HANDBOOK OF COMPOSITESSECOND EDITIONEdited by

S.T. PetersProcess Research, Mountain View, Calfornia, USA

CHAPMAN & HALL

London Weinheim . New York Tokyo Melbourne Madras

-

-

-

-

Published by Chapman & Hall, an imprint of Thomson Science, 2-6 Boundary Row, London SE18HN, UKThomson Science, 2-6 Boundary Row, London SE18HN, UK Thomson Science, 115 Fifth Avenue, New York, NY 10003, USA Thomson Science, Suite 750,400 Market Street, Philadelphia, PA 19106, USA Thomson Science, Pappelallee 3,69469 Weinheim, Germany

First edition 1982 Second edition 1998

0 1998 Chapman & HallThomson Science is a division of International Thomson Publishing Typeset in 10/12 pt Palatino by GreenGate Publishing Services, Tonbridge, England Printed in Great Britain by Cambridge University Press ISBN 0 412 54020 7 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic,,mechanical, photocopying, recording or otherwise, without the prior written permission of the publishers. Applications for permission should be addressed to the rights manager at the London address of the publisher. The publisher makes no representation, express or implied, with regard to the accuracy of the information contained in this book and cannot accept any legal responsibility or liability for any errors or omissions that may be made. A catalogue record for this book is available from the British Library

CONTRIBUTORS

SURESH G. ADVANI Department of Mechanical Engineering, University of Delaware, Spencer Laboratory, Newark, DE 19716, USA MAURICE E AMATEAU Applied Research Laboratory, Pennsylvania State University, PO Box 30, State College, PA 16804, USA

KENNETH R. BERG Riggs Corporation, 837 Agate Street, Medford, OR 97501, USA LARS A. BERGLUND LuleH University of Technology, SE-97187 LuleH, Sweden D. BROWN Boeing Commercial Airplane Group, Douglas Products Division, Mail Stop D001-0018, Long Beach, CA 90846, USA JOHN D. BUCKLEY 23 East Governor Drive, Newport News, V 23602, A USA JERRY L. CADDEN C & S Technologies, 42759 Mountain Shadow, Murrieta, CA 92562, USA ZHONG CAI(deceased) 4180 Berkeley Creek Drive, Duluth, GA 30136, USA

EVER J. BARBER0 315 Engineering Science Building, West Virginia University, Morgantown, WV 26506-6106, USAA.I. BEIL' Institute of Polymer Mechanics, Latvian Academy of Sciences, 23 Aizkraukles Street, Riga LV-1006, Latvia JEROME S. BERG True Temper Sports, 5421 Avenida Encinas, Suite G, Carlsbad, CA 92008, USA

x

Handbook o composites fMIRIA M. FINCKENOR EH12 Bldg 4711, Marshall Space Flight Center, AL 35812, USA LIHWA FONG BLK G 5, Nanyang Avenue, Singapore 63616 HUGH H. GIBBS Polycomp Consulting, Inc., 25 Crestfield Road, Wilmington, DE 19810, USA TIMOTHY GUTOWSKI Department of Mechanical Engineering, Massachusetts Institute of Technology, Bldg 35-234, Cambridge, MA 02139, USA RICHARD N. HADCOCK 6 Sue Circle, Huntington, NY 11743, USA ENAMUL HAQUE Azdel, Inc., Technology Center, 658 Washburn Switch Road, Shelby, NC 28151-2284, USA

FRANK A. CASSIS FAC Associates, 1150 N. Mountain, Suite 1028, Upland, CA 91786, USA LINDA L. CLEMENTS C & C Technologies, PO Box 1089, Dayton, NV 89403, USA DOUGLAS L. DENTON Chrysler Corporation, CIMS 482-00-13, 800 Chrysler Drive, Auburn Hills, MI 48326-2757, USA EDDY A. DERBY Composite Optics, 9617 Distribution Ave, San Diego, CA 92121, USA GEORGE W. DU Principal Engineer, 16331 Bay Vista Drive Cleanvater, FL 34620, USA HARRY W. DURSCH Boeing Defense and Space Group, PO Box 3999, Mail Stop 73-09, Seattle, WA 98124-2846, USA DON 0. EVANS Cincinnati Milacron, 4701 Marburg Avenue, Cincinnati, Ohio 45209, USA

L.J. HART-SMITH Boeing Commercial Airplane Group, Douglas Products Division, Mail Stop D800-0019, 4000 Lakewood Boulevard, Long Beach, CA 90846, USA

Confributors xiJENNIFER HETH Cytec Fiberite, 501 W. Third Street, Winona, MN 55987-2854, USA THOMAS S. JONES Industrial Quality, Inc., 640 E. Diamond Ave., Suite C, Gaithersburg, MD 20877, USA THOMAS JUSKA Naval Surface Warfare Center, Carderock Division, Structures and Composites Department, Bethesda, MD 20084-5000, USA JOHN T. KANNE 2201 Johnson Road, Memphis, TN 38139, USA HARRY S. KATZ Utility Development Corporation, 112 Naylon Avenue, Livingston, NJ 07039, USA VALERY I. KOSTIKOV Niigrafit Institute, 2 Electrodonaya Street, Moscow, 111524, Russia GARY C. KRUMWEIDE Composite Optics, 9617 Distribution Avenue, San Diego, CA 92121, USA V.L. KULAKOV Institute of Polymer Mechanics, Latvian Academy of Sciences, 23 Aizkraukles Street, Riga LV-1006, Latvia KHALID LAFDI Center for Advanced Friction Studies, Southern Illinois University at Carbondale, Carbondale, IL 62901-4343, USA CHRISTY KIRCHNER LAPP 1412 Bellingham Way, Sunnyvale, CA 94087, USA ROBERT A. LATOUR Clemson University, Clemson, SC 29634, USA BURR L. LEACH Cambridge Industries, 1700 Factory Avenue, Marion, IN 46952, USA STEWART N. LOUD Composites Worldwide Inc., 991 Lomas Santa Fe Drive, C469, Solana Beach, CA 92075-2125, USA

