Phenolic laminates with cotton fabric reinforcement are used to manufacture gear wheels are quiet running but must be used at lower working stresses than steel. Phenolic-cotton or phenolic-asbestos laminates have been used as bearings for steel rolling mills to sustain bearing loads as high as 3000 psi (21 MPa). Because of the advent of cheaper thermoplastics, cast phenolic resins (resols) are no longer an important class of plastics materials. 524 Plastics Technology Handbook 4.3.3.2 Urea–Formaldehyde Resins Monomers Urea, formaldehyde Polymerization Major Uses Stepwise polycondensation Particle-board binder resin (60%), paper and textile treatment (10%), molding compound (9%) coatings (7%) Aminoresins or aminoplastics cover a range of resinous polymers produced by reaction of amines or amides with aldehydes [14,46,47]. Two such polymers of commercial importance in the ﬁeld of plastics are the urea-formaldehyde and melamine–formaldehyde resins. Formaldehyde reacts with the amino groups to form aminomethylol derivatives which undergo further condensation to form resinous products. In contras to phenolic resins, products derived from urea and melamine are colorless. Urea and formaldehyde resins are usually prepared by a two-stage reaction. In the ﬁrst stage, urea and formaldehyde (mole ratio in the range 1:1.3–1:1.5) are reacted under mildly alkaline (pH 8) conditions, leading to the production of monomethylol urea (Figure 4.26(I)) and dimethylol urea (Figure 4.26(II)). If the product of the ﬁrst stage, which in practice usually also contains unreacted urea and formaldehyde, is subjected to acid conditions at elevated temperatures (stage 2), the solution increases in viscosity and sets to an insoluble and irreversible gel. The gel eventually converts with evolution of water and formaldehyde to a hard, colorless, transparent, insoluble, and infusible mass having a network molecular structure. The precise mechanisms involved during the second stage are not fully understood. It does appear that in the initial period of the second stage methylol ureas condense with each other by reaction of a –CH2OH group on one molecule with an –NH2 of another molecule, leading to linear polymers of the form shown in Figure 4.26(III). These polymers are relatively less soluble in aqueous media and tend to form amorphous white precipitates on cooling to room temperature. More soluble resins are formed on continuation of heating. This probably involves the formation of pendant methylol groups (Figure 4.26(IV)) by reactions of the –NH– groups with free formaldehyde. These methylol groups and the methylol groups on the chain ends of the initial reaction product can then C NH2 O CH2O C NH CH2OH O C NH2 NH2 NHCH2OH O NHCH2OH II I HO CH2NH CO NH n CH2OH ( III ) CH2O NH N CH2OH ( IV ) N CH2OH HOCH2NH N CH2 O CH2NH CH2O N CH2 (V) FIGURE 4.26 Reactions in the formation of urea-formaldehyde resins. NH Industrial Polymers 525 react with each other to produce ether linkages, or with amine groups to give methylene linkages (Figure 4.26III). The ether linkages may also break down on heating to methylene linkages with the evolution of formaldehyde (Figure 4.26(V)). An idealized network structure of the ﬁnal cross-linked product is shown in Figure 1.25. 4.3.3.2.1 Molding Powder The urea-formaldehyde (UF) Molding powder will contain a number of ingredients. Most commonly these include resin, ﬁller, pigment, accelerator, stabilizer, lubricant, and plasticizer. Bleached wood pulp is employed as a ﬁller for the widest range of bright colors and in slightly translucent moldings. Wood ﬂour, which is much cheaper, may also be used. A wide variety of pigments is now used in UF molding compositions. Their principal requirements are that they should be stable to processing conditions and be unaffected by service conditions of the molding. To obtain a sufﬁciently high rate of cure at molding temperatures, it is usual to add about 0.2–2.0% of an accelerator (hardener)—a latent acid catalyst which decomposes at molding temperatures to yield an acidic body that will accelerate the rate of cure. Many such materials have been described, the most prominent of them being ammonium sulfamate, ammonium phenoxyacetate, trimethyl phosphate, and ethylene sulﬁte. A stabilizer such as hexamine is often incorporated into the molding powder to improve its shelf life. Metal stearates, such as zinc, magnesium, or aluminum stearates are commonly used as lubricants at about 1% concentration. Plasticizers (e.g., monocresyl glycidyl ether) are used in special grades of molding powders. They enable more highly condensed resins to be used in the molding powder and thus reduce curing shrinkage while maintaining good ﬂow properties. In a typical manufacturing process, the freshly prepared UF ﬁrst-stage reaction product is mixed with the ﬁller (usually with a ﬁller-resin dry weight ratio of 1:2) and other ingredients except pigment in a trough mixer at about 60°C for about 2 h. Thorough impregnation of the ﬁller with the resin solution and further condensation of the resin takes place during this process. Next, the wet mix is in a turbine or rotary drier for about 2 h at 100°C or about 1 h in a countercurrent of air at 120–130°C. The drying process reduces the water content from about 40% to about 6% and also causes further condensation of the resin. After it is removed from the drier, the product is ground in a hammer mill and then in a ball mill for 6–9 h. The pigments are added during the ball-milling process, which ensures a good dispersion of the pigment and gives a ﬁne powder that will produce moldings of excellent ﬁnish. The powder, however, has a high bulk factor and needs densiﬁcation to avoid problems of air and gas trappings during molding. There are several methods of densiﬁcation. In one method, the heated powder is formed into strips by passing through the nip of a two-roll mill. The strips are then powdered into tiny ﬂat ﬂakes in a hammer mill. Other processes involve agglomeration of the powder by heating in an internal mixer at about 100°C or by treatment with water or steam and subsequent drying. Continuous compounders, such as the Buss Ko-Kneader, are also used. 4.3.3.2.2 Processing Urea–formaldehyde molding powders have a limited storage life. They should therefore be stored in a cool place and should be used, wherever possible, within a few months of manufacture. Conventional compression and transfer molding are commonly used for UF materials, the former being by far the most important process in terms of tonnage handled. Compression molding pressures usually range from 1 to 4 tn/in.2 (15–60 MPa), the higher pressures being used for deep-draw articles. Molding temperatures from 125°C to 160°C are employed. The cure time necessary depends on the mold temperature and on the thickness of the molding. The cure time for a 1/8 in. thick molding is typically about 55 sec at 145°C. Bottle caps (less than 1/8 in. thick) and similar items, however, are molded industrially with much shorter cure times (∼10–20 sec) at the higher end of the molding temperature range. For transfer molding of UF molding powders, pressures of 4–10 tn/in.2 (60–150 MPa), calculated on the area of the transfer pot, are generally recommended. 526 Plastics Technology Handbook Special injection grades of UF molding powder have been developed for injection-molding applications which call for molding materials with good ﬂow characteristics between 70°C and 100°C, unaffected by long residence time in the barrel but capable of almost instant cure in the mold cavity at a higher temperature. Although the transition from compression molding to injection molding has been extensive for phenolics, the same cannot be said for UF materials, because they are more difﬁcult to mold, possibly because the UF are more brittle than a phenolic resin and so are less able to withstand the stress peaks caused by ﬁller orientation during molding. A combination of compression and injection processes has therefore been developed in which a screw preplasticizing unit delivers preheated and softened material directly to a compression-mold cavity. 4.3.3.2.3 Properties and Applications The wide color range possible with UF molding powders has been an important reason for the widespread use of the material. These moldings have a number of other desirable features: low cost, good electrical insulation properties, and resistance to continuous heat up to a temperature of 70°C. Some typical values of physical properties of UF molding compositions are given in Table 4.21. They do not impart taste and odor to food-stuffs and beverages with which they come in contact and are resistant to detergents and drycleaning solvents. The foregoing properties account for major uses of UF in two applications, namely, bottle caps and electrical ﬁttings. It is also used for colored toilet seats, vacuum ﬂasks, cups and jugs, hair drier housings, toys, knobs, meat trays, switches, lamp shades, and ceiling light bowls. In the latter applications it is important to ensure adequate ventilation to prevent overheating and consequent cracking of the molded articles. However, only about 3% of UF resins are used for molding powders. The bulk (about 85%) of the resins are used as adhesives in the particleboard, plywood, and furniture industries. Resins for these applications are commonly available with U/F molar ratios ranging from 1:1.4 to 1:2.2. To prepare a suitable resin for adhesive applications, urea is dissolved in formalin (initially neutralized to pH 7.5) to give the desired U/F molar ratio. After boiling under reﬂux for about 15 min to give demethylol urea and other low-molecular products, the resin is acidiﬁed, conveniently with formic acid, to pH 4, and reacted for a further period of 5–20 min. The resulting water-soluble resin with approximately 50% solids content is stabilized by neutralizing to a pH 7.5 with alkali. For use as an aqueous solution, as is normally the case, the resin is then partially dehydrated by vacuum distillation to give a 70% solids content. TABLE 4.21 Properties of Urea–Formaldehyde and Melamine–Formaldehyde Moldingsa Property Speciﬁc gravity Urea–Formaldehyde (a-Cellulose Filled) Melamine–Formaldehyde (Cellulose Filled) 1.5–1.6 1.5–1.55 7.5–11.5 52–80 8–12 55–83 0.20–0.35 0.15–0.24 Tensile strength 103 lbf/in.2 MPa Impact strength (ft.-lbf) Dielectric strength (90°C) V/0.001 in. 120–200 160–240 Volume resistivity (ohm-m) 1013–1015 109–1010 Water absorption (mg) 24 h at 20°C 30 min at 100°C 50–130 10–50 180–460 40–110 Source: Brydson, J. A. 1982. Plastics Materials. Butterworth Scientiﬁc, London, UK. a Testing according to BS 2872. 527 Industrial Polymers Phosphoric acid, or more commonly ammonium chloride, is used as a hardener for UF resin adhesives. Ammonium chloride reacts with formaldehyde to produce hexamine and hydrochloric acid, and the latter catalyzes the curing of the resin. In the manufacture of plywood a resin (with U/F molar ratio typically 1:1.8) mixed with hardener is applied to wood veneers, which are then plied together and pressed at 95– 110°C under a pressure of 200–800 psi (1.38–5.52 MPa). The UF resin-bonded plywood is suitable for indoor applications but is generally unsuitable for outdoor use. For outdoor applications phenolformaldehyde, resorcinol–formaldehyde, or melamine–formaldehyde resins are more suitable. Large quantities of UF resin are used in general wood assembly work. For joining pieces of wood the resin-hardener solution is usually applied to the surfaces to be joined and then clamped under pressure while hardening occurs. Alternatively, the resin may be applied to one surface and the hardener to the other, allowing them to come into contact in situ. This method serves to eliminate pot-life problems of the resin-hardener mixture. Gap-ﬁlling resins are produced by incorporating into UF resins plasticizers, such as furfuryl alcohol, and ﬁllers to minimize shrinkage and consequent cracking and crazing. In the manufacture of wood chipboard, which represents one of the largest applications of UF resins, wood chips are mixed with about 10% of a resin-hardener solution and pressed in a multidaylight press at 150°C for about 8 min. Since some formaldehyde is released during the opening of the press, it is necessary to use a resin with a low formaldehyde content. Because it has no grain, a wood chipboard is nearly isotropic in its behavior and so does not warp or crack. However, the water resistance of chipboard is poor. 4.3.3.3 Melamine–Formaldehyde Resins Monomers Melamine (trimerization of cyanamide), formaldehyde Polymerization Major Uses Stepwise polycondensation Dinnerware, table tops, coatings Reaction of melamine (2,4,6-triamino-1,3,5-triazine) with neutralized formalin at about 80–100°C leads to the production of a mixture of water-soluble methylol melamines. The methylol content of the mixture depends on the initial ratio of formaldehyde to melamine and on the reaction conditions. Methylol melamines possessing up to six methylol groups per molecule are formed (Figure 4.27). On further heating, the methylol melamines undergo condensation reactions, and a point is reached where hydrophobic resin separates out. The rate of resiniﬁcation depends on pH. The rate is minimum at about pH 10.0–10.5 and increases considerably both at lower and higher pH. The mechanism of NH2 N H2N NHCH2OH N N N CH2O NH2 HOCH2NH HOCH2 N N CH2O NHCH2OH CH2OH N N HOCH2 HOCH2 FIGURE 4.27 N N N Reactions in the synthesis of formica. Heat N CH2OH CH2OH H2O Formica 528 Plastics Technology Handbook resiniﬁcation and cross-linking is similar to that observed for urea-formaldehyde (Figure 4.26) and involves methylol–amine and methylol–methylol condensations. NH . CH2OH + H2N NH . CH2OH + HO . CH2NH NH . CH2. O . CH2. NH NH . CH2. NH + H2O NH . CH2. O . CH2. NH . + CH2O NH CH2 . NH + H2O In industrial practice, resiniﬁcation is carried out to a point close to the hydrophobe point. This liquid resin is either applied to the substrate or dried and converted into molding powder before proceeding with the ﬁnal cure. In a typical process [14] a jacketed resin kettle ﬁtted with stirrer and reﬂux condenser is charged is charged with 240 parts of 40% w/v formalin (pH adjusted to 8.0–8.5 using a sodium carbonate solution) and 126 parts of melamine (to give a melamine–formaldehyde ratio of 1:3), and the temperature is raised to 85°C. The melamine forms methylol derivatives and goes into solution. This water-soluble A-stage resin may be used for treatment of paper, leather, and fabrics to impart crease resistance, stiffness, shrinkage control, water repellency, and ﬁre retardance. It may be spray dried to give a more stable, watersoluble product. For laminating and other purposes the initial product is subjected to further condensation reactions at about 85°C with continuous stirring for more than 30 min. The hydrophilicity of the resin, as shown by its water tolerance, decreases with increasing condensation. The reaction is usually continued until a stage is reached when addition of 3 cm3 of water will cause 1 cm3 of resin to become turbid. The condensation reactions may be carried out at higher temperatures and lower pH values to achieve this stage more rapidly. In aqueous solutions the hydrophobic resins have a shelf life of just a few days. The resin may be diluted with methylated spirit to about 50% solids content and pH adjusted to 9.0–9.5 to achieve greater stability. The addition of about 0.1% borax (calculated on the weight of the solids content) as an aqueous solution is useful in obtaining this pH maintaining it for several months. The stabilized resin is stored preferably at 20–35°C, because too low a storage temperature will cause precipitation and too high at temperature will cause gelation. Melamine–formaldehyde molding powders are generally prepared by methods similar to those used for UF molding powders. In a typical process an aqueous syrup of MF resin with melamine–formaldehyde ratio of 1:2 is compounded with ﬁllers, pigments, lubricants, stabilizers, and accelerators in a dough-type mixer. The product is then dried and ball-milled by processes similar to those used for UF molding powders. Alpha-cellulose is used as a ﬁller for the more common decorative molding powders. Industrial-grade MF materials use ﬁllers such as asbestos, silica, and glass ﬁber. These ﬁllers are incorporated by dry blending methods. The use of glass ﬁber gives moldings of higher mechanical strength, improved dimensional stability, and higher heat resistance than other ﬁllers. The mineralﬁlled MF moldings have superior electrical insulation and heat resistance and may be used when phenolics and UF compositions are unsuitable. MF moldings are superior to UF products in lower water absorption (see Table 4.21), greater resistance to staining by aqueous solutions such as fruit juices and beverages, better retention of electrical properties in damp conditions, better heat resistance, and greater hardness. Compared with the phenolic resins, MF resins have better color range, track resistance, and scratch resistance. MF resins, are however, more expensive than general-purpose UF and PF resins. MF compositions are easily molded in conventional compression- and transfer-molding equipment. Molding temperatures from 145°C to165°C and molding pressures 2–4 tonf/in.2 (30–60 MPa) are usually employed. In transfer molding, pressures of 5–10 tonf/in.2 (75–150 MPa) are used. The cure time for an 1/8-in.-thick molding is typically 2 1/2 min at 150°C. Largely because of their wide color range, surface hardness, and stain resistance, MF resins are used as molding compositions for a variety of mechanical parts or household goods and as laminating resins for Industrial Polymers 529 tops for counters, cabinets, and tables. The mineral-ﬁlled molding powders are used in electrical application and knobs and handles for kitchen utensils. An interesting application of MF resins in compression molding involves decorative foils made by impregnating a printed or decorated grade of paper with resin and then drying. The foil may be applied to a compression molding shortly before the cure is complete, and the resin in the foil may be cured in that position to produce a bonding. In a typical process of laminating paper layers to make materials useful in electrical applications as well as decorative laminates (best known as formica), kraft paper, about the weight used in shopping bags, is run through a solution of melamine–formaldehyde prepolymer. Drying out water or driving off the solvent leave an impregnated sheet that can be handled easily, since the brittle polymer does not leave the surface sticky. As many as a dozen or more layers are piled up. For decorative purposes a printed rag or decorated cloth paper is put on top and covered with a translucent paper layer. The entire assembly is heated between smooth plates in a high-pressure press to carry out the thermosetting (curing) reaction that binds the sheets together into a strong, solvent-resistant, heat-resistant, and scratch-resistant surfacing material. The laminate, which is only about 1.5-mm thick, can be glued to a plywood base for use in tabletops, countertops, and the like. 4.3.4 Polyurethanes A urethane linkage (–NHCOO–) is formed by the reaction of an isocyanate (–NCO) and an alcohol: RNCO+R′OH!NHCOOR′ [14,48,49]. By the same reaction, polyhydroxy materials will react with polyisocyanates to yield polyurethanes soft thermoplastic elastomers to hard thermoset rigid foams are readily produced from liquid monomers. The basic building blocks for polyurethanes are polyisocyanates and macroglycols, also called polyols. The commonly used polyisocyanates are tolylene-diisocyanate (TDI), diphenylmethane diisocyanate or methylenediphenyl isocyanate (MDI), and polymeric methylenediphenyl isocyanate (PMDI) mixtures manufactured by phosgenating aromatic polyamines derived from the acid-catalyzed condensation of aniline and formaldehyde. MDI and PMDI are produced by the same reaction, and separation of MDI is achieved by distillation. The synthetic routes in the manufacture of commercial polyisocyanates are summarized in Figure 4.28. A number of specialty aliphatic polyisocyanates have been introduced recently in attempts to produce a light-stable polyurethane coating. Triisocyanate made by reacting hexamethylene diisocyanate with water (Figure 4.28c) is reported to impart good light stability and weather resistance in polyurethane coatings and is probably the most widely used aliphatic polyisocyanate. The macroglycols used in the manufacture of polyurethanes are either polyether or polyester based. Polyether diols are low-molecular-weight polymers prepared by ring-opening polymerization of oleﬁn oxides (see also the section on polyethers), and commonly used polyester polyols are polyadipates. A polyol produced by ring-opening polymerization of caprolactone, initiated with low-molecular-weight glycols, is also used. The reactions are summarized in Figure 4.29. Isocyanates are highly reactive materials and enter into a number of reactions with groups or molecules containing active hydrogen, such as water, amine, and also urethane. Isocyanates are also toxic and care should be exercised in their use. Their main effect is on the respiratory system. Major polyurethane products today include cellular materials such as water-blown ﬂexible foams or ﬂuorocarbon-blown rigid foams, elastomers, coatings, and elastic ﬁbers, which are described subsequently. Closely related to polyurethanes is an isocyanate-based product called isocyanurate foam. 4.3.4.1 Polyurethane Rubbers and Spandex Fibers By careful formulations it is possible to produce polyurethane rubbers with a number of desirable properties [50]. The rubbers can be thermoplastic (linear) or thermoset (slightly cross-linked) products. 