V.S. KILIN Niigrafit Institute, 2 Electrodonaya Street, Moscow, 111524, RussiaFRANK K. KO Drexel University, Fibrous Materials Research Laboratory 27-439, Philadelphia, PA 19104, USA KENT E. KOHKONEN Brigham Young University, 435 CTB Technology Department, Provo, UT 84602, USA

xii Handbook o composites f VICKI P. MCCONNELL Ray Publishing, Independence Street, Suite 270, Wheat Ridge, CO 80033, USA ANDREW C. MARSHALL Marshall Consulting, 720 Appaloosa Drive, Walnut Creek, CA 94596, USA ANTHONY MARZULLO 39 Harold Street, COSCob, CT 06807-2132, USA DONALD W. OPLINGER Federal Aviation Administration, Wm. J. Hughes Technical Center AAR-431, Atlantic City, International Airport, NJ 08405, USA HARRY E. PEBLY 198 Center Grove Road, Randolph, NJ 07869, USA LYNN S. PENN Department of Chemical and Materials Engineering, 177 Anderson Hall, University of Kentucky, Lexington, KY 40506-0046, USA S.T. PETERS Process Research, 925 Sladky Avenue, Mountain View, CA 94040-3625, USA NITIN POTDAR Brigham Young University, 435 CTB Technology Department, Provo, UT 84602, USA KENNETH REIFSNIDER Virginia Polytechnic Institute and State University, Patton Hall 120, Blacksburg, VA 24061-0219, USA THEODORE J. REINHART 345 Forrer Boulevard, Dayton, OH 45419-3238, USA PAUL E SADESKY C & S Technologies, 23547 Mountain Court, Murrieta, CA 92562, USA FRANK J. SCHWAN 36671 Montecito Drive, Fremont, CA 94536, USA ANTON L. SEIDL 18941 Mellon Drive, Saratoga, CA 95070, USA JOCELYN M. SENG Owens Corning Science and Technology Center, 2790 Columbus Road, Granville, OH 43023-1200, USA SHALABY W. SHALABY Clemson University, 301 Rhodes Res., Clemson, SC 29634, USA

Contributors xiiiDAVID A. SHIMP PO Box 974, Prospect, KY 40059, USA DONALD R. SIDWELL 44609 Grove Lane, Lancaster, CA 93534-2833, USA BRIAN E. SPENCER Spencer Composite Corporation, 3220 Superior Street, PO Box 4377, Lincoln, NE 68504-0377, USA ROBERT C. TALBOT 7199 Lorine Court, Columbus, OH 43235-5125, USA YU.M. TARNOPOL'SKII Institute of Polymer Mechanics, Latvian Academy of Sciences, 23 Aizkraukles Street, Riga LV-1006, Latvia R.C. TENNYSON University of Toronto, Institute for Aerospace Studies, 4925 Dufferin Street, Downsview, Ontario, Canada M3H 5T6 JAMES L. THRONE Shenvood Technologies, Inc., 158 Brookside Boulevard, Hinckley, OH 44233-9676, USA FRANK TRACESKI Department of Defense, 5203 Leesburg Pike Suite 1403, Falls Church, VA 22041, USA WAYNE C. TUCKER Naval Undersea Warfare Center, PO Box 86, Exeter, RI 02822, USA V V. VASILIEV . Moscow State University, 14-1-110 Podolskih Kursantov Street, Moscow 113545, Russia DENNIS J. VAUGHAN 146 Longview Drive, Anderson, SC 29621, USA H. WANG Department of Chemical and Materials Engineering, 177 Anderson Hall, University of Kentucky, Lexington, KY 40506-0046, USA ANN E WHITAKER EHOl Bldg 4612, Marshall Space Flight Center, AL35812, USA BRIAN A. WILSON Wilson Composite Group, 6611 Folsom-Auburn Road, Suite C, Folsom, CA 95630, USA

S. WONG Boeing Commercial Airplane Group, Douglas Products Division, Mail Stop D001-0018, Long Beach, CA 90846, USA

xiv Handbook of composites MAURICE A. WRIGHT Center for Advanced Friction Studies, Southern Illinois University at Carbondale, Carbondale, IL 62901-4343, USA PHILIP R. YOUNG Emory & Henry College, Department of Chemistry, Emory, V 24327, A USA

ABOUT THE EDITOR

S.T. Peters was previously a fellow engineer with Westinghouse Electric Corporation, Marine Division prior to devoting full time to composite and materials and processing consulting for his own company, Process Research, in Mountain View, CA. He has written many articles on composites and filament winding, a book on filament winding, edited one previous book and holds several patents on winding techniques and composite joints.

He is a private consultant with worldwide clients and has presented tutorials on composites to many audiences, including the US Navy and NASA, several technical societies and two universities. He is a licensed professional engineer in the state of California, a member of ASM, and the composites division of SME and has been elected a fellow of SAMPE.

ACKNOWLEDGEMENTS

As with any large undertaking there is a supporting group of people without whose help the objective would not be met. I wish to acknowledge my wife, Lynn, for her help in deciphering and rewriting some of the articles and for enduing my sometimes uncivil

approach to resolving problems. Thanks also go to Mr Frank Heil and Dr Alvin Nakagawa of Westinghouse Electric, Marine Division (now Norton Grumman) for their editorial and review help. I also wish to thank Dr Linda Clements for her advice and support.

PREFACE

Today, fiber reinforced composites are in use in a variety of structures, ranging from spacecraft and aircraft to buildings and bridges. This wide use of composites has been facilitated by the introduction of new materials, improvements in manufacturing processes and developments of new analytical and testing methods. Unfortunately, information on these topics is scattered in journal articles, in conference and symposium proceedings, in workshop notes, and in government and company reports. This proliferation of the source material, coupled with the fact that some of the relevant publications are hard to find or are restricted, makes it difficult to identify and obtain the up-to-date knowledge needed to utilize composites to their full advantage. This book intends to overcome these difficulties by presenting, in a single volume, many of the recent advances in the field of composite materials. The main focus of this book is on polymeric matrix, metal matrix, and ceramic matrix composites. The book treats a wide range of subjects. The topics, presented in 49 chapters and two appendices include:0

00

0 0 0 00 0

00

properties of different component (fiber, matrix, filler) materials; manufacturing techniques; analysis and design; testing; mechanically fastened and bonded joints; repair; damage tolerance; environmental effects; health, safety, reuse, and disposal; applications in: aircraft and spacecraft; land transportation; marine environments; biotechnology; construction and infrastructure; sporting goods.

Each chapter, written by a recognized expert, is self-contained, and contains many of the 'state-of-the-art' techniques required for practical applications of composites. Thus, this book should serve as a useful source of information for practicing engineers and specialists, as well as for workers new to this field. George S. Springer

overview of composite material systems and products;

CONTENTS

Contributors Preface About the editor Foreword Acknowledgements Introduction, composite basics and road map S.T. Peters1 Overview of composite materials

ixxv

xvi xvii xviii1

21

Theodore J. ReinhartPART ONE: BASIC MATERIALS Polymeric matrix systems2 Polyester and vinyl ester resin Frank A. Cassis and Robert C. Talbot 34 4875

3 Epoxyresins L.S. Penn and H . Wang4 High temperature resins

Hugh H . Gibbs5 Speciality matrix resins David A . Shimp6 Thermoplastic resins Lars A. Berglund

99115

Reinforcements and composites7 Fiberglass reinforcement Dennis J. Vaughan131

vi Handbook of composites 8 Boron, high silica, quartz and ceramic fibers Anthony Marzullo 9 Carbon fibers Khalid Lafdi and Maurice A. Wright 156

169202 242 254 29 1 307 333

10 Organic fibers Linda L. Clements 1 Particulate fillers 1 Harry S. Katz 12 Sandwich construction Andrew C. Marshall13 Metal matrix composites V l . Kostikov and V S . Kilin 14 Ceramic composites M.E Amateau 15 Carbon-carbon composites John D. Buckley

PART TWO: PROCESSING METHODS General composites and reinforced plastics16 Hand lay-up and bag molding D.R. Sidwell 17 Matched metal compression molding of polymer composites Enamul Haque and Burr (Bud) L. Leach 18 Textile preforming Frank K. KO and George W. Du 19 Table rolling of composite tubes John T. Kanne and Jerome S. Berg 20 Resin transfer molding Lihwa Fong and S.G. Advani 21 Filament winding Yu.M. Tarnopolskii, S.T. Peters, A.I. Beil 22 Fiber placement Don 0. Evans 23 Pultrusion Brian A. Wilson 24 Processing thermoplastic composites James L. Throne 352 378 397 425 433 456 476 488

525

Contents viiAdvanced composites25 Tooling for composites Jerry L. Cadden and Paul F. Sadesky 26 Consolidation techniques and cure control Zhong Cui and Timothy Gutowski 27 Composite machining Kent E. Kohkonen and Nitin Potdar 28 Mechanical fastening and adhesive bonding D. W. Oplinger 29 Surface preparations for ensuring that the glue will stick in bonded composite structures L.J. Hart-Smith, D. Brown and S. Wong 556 576 596 610

667

PART THREE: DESIGN AND ANALYSIS30 Laminate design Jocelyn M . Seng 31 Design of structure with composites F.J. Schwan 32 Analysis methods V.V. Vasiliev 33 Design allowables substantiation Christy Kirchner Lapp 34 Mechanical tests Yu.M. Tarnopol'skii and V.L. Kulakov 686709

736 758 778

PART FOUR. ENVIRONMENTAL EFFECTS35 Durability and damage tolerance of fibrous composite systems Ken Reifsnider 36 Environmental effects on composites A n n F. Whitaker, Miria M . Finckenor, Harry W. Dursch, R.C. Tennyson and Philip R. Young37 Safety and health issues

794 810

822 838 857 883

Jennifer A. Heth38 Nondestructive evaluation methods for composites Thomas S. Jones 39 Repair aspects of composite and adhesively bonded aircraft structures Anton L. Seidl40 Reuse and disposal Harry E , Pebly

viii Handbook o composites f

PART FIVE APPLICATIONS41 Land transportation applications Douglas L. Denton 42 Marine applications Wayne C. Tucker and Thomas Juska 43 Commercial and industrial applications of composites Stewart N. Loud 44 Composite biomaterials Shalaby W. Shalaby and Robert A. Latour 45 Scientificapplications of composites Vicki I? McConnell 46 Construction Ever J. Barber0 47 Aerospace equipment and instrument structure G a y C. Krumweide and Eddy A. Derby 905 916 931 957 967 982 1004 1022 1044

48 Aircraft applications Richard N. Hadcock49 Composites in the sporting goods industry Brian E. Spencer

APPENDICESAppendix A Typical properties for advanced composites Kenneth R. Berg Appendix B Specifications and standards for polymer composites Frank T. Traceski Index 1053

1059 1069

INTRODUCTION, COMPOSITE BASICS AND ROAD MAP*S.T. Peters

This is an introduction to composites and will encourage the reader to obtain more information. Only the basic concepts will be covered here; reference will be made to the chapters in the book that expand or follow up and elaborate on these basics. The reader will see that the subjects of this book cover the spectrum of composites and range from the basic and simple to the complex. Thus, there are complicated equations because they are the tools that are used every day to describe real structures; and there will also be the more general, less complicated approaches that are limited in analysis power. These chapters have been developed by the most knowledgeable composite professionals in the world; a blend of academicians and the engineers who fabricate real composite structures. Modern structural composites, frequently referred to as Advanced Composites, are a blend of two or more components, one of which is made up of stiff, long fibers, and the other, a binder or matrix which holds the fibers in place. The fibers are strong and stiff relative to the matrix and are generally orthotropic (having different properties in two different directions). The fiber, for advanced structural composites, is long, with length to diameter ratios of over 100. The fibers strength and stiffness are usually much greater, perhaps several times more, than the matrix material. The matrix material can by polymeric (e.g. polyester resins, epoxies),Handbook of Composites. Edited by S.T. Peters. Published in 1998 by Chapman & Hall, London. ISBN 0 412 54020 7

metallic, ceramic or carbon. When the fiber and the matrix are joined to form a composite they retain their individual identities and both directly influence the composites final properties. The resulting composite will generally be composed of layers (laminae) of the fibers and matrix stacked to achieve the desired properties in one or more directions. The high strength or stiffness to weight ratios of advanced composites are well known, but there are other advantages also (Table 1.1). These advantages translate not only into aircraft, but into everyday activities, such as longer drives with a graphite-shafted golf club (because more of the mass is concentrated at the clubhead) or less fatigue and pain because a graphite composite tennis racquet has mherent damping. Generally, the advantages accrue for any fiber/composite combination and disadvantages are more obvious with some. These advantages have now resulted in many more reasons for composite use as shown in Table 1.2. Proper design and material selection can circumvent many of the disadvantages.1.1 MATERIAL SYSTEMS

An advanced composite laminate can be tailored so that the directional dependence of strength and stiffness matches that of the loading environment. To do that, layers of unidirectional material called laminae are ori* This chapter has been adapted from S.T. Peters, in Handbook o Plastics Elastomers and Composites, 3rd edn, (ed. f C.A. Harper). McGraw-Hill, New York, 1996, and is used with permission of the McGraw-Hill companies.