530 Plastics Technology Handbook CH3 CH3 HNO3 NO2 CH3 H2 CH3 NCO NH2 COCl 2 H2N O2N CH3 OCN + NCO (2,4-TDI) NCO (2,6-TDI) (a) NO2 HNO3 NH2 NH2 NH2 NH2 CH2O HCl H2 CH2 COCl2 CH2 n NCO NCO CH2 CH2 + NCO OCN NCO CH2 n (MDI) (PMDI) (b) 3OCN (CH2)6 NCO + H2O OCN(CH2)6 NH C N O C NH (CH2)6NCO O (CH2)6 + CO2 NCO (c) FIGURE 4.28 Reactions used in the manufacture of commercial isocyanates. (a) TDI. (b) PMDI and MDI. (c) Aliphatic triisocyanate. Polyesters: HO R OH + HOC O R' COH O Polythers: CH CH2 HO R O C R' C O R O O R R O HO CH2 CH O n C R' C O R OH O O R R OCH2 CH n O CH2 CH OH Polycoproloclone CH2 CH2 CH2 + HO R CH2 CH2 OH HO R O C (CH2)5 O O n H O C O FIGURE 4.29 Reactions used in the manufacture of macroglycols. 4.3.4.1.1 Cross-Linked Polyurethane Rubbers The starting point in the preparation of this type of rubber, typiﬁed by Vulkollan rubbers, is a polyester prepared by reacting a glycol such as ethylene or propylene glycol with adipic acid. The glycol is in excess so that the polyester formed has hydroxyl end groups. This polyester macroglycol is then reacted with an excess of a diisocyanate such as 1,5-naphthalene diisocyanate or MDI (Figure 4.28). The molar excess of diisocyanate is about 30%, so the number of polyesters joined together is only about 2–3, and the resulting prepolymer has isocyanate end groups (see Figure 4.30a). 531 Industrial Polymers Prepolymer formation O C N R N C O + HO P OH Di-isocyanate N ( R NH C O P O C NH )n R N C O O O Urethane prepolymer O C Glycol (a) Chain extension of prepolymer (i) With water (in the manufacture of fooms): O NH C NH Urea link NCO + H2O + OCN Prepolymer Prepolymer + CO2 O (ii) With glycols NCO + HO R OH + OCN O NH C O R O C NH Urethane link O (iii) With amines NCO + H2N R NH2 + OCN O NH C NH R NH C NH Double urea link (b) Cross-linking of chain-extented polyurethane NHCONH NCONH OCN OCNH Biurel link NCO NHCOO (c) FIGURE 4.30 HNCO Allophonate link NCOO Equations for preparation, chain extension, and curing of polyurethanes. The prepolymer can be chain extended with water, glycols, or amines which link up prepolymer chains by reacting with terminal isocyanate groups (see Figure 4.30b). (The water reaction liberates carbon dioxide, so it must be avoided in the production of elastomers, but it is important in the manufacture of foams.) The urea and urethane linkages formed in the chain extension reactions also provide sites for branching and cross-linking, since these groups can react with free isocyanate or terminal isocyanae groups to form biuret and allophanate linkages, respectively (see Figure 4.30c). Biuret links, however, predominate since the urea group reacts faster than the urethane groups. The degree of cross-linking can to some extent be controlled by adjusting the amount of excess isocyanate, whereas more highly crosslinked structures may be produced by the use of a triol in the initial polyester. Vulkollan-type rubbers suffer from the disadvantage that the prepolymers are unstable and must be used within a day or two of their production. Moreover, these rubbers cannot be processed with conventional rubber machiners, so the products are usually made by a casting process. Attempts were then made to develop other polyurethane rubbers which could be processed by conventional techniques. One approach was to react the diisocyanate with a slight excess of polyester so that the prepolymer produced has terminal hydroxyl groups. The prepolymers are rubberlike gums and can be compounded with other ingredients on two-roll mills. Final curing can be done by the addition of a diisocyanate or, preferably, a latent diisocyanate, i.e., a substance which produces an active diisocyanate under the conditions of molding. Polyurethane rubbers of this class are exempliﬁed by Chemigum SL (Goodyear), Desmophen A (Bayer), and Daltoﬂex 1 (ICI), which used polyester–amide for the manufacture of prepolymer. Another approach was adopted by DuPont with the product Adiprene C, a polyurethane rubber with unsaturated groups that allow vulcanization with sulfur. Polyurethane rubbers, in general, and the Vulkollan-type rubbers, in particular, possess certain outstanding properties. They usually have higher tensile strengths than other rubbers and possess excellent tear and abrasion resistance. The urethane rubbers show excellent resistance to ozone and oxygen (in 532 Plastics Technology Handbook contrast to diene rubbers) and to aliphatic hydrocarbons. However, they swell in aromatic hydrocarbons and undergo hydrolytic decomposition with acids, alkalis, and prolonged action of water and steam. Though urethane rubbers are more costly than most of other rubbers, they are utilized in applications requiring superior toughness and resistance to tear, abrasion, ozone, fungus, aliphatic hydrocarbons, and dilute acids and bases. In addition, they excel in low-temperature impact and ﬂexibility. Urethane rubbers have found increasing use for forklift tires, shoes soles and heels, oil seals, diaphragms, chute linings, and a variety of mechanical applications in which high elasticity is not an important prerequisite. Emphasis on reaction injection-molding (RIM) technology in the automotive industry to produce automotive exterior parts has created a large potential for thermoset polyurethane elastomers. Reaction injection molding, originally known as reaction casting, is a rapid, one-step process to produce thermoset polyurethane products from liquid monomers. In this process liquid monomers are mixed under high pressure prior to injection into the mold. The polymerization occurs in the mold. Commercial RIM polyurethane products are produced from MDI, macroglycols, and glycol or diamine extenders. The products have the rigidity of plastics and the resiliency of rubber. Later advances include short, glass-ﬁber reinforced, high-modulus (ﬂexural modulus greater than 300,000 psi, i.e., 2070 MPa) polyurethane elastomers produced by the reinforced RIM process. These reinforced high-modulus polyurethane elastomers are considered for automotive door panels, trunk lids, and fender applications. Though originally developed for the automotive industry for the production of car bumpers, the RIM process has found its greatest success in the shoe industry, where semiﬂexible polyurethane foams have proved to be good soling materials. 4.3.4.1.2 Thermoplastic Polyurethane Rubbers The reactions of polyols, diisocyanates, and glycols, as described, do tend to produce block copolymers in which hard blocks with glass transition temperatures well above normal ambient temperature are separated by soft rubbery blocks. These polymers thus resemble the SBS triblock elastomers and, more closely, the polyether–ester thermoplastic elastomers of the Hytrel-type described earlier. In a typical process of manufacturing thermoplastic polyurethane elastomers, a prepolymer is ﬁrst produced by reacting a polyol, such as linear polyester with terminal hydroxyl groups, or a hydroxylterminated polyether, of molecular weights in the range of 800–2500, with an excess of diisocyanate (usually of the MDI type) to give a mixture of isocyanate-terminated polyol prepolymer and free (unreacted) diisocyanate. This mixture is then reacted with a chain extender such as 1,4-butanediol to give a polymer with long polyurethane segments whose block length depends on the extent of excess isocyanate and the corresponding stoichiometric glycol. The overall reaction is shown in Figure 4.31a. Provided that R (in free diisocyanate) and R′ (in glycol) are small and regular, the polyurethane segments will show high intersegment attraction (such as hydrogen bonding) and may be able to crystallize, thereby forming hard segments. In such polymers hard segments with Tg well above normal ambient temperature are separated by polyol soft segments, which in the mass are rubbery in nature. Hard and soft segments alternate along the polymer chain. This structure closely resembles that of polyester–polyether elastomers (Figure 4.11). Similar reactions occur when an amine is used instead of a glycol as a chain extender (see Figure 4.31b). The polymer in this case has polyurea hard segments separated by polyol soft segments. The polymers produced by these reactions are mainly thermoplastic in nature. Though it is possible that an excess of isothiocyanate may react with urethane groups in the chain to produce allophanate cross-links (see Figure 4.30c), these cross-links do not destroy the thermoplastic nature of the polymer because of their thermal lability, i.e., breaking down on heating and reforming on cooling. However, where amines have been used as chain extenders, urea groups are produced (Figure 4.31b), which, on reaction with excess isocyanate, may give the more stable biuret cross-links (see Figure 4.30c). Many of the commercial materials designated as thermoplastic polyurethanes are in reality slightly cross-linked. This cross-linking may be increased permanently by a post-curing reaction after shaping. 533 Industrial Polymers HO P + OH OCN R NCO + HO R´ OH Polyol O C NH R NH C O(R´ O C NH R NH C O)n O O O O P Polyurethane hard segment Polyol soft segment (a) HO OH + OCN R NCO + H2N R´ NH2 P Polyol O C NH (R NH C NH R´ NH C NH)n R NHC O P O Polyol soft segment (b) FIGURE 4.31 O O O Urethane-terminated polyurea hard segment Reactions for the manufacture of polyurethane block copolymers. The polyurethane product Estane (Goodrich) may, however, be regarded as truly thermoplastic. The thermoplastic rubbers have properties similar to those of Vulkollan-type cast polyurethane rubbers, but they have higher values for compression set. Thermoplastic polyurethane elastomers can be molded and extruded to produce ﬂexible articles. Applications include wire insulation, hose, tracks for all-terrain vehicles, solid tires, roller skate wheels, seals, bushing, convoluted bellows, bearings, and small gears for high-load applications. In the automobile industry thermoplastic polyurethanes are used primarily for exterior parts. Their ability to be painted with ﬂexible polyurethane-based paints without pretreatment is valuable. 4.3.4.1.3 Spandex Fibers One particular form of thermoplastic polyurethane elastomers is the elastic ﬁber known as Spandex. The ﬁrst commercial material of this type was introduced by DuPont in 1958 (Lycra). It is a relatively highpriced elastomeric ﬁber made on the principle of segmented copolymers. Here again the soft block is a polyol, and the hard block is formed from MDI (4,4′-diisocyanatodiphenyl methane) and hydrazine. The reactions are shown in Figure 4.32. Note that this ﬁber-forming polymer contains urethane and semicarbazide linkages in the chain. The product is soluble in amide solvent, and the ﬁber is produced by HO OH Polyol (Mol.wt.2000) H2NNH2 + OCN R OCN R NH C (Chain extender) O O Soft segment FIGURE 4.32 O O O C NH R NCO Prepolymer O O C NH R NH C NHHN C NH R NH )n (C O R NCO MDI O O Hard segment CH2 Reactions for the manufacture of segmented elastomeric ﬁber (Lycra). 534 Plastics Technology Handbook dry spinning from a solution. Major end uses of Lycra are in apparel (swimsuits and foundation garments). Subsequently several other similar materials have been introduced, including Dorlastan (Bayer), Spanzelle (Courtaulds), and Vyrene (U.S. Rubber). The polyol component with terminal hydroxyl groups used in the production of the foregoing materials may be either a polyether glycol or a polyester glycol (see Figure 4.29). For example, DuPont uses polytetrahydrofuran (a polyether glycol) for Lycra. U.S. Rubber originally used a polyester of molecular weight of about 2000, obtained by condensation of adipic acid with a mixture of ethylene glycol and propylene glycol, and a polyether-based mixture was used for Vyrene 2, introduced in 1967. These polyols are reacted with an excess of diisocyanate to yield an isocyanate-terminated prepolymer which is then chain extended by an amine such as hydrazine (NH2NH2) or ethylenediamine (see Figure 4.32). Fibers are usually spun from solution in dimethylformamide. Possessing higher modulus, tensile strength, resistance to oxidation, and ability to be produced at ﬁner deniers, spandex ﬁbers have made severe inroads into the natural rubber latex thread market. Major end uses are in apparel. Staple ﬁber blends with nonelastic ﬁbers have also been introduced. 4.3.4.2 Flexible Polyurethane Foam Although polyurethane rubbers are specialty products, polyurethane foams are well known and widely used materials [51]. About half of the weight of plastics in modern cars is accounted for by such foams. Flexible urethane foam is made in low densities of 1–1.2 lb/ft3 (0.016–0.019 g/cm3), intermediate densities of 1.2–3 lb/ft3 (0.019–0.048 g/cm3), high resilience (HR) foams of 1.8–3 lb/ft3 (0.029–0.048 g/cm3), and semiﬂexible foams of 6–12 lb/ft3 (0.096–0.192 g/cm3). Filled foams of densities as high as 45 lb/ft3 (0.72 g/cm3) have also been made. In many respects the chemistry of ﬂexible urethane foam manufacture is similar to that of the Vulkollan-type rubbers except that gas evolution reactions are allowed to occur concurrently with chain extension and cross linking (see Figure 4.30). Most ﬂexible foams are made from 80/20 TDI, which refers to the ratio of the isomeric 2,4-tolylene-diisocyanate to 2,6-tolylene-diisocyanate. Isocyanates for HR foams are about 80% 80/20 TDI and 20% PMDI, and those for semiﬂexible foams are usually 100% PMDI. Polyols usually are polyether based, since these are cheaper, give greater ease of foam processing, and provide more hydrolysis resistance than polyesters. Polyether polyols can be diols, such as polypropylene glycols with molecular weight of about 2000, or triols with molecular weight of 3000–6000. The most common type of the latter is triol adduct of ethylene oxide and propylene oxide with glycerol with a molecular weight of 3000. For HR foams polyether triols with molecular weight of 4500–6000 made by reacting ethylene oxide with polypropylene oxide based triols, are used. The reaction of isocyanate and water that evolves carbon dioxide (see Figure 4.30b) is utilized for foaming in the production of ﬂexible foams. The density of the product, which depends on the amount of gas evolved, can be reduced by increasing the isocyanate content of the reaction mixture and by correspondingly increasing the amount of water to react with the excess isocyanate. For greater softness and lower density some ﬂuorocarbons (F-11) may be added in addition to water. The processes for producing ﬂexible polyurethane foams have been described in Chapter 2. 4.3.4.2.1 Applications The largest-volume use of ﬂexible foam is furniture and bedding, which account for nearly 47% of the ﬂexible polyurethane foam market. Almost all furniture cushioning is polyurethane. Automobile seating is either made from ﬂexible slabstock or poured directly into frames, so-called deep seating, using HR chemical formulations. Semiﬂexible polyurethane foams ﬁnd use in crash pads, arm and head rests, and door panels. Much ﬂexible foam is used in carpet underlay. About 43% is virgin foam, and 57% is scrap rebonded with adhesives under heat and pressure. Industrial Polymers 535 Thermal interlining can be made by ﬂame bonding thin-sliced, low-density (1–1.5 lb/ft3) polyesterbased fabric. The ﬂame melts about one-third of the foam thickness, and the molten surface adheres to the fabric. Flexible foams also ﬁnd use in packaging applications. Die-cut ﬂexible foam is used to package costly goods such as delicate instruments, optical products, and pharmaceuticals. Semiﬂexible foam lining is used for cart interiors to protect auto and machine parts during transportation. 4.3.4.3 Rigid and Semirigid Polyurethane Foams Most rigid foam is made from polymeric isocyanate (PMDI) and difunctional polyether polyols. PMDI of functionality 2.7 (average number of isocyanate groups per molecules) is used in insulation foam manufacturing. Functionalities greater than two contribute rigidity through cross-linking. Higherfunctionality, low-molecular-weight polyols are sometimes added because they contribute rigidity by cross-linking and short chain length. Such polyols are made by reaction of propylene oxide with sucrose, pentaerythritol, or sorbitol. Rigid insulation foams are usually hydrocarbon blown to produce a closed-cell foam with excellent insulation properties. High-density foams are water blown where structural and screw-holding strength is needed, and halocarbons are used as blowing agents when high-quality, decorative surfaces are required. A compromise between these two aims can be achieved by the addition of water to halocarbon-blown formulations. Formulation, processing methods, properties, and applications of rigid and semirigid polyurethane foams have been described in have been described in Chapter 2. Major use of these foams are in building and construction (56%), transportation (12%), furniture, and packaging. Low-density, rigid foam is the most efﬁcient thermal insulation commercially available and is extensively used in building construction. In transportation, urethane insulation is used in rail cars, containers, truck trailer bodies, and in ships for transporting liqueﬁed natural gas. This requires thinner insulation and so yields more cargo space. Rigid foam is also used to give ﬂotation in barge compartments. High-density rigid foam, 5–15 lb/ft3 (0.08–0.24 g/cm3), is used for furniture items such as TV and stereo cabinets, chair shells, frames, arms, and legs, cabinet drawer fronts and doors, and mirror and picture frames. RIM-molded, integral-skin, high-density foams with core densities of 10–20 lb/ft3 (0.16– 0.32 g/cm3) and skin densities of 55–65 lb/ft3 (0.88–1.04 g/cm3 are used in electronic, instrument, and computer cabinets. Industrial uses of rigid foams include commercial refrigeration facilities as well as tanks and pipelines for cryogenic transport and storage. 4.3.4.3.1 Polyisocyanurates Though closed-cell rigid polyurethane foams are excellent thermal insulators, they suffer from the drawback of unsatisfactory ﬁre resistance even in the presence of phosphorus- and halogen-based ﬁre retardants. In this context, polyisocyanurates, which are also based on isocyanates, have shown considerable promise. Isocyanurate has greater ﬂame resistance then urethane. Although rigid polyurethane is speciﬁed for temperatures up to 200°F (93°C), rigid polyisocyanurate foams, often called trimer foams, withstand use temperatures to 300°F (149°C). Physical properties and insulation efﬁciency are similar for both types. The underlying reaction for polyisocyanurate formation is trimerization of an isocyanate under the inﬂuence of speciﬁc catalysts (Figure 4.33a). The most commonly used is a polymeric isocyanate (PMDI) prepared by reacting phosgene with formaldehyde–aniline condensates, as shown in Figure 4.28b. PMDIs are less reactive than monomeric diisocyanate but are also less volatile. The polyisocyanurate produced from this material will be of type shown in Figure 4.33b. Some of the catalysts used for the polytrimerization reactions are alkali-metal phenolates, alcoholates and carboxylates, and compounds containing o-(dimethylaminomethyl)-phenol groups. 