2 Introduction, composite basics and road mapTable 11 Advantages/disadvantages of advanced composites . Advantages___-

Disadvantages

Weight reduction High strength or stiffness to weight ratio Tailorable properties Can tailor strength or stiffness to be in the load direction Redundant load paths (fiber to fiber) Longer life (no corrosion)Lower manufacturing costs because of less part count

Cost of raw materials and fabrication Transverse properties may be weak

Matrix is weak, low toughness

Reuse and disposal may be difficult Difficult to attach

Inherent damping Increased (or decreased) thermal or electrical conductivity ented to satisfy the loading requirements. These laminae contain fibers and a matrix. Because of the use of directional laminae, the tensile, flexural and torsional shear properties of a structure can be disassociated from one another to some extent and a golf shaft, for example, can be changed in torsional stiffness without changing the flexural or tensile stiffness. Fibers can be of the same material within a lamina or several fibers mixed (hybrid). The common commercially available fibers are as follows:

Analysis is difficultMatrix subject to environmentaldegradation

Carbon/graphite fibers (Chapter 9) have demonstrated the widest variety of strengths and modulii and have the greatest number of suppliers. The fibers begin as an organic fiber, rayon, polyacrylonitrile or pitch which is called the precursor. The precursor is then stretched, oxidized, carbonized and graphitized. There are many ways to produce these fibers, but the relative amount of exposure at temperatures from 2500-3000C results in greater or less graphitization of the fiber. Higher degrees of graphitization usually result in a stiffer fiber (higher modulus) with 0 fiberglass; greater electrical and thermal conductivities 0 graphite; and usually higher cost. 0 aramid; The organic fiber Kevlar 49, (Chapter 10) 0 polyethylene; also called aramid, essentially revolutionized 0 boron; pressure vessel technology because of its great 0 silicon carbide; tensile strength and consistency coupled with 0 silicon nitride, silica, alumina, alumina silica. low density, resulting in much more weight The advantages of fiberglass (Chapter 7) are its effective designs for rocket motors. Aramid high tensile strength and strain to failure, but composites are still widely used for pressure heat and fire resistance, chemical resistance, vessels but have been largely supplanted by moisture resistance and thermal and electrical the very high strength graphite fibers. Aramid properties are also cited as reasons for its use. composites have relatively poor shear and It is by far the most widely used fiber, primar- compression properties; careful design is ily because of its low cost; but its mechanical requires for their use in structural applications properties are not comparable with other that involve bending or compression. structural fibers.

Material systems 3Table 1.2 The reasons for using composites

Reason for useLighter, stiffer stronger

Material selectedBoron, all carbodgraphites, some aramid Very high modulus carbon/graphite Fiberglass, vinyl esters, bisphenol A fumarates, chlorendic resins High strength carbon/graphite, epoxy

___

Appl ica t ion/driver

Military aircraft, better performance Commercial aircraft, operating costs Spacecraft with high positional accuracy requirements for optical sensors Tanks and piping, corrosion resistance to industrial chemicals, crude oil, gasoline at elevated temperatures Industrial rolls, for paper, films

Controlled or zero thermal expansion Environmental resistance

Lower inertia, faster startups, less deflection Lightweight, damage tolerance

High strength carbon/graphite, CNG tanks for greencars, trucks fiberglass, (hybrids), epoxy and busses to reduce environmental pollution High strength or high modulus carbon graphite/ epoxy Carbon/graphite/epoxy High-speed aircraft. Metal skins cannot be formed accurately Tennis, squash and racquetball racquets. Metallic racquets are no longer available Laminated new growth wooden support beams with high modulus fibers incorporated Cooling tower driveshafts

More reproducible complex surfaces Less pain and fatigue

Reduces logging in old growth forests Reduces need for intermediate support and resists constant 100% humidity atmosphere Tailorability of bending and twisting response Transparency to radiation Crashworthiness

Aramid, carbon/graphite

High strength carbon/graphite-epoxy Carbon/graphite-epoxy Carbon/ graphite-epoxy Carbon/ graphite-epoxy

Golf shafts, fishing rods X-ray tables Racing cars Automotive and industrial driveshafts Commercial boats Freeway support structure repair after earthquake

Higher natural frequency, lighter Carbon/ graphite-epoxy Water resistance Ease of field application Fiberglass (woven fabric), polyester or isopolyester Carbon/graphite, fiberglass- epoxy, tape and fabric

The polyethylene fibers have the same property drawbacks as aramids, but also suffer from low melting temperature which limits

their use to composites that cure or operate below 149C (300F) and a susceptibility to degradation by ultraviolet light exposure.

4 lntvodmction, composite basics and road map

Both of these types of fibers have wide usage in personal protective armor. In spite of the drawbacks, production of both of these fibers is enjoying strong worldwide growth. Boron fibers (Chapter 8), the first advanced composite fibers to be used on production aircraft, are produced as individual monofilaments upon a tungsten or carbon substrate by pyrolytic reduction of boron trichloride (BC1,) in a sealed glass chamber. The relatively large cross section fiber is used today primarily in metal matrix composites which are processed at temperatures which would attack carbon/graphite fibers.

1.2 MATRIX SYSTEMS

If parallel and continuous fibers are combined with a suitable matrix and cured properly, unidirectional composite properties such as those shown on Table 1.3 are the result. The functions and requirements of the matrix are to:0

00

0

0

keep the fibers in place in the structure; help to distribute or transfer loads; protect the filaments, both in the structure and before and during fabrication; control the electrical and chemical properties of the composite; carry interlaminar shear.

Table 1.3 Properties of typical unidirectional graphite/epoxy composites (Fiber volume fraction, V , = 0.62)

High strength

High modulusGPa (psi x IO6)220 (32) 6.9 (1.0) 4.8 (0.7) 0.30

Elastic constants~~

GPa (psi x I O 6 )145 (21) 9.6 (1.4) 5.8 (0.85) 0.30~~

Longitudinal modulus, E, Transverse modulus, E , Shear modulus, G , Poissons ratio (dimensionless)u ~ ,~~~ ~~

.~

~

~~

~~

Strength properties~ ~~~ ~~~~ ~ ~

MPa ( Z 0 3 psi)2139 (310) 54 (7.8) 1724 (250) 76 (11) 87 (12.6) 128 (18.5)

MPa (lo3 psi)760 (110) 28 (4) 690 (100) 170 (25) 70 (10) 70 (10)%

Longitudinal tension, Ft, Transverse tension, FtUT Longitudinal compression, FCUL Transverse compression, FCUT Inplane shear, PLT Interlaminar shear, F,

Ultimate strains-

-~

Longitudinal tension, Transverse tension, Longitudinal compression, ECUL Transverse compression, EC1lT Inplane shear

1.4 0.67 0.9 3.6 2.0

0.3 0.4 0.3 2.8-

Physical propertiesDensity, kg/m3 (Ib/in3) Longitudinal CTE, ye/K (pe/OF) Transverse CTE ~ E / (pe/OF) K

1600 (0.056) -0.079 (-0.044) 21.6 (12)

1700 (0.058)

-0.54 (-0.3) 58 (32)

From References 1, 2 and 3; CTE = coefficient of thermal expansion

Matrix systems 5The needs, or desired properties of the matrix, that depend on the purpose of the structure are:0 0 0 0 0

composite. The common thermoset matrices for composites include the following:0

0 0

0

0

0

0

minimize moisture absorption; have low shrinkage; Must wet and bond to fiber; low coefficient of thermal expansion; must flow to penetrate the fiber bundles completely and eliminate voids during the compacting/curing process; have reasonable strength, modulus and elongation (elongation should be greater than fiber); must be elastic to transfer load to fibers; have strength at elevated temperature (depending on application); have low temperature capability (depending on application); have excellent chemical resistance (depending on application); be easily processable into the final composite shape; have dimensional stability (maintain its shape).

0 0 0

polyester and vinylesters (Chapter 2); epoxy (Chapter 3); bismaleimide (BMI) (Chapter 4); polyimide (Chapter 4); cyanate ester and phenolic triazine (Chapter 5).

There are many matrix choices available; each type has impact o n the processing technique, physical and mechanical properties and environmental resistance of the finishedTable 1 4 Selection criteria for epoxy resin systems .

Each of the resin systems has some drawbacks, which must be accounted for in design and manufacturing plans. Polyester matrices have been in use for the longest period, and are used in the widest range and greatest number of structures. The usable polymers may contain up to 50% by weight of unsaturated monomers and solvents such as styrene. Polyesters cure via a catalyst (usually a peroxide) resulting in an exothermic reaction, which can be initiated at room temperature. The most widely used matrices for advanced composites have been the epoxy resins. These resins cost more than polyesters and do not have the high temperature capability of the bismaleimides or polyimides, but because of the advantages shown in Table 1.4 they are widely used.

AdvantagesAdhesion to fibers and to resin No by-products formed during cure Low shrinkage during cure High or low strength and flexibility Solvent and chemical resistance Resistance to creep and fatigue Solid or liquid resins in uncured state Wide range of curative options Adjustable curing rate Good electrical properties

DisadvantagesResins and curatives somewhat toxic in uncured form Absorb moisture Heat distortion point lowered by moisture absorption Change in dimensions and physical properties due to moisture absorption Limited to about 200C upper temperature use (dry) Difficult to combine toughness and high temperature resistance High thermal coefficient of expansion High degree of smoke liberation in a fire May be sensitive to ultraviolet light degradation Slow curing

6 Introduction, composite basics and road map There are two resin systems in common use for higher temperatures, bismaleimides and polyimides. New designs for aircraft demand a 177C (350F) operating temperature not met by the other common structural resin systems. The primary bismaleimide in use is based on the reaction product from methylene dianiline (MDA) and maleic anhydride: bis (4-maleimidophenyl) methane (MDA BMI). Two newer resin systems have been developed and have found applications in widely diverse areas. The cyanate ester resins, marketed by Ciba-Geigy, have shown superior dielectric properties and much lower moisture absorption than any other structural resin for composites. The dielectric properties have enabled their use as adhesives in multilayer microwave printed circuit boards, and the low moisture absorbance have caused them to be the resin of universal choice for structurallystable spacecraft components. The phenolic triazine (PT) resins also have superior elevated temperature properties, along with excellent properties at cryogenic temperatures. Their resistance to proton radiation under cryogenic conditions was a prime cause for their choice for use in the superconducting supercollider, subsequently canceled by the US Congress. Polyimides are the highest temperature polymer in general advanced composite use with a long term upper temperature limit of 232C (450F) or 316C (600F). Two general types are: condensation polyimides, that release water during the curing reaction, and addition type polyimides with somewhat easier process requirements.1.3 FIBER MATRIX SYSTEMS

The end user sees a composite structure. Someone else, probably a prepregger, combined the fiber and the resin system and someone else caused the cure and compaction to result in a laminated structure. A schematic of the steps to arrive at a finished composite from the initial fiber is shown in Fig. 1.1. In many cases, the end user of the structure has fabricated the composite from prepreg, which is a low-temperature-stable combination

ROVINGN?

WEAUE?

COLLIMRTE

UNI TRPE

A

Fig. 11 Manufacturing steps in composite structure. .

Fiber matrix systems 7of the resin, its curing agents and the fiber. The three types of continuous fibers, roving, tape and woven fabric available as prepregs give the end user many options in terms of design and manufacture of a composite structure. Although the use of dry fibers and impregnation at the work (i.e. filament winding, pultrusion or hand lay-up) is very advantageous in terms of costs; there are many advantages to the use of prepregs as shown in Table 1.5, particularly for the manufacture of modem composites. The prepreg process for thermoset matrices can be accomplished by feeding the fiber continuous tape, woven fabric or roving through a resin-rich solvent solution and then removing the solvent by hot tower drying. The excess resin is removed via a doctor blade or metering rolls and then the product is staged to the cold-stable prepreg form, (B stage) (Fig. 1.2). The newer hot melt procedure for prepregs is gradually replacing the solvent method because of environmental concerns. A film of resin that has been cast hot onto release paper

Table 1.5 Advantages of prepregs over wet impregnation

Prepregs reduce the handling damage to dry fibers Improve laminate properties by better dispersion of short fibers Prepregs allow the use of hard-to-mix or proprietary resin systems Allow more consistency because there is a chance for inspection before use Heat curing provides more time for the proper laydown of fibers and for the resin to move and degas before cure Increased curing pressure reduces voids and improves fiber wettingMost prepregs have been optimized as individual systems to improve processing

Release Poly

Unwind

Prepreg Wind

Pump and Reservoir

Unwind

Fig. 1.2 Schematic of typical solvent prepregging process. (Adapted from Reference 2.)