536 Plastics Technology Handbook O N R N C O R C R N C C O N O R (a) N N X X CH O 2 CH 2 N C N N O C C O N CH2 X N N (b) FIGURE 4.33 (a) Trimerization of an isocyanate. (b) Structure of polyisocyanurate produced from polymeric MDIs. To produce foams, ﬂuorocarbons such as trichloroﬂuoromethanes are use as the sole blowing agents. Polyisocyanurate foams may be prepared by using standard polyurethane foaming equipment and a twocomponent system, with isocyanate and ﬂuorocarbon forming one component and the activator or activator mixture forming the second component. Because of the high cross-link density of polyisocyanurates, the resultant foam tends to be brittle. Consequently, there has been a move toward making polyisocyanurate–polyurethance combinations. For example, the isocyanate trimerization reaction has been carried out with isocyanate end-capped TDIbased prepolymers to make isocyanurate-containing polyurethane foams. Isocyanate trimerization in the presence of polyols of molecular weights less than 300 has also been employed to produce foams by both one-shot and prepolymer methods. 4.3.4.4 Polyurethane Coatings Polyurethane systems are also formulated for surface-coating applications. A wide range of such products has become available. These include simple simple solutions of ﬁnished polymer (linear polyurethanes), One component systems containing blocked isocyanates, two-component systems based on polyester and isocyanate or polyether and isocyanate, and a variety of prepolymer and adduct systems. Coatings based reaon TDI and MDI gradually discolor upon exposure to light and oxygen. In contrast, aliphatic diisocyanates such as methylene dicyclohexyl isocyanate, hexamethylene diisocyanate derivatives, and isophorone diisocyanate all produce yellowing resistant, clear, or color-stable pigmented 537 Industrial Polymers TABLE 4.22 Blocked Isocyanates for One-Component System Unblocking Temperature Blocking Agent °C °F Phenol 160 320 m-Nitrophenol 130 266 180 130–140 356 266–284 Acetone oxime Diethyl malonate Caprolactam Hydrogen cyanide Δ HN N O N O O 160 320 120–130 248–266 O CNH(CH2)3NCO O OCN(CH2)3 NHC NH O HO R OH O O C NH(CH2)3 NHC O O CNH(CH2)3NH C O R O n FIGURE 4.34 Reactions of macrocyclic ureas used as masked diisocyanates. coatings. Of the polyols used, half are polyester type and half are polyehter types. The coatings can vary considerably in hardness and ﬂexibility, depending on formulation. Polyurethane coatings are used wherever applications require toughness, abrasion resistance, skin ﬂexibility, fast curing, good adhesion, and chemical resistance. Use include metal ﬁnishes in chemical plants, wood ﬁnishes for boats and sports equipment. ﬁnishes for rubber goods. and rainerosion resistant coatings for aircraft. The one-component coating systems require blocking of the isocyanate groups to prevent polymerization in the container. Typical blocking agents are listed in Table 4.22. Generation of the blocking agent upon heating to cause polymerization is a disadvantage of blocked one-component systems. This problem cab be overcome by using a masked aliphatic diisocyanate system, as shown in Figure 4.34. The cyclic bisurea derivative used in such a system is stable in the polyol or in water emulsion formulated with the polyol at ordinary temperatures. Upon heating, ring opening occurs, generating the diisocyanate, which reacts instantaneously with the macroglycol to form a polyurethane coating. Polyurethane adhesives involving both polyols and isocyanates are used. These materials have found major uses in the boot and shoe industry. 4.3.5 Ether Polymers For the purposes of this chapter ether polymers or polyethers are deﬁned as polymers which contain recurring ether groupings in their backbone structure: C O C Polyethers are obtained from three different classes of monomers, namely, carbonyl compounds, cyclic ethers, and phenols. They are manufactured by a variety of polymerization processes, such as ionic 538 Plastics Technology Handbook polymerization (polyacetal), ring-opening polymerization (polyethylene oxide, polyprophylene oxide, and epoxy resins), oxidative coupling (polyphenylene oxide), and polycondensation (polysulfone). Polyacetal and polyphenylene oxide are widely used as engineering thermoplastics, and epoxy resins are used in adhesive and casting application. The main uses of poly(ethylene oxide) and poly(propylene oxide) are as macroglycols in the production of polyurethanes. Polysulfone is one of the high-temperatureresistant engineering plastics. 4.3.5.1 Polyacetal OCH2 Monomers Formaldehyde, trioxane n Polymerization Major Uses Cationic or anionic chain polymerization Appliances, plumbing and hardware, transportation Polyoxymethylene (polyacetal) is the polymer of formaldehyde and is obtained by polymerization of aqueous formaldehyde or ring-opening polymerization of trioxane (cyclic trimer of formaldehyde, melting point 60–60°C), the latter being the preferred method [52]. This polymerization of trioxane is conducted in bulk with cationic initiators. In contrast, highly puriﬁed formaldehyde is polymerized in solution using either cationic or anionic initiators. Polyacetal strongly resembles polyethylene in structure, both polymers being linear with a ﬂexible chain backbone. Since the structures of both the polymers are regular and the question of tacticity does not arise, both polymers are capable of high degree of crystallization. However, the acetal polymer molecules have a shorter backbone (–C–O–) bond and so pack more closely together than polyethylene molecules. The acetal polymer is thus harder and has a higher melting point (175°C for the homopolymer). Being crystalline and incapable of speciﬁc interaction with liquids, acetal homopolymer resins have outstanding resistance to organic solvents. No effective solvent has yet been found for temperatures blow 70°C (126°F). Above this temperature, solution occurs in a few solvents such as chlorophenols. Swelling occurs with solvents of similar solubility parameter to that of the polymer [d = 11.1 (cal/cm3)1/2 = 22.6 MPa1/2]. The resistance of polyacetal to inorganic reagents in not outstanding, however. Strong acids. strong alkalis, and oxidizing agents cause a deterioration in mechanical properties. The ceiling temperature for the acetal polymer is 127°C. Above this temperature the thermodynamics indicate that depolymerization will take place. Thus it is absolutely vital to stabilize the polyacetal resin sufﬁciently for melt processing at temperature above 200°C. Stabilization is accomplished by capping the thermolabile hydroxyl end groups of the macromolecule by etheriﬁcation or esteriﬁcation, or by copolymerizing with small concentrations of ethylene oxide. These expedients retard the initiation or propagation steps of chain reaction that could cause the polymer to unzip to monomer (formaldehyde). End-group capping is more conveniently achieved by esteriﬁcation using acetic anhydride. OCH2 HOCH2 CH3C O O CH2 n O OCH2 CH2OH n O Ac2O CH2 O CCH3 O If formaldehyde is copolymerized with a second monomer, which is a cyclic ether such as ethylene oxide and 1,3-dioxolane, end-group capping is not necessary. The copolymerization results in occasional incorporation of molecules containing two successive methylene groups, whereby the tendency of the molecules to unzip is markedly reduced. This principle is made use of in the commercial products marketed as Celcon (Celanese), Hostaform (Farbwerke Hoechst), and Duracon (Polyplastic). 539 Industrial Polymers Degradation of polyacetals may also occur by oxidative attack at random along the chain leading to chain scission and subsequent depolymerization (unzipping). Oxidative chain scission is reduced by the use of antioxidants (see Chapter 1), hindered phenols being preferred. For example, 2,2′-methylene-bis(4methyl-6-t-butylphenol) is used in Celcon (Celanese) and 4,4′-butylidene bis(3-methyl-6-t-butylphenol) in Delrin (DuPont). Acid-catalyzed cleavage of the acetal linkage can also cause initial chain scission. To reduce this acid acceptors are believed to be used in commercial practice. Epoxides, nitrogen-containing compounds, and basic salts are all quoted in the patent literature. Polyacetal is obtained as a linear polymer (about 80% crystalline) with an average molecular weight of 30,000–50,000. Comparative values for some properties of typical commercial products are given in Table 4.23. The principal features of acetal polymers which render them useful as engineering thermoplastics are high stiffness, mechanical strength over a wide temperature range, high fatigue endurance, resistance to creep, and good appearance. Although similar to nylons in many respects, acetal polymers are superior to them in fatigue resistance, creep resistance, stiffness, and water resistance (24-h water absorption at saturation being 0.22% for acetal copolymer vs. 8.9% for nylon-6,6). The nylons (except under dry conditions) are superior to acetal polymers in impact toughness. Various tests indicate that the acetal polymers are superior to most other plastics and die cast aluminum. The electrical properties of the acetal polymers may be described as good but not outstanding. They would thus be considered in applications where impact toughness and rigidity are required in addition to good electrical insulation characteristics. The end-group capped acetal homopolymer and the trioxane-based copolymers are generally similar in properties. The copolymer has greater thermal stability, easier moldability, better hydrolytic stability at elevated temperatures, and much better alkali resistance than the homopolymer. The homopolymer, on the other hand, has slightly better mechanical properties, e.g., higher ﬂexural modulus, higher tensile strength, and greater surface hardness. TABLE 4.23 Typical Values for Some Properties of Acetal Homopolymers and Copolymers Property Speciﬁc gravity Crystalline melting point (°C) Acetal Homopolymer Acetal Copolymer 1.425 175 1.410 163 10,000 70 8500 58 410,000 2800 360,000 2500 100 170 110 158 Tensile strength (23°C) lbf/in.2 MPa Flexural modulus (23°C) lbf/in.2 MPa Deﬂection temperature °C) at 264 lbf/in.2 at 66 lbf/in.2 Elongation at break (23°C) (%) 15–75 23–35 Impact strength (23°C) ft.-lbf/in. notch 1.4–2.3 1.1 Hardness, Rockwell M 94 80 Coefﬁcient of friction Water absorption (%) 0.1–0.3 0.2 24-h immersion 0.4 0.22 50% RH equilibrium Continuous immersion equilibrium 0.2 0.9 0.16 0.8 Source: From Brydson, J. A. 1982. Plastics Materials. Butterworth Scientiﬁc, London, UK. 540 Plastics Technology Handbook Acetal polymers and copolymers are engineering materials and are competitive with a number of plastics materials, nylon in particular, and with metals. Acetal resins are being used to replace metals because of such desirable properties as low weight (sp. gr. 1.41–1.42), corrosion resistance, resistance to fatigue, low coefﬁcient of friction, and ease of fabrication. The resins may be processed without difﬁculty on conventional injection-molding, blow-molding, and extrusion equipment. The acetal resins are used widely in the molding of telephone components, radios, small appliances, links in conveyor belts, molded sprockets and chains, pump housings, pump impellers, carburetor bodies, blower wheels, cams, fan blades, check valves, and plumbing components such as valve stems and shower heads. Because of their light weight, low coefﬁcient of friction, absence of slipstick behavior, and ability to be molded into intricate shapes in one piece, acetal resins ﬁnd use as bearings. The lowest coefﬁcient of friction and wear of acetal resin are obtained against steel. Though counted as one of the engineering plastics, acetal resins with their comparatively high cost cannot, however, be considered as general-purpose thermoplastics in line with polyethylene, polypropylene, PVC, and polystyrene 4.3.5.2 Poly(Ethylene Oxide) OCH2CH2 Monomer Ethylene oxide Polymerization Major Uses Ring-opening polymerization Molecular weights 200–600—surfactants, humectants, lubricants Molecular weights>600 pharmaceutical and cosmetic bases, lubricants, mold release agents Molecular weights 105 to 5 × 106—watersoluble packaging ﬁlms and capsules Poly(ethylene oxide) of low molecular weight, i.e., below about 3000, are generally prepared by passing ethylene oxide into ethylene glycol at 120–150°C (248–302 F) and about 3 atm pressure (304 kPa) by using an alkaline catalyst such as sodium hydroxide [53]. Polymerization takes place by an anionic mechanism. CH2 CH2 NaOK HO ( CH2 CH2 O (n CH2 CH2 O– Na+ O HO ( CH2 CH2 O ( CH2 CH2 O– Na+ + H2O n HO ( CH2 CH2 O (n CH2 CH2 OH + Na OH The polymers produced by these methods are thus terminated mainly by hydroxyl groups and are oftern referred to as polyethylene glycols (PEGs). Depending on the chain length, PEGs range in physical form at room temperature from water white viscous liquids (mol. wt. 200–700), through waxy semisolids (mol. wt. 1000–2000), to hard, waxlike solids (mol. wt. 3000–20,000 and above). All are completely soluble in water, bland, nonirritating, and very low in toxicity; they possess good stability, good lubricity, and wide compatibility. Since PEGs form a homologous series of polymers, many of their properties vary in a continuous manner with molecular weight. The freezing temperature, which is less than −10°C for PEG of molecular weight 300, rises ﬁrst rapidly with molecular weight through the low-molecular-weight grades, then Industrial Polymers 541 increases more gradually through the solids while approaching 66°C, the true crystalline melting point for very high-molecular-weight poly(ethylene oxide) resins. Other examples of such continuous variations are the increase in viscosity, ﬂash points, and ﬁre points with an increase in molecular weight and a slower increase in speciﬁc gravity. In a reverse relationship, hygroscopicity decreases as molecular weight increases, as does solubility in water. In 1958, commercial poly(ethylene oxide)s of very high molecular weight became available from Union Carbide (trademark Polyox). Two Japanese companies, Meisei Chemical Works, Ltd., and Seitetsu Kagaku Company, Ltd., began producing poly(ethylene oxide) under the trademarks Alkox and PEO, respectively. Heterogeneous catalyst systems, which are mainly of two types, namely, alkaline earth compounds (e.g., oxides and carbonates of calcium, barium, and strontium) and organometallic compounds (e.g., aluminum and zinc alkyls and alkoxides, usually with cocatalysts) are used in their manufacture. Commercial poly(ethylene oxide) resins supplied in the molecular weight range 1×105–5×106 are dry, free-ﬂowing, white powders soluble in an unusually broad range of solvents. The resins are soluble in water at temperatures up to 98°C and also in a number of organic solvents, which include chlorinated hydrocarbons such as carbon tetrachloride and methylene chloride, aromatic hydrocarbons such as benzene and toluene, ketones such as acetone and methyl ethyl ketone, and alcohols such as methanol and isopropanol. 4.3.5.3 Applications 4.3.5.3.1 Polyethylene Glycol The unusual combination of properties of PEGs has enabled them to ﬁnd a very wide range of commercial uses as cosmetic creams, lotions, and dressings; textile sizes; paper-coating lubricants; pharmaceutical sales, ointments, and suppositories, softeners and modiﬁers; metal-working lubricants; detergent modiﬁers; and wood impregnates. In addition, the chemical derivatives of PEGs, such as the mono- and diesters of fatty acids, are widely used as emulsiﬁers and lubricants. The PEGs themselves show little surface activity, but when converted to mono- and diesters by reaction with fatty acids, they form a series of widely useful nonionic surfactants. The required balance of hydrophilic–hydrophobic character can be achieved by suitable combination of the molecular weight of the PEG and the nature of the fatty acid. For large-volume items a second production route of direct addition of ethylene oxide to the fatty acids is often preferred. End uses for the fatty esters are largely as textile lubricants and softeners and as emulsiﬁers in food products, cosmetics, and pharmaceuticals. The PEGs have found a variety of uses in pharmaceutical products. Their water solubility, blandness, good solvent action for many medicaments, pleasant and nongreasy feel on the skin, and tolerance of body ﬂuids are the reasons why they are frequently the products of choice. Blends of liquid and solid grades are often selected because of their desirable petrolatum like consistency. An especially important example of pharmaceutical application of PEGs is as bases for suppositories, where the various molecular grades can be blended to provide any desired melting point, degree of stability, and rate of release of medication. The fatty acid esters of the PEGs are often used in pharmaceuticals as emulsiﬁers and suspending agents because of their nonionic, blandness, and desirable surface activity. The solid PEGs also ﬁnd use a lubricants and binders in the manufacture of medicinal tablets. The PEGs, providing they contain not over 0.2% ethylene and diethylene glycols, are permitted as food additives. The PEG fatty acid esters are especially useful emulsifying agents in food products. For many of the same reasons which account for their use in pharmaceuticals, the PEGs and their fatty acid esters ﬁnd many applications in cosmetics and toiletries. Their moisturizing, softening, and skinsmoothing characteristics are especially useful. Typical examples of applications are shaving creams, vanishing creams, toothpastes, powders, shampoos, hair rinses, suntan lotions, pomades and dressings, deodorants, stick perfumes, rouge, mascara, and so on. The nonionic surface-active PEG fatty acid esters ﬁnd use in a variety of detergents and cleaning compositions. Liquid PEGs, solid PEGs, or their solutions are often used in a variety of ink preparations, such as thixotropic inks for ballpoint pens, water-based stencil inks, steam-set printing inks, and stamp-pad inks. 542 Plastics Technology Handbook There are numerous other industrial uses for the PEGs in which they serve primarily as processing aids and do not remain as integral components of the products. The PEGs add green strength and good formability to various ceramic components of the products. The PEGs add green strength and good formability to various ceramic compositions to be stamped, extruded, or molded. They can be burned out cleanly during subsequent ﬁring operations. In electroplating baths small amounts of PEGs improve smoothness and grain uniformity of the deposited coatings. Solid PEGs are effective lubricants in paper-coating compositions and promote better gloss and smoothness in calendaring operations. PEGs, and more particularly their fatty acid esters, are quite widely used as emulsiﬁers, lubricants, and softeners in textile processing. PEGs and their esters ﬁnd use as components of metal corrosion inhibitors in oil wells where corrosive brines are present. Fatty acid esters of PEGs are useful demulsiﬁers for crude oil–water separation. The water solubility, nonvolatility, blandness, and good lubricating abilities are some of the reasons for the use of PEGs in metal-working operations. Metal-working lubricants for all but the most severe forming operations are made with the PEGs or with their esters or other derivatives. 