8 Introduction, composite basics and road map

0

0Doctor Plate 1 Impregnation Zone

Pauer Paper

Creel

Paper

TChill Plate

Take-up Prepreg Windup

Plate 2

Fig. 1.3 Schematic of typical film impregnating process. (Adapted from Reference 2.)

is fed, along with the reinforcement, through a series of heaters and rollers to force the resin into the reinforcement. Two layers of resin are commonly used so that a resin film is on both sides of the reinforcement; one of the release papers is removed and the prepreg is then trimmed, rolled and frozen (Fig. I.3)2.The solvent technique has been largely replaced for advanced fibers because of environmental pollution concerns and a need to exert better control over the amount of resin on the fiber.1.3.1 UNIDIRECTIONAL PLY PROPERTIES

Wf = weight fraction of fiber wf = weight of fiber wc= weight of composite pf = density of fiber p, = density of composite uf = volume of fiber u, = volume of composite Vf = volume fraction of fiber V, = volume fraction of matrix p, = density of matrix. A percentage fiber that is easily achievable and repeatable in a composite and convenient for reporting mechanical and physical properties for several fibers is 60%. The properties of unidirectional fiber laminates are shown in Table 1.3 for carbon/graphite/epoxy. Values for the other fibers can be seen in their respective chapters. These values are for individual lamina or for a unidirectional composite, and they represent the theoretical maximum (for that fiber volume) for longitudinal in plane properties. Transverse, shear and compression properties will show maxima at different fiber volumes and for different fibers, depending on how the matrix and fiber interact. These values can be used to calculate the properties of a laminate which has fibers oriented in several directions. To do that, the methods of description for ply orientation must be introduced.

The manufacturer of the prepreg reports an areal weight for the prepreg and a resin percentage, by weight. Each of the different fibers has a different density, resulting in a composite of different density at the same fiber volume percentage. Since fiber volume is used to relate the properties of the manufactured composites, the following equations can be used to convert between weight fraction and fiber volume.

where:

Quasi-isotropic laminate 91.4 PLY ORIENTATIONS, SYMMETRY AND

BALANCE1.4.1 PLY ORIENTATIONS

lined to indicate that half of it lies on either side of the plane of symmetry (Fig. 1.4(f)).1.4.2 SYMMETRY

One of the advantages of using a modern composite is the potential to orient the fibers to respond the load requirements. This means that the composite designer must show the material, the fiber orientations in each ply, and how the plies are arranged (ply stackup). A 'shorthand' code for ply fiber orientations has been adapted for use in layouts and studies. Each ply (lamina)is shown by a number representing the direction of the fibers in degrees, with respect to a reference ( x ) axis. 0" fibers of both tape and fabric are normally aligned with the largest axial load (axis) (Fig. 1.4(a)). Individual adjacent plies are separated by a slash in the code if their angles are different (Fig. 1.4@)). The plies are listed in sequence, from one laminate face to the other, starting with the ply first on the tool and indicated by the code arrow with brackets indicating the beginning and end of the code. Adjacent plies of the same angle of orientation are shown by a numerical subscript (Fig. 1.4(c)). When tape plies are oriented at angles equal in magnitude but opposite in sign, (+) and (-) are used. Each (+) or (-) sign represents one ply. A numerical subscript is used only when there are repeating angles of the same sign. Positive and negative angles should be consistent with the coordinate system chosen. An orientation shown as positive in one right handed coordinate system may be negative in another. If the y and z axis directions are reversed, the f 45 plies are reversed (Fig. 1.4(d)). Symmetric laminates with an even number of plies are listed in sequence, stating at one face and stopping at the midpoint. A subscript 'S' following the bracket indicates only one half of the code is shown (Fig. 1.4(e)). Symmetric laminates with an odd number of plies are coded as a symmetric laminate except that the center ply, listed last, is over-

The geometric midplane is the reference surface for determining if a laminate is symmetrical. In general, to reduce out-ofplane strains, coupled bending and stretching of the laminate and complexity of analysis, symmetric laminates should be used. However, some composite structures (e.g. filament wound pressure vessels) can achieve geometric symmetry so that symmetry through a single laminate wall is not necessary, if it constrains manufacture. To construct a midplane symmetric laminate, for each layer above the midplane there must exist an identical layer (same thickness, material properties, and angular orientation) below the midplane (Fig. 1.4(e)).1.4.3 BALANCE

All laminates should be balanced to achieve inplane orthotropic behavior. To achieve balance, for every layer centered at some positive angle +e there must exist an identical layer oriented at -8 with the same thickness and material properties. If the laminate contains only 0" and/or 90" layers it satisfies the requirements for balance. Laminates may be midplane s p metic but not balanced and vice versa. Figure 1.4(e) is symmetric and balanced whereas Fig. 1.4(g)is balanced but unsymmetric .1.5 QUASI-ISOTROPICLAMINATE

The goal of composite design is to achieve the lightest, most efficient structure by aligning most of the fibers in the direction of the load. Many times there is a need, however, to produce a composite which has some isotropic properties, similar to metal, because of multiple or undefined load paths. A 'quasi-isotropic' laminate lay-up accomplishes this for the x and y planes only; the z or through-the-laminate-

10 Introduction, composite basics and road map90"

-

Reference Axis

90"

l; z9' -45' 00 "

Tape Laminate

,/

.-,

Tool side0"45'

I I P

1I\ \

[0/9O]s Typical Callout

0"

-450

w

+450

0"

L

[0/903/0]

P

I

90" +45" -45" -45" +45" 90"V I

Line of Symmetry

T

[0/90/*45]s

Typical Callout

Tape and Fabric Laminate[ 0/f45/To1 s. Typical Callout

Line of Symmetry

0

Fig. 1.4 Ply orientations, symmetry and balance. (Continued on next page)

Methods of analysis 11

p,+45" -45"-45"

Tape Laminate

[0/90/f45/ 452/ 0/ 01 i 9Typical Callout

+45"

Fabric Laminate

[(0,90)/(~45)/(0,90)]Typical Callout

I

0".90" I

j

0",90"

I

h)