4.3.5.3.2 Poly(Ethylene Oxide) Since their commercialization in 1958, the reported and established applications for high-molecularweight poly(ethylene oxide) resins have been numerous and diversiﬁed. Table 4.24 summarizes in alphabetical order the main applications of these resins in various industries. TABLE 4.24 Applications of Poly(Ethylene Oxide) Resins Industry Agriculture Applications Water-soluble seed tapes Water-soluble packages for agricultural chemicals Hydrogels as soil amendments to increase water retention Soil stabilization using association complexes with poly(acrylic acid) Ceramics and glass Drift control agent for sprays Binders for ceramics Size for staple glass-ﬁber yarns Chemical Electrical Dispersant and stabilizer in aqueous suspension polymerization Water-soluble, fugitive binder for microporous battery and fuel-cell electrodes Metals and mining Flocculant for removal of silicas and clays in hydrometallurgical processes Paper Flocculant for clariﬁcation of efﬂuent streams from coal-washing plants Filler retention and drainage aid in the manufacture of paper Flocculant for clariﬁcation of efﬂuent water Personal-care products Lubricant and toothpaste Tickener in preparation of shaving stick Opthalmic solution for wetting, cleaning, cushioning, and lubricating contact lenses Adhesion and cushioning ingredient in denture ﬁxatives Petroleum Hydrogels as adsorptive pads for catamenial devices and disposable diapers Thickener for bentonite drilling muds Thickener for secondary oil-recovery ﬂuids in waterﬂooding process Pharmaceutical Water-soluble coating for tablets Suspending agent to inhibit settling of ceramic lotion Printing Microencapsulation of inks Soap and detergent Textile Emollient and thickener for detergent bars and liquids Additive to improve dyeability and antistatic properties of polyoleﬁn, polyester, and polyamide ﬁbers Industrial Polymers 543 In addition to their water solubility and blandness, the main functions and effects of poly(ethylene oxide) resins which lead to these diverse applications are lubrication, ﬂocculation, thickening, adhesion, hydrodynamic drag reduction, and formation of association complexes. The resins are relatively nontoxic and have a very low level of biodegradability (low BOD). Poly (ethylene oxide) resins with molecular weights from 1×105 to 1×107 have a very low level of oral toxicity and are not readily absorbed from the gastrointestinal tract. The resins are relatively nonirritating to the skin and have a low sensitizing potential. Poly(ethylene oxide) resins can be formed into various shapes by using conventional thermoplastic processing techniques. Commercially, thermoplastic processing of these resins has been, however, limited almost exclusively to the manufacture of ﬁlm and sheeting. Generally, the medium-molecular-weight resins (4×105–6×105) possess melt rheology best suited to thermoplastic processing. The ﬁlms are produced by calendering or blown-ﬁlm extrusion techniques. Usually produced in thicknesses from 1 to 3 mils, the ﬁlms have very good mechanical properties combined with complete water solubility. Poly(ethylene oxide) ﬁlms have been used to produce seed tapes, which consist of seeds sandwiched between two narrow strips of ﬁlm scaled at the edges. When the seed tape is planted, water from the soil dissolves the water-soluble ﬁlm within a day or two, releasing the seed for germination. Because the seeds are properly spaced along the tape, the process virtually eliminates the need for thinning of crops. Films of poly(ethylene oxide) are also used to manufacture water-soluble packages for preweighed quantities of fertilizers, pesticides, insecticides, detergents, dyestuffs, and the like. The packages dissolve quickly in water, releasing the contents. They eliminate the need for weighing and offer protection to the user from toxic or hazardous substances. Poly(ethylene oxide) forms a water-insoluble association complex with poly(acrylic acid). This is the basis of microencapsulation of nonaqueous printing inks. Dry, free-ﬂowing powders obtained by this process can be used to produce “carbonless” carbon papers (see “Microencapsulation” in Chapter 5). When pressure is applied to the paper coated with the microencapsulated ink, the capsule wall ruptures and the ink is released. The formation of a water-insoluble association complex of poly(ethylene oxide) and poly(acrylic acid) is also the basis for a soil stabilization process to prevent erosion of soil on hillsides and river banks. A variety of different types of adhesives can be produced by forming association complexes of poly (ethylene oxide) with tannin or phenolic resins. Examples include wood glue, water-soluble quickset adhesive, and pressure-sensitive adhesives. High-molecular-weight poly(ethylene oxide) resins are effective ﬂocculants for many types of clays, coal suspensions, and colloidal silica, and so ﬁnd application as process aids in mining and hydrometallurgy. The turbulent ﬂow of water through pipes and hoses or over surfaces causes the effect known as hydrodynamic drag. High-molecular-weight poly(ethylene oxide) resins are most effective in reducing the hydrodynamic drag and thus ﬁnd use in ﬁre ﬁghting, where small concentrations of these resins (50– 100 ppm) reduce the pressure loss in ﬁre hoses and make it possible to deliver as much as 60% more water through a standard 2.5-in. (6.35-cm)-diameter ﬁre hose. The Union Carbide product UCAR Rapid Water Additive, which contains high-molecular-weight poly(ethylene oxide) as the active ingredient, is a hydrodynamic-drag-reducing additive for this application. The ability of poly(ethylene oxide) to reduce hydrodynamic drag has also led to its use in ﬂuid-jet systems used for cutting soft goods, such as textiles, rubber, foam, cardboard, etc. In these systems specially designed nozzles produce a very-small-diameter water jet at a pressure of 30,000–60,000 psi (200–400 MPa). Although a plain water disperses signiﬁcantly as it leaves the nozzle, with poly(ethylene oxide) addition the stream becomes more cohesive and maintains its very small diameter up to 4 in. (10 cm) from the nozzle. Chemical or irradiation cross-linking of poly(ethylene oxide) resins yield hydrogels, which are not water soluble but water absorptive, capable of absorbing 25–100 times their own weight of water. These 544 Plastics Technology Handbook hydrogels are reportedly useful in the manufacture of absorptive pads for catamenial devices and disposable diapers. The water absorbed by the hydrogels is also readily desorbed by drying the hydrogel. This characteristics is the basis for the use of these hydrogels as so-called soil amendments. When mixed with ordinary soil in a concentration of about 0.001–5.0 wt% of the soil, these hydrogels will reduce the rate of moisture loss due to evaporation but will still release water to the plants and thus eliminate the need for frequent watering of the soil. 4.3.5.4 Poly(Propylene Oxide) OCH2 CH CH3 Monomer Propylene oxide n Polymerization Major Uses Base-catalyzed ring-opening polymerization Polyols for polyurethane foams, surfactants, lubricants cosmetic bases Propylene oxide is polymerized by methods similar to those described in the preceding section for poly (ethylene oxide). Like the latter, low- and high-molecular-weight polymers are of commercial interest. Poly(propylene oxide)s of low molecular weight (i.e., from 500 to 3500), often referred to as polypropylene glycols (PPGs), are important commercial materials mainly because of their extensive use in the production of polyurethane foams (see Chapter 2 and Chapter 4). PPGs are less hydrophilic and lower in cost and may be prepared by polymerizing propylene oxide in the presence of propylene glycol as an initiator and sodium hydroxide as a catalyst at about 160°C. The polymers have the general structure HO CH CH3 CH3 CH3 CH2 O CH2 CH O n CH2 OH CH The end hydroxy groups of the polymer are secondary groups and are ordinarily rather unreactive in the urethane reaction. Initially, this limitation was overcome by the preparation of isocyanate-terminated prepolymer and by the use of block copolymers with ethylene oxide. The latter products are known as tipped polyols and are terminated with primary hydroxy groups of enhanced activity. CH3 HO CH2 CH2 O x CH2 CH O y CH2 CH2 O z CH2 CH2 OH (Note that straight PEG is not satisfactory for polyurethane foam production due to its water sensitivity and tendency to crystallize.) The later advent of more powerful catalysts, however, made it possible for straight PPG to be used in the preparation of ﬂexible polyurethane foams (see one-shot processes in Chapter 2) without recourse to the foregoing procedures. Also, today the bulk of the polyethers used are triols rather than diols, since these lead to slightly cross-linked ﬂexible foams with improved load-bearing characteristics. The polyether triols are produced by polymerizing propylene oxide by using 1,1,1-trimethylolpropane, 1,2,6-hexane triol, or glycerol as the initiator. The use of, for example, trimethylolpropane leads to the following polyether triol. 545 Industrial Polymers CH3 CH2 C CH2 O CH2 CH(CH3) O CH2 O CH2 CH(CH3) O CH2 O CH2 CH(CH3) O x CH2 CH(CH3) OH y CH2 CH(CH3) OH CH2 CH(CH3) OH z For ﬂexible polyurethane foams, polyether triols of molecular weights from 3000 to 3500 are normally used because they give the best balance of properties. For the production of rigid foams, polyether triols of lower molecular weight (about 500) are used to increase the degree of cross-linking. Alternatively, polyether polyols of higher functionality, such as produced by polymerizing propylene oxide with pentaerythritol or sorbitol, may be used. Copolymerization of ethylene oxide and propylene oxide yields quite valuable functional ﬂuids of various sorts. The random copolymers of ethylene and propylene oxides of relatively low molecular weights are water soluble when the proportion of ethylene oxide is at least 40–50% by weight. The block copolymers consist of sequences of “blocks” of all oxypropylene or all-oxyethylene groups, as shown. O O CH2 HO CH2 CH2 CH2 HO CH O y CH2 O CH2 CH2 x O H y CH2 CH CH3 CH CH2 O z H CH3 CH3 Properties vary considerably, depending on the lengths and arrangements of these blocks. The block copolymers comprise unique and valuable surface-active agents. They can act as breakers for water-in-oil emulsions, as defoamers, and as wetting and dispersing agents. 4.3.5.5 Epoxy Resins O CH2 CH CH2 CH3 O C n CH3 Monomers Bisphenol A, epichlorohydrin O CH3 OH O CH2 CH CH2 O C O CH2 CH CH2 CH3 Polymerization Major Uses Condensation and ringopening polymerization Surface coating (44%), laminates and composites (18%, moldings (9%), ﬂooring (6%), adhesives (5%) Epoxide or epoxy resins [14,54,55] contain the epoxide group, also called the epoxy, oxirane, or ethoxyline group, which is a three-membered oxide ring: C C C (The simplest compound in which the epoxy group is found is ethylene oxide.) In the uncured stage epoxies are polymers with a low degree of polymerization. They are most often used as thermosetting resins which cross-link to form a three-dimensional nonmelting matrix. A curing agent (hardener) is generally used to achieve the cross-linking. In room-temperature curing the hardener is generally an amine such as diethylene triamine or triethylenetetramine. For elevated temperature curing a number of different curing agents could be utilized, including aromatic amines and acid anhydrides. 546 Plastics Technology Handbook Epoxy resins ﬁrst developed commercially and still completely dominating the worldwide markets are those based on 2,2-bis(4′-hydroxyphenyl) propane, more commonly known as bisphenol A (as it is produced by condensation of phenol with acetone) and 1-chloro-2,3-epoxy-propane, also known as epichlorohydrin. It can be seen from the general formula of these resins that the molecular species concerned is a linear polyether with terminal glycidyl ether group O O CH2 CH CH2 and secondary hydroxyl groups occurring at regular intervals along the length of the macromolecule. The number of repeating units (n) depends essentially on the molar ratio of bisphenol A and epichlorohydrin. When n = 0, the product is diglycidyl ether and the molecular weight is 340. When n = 10, the molecular weight is about 3000. Commercial liquid epoxy resins based on bisphenol A and epichlorohydrin have average molecular weights from 340 to 400, and it is therefore obvious that these materials are composed largely of diglycidyl ether. The liquid resin is thus often referred to as DGEBA (diclycidyl ether of bisphenol A). Similarly, the epoxy resin based on bis(4-hydroxyphenyl)methane, more commonly as bisphenol F (as it is produced by the condensation of phenol with formaldehyde) and epichlorohydrin is referred to as DGEBF. DGEBA may be reacted with additional quantities of bisphenol A in an advancement reaction. This advancement produces higher-molecular-weight solid resins possessing a higher melting point (>90°C). Advancement generally increases ﬂexibility, improves salt fog corrosion resistance, and increases hydroxyl content, which can be utilized later for cross-linking. Possessing generally low functionality (number of epoxy groups), their major use is in coatings. They provide outstanding adhesion and good salt fog corrosion resistance. Commercial solid epoxy resins seldom have average molecular weights exceeding 4000, which corresponds to an average value of n of about 13. Resins with molecular weights above 4000 are of limited use since their high viscosity and low solubility make subsequent processing difﬁcult. 4.3.5.5.1 Resin Preparation The molecular weights of epoxy resins depend on the molar ratio of epichlorohydrin and bisphenol A used in their preparation (see Table 4.25). In a typical process for the production of liquid epoxy resins, epichlorohydrin and bisphenol A in the molar ratio of 10:1 are added to a stainless steel kettle ﬁtted with a powerful anchor stirrer. The water content of the mixture is reduced to below 2% by heating the mixture until the epichlorohydrin–water distils off. After condensation, the epichlorohydrin layer is returned to the kettle, the water being discarded. When the necessary water content of the reaction mixture is reached, the reaction is started by the slow addition of sodium hydroxide (2 mole per mole of bisphenol A) in the form of 40% aqueous solution, the temperature being maintained at about 100°C. The water content of the reaction mixture should TABLE 4.25 Effect of Reactant Ratio on Molecular Weight of Epoxy Resins Molar Ratio Epichlorohydrin/Bisphenol A Molecular Weight Epoxide Equivalenta Softening Point (°C) 10:1 2:1 370 451 192 314 9 43 1.4:1 791 592 84 1.33:1 1.25:1 802 1133 730 862 90 100 1.2:1 1420 1176 112 a This is the weight of resin (in grams) containing one epoxide equivalent. For a pure diglycidyl ether (mol. wt. 340) with two epoxy groups per molecule, epoxide equivalent = 340/2 = 170. 547 Industrial Polymers TABLE 4.26 Typical Properties of Some Commercial Glycidyl Ether Resins Resin Average Molecular Weight Epoxide Equivalent Melting Point (°C) – A 340–400 175–210 B 450 225–290 – C D 700 950 300–375 450–525 40–50 64–75 E 1400 870–1025 95–105 F G 2900 3800 1650–2050 2400–4000 125–132 145–155 Source: From Brydson, J. A. 1982. Plastics Materials. Butterworth Scientiﬁc, London, UK. be maintained between 0.31% and 2% by weight throughout the reaction if high yields (90–95%) are to be obtained. (No reaction occurs under anhydrous conditions, and undesired by-products are formed if the water content is greater than 2%). Besides distilling off the water as an azeotrope with the epichlorohydrin, the water content can also be partly controlled by the rate of addition of the caustic soda solution. When all the alkali has been added, which may take 2–3 h, the excess epichlorohydrin is recovered by distillation at reduced pressure. A solvent such as toluene or methyl isobutyl ketone is then added to the cooled reaction product to dissolve the resin and leave the salt formed in the reaction. The resin solution is washed with hot water, and after ﬁltration and further washing, the solvent is removed by distillation and the resin is dried by heating under vacuum. Though the pure diglycidyl ether of bisphenol A is a solid (m.p. 43°C), the commercial grades of the resin which contain a proportion of high-molecular-weight materials are supercooled liquid with viscosities of about 100–140 P at room temperature. The high-molecular-weight epoxy resins usually manufactured have values of n in the general formula ranging from 2 to 12. These resins are synthesized by allowing epichlorohydrin and bisphenol A to interact in the presence of excess sodium hydroxide. In the taffy process usually employed, a mixture of bisphenol A and epichlorohydrin (the molar ratio of the reactants used depends on the resin molecular weight required; see Table 4.25) is heated to 100°C and aqueous sodium hydroxide (NaOH–epichlorohydrin molar ratio 1.3:1) is added slowly with vigorous stirring, the reaction being completed in 1–2 h. A white puttylike taffy (which is an emulsion of about 30% water in resin and also contains salt and sodium hydroxide) rises to the top of the reaction mixture. The lower layer of brine is removed; the resinous layer is coagulated and washed with hot water until free from alkali and salt. The resin is then dried by heating and stirring at 150°C under reduced pressure, poured into cooling pans, and subsequently crushed and bagged. To avoid the difﬁculty of washing highly viscous materials, higher-molecular-weight epoxy resins may also be prepared by a two-stage process (advancement reaction, described earlier). This process consists of fusing a resin of lower molecular weight with more bisphenol A at about 190°C. The residual base content of the resin is usually sufﬁcient to catalyze the second-stage reaction between the phenolic hydroxyl group and the epoxy group. Typical data of some commercial glycidyl ether resins are given in Table 4.26. Solid resins have been prepared with narrow molecular-weight distributions. These resins melt sharply to give low-viscosity liquids, which enables incorporation of larger amounts of ﬁllers with a consequent reduction in cost and coefﬁcient of expansion. These resins are therefore useful in casting operations. 4.3.5.5.2 Curing To convert the epoxy resins into cross-linked structures, it is necessary to add a curing agent. Most of the curing agents in common use can be classiﬁed into three groups: tertiary amines, polyfunctional amines, and acid anhydrides. 548 Plastics Technology Handbook Examples of tertiary amines used as curing agents for epoxy resins include: CH3 CH2 CH2 CH3 CH2 CH3 N Triethylamine (TEA) CH3 CH2 N CH3 Bezyldimethylamine (BDA) OH CH3 CH2 N CH3 Dimethylaminomethylphenol (DMAMP) OH H3C CH3 N CH2 CH2 N H3C CH3 CH2 N CH3 CH3 Tri(dimethylaminomethyl) Phenol (TDMAMP) C2H5 X[HO C CH2 CH CH2 CH2 CH3]3 O Tri-2-ethylhexoate salt of TDMAMP (denoted as X) Tertiary amines are commonly referred to as catalytic curing agents since they induce the direct linkage of epoxy groups to one another. The reaction mechanism is believed to be as follows: O + R3N CH2 CH + R3N CH2 CH O O + R3N CH2 CH O– + CH2 – + CH R3N CH2 CH O CH2 CH O– 549 Industrial Polymers Since the reaction may occur at both ends of the diglycidyl ether molecule, a cross-linked structure will be built up. The overall reaction may, however, be more complicated because the epoxy group, particularly when catalyzed, also reacts with hydroxyl groups. Such groups may be present in the highermolecular-weight homologues of the diglycidyl ether of bisphenol A, or they may be formed as epoxy rings are opened during cure. In contrast to tertiary amine hardeners, which, as shown, cross-link epoxide resins by a catalytic mechanism, polyfunctional primary and secondary amines act as reactive hardeners and cross-link epoxy resins by bridging across epoxy molecules. An amine molecule with two active hydrogen atoms can link across two epoxy molecules, as shown: R O N H + CH2 CH2 + H CH OH O CH CH OH R CH2 N CH2 CH With polyfunctional aliphatic acid aromatic amines having three or more active hydrogen atoms in amine groups, this type of reaction results in a network polymer (see Figure 1.20). Generally speaking, aliphatic amines provide fast cures and are effective at room temperature, whereas aromatic amines are somewhat less reactive but give products with higher heat-distortion temperatures. Polyfunctional amines are widely used in adhesive, casting, and laminating applications. Diethylenetriamine (DETA), H2N–CH2–CH2–NH–CH2–CH2–NH2, and triethylenetetramine (TETA), H2N–CH2–CH2–NH–CH2–CH2–NH–CH2–CH2–NH2, are highly reactive primary aliphatic amines with ﬁve and six active hydrogen atoms available for cross-linking, respectively. Both materials will cure (harden) liquid epoxy resins at room temperature and produce highly exothermic reactions. With DETA the exothermic temperature may reach as high as 250°C in 200-g batches. With this amine used in the stoichiometric quantity of 9–10 pts phr (parts per hundred parts resin), the room temperature pot life is less then an hour. The actual time, however, depends on the ambient temperature and the size of the batch. With TETA, 12–13 pts phr are required. Dimethylaminopropylamine (DMAPA), (CH3)2N–CH2–CH2–CH2–NH2, and diethylaminoropylamine (DEAPA), (C2H5)2N–CH2–CH2–CH2–NH2, are slightly less reactive and give a pot life of about 140 min (for a 500-g batch) and are sometimes preferred. Note that both DMAPA and DEAPA have less than three hydrogen atoms necessary for cross-linking by reaction with epoxy groups. Other examples of amine curing agents with less than three hydrogen atoms are diethanolamine (DEA), NH(CH2CH2OH)2, and piperidine (see later). These curing agents operate by means of two-part reaction. Firstly, the active hydrogen atoms of the primary and secondary amine groups are utilized in the manner already described. Thereafter, the resulting tertiary amines, being sufﬁciently reactive to initiate polymerization of epoxy groups, function as catalytic curing agents, as described previously. As a class, the amines usually suffer from the disadvantage that they are pungent, toxic, and skin sensitizers. To reduce toxicity, the polyfunctional amines are often used in the form of adducts. A number of such modiﬁed amines have been introduced commercially. For example, reaction of the amine with a mono- or polyfunctional glycidyl material will give a higher-molecular-weight product with less volatility. O CH2 O CH CH2 O R O CH2 CH2 + 2H2N CH OH H2N R‘ NH CH2 CH R‘ NH2 OH CH2 O (I) R O CH2 CH CH2 NH R‘ NH2 An advantage of such modiﬁed amines is that because of higher molecular weight larger quantities are required for curing, and this helps to reduce errors in metering the hardener. These hardeners are also extremely active, and the pot life for a 500-g batch may be as little as 10 min. 550 Plastics Technology Handbook The glycidyl adducts are, however, skin irritants, being akin to the parent amines in this respect. Substitution of the hydroxyethyl group and its alkyl and aryl derivatives at the nitrogen atom is effective in reducing the skin sensitization effects of primary aliphatic amines. Both ethylene and propylene oxides have thus been used in the preparation of adducts from a variety of amines, including ethylene diamine and diethylenetriamine. O H2N R NH2 + CH2 CH2 H2N R NH CH2 CH2 O CH2 HO CH2 CH2 NH NH R CH2 (III) CH2 OH (II) CH2 OH Such adducts from diethylenetriamine appear free of skin sensitizing effects. A hardener consisting of a blend of the reaction products II and III is a low-viscosity liquid giving a pot life (for a 500-g batch) of 16–18 min at room temperature. Modiﬁcation of primary amines with acrylonitrile results in hardeners with reduced activity. H2N R NH2 + CH2 CH CN H2N R NH CH2 CH2 CH2 CH CN CN (IV) CN CH2 CH2 NH R NH CH2 (V) CH2 CN Commercial hardeners are mixtures of the addition compounds IV and V. Since accelerating hydroxyl groups are not present in IV and V (in contrast to I, II, and III), these hardeners have reduced activity and so give longer pot lives. A number of aromatic amines also function as epoxy hardeners. Since they introduce the rigid benzene ring structure into the cross-linked network, the resulting products have signiﬁcantly higher heat-distortion temperatures than are obtainable with the aliphatic amines (see Table 4.27). For example, metaphenylenediamine (MPDA) (shown below) gives cure resins with a heat distortion temperature of 150°C and very good chemical resistance. The hardener ﬁnds use in the manufacture of chemical resistance laminates. NH2 (MPDA) NH2 Diaminodiphenylmethane (DADPM) (below) H2N CH2 NH2 and diaminodiphenyl sulfone (DADPS) (below) O H2N S NH2 O used in conjunction with an accelerator provide even higher heat-distortion temperatures but at some expense to chemical resistance. 551 Industrial Polymers TABLE 4.27 Characteristics of Amine Hardeners Used in Low-Molecular-Weight Bisphenol A-Based Epoxy Resins Parts used per 100 Parts Resin Pot Life (500-g Batch) Typical Cure Schedule Max HDT of Cured Resin (°C) Applications 10–11 7 20 min 140 min Room temperature Room temperature 110 97 General purpose General purpose DETA-glycidyl adduct 25 10 min Room temperature 75 Adhesive laminating, fast cure DETA-ethylene oxide adduct 20 16 min Room temperature 92 – 14–15 6h 4–6 h at 150°C 150 Laminates, chemical resistance Hardenera DETA DEAPA MPDA DADPM 28.5 – 4–6 h at 165°C 160 Laminates DADPS Piperidine 30 5–7 – 8h 8 h at 160°C 3 h at 100°C 175 75 Laminates General purpose Triethylamine 10 7h Room temperature – Adhesives BDA TDMAMP 15 6 75 min 30 min Room temperature Room temperature – 64 10–14 3–6 h – – Adhesives Adhesives, coatings Encapsulation 2-Ethyl hexoate salt of TDMAMP Source: Brydson, J. A. 1982. Plastics Materials. Butterworth Scientiﬁc, London, UK. For chemical formulas see text. a Piperidine, a cyclic aliphatic amine H2C H2C H2C CH2 NH CH2 has been in use as an epoxy hardener since the early patents of Castan. Although a skin irritant, this hardener is still used for casting larger masses than are possible with primary aliphatic amines. Note that because it has only one active hydrogen atom in the amine group, piperidine operates by a two-part reaction in curing. Typical amine hardeners and their characteristics are summarized in Table 4.27. Cyclic acid anhydrides are widely employed as curing agents for epoxy resins. Compared with amines they are less skin sensitive and generally give lower exotherms in curing reaction. Some acid curing agents provide cured resins with very high heat-distortion temperatures and with generally good mechanical, electrical, and chemical properties. The acid-cured resins, however, show less alkali resistance than aminecured resins because of the susceptibility of ester groups to hydrolysis. Anhydride curing agents ﬁnd use in most of the important applications of epoxy resins, particularly in casting and laminates. In practice, acid anhydrides are preferred to acids since the latter are generally less soluble in the resin and also release more water on cure, leading to foaming of the product. Care must be taken, however, during storage since the anhydrides in general are somewhat hygroscopic. Examples of some anhydrides which are used are shown in Figure 4.35. The mechanism of hardening by anhydride is complex. In general, however, two types of reactions occur: (1) opening of the anhydride ring with the formation of carboxy groups, and (2) opening of the epoxy ring. The most important reactions which may occur are shown in what follows, with phthalic anhydride as an example. 552 Plastics Technology Handbook O CH C CH C O O (a) CH3 CH3 CH2 CH2 CH CH3 CH3 CH2 CH C O C CH C CH3 CH2 C (b) O H2 C O C O (c) O H2C CH C H2C CH C O O C O O C H2 (d) O O C C O O C O C C O O (e) H C HO (f ) O HC CH C CH HC CH O O C Cl C O Cl C O CH C CH C O CCI2 C Cl C C H O C O C Cl CH3 (g) O (h) FIGURE 4.35 Anhydride curing agents. (a) Maleic anhydride. (b) Dodecenylsuccinic anhydride. (c) Phthalic anhydride. (d) Hexahydrophthalic anhydride. (e) Pyromellitic anhydride. (f) Trimellitic anhydride. (g) Nadic methyl anhydride. (h) Chlorendic anhydride. 1. The ﬁrst stage of the interaction between an acid anhydride and an epoxy resin is believed to be the opening of the anhydride ring by (a) water (traces of which may be present in the system) or (b) hydroxy groups (which may be present as pendant groups in the original resin or may result from reaction (2a). O = = O = O O O + CH2 O = OH CH CH2 C OH C O = C C O = (b) C OH C OH H2O = + = (a) C O C O O CH2 CH CH2 (4.1) 553 Industrial Polymers 2. The epoxy ring may then be opened by reaction with (a) carboxylic groups formed by reactions (1a, b) or (b) hydroxyl groups [see (1)]. C OH C O + CH2 C O CH2 CH C O CH = = O = (a) O CH2 CH CH2 O CH2 CH CH2 (b) (4.2) O OH CH2 CH CH2 OH O = O CH2 OH cat CH O CH2 CH CH2 CH CH2 The reaction between an epoxy resin and an anhydride is rather sluggish. In commercial practice, the curing is accelerated by the use of organic bases to catalyze the reaction. These are usually tertiary amines such as a-methylbenzyl-dimethylamine and butylamine. The tertiary amine appears to react preferentially with the anhydride to generate a carboxy anion (–COO−). This anion then opens an epoxy ring to give an alkoxide ion ( C O– ), which forms another carboxy anion from a second anhydride molecule, and so on. Phthalic anhydride is the cheapest anhydride curing agent, but it has the disadvantage of being rather difﬁcult to mix with the resin. Liquid anhydrides (e.g., dodecenylsuccinic anhydride and nadic methyl anhydride), low-melting anhydrides (e.g., hexahydrophthalic anhydride), and eutectic mixtures are more easily incorporated into the resin. Since maleic anhydride produces brittle products, it is seldom used by itself and is used as a secondary hardener in admixture with other anhydrides. Dodecenylsuccinic anhydride imparts ﬂexibility into the casting, whereas chlorendic anhydride confers ﬂame resistance. Anhydride-cured resins generally have better thermal stability. Pyromellitic dianhydride with higher functionality produces tightly cross-linked products of high heat-distortion temperatures. Heat-distortion temperatures as high as 290°C have been quoted. Table 4.28 summarizes the characteristics of some of the anhydride hardeners. In addition to the amine and anhydride hardeners, many other curing agents have been made available. Among them are the so-called fatty polyamides. These polymers are of low molecular weight (2000–5000) TABLE 4.28 Properties of Some Anhydride Hardeners used in Low-Molecular-Weight Bisphenol A-Based Epoxy Resins Hardener (Anhydride) Phthalic Hexahydrophthalic (+accelerator) Parts Used per 100 Parts Resin Typical Cure Schedule Max HDT of Cured Resin (°C) Application 35–45 24 h at 120°C 110 Casting 80 24 h at 120°C 130 Casting Pyromellitic 26 20 h at 220°C 290 High HDT Nadic methyl Dodecenylsuccinic (+accelerator) Chlorendic 80 16 h at 120°C 2 h at 100°C + 2 h at 150°C 24 h at 180°C 202 38 High HDT Flexibilizing 180 Flame retarding 100 Source: From Brydson, J. A. 1982. Plastics Materials. Butterworth Scientiﬁc, London, UK. 554 Plastics Technology Handbook and are prepared by treating dimer acid (which is a complex mixture consisting of 60–75% dimerized fatty acids together with lesser amounts of trimerized acids and higher polymers) with stoichiometric excess of ethylenediamine or diethylenetriamine so that the resultant amides have free amine groups. Fatty polyamides are used to cure epoxy resins where a more ﬂexible product is required, particularly in adhesive and coating applications. An advantage of the system is that roughly similar quantities of hardener and resin are required and, because it is not critical, metering can be done visually without the need of measuring aids. They thus form the basis of some domestic adhesive systems. Also used with epoxy resins for adhesives is dicyanodiamide, H2N–C(═NH)NH–CN. It is insoluble in common resins at room temperature but is soluble at elevated temperatures and thus forms the basis of a one-pack system. 4.3.5.5.3 Other Epoxies Resins containing the glycidyl ether group O ( O CH2 CH CH2) result from the reaction of epichlorohydrin and hydroxy compounds. Although bisphenol A is the most commonly used hydroxyl compound, a few glycidyl ether resins based on other hydroxy compound are also commercially available, including novolac epoxies, polyglycol epoxies, and halogenated epoxies. A typical commercial novolac epoxy resin (VI) produced by epoxidation of phenolic hydroxyl groups of novolac (see the section on phenolformaldehyde resins) by treatment with epichlorohydrin has an O CH2 O CH2 CH2 O CH CH CH CH2 CH2 CH2 O O CH2 O CH2 n (VI) average molecular weight of 650 and contains about 3.6 epoxy groups per molecule. Because of higher functionality, the novolac epoxy resins give, on curing, more hightly cross-linked products than the bisphenol A-based resins. This results in greater thermal stability, higher heatdeﬂection temperatures, and improved chemical resistance. Their main applications have been in hightemperature adhesives, heat-resistant structural laminates, electrical laminates resistant to solder baths, and chemical-resistant ﬁlament-wound pipes. The use of novolac epoxies has been limited, however, by their high viscosity and consequent handling difﬁculties. Polymeric glycols such as polypropylene glycol may be epoxidized through the terminal hydroxy groups to give diglycidyl ethers (VII). CH2 CH3 CH3 O CH CH2 O CH CH2 O CH2 CH2 O CH CH CH2 O CH2 n O C CH3 (VII) In commercial products n usually varies from 1 to 6. Alone, these resins give soft products of low strength on curing. So they are normally used in blends with bisphenol A- or novolac-based resins. Added 555 Industrial Polymers to the extent of 10–30%, they improve resilience without too large a loss in strength and are used in such applications as adhesives and encapsulations. Epoxies containing halogens have ﬂame-retardant properties and may be prepared from halogenated hydroxy compounds. Halogenated epoxies are available based on tetrabromobisphenol A and tetrachlorobisphenol A (VIII). X CH3 HO X OH C CH3 X X (VIII) The brominated resins are more effective than the chlorinated resins and have become more predominant commercially. The ability of the resins to retard or extinguish burning is due to the evolution of hydrogen halide at elevated temperatures. Brominated epoxy resins are generally blended with other epoxy resins to impart ﬂame retardance in such applications as laminates and adhesives. Nonglycidyl ether epoxy resins are usually prepared by treating unsaturated compounds with peracetic acid. O C + CH3 C C O O O OH C C + CH3 C OH Two types of nonglycidyl ether epoxy resins are commercially available: cyclic aliphatic epoxies and acyclic aliphatic epoxies. Cyclic aliphatic epoxy resins were ﬁrst introduced in the United States. Some typical examples of commercial materials are 3,4-epoxy-6-methylcyclohexylmethyl-3,4-epoxy-6-methylcyclohexane carboxylate (Unox epoxide 201, liquid) (IX), vinylcyclohexene dioxide (Unox epoxide 206, liquid) (X), and dicyclopentadiene dioxide (Unox epoxide 207, solid) (XI). O O O C O H2C CH3 (IX) O O CH CH2 O CH2 O H3C (X) (XI) Generally, acid anhydrides are the preferred curing agents since amines are less effective. A hydroxy compound, such as ethylene glycol, is often added as initiator. Because of their more compact structure, cycloaliphatic resins produce greater density of cross-links in the cured products than bisphenol A-based glycidyl resins. This generally leads to higher heat-distortion temperatures and to increased brittleness. The products also are clearly superior in arc resistance and are track resistant. Thus, although bisphenol A-based epoxies decompose in the presence of a high-temperature arc to produce carbon which leads to tracking and insulation failure, cycloaliphatic epoxies oxidize to volatile products which do not cause tracking. This has led to such applications as heavy-duty electrical castings and laminates, tension insulators, rocket motor cases, and transformer encapsulation. Acyclic aliphatic resins differ from cyclic aliphatic resins in that the basic structure of the molecules in the former is a long chain, whereas the latter, as shown, contains ring structures. Two types of acyclic aliphatic epoxies are commercially available, namely, epoxidized diene polymers and epoxidized oils. Typical of the epoxidized diene polymers are the products produced by treatment of polybutadiene with peracetic acid. Epoxidized diene polymers are not very reactive toward amines but may be crosslinked with acid hardeners. Cured resins have substantially higher heat-distortion temperatures (typically, 250°C) than do the conventional amine-cured diglycidylether resins. 556 Plastics Technology Handbook Epoxidized oils are obtained by treatment of drying and semidrying oils (unsaturated), such as linseed and soybean oils, with peracetic acid. Epoxidized oils ﬁnd use primarily as plasticizers and stabilizers for PVC. 4.3.5.5.4 Applications Epoxy resins have found a wide range of applications and a steady rate of growth over the years mainly because of their versatility. Properties of the cured products can be tailored by proper selection of resin, modiﬁer, cross-linking agent, and the curing schedule. The main attributes of properly cured epoxy systems are outstanding adhesion to a wide variety of substrates, including metals and concrete; ability to cure over a wide temperature range; very low shrinkage on cure; excellent resistance to chemicals and corrosion; excellent electrical insulation properties; and high tensile, compressive, and ﬂexural strengths. In general, the toughness, adhesion, chemical resistance, and corrosion resistance of epoxies suit them for protective coating applications. It is not surprising that about 50% of epoxy resins are used in protective coating applications. Two types of epoxy coatings are formulated: those cured at ambient temperature and those that are heat cured. The ﬁrst type uses amine hardening systems, fatty acid polyamides, and polymercaptans as curing agents. Very high cure rates may be achieved by using mercaptans. Heat-cured types use acid anhydrides and polycarboxylic acids as well as formaldehyde resins as curing agents. Typical coating applications for phenol-formaldehyde resin-modiﬁed epoxies include food and beverage can coatings, drum and tank liners, internal coatings for pipes, wire coating, and impregnation varnishes. Ureaformaldehyde resin-modiﬁed epoxies offer better color range and are used as appliance primers, can linings, and coatings for hospital and laboratory furniture. Environmental concerns have prompted major developmental trends in epoxy coating systems. Thus epoxy coating systems have been developed to meet the high-temperature and hot-acid environments to which SO2 scrubbers and related equipment are subjected. Ambient curable epoxy systems have been developed to provide resistance to concentrated inorganic acids. In the development of maintenance and marine coatings, the emphasis has been on the development of low-solvent, or solvent-free, coatings to satisfy EPA volatile organic content standards. Thus liquid epoxy systems based on polyamidamines have been developed. Epoxies have also been formulated as powdered coatings, thus completely eliminating solvents. Pipe coatings still represent a major market for epoxies. High-molecular-weight powdered formulations are used in this application. The important maintenance coating area, particularly for pipe and tank coatings, is served by epoxy systems cured with polyamine or polyamidamines. Two-component, air-dried, solventless systems used in maintenance coatings provide tough, durable, nonporous surfaces with good resistance to water, acids, alkalies, organic solvents, and corrosion. Emulsiﬁable epoxies of varying molecular weights, water-dilutable modiﬁed epoxies, and dispersions of standard resins represent promising developments for coating applications. Can coatings are an important waterborne resin application. Two-component waterborne systems are also ﬁnding use as architectural coatings. For electrical and electronic applications epoxy formulations are available with low or high viscosity, unﬁlled or ﬁlled, slow or fast curing at low or high temperatures. Potting, encapsulation, and casting of transistors, integrated circuits, switches, coils, and insulators are a few electrical applications of epoxies. With their adhesion to glass, electrical properties, and ﬂexural strength, epoxies provide high-quality printed circuit boards. Epoxies have been successfully used in Europe for outdoor insulators, switchgear, and transformers for many years. In these heavy electrical applications, the advantage of cycloaliphatic epoxies over porcelain has been demonstrated. In a relatively new development, epoxy photopolymers have been used as solder masks and photoresists in printed circuit board fabrications. 