Fig. 1.4 Ply orientations, symmetry and balance. (Continued)

thickness plane is quite different and lower. 1. arrive at quick values to determine if a comMost laminates produced for aircraft applicaposite is feasible; tions have been, with few exceptions, 2. arrive at values for insertion into computer 'quasi-isotropic'. As designers become more programs for laminate analysis or finite eleconfident and have access to a greater database ment analysis; with fiber-based structures, more applications 3. check on the results of computer analysis. will evolve. For a quasi-isotropic laminate, the The rule of mixtures holds for composites. The following are requirements: micromechanics formula to arrive at the 0 It must have three layers or more. Young's modulus for a given composite is: 0 Individual layers must have identical stiffEc = V, + Vm Em E, ness matrices and thicknesses. 0 The layers must be oriented at equal angles. and v,+ vm= 1 For example, if the total number of layers is = V ,E , + Em (1- V,) (1.3) M , the angle between two adjacent layers should be 360"ln. If a laminate is con- where structed from identical sets of three or more Ec = composite or ply Young's modulus in layers each, the condition on orientation tension for fibers oriented in direction of must be satisfied by the layers in each set, applied load for example: ( O o / + 60"), or ( O o / + 45"/90)s. V = volume fraction of fiber ( f ) or matrix (m) E = Young's modulus of fiber ( f ) or matrix 1.6 METHODS OF ANALYSIS (m). There are a number of methods in common But, since the fiber has much higher use for the analysis of composite laminates. Young's modulus than the matrix, the second The use of micromechanics, i.e. the application part of the equation can be ignored. of the properties of the constituents to arrive at E, >> Em the properties of the composite ply can be Ec = E,V, (1.4) used to:

12 Introduction, composite basics and road mapThis is the basic rule of mixture and represents the highest Youngs modulus composite, where all fibers are aligned in the direction of load. The minimum Youngs modulus for a reasonable design (other than a preponderance of fibers being orientated transverse to the load direction) is the quasi-isotropic composite and can be approximated by: appropriate for a particular application. Figure 1.5 shows the progression of physical properties for Youngs modulus in tension, E, (fiber), E, (lamina) and Ex,, (laminate), longitudinal tensile strength, and coefficient of thermal expansion a where the subscripts L and X stand for in-plane in the principal fiber direction and t and Y stand for the transverse direction for a theoretical high strength (from Ec = (3/8) E,V, (1.5) Table 1.3) carbon/graphite fiber composite The quasi-isotropic modulus, E, of a composite from the fiber to the laminate. The values laminate is (3/8)E,+(5/8)EZ where E,, is the decrease or are translatedin a logical fashion modulus of the lamina in the fiber direction and and reflect the law of mixtures. The analysis is E, is the transverse modulus of the lamina3. relatively simple for modulus dominated The transverse modulus for polymeric-based properties but strength-dominated values composites is a small fraction of the longitudinal must be treated in light of one of several failmodulus (see E, in Table 1.3)and can be ignored, ure theories and changes in the thermal for preliminary estimates, resulting in a slightly coefficient of expansion are not predictable lower-than-theoretical value for Ec for a quasi- from laws of mixtures. Other factors which isotropic laminate. This approximate value for enter into the translation efficiency are: comthe quasi-isotropic modulus represents the patibility of the resin system with the fiber and lower limit of composite modulus. It is useful in the fiber finish, strain-to-failure of the resin comparing of composite properties to those of system and the damage the fiber undergoes metals and in establishing if a composite is during impregnation, laydown and cure.

.6 GPa, FT =54 MPa

E x = 76 GPa a x = 4.98peK

ra2

E y = 76 GPa a y = 4.98~ E K

>ay

Fig. 1.5 The anatomy of a composite laminate.

Composite fabrication techniquesTable 1.6 High-strength carbon/graphite laminate

13

propertiesLaminate Longitudinal modulus E,, (GPa)76.5 76.5 98.5 81.3 55.0 41.34

Bending modulus, E , (GPa)126.8 26.3 137.8 127.5 89.6 41.34

Shear modulus, G,, (GPa)5.24 5.24 5.24 21.0 21.0 27.56

(0/90,/0) (90/0,/90) (02/902/OJ (0,/~45,/0,) (0/+45/90)>

Aluminum

competitor, so vendor values in a generic class may differ widely. 4. Most tables of values are presented as 'typical values'. Those values and the values that are part of the menu of many computer analysis programs should be used with care. Each user must find their own set of values for design, develop useful design allowables, and apply appropriate 'knock down' factors, based on the operating environments expected in service. (Chapter 33 and Appendix A give guidelines.)1.7 COMPOSITE FABRICATION TECHNIQUES

Table 1.6 shows mechanical values for several composite laminates with the fiber of Table 1.3 and a typical resin system. The first and second entries are for simple 0/90 laminates and show the effect of changing the position of the plies. The effect of increasing the number of 0 plies is shown next and the final two laminates demonstrate the effect of +45 plies on mechanical properties, particularly the shear modulus. The last entry is a quasi-isotropic laminate. These laminates are then compared to a typical aluminum alloy. When employing the data extracted from tables, some caution should be observed by the reader. The values seen in many tables of data may not always be consistent for the same materials or the same group of materials from several sources for the following reasons:1. Manufacturers have been refining their production processes so that newer fibers may have greater strength or stiffness. These new data may not be reflected in the compiled data. 2. The manufacturer may not be able to change the value quoted for the fiber because of government or commercial restrictions imposed by the specification process of his customers. 3. There are many different high-strength fibers commercially available. Each manufacturer has optimized their process to maximize their mechanical properties and each process may differ from that of the

The goals of the composite manufacturing process are to:0

00 0

achieve a consistent product by controlling fiber thickness; - fiber volume; - fiber directions; minimize voids; reduce internal residual stresses; process in the least costly manner.-

The procedures to reach these goals involve iterative processes to select the three key components:0 0 0

composite material and its configuration; tooling; process.

Once material selection has been completed, the first step leading to the acceptable composite structure is the selection of tooling, which is intimately tied to process and material. For all curing techniques the tool must be:0

0

0

0

strong and stiff enough to resist the pressure exerted during cure; dimensionally stable through repeated heating and cooling cycles; light enough to respond reasonably quickly to the changes in cure cycle temperature and to be moved in the shop; leakproof so that the vacuum and pressure cycles are consistent.