557 Industrial Polymers Adhesion properties of epoxies, complete reactivity with no volatiles during cure, and minimal shrinkage make the materials outstanding for adhesives, particularly in structural applications. The most commonly known adhesive applications involve the two-component liquids or pastes, which cure at room or elevated temperatures. A novel, latent curing system, which gives more than one year pot life at room temperature, has increased the use of epoxies for specialty adhesives and sealants, and for vinyl plastisols. The one-pack system provides fast cure when heated—for example 5 min at 100°C. Epoxies are used in ﬁber-reinforced composites, providing high strength-to-weight ratios and good thermal and electrical properties. Filament-would epoxy composites are used for rocket motor casings, pressure vessels, and tanks. Glass-ﬁber reinforced epoxy pipes are used in the oil, gas, mining, and chemical industries. Sand-ﬁlled epoxies are used in industrial ﬂooring. Resistance to a wide variety of chemicals and solvents and adhesion to concrete are key properties responsible for this use. Epoxies are also used in patching concrete highways. Decorative ﬂooring and exposed aggregate systems make use of epoxies because of their low curing shrinkage, and the good bonding of glass, marble, and quartz chip by the epoxy matrix. 4.3.5.6 Poly(Phenylene Oxide) CH3 O CH3 Monomer 2,6-Dimethylphenol n Polymerization Major Uses Condensation polymerization by oxidative coupling Automotive, appliances, business machine cases, electrical components Poly(2,6-dimethyl-1,4-phenylene oxide), commonly called poly(phenylene oxide) or PPO, was introduced commercially in 1964. PPO is manufactured by oxidation of 2,6-dimethyl phenol in solution using cuprous chloride and pyridine as catalyst. The monomer is obtained by the alkylation of phenol with methanol. End-group stabilization with acetic anhydride improves the oxidation resistance of PPO. PPO is counted as one of the engineering plastics. The rigid structure of the polymer molecules leads to a material with a high Tg of 208°C. It is characterized by high tensile strength, stiffness, impact strength, and creep resistance, and low coefﬁcient of thermal expansion. These properties are maintained over a broad temperature range (−45°C to 120°C). One particular feature of PPO is its exceptional dimensional stability among the so-called engineering plastics. The polymer is self-extinguishing. PPO has excellent resistance to most aqueous reagents and is unaffected by acids, alkalis, and detergents. The polymer has outstanding hydrolytic stability and has one of the lowest water absorption rates among the engineering thermoplastics. PPO is soluble in aromatic hydrocarbons and chlorinate solvents. Several aliphatic hydrocarbons cause environmental stress cracking. PPO has low molding shrinkage. The polymer is used for the injection molding of such items as pump components, domestic appliance and business machines, and electrical parts such as connectors and terminal blocks. The high price of PPO has greatly restricted its application and has led to the introduction of the related and cheaper thermoplastic materials in 1966 under the trade name Noryl by General Electric. If PPO (Tg = 208°C) is blended with polystyrene (Tg∼90°C) in equal quantities, a transparent polymer is obtained with a single Tg of about 150°C, which apparently indicates a molecular level of mixing. Noryl thermoplastics may be considered as being derived from such polystyrene–PPO blends. Since the 558 Plastics Technology Handbook electrical properties of the two polymers are very similar, the blends also have similar electrical characteristics. In addition to Noryl blends produced by General Electric, grafts of styrene onto PPO are also available (Xyron by Asahi-Dow). The styrenic component in polystyrene–PPO blends may not necessarily be straight polystyrene (PS) but instead high-impact polystyrene (HIPS) or some other related material. The most widely used blend is the blend of PPO and HIPS. Like polystyrene, these blends have the following useful characteristics: (1) good dimensional stability and low molding shrinkage, thus allowing close dimensional tolerance in the production of moldings; (2) low water absorption; (3) excellent resistance to hydrolysis; and (4) very good dielectric properties over a wide range of temperature. In addition, unlike polystyrene, the blends have heat-distortion temperatures above the boiling point of water, and in some grades this is as high as 160°C. The range of Noryl blends available comprises a broad spectrum of materials superior in many respects, particularly heat deformation resistance, to the general purpose thermoplastics but at a lower price than the more heat resistant materials such as polycarbonates, polyphenylene sulﬁdes, and polysulfones (discussed later). The materials that come close to them in properties are the ABS/polycarbonate blends. Noryl is also characterized by high dielectric strength (192 V/mil) and low dissipation factors (4.7 × 103 at 100 Hz and 3.9 × 103 at 106 Hz). In common with other engineering thermoplastics, there are four main groups of modiﬁed PPOs available. They are: (1) non-self-extinguishing grades with a heat-distortion temperature in the range 110– 106°C and with a notched Izod impact strength at 200–500 J/m; (2) self-extinguishing grades with slightly lower heat-distortion temperatures and impact strengths; (3) non-self-extinguishing glass-reinforced grades (10%, 20%, 30% glass ﬁber) with heat-distortion temperatures in the range of 120–140°C; and (4) self-extinguishing glass-reinforced grades. Among the special grades that should be mentioned are those containing blowing agents for use in the manufacture of structural foams (see Chapter 2). Noryls may be extruded, injection molded, and blow-molded without undue difﬁculty. Processing conditions depend on the grade used but in injection molding a typical melt temperature would be in the range 250–300°C. The introduction of self-extinguishing, glass-reinforced, and structural foam grades has led to steady increase in the use of these materials in ﬁve main application areas. These are (1) the automotive industry; (2) the electrical industry; (3) radio and television; (4) business machines and computer housings; and (5) pumps and other plumbing applications. Use in the automotive industries largely arises from the availability of high impact grades with heatdistortion temperatures above those of the general purpose thermoplastics. Speciﬁc uses include instrument panels, steering column cladding, central consoles, loudspeaker housings, ventilator grilles and nozzles, and parcel shelves. In cooling systems, glass-reinforced grades have been used for radiator and expansion tanks. Several components of car heating systems are also produced from modiﬁed PPOs. The materials have been increasingly used for car exterior trim such as air inlet and outlet grills and outer mirror housings. In the electrical industry, well-known applications include fuse boxes, switch cabinets, housing for small motors, transformers, and protective circuits. Uses in radio and television arise largely from the ability to produce components with a high level of dimensional accuracy coupled with good dielectric properties, high heat-distortion temperatures, and the availability of self-extinguishing grades. Speciﬁc uses include coil formers, picture tube deﬂection yokes, and insert card mountings. In the manufacture of business machine and computer housings, structural foam grades have found use and moldings weighing as much as 50 kg have been reported. Glass-reinforced grades have widely replaced metals in pumps and other functional parts in washing equipment and central heating systems. Another PPO, called PPE, is produced by the oxidative coupling of a mixture of 2,6-dimethylphenol and 2,3,6-trimethylphenol. This stiff polymer, like Noryl, is available from General Electric. It is usually modiﬁed by blending with PS or HIPS. 559 Industrial Polymers TABLE 4.29 Properties of Typical PPO–PA Blends (Noryl GTX 810) Property Value Tensile strength 103 lbf/in.2 13 MPa Tensile modulus 90 105 lbf/in.2 2.0 GPa Elongation at break (%) 1.4 10 Flexural strength 103 lbf/in.2 MPa 19 131 Flexural modulus 105 lbf/in.2 GPa 2.25 1.6 Impact strength, notched Izod ft.-lb/in. J/m 1.5 80 Blends of PPO and polyamide (PA, nylon) are incompatible but good properties can be obtained through the use of compatibilizing agents. The PPO is dispersed in a continuous nylon matrix in these blends. Because of the incompatibility of the two phases, the modulus decreases very little at the Tg of PA (71°C) and is maintained up to the Tg of the PPO phase (208°C). The PPO–PA blends, which are sold by General Electric under the trade name Noryl GTX, can be baked and painted at 190°C without noticeable warpage or distortion, and have been used for producing automobile fenders. The PPO–PA blends show a mold shrinkage of 0.001 in./in. The molding pressure is 15 × 103 lbf/in.2 (103 MPa) and the processing temperature 260°C. The heat-deﬂection temperature of the molded specimen under ﬂexural load of 264 lbf/in.2 is typically 190°C and the maximum resistance to continuous heat is 175°C. The coefﬁcient of linear expansion is 10−5 cm/cm °C. The mechanical properties of the PPO–PA blends are shown in Table 4.29. 4.3.6 Cellulosic Polymers Cellulose is a carbohydrate with molecular formula (C6H10O5)n, where n is a few thousand. Complete hydrolysis of cellulose by boiling with concentrated hydrochloric acid yields D-glucose, C6H12C6, in 95– 96% yield [56]. Cellulose can thus be considered chemically as a polyanhydroglucose. The structure of cellulose is CH2OH 4 OH OH H H O H –O H H OH 1 H H H O H OH CH2OH H O CH2OH OH OH H H O H O H H OH H H H H O OH H O O CH2OH The regularity of the cellulose chain and extensive hydrogen bonding between hydroxyl groups of adjacent chains cause cellulose to be a tightly packed crystalline material which is insoluble and infusible. 560 Plastics Technology Handbook As a result, cellulose cannot be processed in the melt or in solution. However, cellulose derivatives in which there is less hydrogen bonding are processable. The most common means of preparing processable cellulose derivatives are esteriﬁcation and etheriﬁcation of the hydroxyl groups. In the following, the more important commercial cellulosic polymers are described. 4.3.6.1 Regenerated Cellulose As mentioned, many derivatives of cellulose are soluble, though cellulose itself is insoluble. A solution of a cellulose derivative can be processed (usually by extrusion) to produce the desired shape (commonly) ﬁber or ﬁlm) and then treated to remove the modifying groups to reform or regenerated unmodiﬁed cellulose. Such material is known as regenerated cellulose. Modern methods of producing regenerated cellulose can be traced to the discovery in 1892 by Cross, Bevan, and beadle that cellulose can be rendered soluble by xanthate formation by treatment with sodium hydroxide and carbon disulﬁde and regenerated by acidiﬁcation of the xanthate solution. This process is known as the viscose process. The reactions can be indicated schematically as S R OH NaOH Cellulose R ONa Alkali cellulose CS2 R O Cellulose xanthate C S Na H+ R OH Regenerated cellulose The viscose process is used for the production of textile ﬁbers, known as viscose rayon, and transparent packaging ﬁlm, known as cellophane (the name is coined from cellulose and diaphane, which is French for transparent). A suitably aged solution of cellulose xanthate, known as viscose, is fed through spinnerets with many small holes (in the production of ﬁber), through a slot die (in the production of ﬁlm), or through a ring die (in the production of continuous tube used as sausage casing) into a bath containing 10–15% sulfuric acid and 10–20% sodium sulfate at 35–40°C, which coagulates and completely hydrolyzes the viscose. Cellulose is thus regenerated in the desired shape and form. It is possible to carry out a drawing operation on the ﬁber as it passes through the coagulating bath. The stretching (50–150%) produces crystalline orientation in the ﬁber. The product, known as hightenacity rayon, has high strength and low elongation and is used in such application as tire cord and conveyor belting. For the production of cellophane, the regenerated cellulose ﬁlm is washed, bleached, plasticized with ethylene glycol or glycerol, and then dried; sometimes a coating of pyroxylin (cellulose nitrate solution) containing dibutyl phthalate as plasticizer is applied to give heat sealability and lower moisture permeability. Cellophane has been extensively and successfully used as a wrapping material, particularly in the food and tobacco industries. However, the advent of polypropylene in the early 1960s has produced a serious competitor to this material. 4.3.6.2 Cellulose Nitrate Cellulose nitrate or nitrocellulose (as it is often erroneously called) is the doyen of cellulose ester polymers. It is prepared by direct nitration with nitric and sulfuric acid mixtures at about 30–40°C for 20– 60 min. Complete substitution at all three hydroxyl groups on the repeating anhydroglucose unit will give cellulose trinitrate containing 14.14% nitrogen: [C6H7O2(OH)3]n HNO3/H2SO4 [C6H7O2(ONO2)3]n + 3nH2O This material is explosive and is not made commercially, but products with lower degrees of nitration are of importance. The degree of nitration may be regulated by the choice of reaction conditions. 561 Industrial Polymers Industrial nitrocelluloses have a degree of substitution somewhere between 1.9 and 2.7 and are generally characterized for their various uses by their nitrogen content, usually about 11% for plastics, 12% for lacquer and cement base, and 13% for explosives. The largest use of cellulose nitrate is as a base for lacquers and cements. Butyl acetate is used as a solvent. Plasticizers such as dibutyl phthalate and tritolyl phosphate are necessary to give ﬁlms of acceptable ﬂexibility and adhesion. For use as plastic in bulk form, cellulose nitrate is plasticized with camphor. The product is known as celluloid. In a typical process alcohol-wet cellulose nitrate is kneaded at about 40°C with camphor (about 30%) to form a viscous plastic mass. Pigments or dyes may be added at this stage. The dough is then heated at about 80°C on milling rolls until the alcohol content is reduced to about 15%. The milled product is calendered into sheets about 1/2-in. (1.25-cm) thick. A number of sheets are laid up in a press and consolidated into a block. The block is sliced into sheets of thickness 0.005–1 in. (0.012–2.5 cm), which are then allowed to season for several days at about 50°C so that the volatile content is reduced to about 2%. Celluloid sheet and block may be machined with little difﬁculty if care is taken to avoid overheating. The high inﬂammibility and relatively poor chemical resistance of celluloid severely restrict its use in industrial applications. Consequently the material is used because of its desirable characteristics, which include rigidity, dimensional stability, low water absorption, reasonable toughness, after-shrinkage around inserts, and ability of forming highly attractive colored sheeting. Today the principal outlets of celluloid are knife handles, table-tennis balls, and spectacle frames. Celluloid is marketed as Xylonite (BX Plastics Ltd.) in UK. 4.3.6.3 Cellulose Acetate The acetylation of cellulose is usually carried out with acetic anhydride in the presence of sulfuric acid as catalyst. It is not practicable to stop acetylation short of the essentially completely esteriﬁed triacetate. Products of lower acetyl content are thus produced by partial hydrolysis of the triacetate to remove some of the acetyl groups: [C6H7O2(OH)3]n Cellulose HOAc, Ac2O [C6H7O2(OAc)3]n Triacetate (44.8%acetyl) H2O Heat [C6H7O2(OAc)2OH)n Diacetate (34.9%acetyl) Cellulose triacetate is often known as primary cellulose acetate, and partially hydrolyzed material is called secondary cellulose acetate. Many physical and chemical properties of cellulose acetylation products are strongly dependent on the degree of esteriﬁcation, which is measured by the acetyl content (i.e., the weight of acetyl radical (CH3CO–) in the material) or acetic acid yield (i.e., the weight of acetic acid produced by complete hydrolysis of the ester). The commercial products can be broadly distinguished as cellulose acetate (37–40% acetyl), high-acetyl cellulose acetate (40–42% acetyl), and cellulose triacetate (43.7–44.8% acetyl). Cellulose acetate containing 37–40% acetyl is usually preferred for use in general-purpose injectionmolding compounds. Cellulose acetate, however, decomposes below its softening point, and it is necessary to add a plasticizer (e.g., dimethyl phthalate or triphenyl phosphate), usually 25–35%, to obtain a moldable material. The use of cellulose acetate for molding and extrusion is now small owing largely to the competition of polystyrene and other polyoleﬁns. At the present time the major outlets of cellulose acetate are in the fancy goods trade as toothbrushes, combs, hair slides, etc. Cellulose acetate with a slightly higher degree of esteriﬁcation (38.7–40.1% acetyl) is usually preferred for the preparation of ﬁbers, ﬁlms, and lacquers because of the greater water resistance. A signiﬁcant application of cellulose acetate ﬁlm has been found in sea-water desalination by reverse osmosis. High-acetyl cellulose acetate (40–42% acetyl) has found occasional use in injection-molding compounds where greater dimensional stability is required. However, processing is more difﬁcult. 562 Plastics Technology Handbook Cellulose triacetate (43.7–44.8% acetyl) ﬁnds little use in molding compositions because its very high softening temperature is not greatly reduced by plasticizers. It is therefore processed is solution. A mixture of methylene chloride and methanol is the commonly used solvent. The sheeting and ﬁbers are made from cellulose triacetate by casting or by extruding a viscous solution and evaporating the solvent. The sheeting and ﬁlm are grainless, have good gauge uniformity, and good optical clarity. The products have good dimensional stability and are highly resistant to water, grease, oils, and most common solvents such as alcohol and acetone. They also have good heat resistance and high dielectric constant. Sheeting and ﬁlms of cellulose triacetate are used in the production of visual aids, graphic arts, greeting cards, photographic albums, and protective folders. Cellulose triacetate is extensively used for photographic, x-ray, and cinematographic ﬁlms. In these applications cellulose triacetate has displaced celluloid mainly because the triacetate does not have the great inﬂammability of celluloid. 4.3.6.4 Other Cellulose Esters Homologues of acetic acid have been employed to make other cellulose esters. Of these, cellulose propionate, cellulose acetate-propionate, and cellulose acetate-butyrate are produced on a commercial scale. The are produced in a manner similar to that described previously for cellulose acetate. The propionate and butyrate esters are made by substituting propionic acid and propionic anhydride or butyric acid and butyric anhydride for some of the acetic acid and acetic anhydride. Cellulose acetate-butyrate (CAB) has several advantages in properties over cellulose acetate: lower moisture absorption, greater solubility and compatibility with plasticizer, higher impact strength, and excellent dimensional stability. CAB used in plastics has about 13% acetyl and 37% butyryl content. It is an excellent injection-molding material (Tenite Butyrate by Kodak, Cellidor B by Bayer). Principal end products of CAB have been for tabulator keys, automobile parts toys, pen and pencil barrels, steering wheels, and tool handles. In the United States CAB has been sued for telephone housings, and extruded CAB piping has been used for conveying water, oil, and natural gas. CAB sheet is readily vacuum formed and is especially useful for laminating with thin-gauge aluminum foil. It also serves particularly well for vacuum metallizing. Cellulose propionate (Forticel by Celanese) is very similar in both cost and properties to CAB. It has been used for similar purposes as CAB. Cellulose acetate propionate (Tenite Propionate by Kodak) is similar to cellulose propionate. It ﬁnd wide use in blister packages and formed containers, safety goggles, motor covers, metallized ﬂash cubes, brush handles, steering wheels, face shields, displays, and lighting ﬁxtures. 4.3.6.5 Cellulose Ethers Of cellulose ethers only ethyl cellulose has found application as a molding material. Methyl cellulose, hydroxyethyl cellulose, and sodium carboxymethyl cellulose are useful water-soluble polymers. The ﬁrst step in the manufacture of each of these materials is the preparation of alkali cellulose (soda cellulose) by treating cellulose with concentrated sodium hydroxide. Ethyl cellulose is made by reacting alkali cellulose with ethyl chloride. [[R(OH)3]n + NaOH, H2O] + CI CH2CH3 90–150°C 6–12 h where R = C6H7O2 [R(OCH2CH3)m(OH)3–m]n m = 2.4 - 2.5 Ethyl cellulose is produced in pellet form for molding and extrusion and in sheet form for fabrication. It has good processability, is tough, and is moderately ﬂexible; its outstanding feature is its toughness at low temperatures. The principal uses of ethyl cellulose moldings are thus in those applications where good impact strength at low temperatures is required, such as refrigerator bases and ﬂip lids and ice-crusher parts. Ethyl cellulose is often employed in the form of hot melt for strippable coatings used for protection 563 Industrial Polymers of metal parts against corrosion and marring during shipment and storage. A recent development is the use of ethyl cellulose gel lacquers for permanent coatings. Methyl cellulose is prepared by reacting alkali cellulose with methyl chloride at 50–100°C (cf. Ethyl cellulose). With a degree of substitution of 1.6–1.8, the resultant either is soluble in cold water but not in hot water. It is used as a thickening agent and emulsiﬁer in cosmetics, pharmaceuticals, ceramics, as a paper size, and in leather-tanning operations. Hydroxyethyl cellulose, produced by reacting alkali cellulose with ethylene oxide, is employed for similar purposes. Reaction of alkali cellulose with sodium salt of chloroacetic acid yields sodium carboxymethyl cellulose (SCMC), [[R(OH)3]n + NaOH, H2O] + CI CH2 CO ONa 40-50°C 2-4 hr [R(OCH2CO ONa)m(OH)3–m]n where R = C6H7O2 and m = 0.65–1.4. SCMC appears to be physiologically inert and is very widely used. Its principal application is as a soil-suspending agent in synthetic detergents, but it is also used as a sizing and ﬁnishing agent for textile, as a surface active agent, and as a viscosity modiﬁer in emulsion and suspension. Puriﬁed grades of SCMC are used in ice cream to provide a smooth texture and in a number of cosmetic and pharmaceutical products. SCMC is also the basis of a well-known proprietary wallpaper adhesive. 4.3.7 Sulﬁde Polymers 4.3.7.1 Polysulﬁdes Monomers Aliphatic dihalide, sodium sulﬁde Polymerization Major Uses Polycondensation Sealing, caulks, gaskets Polysulﬁde elastomers are produced by the reaction of an aliphatic dihalide, usually bis(2chloroethyl) formal, with sodium polysulﬁde under alkaline conditions: CICH2CH2OCH2OCH2CH2CI + Na2Sx [ CH2CH2OCH2OCH2CH2Sx ] + NaCI The reaction is carried out with the dihalide suspended in an aqueous magnesium hydroxide phase. The value of x is slightly above 2. A typical polymerization system also includes up to 2% 1,2,3-trichloro propane. The polymerization occurs readily yielding a polymer with a very high molecular weight. High molecular weight, however, is not desirable until its end-use application. The molecular weight is therefore lowered, and the polysulﬁde rank (value of x) is simultaneously brought close to 2, by reductive treatment with NaSH and Na2SO3 followed by acidiﬁcation. The result is a liquid, branched polysulﬁde with terminal thiol end group and a molecular weight in the range of 1000–8000. Curing to a solid elastomer is accomplished by oxidation of thiol to disulﬁde links by oxidants such as lead dioxide and p-quinone dioxime: S S HS SH [0] S S SH S S S S S S S S 564 Plastics Technology Handbook These materials are widely used as sealants, binders for solid propellants, caulking materials, and cements for insulating glass and fuel tanks. Polysulﬁdes, often referred to as thiokols, are produced at low volumes as specialty materials geared toward a narrow market. The advantages and disadvantages of polysulﬁdes both reside in the disulﬁde linkage. Thus they possess low-temperature ﬂexibility and very good resistance to ozone, oil, solvent (hydrocarbons as well as polar solvents such as alcohols, ketones, esters, etc.), and weathering. However, polysulﬁdes have poor thermal stability and creep resistance, have low resilience, and are malodorous. Thiokols are amorphous polymers which do not crystallize when stretched and hence reinforcing ﬁllers, such as carbon black, must be added to obtain relatively high tensile strengths. Thiokol may be vulcanized in the presence of zinc oxide and thiuram accelerators, such as tetramethyl thiuram disulﬁde (Tuads). The accelerators modify the sulfur links and serve as chemical plasticizers. A typical thiokol with 60 parts carbon black per 100 of polymer has a tensile strength of 1200 lbf/in.2, an elongation of 300%, a speciﬁc gravity of 1.25, and a Shore A hardness of 68. Thiokol has excellent resistance to ozone (O3) and ultraviolet radiation. It has a low permeability to solvents, such as gasoline; esters, such as ethyl acetate; and ketones, such as acetone. The principal application of solid Thiokol elastomers is as gaskets, O-rings, gasoline, and fuel hose lines, gas meter diaphragms, and as rollers, which are used for lacquering cans. 4.3.7.2 Poly(Phenylene Sulﬁde) S n Monomers p-Dichlorobenzene, sodium sulﬁde Polymerization Major Uses Polycondensation Electrical components, mechanical parts Poly(phenylene sulﬁde) (PPS) is the thio analogue of poly(phenylene oxide) (PPO) [57]. The ﬁrst commercial grades were introduced by Phillips Petroleum in 1968 under the trade name Ryton. Other manufacturers also have introduced PPS (e.g., Tedur by Bayer). The commercial process involves the reaction of p-dichlorobenzene with sodium sulﬁde in a polar solvent. PPS is an engineering plastic. The thermoplastic grades of PPS are outstanding in heat resistance, ﬂame resistance, chemical resistance, and electrical insulation characteristics. The linear polymers are highly crystalline with melting point in the range of 285–295°C and Tg of 193–204°C. The material is soluble only above 200°C in aromatic and chlorinated aromatic solvents. It has the ability to cross-link by air-oxidation at elevated temperatures, thereby providing an irreversible cure. Thermogravimetric analysis shows no weight loss below 500°C in air but demonstrates complete decomposition by 700°C. It is found to retain its properties after four months at 233°C (450°F) in air. Signiﬁcant increases in mechanical properties can be achieved with glass-ﬁber reinforcement. In the unﬁlled form the tensile strength of the material is 64–77 MPa at 21°C, 33 MPa at 204°C, and the ﬂexural modulus is 4200 MPa at 21°C. The corresponding values for PPS–glass ﬁber (60:40) composites are 150, 33, and 15,500 MPa. Although rigidity and tensile strength are similar to those of other engineering plastics, PPS does not possess the toughness of amorphous materials, such as the polycarbonates and the polysulfones (described later), and are somewhat brittle. On the other hand, PPS does show a high level of resistance to environmental stress cracking. Being one of the most expensive commercial moldable thermoplastics, the use of PPS is heavily dependent on its particular combination of properties. Good electrical insulation characteristics, including 565 Industrial Polymers good arcing and arc-tracking resistance ha led to PPS replacing some of the older thermosets in electrical parts. Thee include connectors, terminal blocks, relay components, brush holders, and switch components. PPS is used in chemical process plants for gear pumps. It has found application in the automotive sector, in such speciﬁc uses as carburetor parts, ignition plates, ﬂow control values for heating systems, and exhaust-gas return valves to control pollution. The material also ﬁnds uses in sterilizable medical, dental, and general laboratory equipment, cooking appliance, and hair dryer components. Injection-molded products of PPS include high-temperature lamp holders and reﬂectors, pump parts, valve, and, especially when ﬁlled for example with PTFE or graphite, bearings. Processing temperatures are 300–350°C with mold temperatures of up to 200°C. PPS is also used for encapulation of electronic components and as a high temperature surface coating material. PPS is resistant to neutron and gamma radiation. In nuclear installations, its ﬂexural strength and modulus are essentially unchanged when it is exposed to gamma radiation of 5 × 109 rad and neutron radiation of 1 × 109 rad. 4.3.8 Polysulfones Polysulfones are a family of engineering thermoplastics with excellent high-temperature properties. The simplest aromatic polysulfone, poly(p-phenylene sulfone) SO2 n does not show thermoplastic behavior, melting with decomposition above 500°C. Hence, to obtain a material capable of injection molding in conventional machines, the polymer chain is made more ﬂexible by incorporating ether links into the backbone. The structures and glass transition temperatures of several commercial polysulfones are listed in Table 4.30. The polymers have different degrees of spacing between the p-phenylene groups and thus have a TABLE 4.30 Commercial Polysulfones Type of Structure CH3 I O O S CH3 190 Udel (Union Carbide) 285 Astrel (3M Corp.) 230 Victrex (ICI) 250 Polyether-Sulfone 720 P (ICI) – Rodel (Union Carbide) O O O S (a) Trade Name O C II Tg (°C) S O O O (b) (a) Predominates O III S O O O IV O S O S O O (a) (b) (b) Predominates O V S O O O 566 Plastics Technology Handbook spectrum of glass transition temperatures which determine the heat-distortion temperature (or deﬂection temperature under load), since the materials are all amorphous. The ﬁrst commercial polysulfone (Table 4.30I) was introduced in 1965 by Union Carbide. This material, now known as Udel, has a continuous-use temperature of 150°C and a maximum-use temperature of 170°C, and it can be fabricated easily by injection molding in conventional machines. In 1967, Minnesota Mining and Manufacturing (3M) introduced Astrel 360 (Table 4.30II), an especially high-performance thermoplastic, which requires specialized equipment with extra heating and pressure capabilities for processing. ICI’s polyether sulfones, introduced in 1972—Victrex (Table 4.30III) and polyethersulfone 720P (Table 4.30IV)—are intermediate in performance and processing. In the late 1970s, Union Carbide introduced Radel (Table 4.30V), which has a higher level of toughness. Note that all of the commercial materials mentioned in Table 4.30 may be described as polysulfones, polyarylsulfones, polyether sulfones, or polyaryl ether sulfones. In principle, there are two main routes to the preparation of polysulfones: (1) polysulfonylation and (2) polyetheriﬁcation. Polysulfonylation reactions are of the following general types: H H + CISO2 Ar ( Ar H Ar Ar´ SO2 SO2CI SO2 )n + HCI Ar´ ( Ar SO2CI Friedel−Crafts Catalyst SO2 ) n + HCI The Ar and/or Ar′ group(s) contain an ether oxygen, and if Ar = Ar′, then basically identical products may be obtained by the two routes. In the polyetheriﬁcation route the condensation reaction proceeds by reactions of types HO Ar OH HO Ar Cl Cl NaOH Cl Ar´ O Ar NaOH O Ar O Ar´ n NaCl NaCl n The Ar and/or Ar′ group(s) contain sulfone groups, and if Ar = Ar′, then identical products may be obtained by the two routes. Polyetheriﬁcation processes form the basis of commercial polysulfone production methods. For example, the Udel-type polymer (Union Carbide) is prepared by reacting, 4,4′-dichlorodiphenylsulfone with an alkali salt of bisphenol A. The polycondensation is conducted in highly polar solvents, such as dimethylsulfoxide or sulfolane. CH3 HO C O OH + Cl S Cl NaOH O CH3 CH3 C CH3 O O S O O n 4.3.8.1 Properties In spite of their linear and regular structure the commercial polysulfones are amorphous. This property might be attributed to the high degree of chain stiffness of polymer molecules which make crystallization difﬁcult. Because of their high in-chain aromaticity and consequent high chain stiffness, the polymers have high values of Tg (see Table 4.30), which means that the processing temperatures must be above 300°C. Commercial polymers generally resist aqueous acids and alkalis but are attacked by concentrated sulfuric acid. Being highly polar, the polymer is not dissolved by aliphatic hydrocarbons but dissolves in dimethyl formamide and dimethyl acetamide. 567 Industrial Polymers 150 Polysulfone PPS PPO, PC PBT 50 Polyacetol Nylon-6,6 100 ABS Use temperature (°C) 200 0 FIGURE 4.36 Use temperatures of major engineering thermoplastics. In addition to the high heat-deformation resistance, the polymers also exhibit a high degree of chemical stability. This has been ascribed to an enhanced bond strength arising from the high degree of resonance in the structure. The polymers are thus capable of absorbing a high degree of thermal and ionizing radiation without cross-linking. The principal features of commercial polysulfones are their rigidity, transparency, self-extinguishing characteristics, exceptional resistance to creep, especially at somewhat elevated temperatures, and good high-temperature resistance. The use temperatures of the major engineering thermoplastics are compared in Figure 4.36. Polysulfones are among the higher-priced engineering thermoplastics and so are only considered when polycarbonates or other cheaper polymers are unsuitable. In brief, polysulfones are more heat resistant and have greater resistance to creep, whereas polycarbonates have a somewhat higher Izod and tensile impact strength besides being less expensive. In many ﬁelds of use polysulfones have replaced or are replacing metals, ceramics, and thermosetting plastics, rather than other thermoplastics. Since commercial polysulfones can be injection molded into complex shapes, they avoid costly machining and ﬁnishing operations. Polysulfones can also be extruded into ﬁlm and foil. The latter is of interest for ﬂexible printed circuitry because of its high-temperature performance. Polysulfones have found widespread use where good dimensional stability at elevated temperatures is required and fabrication is done by injection molding. Some products made from polysulfones are electrical components, connectors, coil bobbins, relays, and appliances operating at high temperatures (e.g., hair driers, fan heaters, microwave ovens, lamp housings and bases). Polysulfones are transparent (though often slightly yellow), have low ﬂammability (limiting oxygen index typically 38), and burn with little smoke production. Typical properties of some of the commercial polysulfones are shown in Table 4.31. 4.3.9 Polyether Ketones The chemistry and technology of aromatic polyether ketones may be considered as an extension to those of the polysulfones [58]. The two polymer classes have strong structural similarities, and there are strong parallels in preparative methods. Preparations reported for aromatic polyether ketones are analogous to the polysulfonylation and polyetheriﬁcation reactions for the polysulfones. Several aromatic polyether ketones have been prepared. The polyether ether ketone (PEEK) was test marketed in 1978 by ICI. C O O O n 568 Plastics Technology Handbook TABLE 4.31 Properties of Aromatic Polysulfones Property Udel (Union Carbide) Victrex (ICI) Astrel (3M) 9.3 12.2 13.0 64 84 90 3.6 3.5 3.8 2.5 50–100 2.4 40–80 2.6 10 3.0 2.1 3.8 2.6 4.0 2.8 174 203 274 – – 1.9 100 – – – 38 – Tensile strength 103 lbf/in.2 MPa Tensile modulus 105 lbf/in.2 GPa Elongation at break (%) Flexural modulus 105 lbf/in.2 GPa Heat distortion temperature (°C) Impact strength, notched Izod ft.-lb/in. J/m Limiting oxygen index (%) PEEK is made by the reaction of the potassium salt of hydroquione with diﬂuorobenzophenone in a high boiling solvent, diphenylsulfone, at temperatures close to the melting point of the polymer: KO OK Diphenyl sulphone + C C O 280–340°C O F O O n F PEEK is semicrystalline with a melting temperature (Tm) of 335°C and a glass transition temperature (Tg) of 145°C. The degree of crystallinity can vary from 40% (slow cooling) to essentially amorphous (quenching), but is usually about 35%. PEEK is a high-temperature-resistant thermoplastic suitable for wire coating, injection molding, ﬁlm and advanced structural composite fabrication. The wholly aromatic structure of PEEK contributes to its high-temperature performance. The polymer exhibits very low water absorption, very good resistance to water at 125°C (under which conditions other heat-resisting materials, such as aromatic polyamides, are liable to fail), and is resistant to attack over a wide pH range, from 60% sulfuric acid to 40% sodium hydroxide at elevated temperatures, although attack can occur with some concentrated acids. PEEK has outstanding resistance to both abrasion and dynamic fatigue. Its tensile strength decreases less than 10% after 107 cycles at 23°C. It has low ﬂammability with a limiting oxygen index of 35% and generates an exceptionally low level of smoke. Other speciﬁc features are excellent resistant to gamma radiation and good resistance to environmental stress cracking. PEEK has greater heat resistance compared to poly(phenylene sulﬁde) and is also markedly tougher (and markedly more expensive). PEEK is melt processable and may be injection molded and extruded on conventional equipment. Typical applications of PEEK include coating and insulation for high-performance wiring, particularly for the aerospace and computer industries, military equipment, nuclear plant applications, oil wells, and compressor parts. Since it is a crystalline polymer, the strength and thermal resistance of PEEK are increased dramatically by incorporation of reinforcing agents. Composite prepegs with carbon ﬁbers have been developed for structural aircraft components. Typical properties of PEEK are shown in Table 4.32. 569 Industrial Polymers TABLE 4.32 Properties of Polyether Ether Ketone Property Unﬁlled 30% Glass Fiber Filled Tensile strength 103 lbf/in.2 13.3 22.8 92 4.9 (yield) 157 2.2 5.4 3.7 14.9 10.3 1.6 83 1.8 96 Heat distortion temperature (°C) 140 315 Limiting oxygen index (%) 35 – MPa Elongation at break (%) Flexural modulus 105 lbf/in.2 GPa Impact strength, notched Izod ft.-lb/in. J/m 4.3.10 Polybenzimidazole N N N H N H Monomers Tetraaminobiphenyl, terephthalic acid n Polymerization Major Uses Polycondensation Fiber Polybenzimidazole (PBI) is the most well-known commercial example of aromatic heterocycles used as high-temperature polymers. The synthesis of PBI is carried out as follows (see also Figure 1.36). The tetraaminobiphenyl required for the synthesis of PBI is obtained from 3,3′-dichloro-4,4′diaminodiphenyl (a dye intermediate) and ammonia. Many other tetraamines and dicarboxylic acids have been condensed to PBI polymeric systems. H2N H2N NH2 NH2 + HOC O N N N H N H COH O n The high thermal stability of PBI (use temperature about 400°C compared to about 300°C for polyimides) combined with good stability makes it an outstanding candidate for high-temperature application despite its relatively high cost. Fibers have been wet spun from dimethylacetamide solution, and a deep gold woven cloth has been made from this ﬁber by Celanese. The cloth is said to be more comfortable than cotton (due to high moisture retention) and has greater ﬂame resistance than Nomex (oxygen index of 29% for PBI compared to 17% for Nomex). The U.S. Air Force has tested ﬂight suits of PBI and found them superior to other materials. Other applications of PBI are in drogue parachutes and lines for military aircraft as well as ablative heat shields. The PBI ﬁbers have also shown promise as reverse osmosis membranes and in graphitization to 570 Plastics Technology Handbook high-strength, high-modulus ﬁbers for use in composites. The development of ultra-ﬁne PBI ﬁbers for use in battery separator and fuel cell applications has been undertaken by Celanese. A new technique of simple precipitation has been used to process PBI polymers into ﬁlms. Highstrength molecular-composite ﬁlms have been produced with tensile strength in the region of 20,000 psi (137 MPa). The PBI polymer has also been fabricated as a foam. The material provides a low-weight, highstrength, thermally stable, machinable insulation, much needed in the aerospace industry. PBIs exhibit good adhesion as ﬁlms when cast from solution onto glass plates. This property leads to their use in glass composites, laminates, and ﬁlament-wound structures. 4.3.11 Silicones and Other Inorganic Polymers The well-known thermal stability of minerals and glasses, many of which are themselves polymeric, has led to intensive research into synthetic inorganic and semi-inorganic polymers [14,59,60]. Numerous such polymers have been synthesized, but only a few have found industrial acceptance, due to the difﬁculties encountered in processing them. 4.3.11.1 Silicones R O Si R n Monomer Chlorosilanes Polymerization Major Uses Polycondensation Elastomer, sealants, and ﬂuids Silicones are by far the most important inorganic polymers and are based on silicon, an element abundantly available on our planet. The silicone polymers are available in a number of forms, such as ﬂuids, greases, rubbers, and resins. Because of their general thermal stability, water repellency, antiadhesive characteristics, and constancy of properties over a wide temperature range, silicones have found many and diverse applications. [The structure used as the basis of the nomenclature of the silicon compounds is silane SiH4, corresponding to methane CH4 Alkyl-, aryl-, alkoxy-, and halogen-substituted silanes are referred to by preﬁxing silane by the speciﬁc group present. For example, (CH3)2SiH2 is dimethyl silane, and CH3SiCl3 is trichloromethylsilane. Polymers in which the main chain consists of repeating –Si–O– units together with predominantly organic side groups are referred to as polyorganosiloxanes or, more loosely, as silicones.] The commercial production of a broad variety of products from a few basic monomers followed the development of an economically attractive direct process for chlorosilanes, discovered by E. G. Rochow in 1945 at the G. E. Research Laboratories. The process involves reaction of alkyl or aryl halides with elementary silicon in the presence of a catalyst, e.g., copper for methyl- and silver for phenylchlorosilanes. The basic chemistry can be described as SiO2 + C ! Si + 2CO Si + RX ! Rn SiX4−n (n = 0 – 4) In the alkylation of silicon with methyl chloride, mono-, di-, and trimethyl chlorosilanes are formed. The reaction products must then be fractionated. Because the dimethyl derivative is bifunctional, it produces linear methylsilicone polymers on hydrolysis. 571 Industrial Polymers TABLE 4.33 Major Applications of Silicones Mold-release agents Greases and waxes Water repellants Cosmetics Antifoaming agents Glass-sizing agents Insulation Dielectric encapsulation Heat-exchange ﬂuids Caulking agents (RTV) Hydraulic ﬂuids Surfactants Gaskets and seals Laminates Coupling agents Biomedical devices Cl Si CH3 CH3 CH3 Cl H2O HO CH3 Si Si OH CH3 CH3 O n Since monomethyl trichlorosilane has a functionality of 3, the hydrolysis leads to the formation of a highly cross-linked gel. CH3 Cl Si CH3 Cl H2O Cl Si O O Network polymer Since the trimethyl monochlorosilane is monofunctional, it forms only a disiloxane. H2 O (CH3 )3 SiCl⟶(CH3 )3 Si – O – Si(CH3 )3 Products of different molecular-weight ranges and degrees of cross-linking are obtained from mixtures of thee chlorosilanes in different ratios. In characterizing commercial branched and network structures, the CH3/Si ratio (or, generally, R/Si ratio) is thus a useful parameter. For example, the preceding three idealized products have CH3/Si ratios of 2:1, 1:1, and 3:1, respectively. A product with a CH3/Si ratio of 1.5:1 will thus be expected to have a moderate degree of cross-linking. Many different silicon products are available today. The major applications are listed in Table 4.33. 4.3.11.1.1 Silicone Fluids The silicone ﬂuids form a range of colorless liquids with viscosities from 1 to 1,000,000 centistokes (cs). The conversion of chlorosilane intermediates into polymer is accomplished by hydrolysis with water, which is followed by spontaneous condensation. In practice, the process involves three important stages: (1) hydrolysis, condensation, and neutralization (of the HCl evolved on hydrolysis); (2) catalytic equilibration; and (3) devolatilization. The product after the ﬁrst stage consists of an approximately equal mixture of cyclic compounds, mainly the tetramer, and linear polymer. To achieve a more linear polymer and also to stabilize the viscosity, it is common practice to equilibrate the products of hydrolysis by heating with a catalyst such as dilute sulfuric acid. For ﬂuids of viscosities below 1000 cs, this equilibrium reaction is carried out for hours at 100–150°C. After addition of water, the oil is separated from the aqueous acid layer and neutralized. To produce nonvolatile silicone ﬂuids, volatile low-molecular products are removed by using a vacuum still. Commercial nonvolatile ﬂuids have a weight loss of less than 0.5% after 24 h at 150°C. 572 Plastics Technology Handbook Dimethylsilicone ﬂuids ﬁnd a wide variety of applications mainly because of their water repellency, lubrication and antistick properties, low surface tension, a high order of thermal stability, and a fair constancy of physical properties over a wide range of temperature (−70°C to 200°C). As a class the silicone ﬂuids have no color or odor, have low volatility, and are nontoxic. The ﬂuids have reasonable chemical resistance but are attacked by concentrated mineral acids and alkalis. They are soluble in aliphatic, aromatic, and chlorinated hydrocarbons. A well-known application of the dimethylsilicone ﬂuids is as a polish additive. The value of the silicone ﬂuids in this application is due to its ability to lubricate, without softening, the microcrystalline wax plates. Dilute solutions or emulsions containing 0.5–1% of a silicone ﬂuid are extensively used as a release agent for rubber molding. However, their use has been restricted with thermoplastics because of the tendency of the ﬂuids to cause stress cracking in polymers. Silicone ﬂuids are used in shock absorbers, hydraulic ﬂuids, dashpots, and in other damping systems in high-temperature operations. The silicones have established their value as water-repellent ﬁnishes for a range of natural and synthetic textiles. Techniques have been developed which result in the pickup of 1–3% of silicone on the cloth. Leather also may be made water repellent by treatment with solutions or emulsions of silicone ﬂuids. These solutions are also used for paper treatment. Silicone ﬂuids and greases are useful as lubricants for high-temperature operations for applications depending on rolling friction. Grease may be made by blending silicone with an inert ﬁller such as ﬁne silicas, carbon black, or a metallic soap. The silicone/silica greases are used as electrical greases for such applications as aircraft and car ignition systems. Silicone greases have also found uses in the laboratory for lubricating stopcocks and for high-vacuum work. Silicone ﬂuids are used extensively as antifoams, although concentration needed is normally only a few parts per million. The ﬂuids have also found a number of uses in medicine. Barrier creams based on silicone ﬂuids are particularly useful against cutting oils used in metal machinery processes. High-molecular-weight dimethylsilicone ﬂuids are used as stationary phase for columns in vaporphase chromatographic apparatus. Surfactants based on block copolymers of dimethylsilicone and poly(ethylene oxide) are unique in regulating the cell size in polyurethane foams. One route to such polymers used the reaction between a polysiloxane and an allyl ether of poly(ethylene oxide), CH3 CH3 CH3 ( Si O )m Si CH3 CH3 CH3 CH3 CH3 ( Si CH3 O )m Si H + CH2 CH CH2 ( OCH2CH2 )n OH Pt CH2CH2CH2 ( OCH2CH2 ) n OH CH3 where m = 2–5 and n = 3–20. Increasing the silicone content makes the surfactant more lipophilic, whereas a higher poly(ethylene oxide) content makes it more hydrophilic. 4.3.11.1.2 Silicone Resins Silicone resins are manufactured batchwise by hydrolysis of a blend of chlorosilanes. For the ﬁnal product to be cross-linked, a certain amount of trichlorosilane must be incorporated into the blend. (In commercial practice, R/Si ratios are typically in the range of 1.2:1–1.6:1) The cross-linking of the resin is, of course, not carried out until it is in situ in the ﬁnished product. The cross-linking takes place by heating the resin at elevated temperatures with a catalyst, several of which are described in the literature (e.g., triethanolamine and metal octoates). 573 Industrial Polymers The resins have good heat resistance but are mechanically much weaker than cross-linked organic plastics. The resins are highly water repellent and are good electrical insulators particularly at elevated temperatures and under damp conditions. The properties are reasonably constant over a fair range of temperature and frequency. Methyl phenyl silicone resins are used in the manufacture of heat-resistant glass-cloth laminates particularly for electrical applications. These are generally superior to PF and MF glass-cloth laminates. The dielectric strength of silicon-bonded glass-cloth laminates is 100–120 kV/cm compared to 60– 80 kV/cm for both PF and MF laminates. The insulation resistance (dry) of the former (500,000 W) is signiﬁcantly greater than those for the PF and MF laminates (10,000 and 20,000 W, respectively). The corresponding values after water immersion are 10,000, 10, and 10 W. Silicone laminates are used principally in electrical applications such as printed circuit boards, transformers, and slot wedges in electric motors, particularly class H motors. Compression-molding powders based on silicone resins are available and have been used in the molding of switch parts, brush ring holders, and other electrical applications that need to withstand high temperatures. 4.3.11.1.3 Silicone Rubbers Silicone elastomers are either room-temperature vulcanization (RTV) or heat-cured silicone rubbers, depending on whether cross-linking is accomplished at ambient or elevated temperature. [The term vulcanization (see Chapter 1 and Chapter 2) is a synonym for cross-linking. While curing is also a synonym for cross-linking, it often refers to a combination of additional polymerization plus crosslinking.] RTV and heat-cured silicone rubbers typically involve polysiloxanes with degrees of polymerizations of 200–1500 and 2500–11,000, respectively. While the lower-molecular-weight polysiloxanes can be synthesized by the hydrolytic step polymerization process, the higher-molecular-weight polymers are synthesized by ring-opening polymerization using ionic initiators: H3C CH3 Si H3C H3C CH3 O O Si Si O CH3 O H3C CH3 O Si CH3 Si CH3 The cyclic tetramer (octamethylcyclotetrasiloxane) is equilibrated with a trace of alkaline catalyst for several hours at 150–200°C, the molecular weight being controlled by careful addition of monofunctional siloxane. The product is a viscous gum with no elastic properties. Before fabrication it is necessary to compound the gum with ﬁllers, a curing agent, and other special additives on a two-roll mill or in an internal mixer (see “Rubber Compounding,” Chapter 2). Unﬁlled polymers have negligible strength, whereas reinforced silicone rubbers may have strengths up to 2000 psi (14 MPa). Silica ﬁllers are generally used with silicone rubbers. These materials with particle sizes in the range 0.003–0.03 mm are prepared by combustion of silicon tetrachloride (fume silicas), by precipitation, or as an aerogel. Heat-curing of silicone rubbers usually involve free-radical initiators such as benzoyl peroxide, 2,4dichlorobenzoyl peroxide, and t-butyl per-benzoate, used in quantities of 0.5–3%. These materials are stable in the compounds at room temperature for several months but will start to cure at about 70°C. The curing (cross-linking) is believed to take place by the sequence of reactions shown in Figure 4.37. The process involves the formation of polymer radicals via hydrogen abstraction by the peroxy radicals 574 Plastics Technology Handbook R Heat R Peroxide 2R • Radical CH3 CH3 Si Si O CH3 CH2 2R • + 2RH CH3 CH2 O Si O Si CH3 FIGURE 4.37 O CH3 Peroxide curing of silicone rubbers. formed from the thermal decomposition of the peroxide and subsequent cross-linking by coupling of the polymer radicals. The cross-linking efﬁciency of the peroxide process can be increased by incorporating small amounts of a comonomer containing vinyl groups into the polymer, e.g., by copolymerization with small amounts of vinyl-methyl silanol: CH2 CH2 CH HO Si CH3 OH + HO CH3 Si OH O CH3 CH CH3 Si O Si O CH3 CH3 Dimethylsilicone rubbers show a high compression set. (For example, normal cured compounds have a compression set of 20–50% after 24 h at 150°C.) Substantially reduced compression set values may be obtained by using a polymer containing small amounts of methylvinylsiloxane. Rubbers containing vinyl groups can be cross-linked by weaker peroxide catalysts. Where there is a high vinyl content (4–5% molar), it is also possible to vulcanize with sulfur. Room-temperature vulcanizing silicone rubbers (RTV rubbers) are low-molecular-weight liquid silicones with reactive end groups and loaded with reinforcing ﬁllers. Several types are available on the market. “One-component” RTV rubbers consist of an air-tight package containing silanol-terminated polysiloxane, cross-linking agent (methylacetoxysilane), and catalyst (e.g., dibutyltin laurate). Moisture from the atmosphere converts the cross-linking agent to the corresponding silanol (acetic acid is a byproduct), CH3Si(OH)3, which brings about further polymerization combined with cross-linking of the polysiloxane, 3 SiR2 OH CH3Si(OH)3 –H2O CH3 SiR2 O Si O SiR2 O SiR2 Two-component RTV formulations involve separate packages for the polysiloxane and cross-linking agent. A typical two-component RTV formulation cures by reaction of silanol end groups with silicate esters in the presence of a catalyst such as tin octoate or dibutyltin dilaurate (Figure 4.38). 575 Industrial Polymers CH3 CH3 Si HO OH RO CH3 Si OR CH3 OR CH3 Si CH3 Si RO HO OH Si CH3 CH3 CH3 CH3 Si Si CH3 O O CH3 O CH3 Si CH3 O Si Si CH3 CH3 FIGURE 4.38 Curing of RTV rubbers by reaction of silanol end groups with silicate esters in the presence of a catalyst such as tin octoate or dibutyltin dilaurate. Another two-pack RTV formulation cures by hydrosilation, which involves the addition reaction between a polysiloxane containing vinyl groups (obtained by including methylvinyldichlorosilane in the original reaction mixture for synthesis of polysiloxane) and a siloxane cross-linking agent that contains Si–H functional groups, such as Si[OSi(CH3)2H]4. The reaction is catalyzed by chloroplatinic acid or other soluble platinum compounds. SiR O CH = CH2 + Si[OSi(CH3)2H]4 Si [ OSi(CH3)2 CH2CH2SiR O ]4 Hydride functional siloxanes can also cross-link silanol-terminated polysiloxanes. The reaction is catalyzed by tin salts and involves elimination of H2 between Si–H and Si–O–H groups. RTV rubbers have proved to be of considerable value as they provide a method for producing rubbery products with the simplest equipment. These rubbers ﬁnd use in the building industry for caulking and in the electrical industry for encapsulation. Nontacky self-adhesive rubbers (fusible rubbers) are obtained if small amounts of boron (∼1 boron atom per 300 silicon atoms) are incorporated into the polymer chain. They may be obtained by condensing dialkylpolysiloxanes end-blocked with silanol groups with boric acid or by reacting ethoxyl endblocked polymers with boron triacetate. Bouncing putty is somewhat similar in that the Si–O–B bond occurs occasionally along the chain. It is based on a polydimethylsiloxane polymer modiﬁed with boric acid, additives, ﬁllers, and plasticizers to give a material that shows a high elastic rebound when small pieces are dropped on a hard surface but ﬂows like a viscous ﬂuid on storage or slow application of pressure. 576 Plastics Technology Handbook The applications of the rubbers stem from their important properties, which include thermal stability, good electrical insulation properties, nonstick properties, physiological inertness, and retention of elasticity at low temperatures. The temperature range of general-purpose material is approximately −50°C to +250°C, and the range may be extended with special rubbers. Silicone rubbers are, however, used only as special-purpose materials because of their high cost and inferior mechanical properties at room temperature as compared to conventional rubbers (e.g., natural rubber and SBR). Modern passenger and military aircraft each use about 500 kg of silicone rubber. This is to be found in gaskets and sealing rings for jet engines, vibration dampers, ducting, sealing strips, and electrical insulators. Silicone cable insulation is also used extensively in naval craft since the insulation is not destroyed in the event of ﬁre but forms an insulating layer of silica. The rubbers ﬁnd use in diverse other applications which include electric iron gaskets, domestic refrigerators, antibiotic container closures, and for nonadhesive rubber-covered rollers for handling such materials as confectionary and adhesive tape. Due to their relative inertness, new applications have emerged in the biomedical ﬁeld. A silicone rubber ball is used in combination with a ﬂuorocarbon seal to replace defective human heart valves. Silicone rubber has had many applications in reconstructive surgery on or near the surface of the body. Prosthetic devices are very successfully used in all parts of the body. The cold-curing silicone rubbers are of value in potting and encapsulation. Liquid silicone rubbers may be considered as a development from the RTV silicone rubbers but they have a better plot life and improved physical properties, including heat stability (in the cured state) similar to that of conventional silicone elastomers. Liquid silicone rubbers range from a ﬂow consistency to a paste consistency and are usually supplied as a two-pack system, which requires simple blending before use. The materials cure rapidly above 110°C. In injection molding of small parts at high temperatures (200–250°C), cure times may be as small as a few seconds. One example of application is in baby bottle nipples, which, although more expensive, have a much longer working life. Liquid silicone rubbers have also been used in some extruded applications. Vulcanization of the extruded material may be carried out by using infrared heaters or circulated hot air. The process has been applied to wire coating, ignition cables, optical ﬁbers, various tapes, and braided glass-ﬁber sleeving, as well as for covering delicate products. 4.3.11.2 Polyphosphazenes OCH2CF2CHF2 N P OCH2CF2CHF2 n Monomers Phosphorus pentachloride, ammonium chloride, ﬂuorinated alcohols Polymerization Major Uses Polycondensation followed by nucleophilic replacement of chloro-groups Aerospace, military, oil exploration applications Polyphosphazenes containing nitrogen and phosphorus have been synthesized by replacing the chlorine atoms on the backbone chain of polymeric phosphonitrilic chloride (dichlorophosphazene) by alkoxy or ﬂuoroalkoxy groups. These derivative polymers do not exhibit the hydrolytic instability of the parent polymer. The general synthesis scheme is 577 Industrial Polymers OR Cl [N PCI5 + NH4Cl P ]n RONa [N P ]n OR Cl With mixtures o

Log In

Plastics Technology Handbook Fifth Edition

Plastics Technology Handbook Fifth Edition

Related Papers

RELATED PAPERS