14 Introduction, composite basics and road map

The tool face is commonly the surface imparted to the outer surface of the composite and must be smooth, particularly for aerodynamic surfaces. The other surface frequently may be of lower finish quality and is imparted by the disposable or reusable vacuum bag. This surface can be improved by the use of a supplemental metal tool known as a caul plate. (Press curing, resin transfer molding, injection molding and pultrusion require a fully closed or two sided mold). Figure I. 6 shows the basic components of the tooling for vacuum bag or autoclave processed components and Table 1.7 shows the function of each part of the system. Tooling options have been augmented by2

the introduction of elastomeric tooling wherein the thermal expansion of an elastomer provides some or all of the pressure curing cure, or a rubber blanket is used as a reusable vacuum bag. The volumetric expansion of an elastomer can be used to fill a cavity between the uncured composite and an outer mold. The use of elastomeric tooling can provide the means for fabricating complex box-like structures such as integrally stiffened skins with a co-cured substructure in a single curing operation. Tooling (Chapter 25) and the configuration of the reinforcement have a great influence on the curing process selected and vice-versa. The5 67

3

4

8

9

12

13

14

9

10

11

Fig. 1.6 Typical vacuum bag lay-up components. Table 1.7 Functions of vacuum bag components

Component *-

FunctionsTemporarily bonds vacuum bag to tool Exhausts air, provides convenient connection to vacuum pump Encloses part, allows for vacuum and pressure Allows air or vacuum transfer to all of part Holds other components of bag in place Holds components in place Imparts desired contour and surface finish to composite Allows flow of resin or air without adhesion Prevents adhesion of laminate resin to tool surface Imparts a bondable surface to cured laminate Allows transfer of air or vacuum Soaks up excess resin Forces excess resin to flow vertically, increasing fluid pressure

1 2 3 4 5 6 7 8 910 1 1 12

1314

Bag sealant Vacuum fitting and hardware Bagging film Open weave breather mat Polyester tape (wide) Polyester tape (narrow) Caul sheet Perforated release film Non-perforated release film Peel ply Laminate 1581-styleglass breather manifold 1581 style glass bleeder ply Stacked silicone edge dam

* numbers refer to Fig. 1.6

Composite fabrication techniques 15probable reinforcement configuration that facilitates the completion of the finished composite is shown on Table 1.8. The choice between unidirectional tape and woven fabric has frequently been made on the basis of the greater strength and modulus attainable with the tape particularly in applications which compression strength is important. There are other factors that should be included in the trade, as shown in Table 1.9.1.7.1 LAY-UP TECHNIQUE

Table 1.8 Common reinforcement configuration for the manufacturing process

Reinforcement Prepreg Prepreg Prepreg Other, configuration tape or (dry) or (dry) woven tow woven preforms, or non- chopped woven fibers fabricHandlay-up Automatic tape laydown Filament winding Resin transfer molding Pultrusion Fiber placement

X X

x, (XI x, (X)(XI

x

xm(X)

xmX

Lay-up techniques along with composite cure control have received the greatest attention for processing. In efforts to reduce labor costs of composite fabrication, to which lay-up (Chapter 16) has traditionally been the largest contributor, mechanically assisted, controlled tape laying and automated integrated manuTable 1.9 Fabric compared with tape reinforcement

(X) X X

X

Tape advantages

_

_

_

_

_

~

-

~

Tape disadvantagesPoor drape on complex shapes Cured composite more difficult to machine Lower impact resistance Multiple plies required for balance and symmetry Higher labor costs for hand lay-up

Best modulus and strength efficiency High fiber volume achievable

Low scrap rateNo discontinuities Automated lay-up possible Available in thin plies Lowest cost prepreg form Less tendency to trap volatiles

Fabric advantages~

_ _ ~ _ _ _ - - ~ _ _

Fabric disadvantages

Better drape for complex shapes Single ply is balanced and may be essentially symmetric Can be laid up without resin Plys stay in line better during cure Cured parts easier to machine Better impact resistance Many forms available

Fiber discontinuities (splices) Less strength and modulus efficient Lower fiber volume than tape More costly than tape Greater scrap rates Warp and fill properties differ Fabric distortion can cause part warping

16 Introduction, composite basics and road m a pfacturing systems have been developed. Table Generally, the percent matrix weight is higher 1.10 shows some of the considerations for before cure initiation; the matrix flows out of choosing a lay-up technique. the laminate and takes the excess resin with void limit In addition to any cost savings by the use of the potential voids. An arbitrary 1% an automated technique for long production has been adopted for most autoclaved comruns, there are two key quality assurance fac- posites; filament wound and pultruded tors which validate the automated techniques. composites will have higher void volumes They are: greatly reduced chance that release depending upon the application. paper or film could be retained, which would An autoclave is essentially a closed, presdestroy shear and compressive strength if surized oven; many common epoxy laminates undetected, and reduced probability of the are cured at an upper temperature of 177C addition or loss of an angle ply which would (350F) and 6 MPa (100 psi). Autoclaves are cause warping due to the laminates lack of still the primary tool in advanced composite symmetry and balance. processing and have been built up to 16 m (55 All curing techniques use heat and pressure feet) long at 6.1 m (20 feet) diameter. Since to cause the matrix to flow and wet out all the autoclaves are expensive to build and operate, fibers before the matrix solidifies (Chapter 26). many other methods of curing, compactingTable 1.10 Considerations in composite lay-up technique

Considerat ionOrientation accuracy

ManualLeast accurate

Flat tapeAutomatic

Contoured tapeSomewhat dependent on tape accuracy and computer program Program records Automatic removal Additional improvement Approximately 1M$ or greater Approximately same as flat tape Complex program and machine make this a consideration Difficulty in changing Longer tape is more economical Least scrap due to back and forth laydown Least voids Necessary

Ply count Release film retention Labor costs Machine costs Production rate Machine up time

Dependent on operator, count Mylars Up to operator High N/A Low (1.5 Ib/h) N/A

Dependent on operator Automatic86% improvement quoted

Some costs10 lb/h

Not a consideration

Varying tape widths Tape lengths Cutting waste Compaction pressure Programming

Not a concern Longer tapes more difficult Scrap on cuttingNo pressure

Easily changed Longer is more economical Less scrap Less voids N/A

N/A

Compositefabrication techniques 17composites have been developed. The two newest and most attractive methods are fiber placement and resin transfer molding.1.7.2 RESIN TRANSFER MOLDING

Fiber placement, initially developed by Hercules Aerospace Co., is a cross between filament winding and automatic tape laydown, 1.7.3 FILAMENT WINDING retaining many of the advantages of both. The Filament winding is a process by which con- natural outgrowth of adding multiple axes of tinuous reinforcements in the form of rovings control to filament winding machines results or tows (gathered, untwisted strands of fiber) in control of the fiber laydown so that non axiare wound over a rotating mandrel. The man- symmetric surfaces can be wound. This drel can be cylindrical, round or any other involves the addition of a modified tape layshape as long as it does not have re-entrant down head to the filament winding machine curvature. Special machines (Fig. 1.7) traversing and much more. The Cincinnati-Milacron machine additions include in-process compaction, individual tow cut/start capabilities, a resin tack control system, differential tow payout, low tension on fiber and enhanced offf 1 iine programming (Chapter 22).

Previous discussions have centered on moving resin out of the laminate to reduce voids. Resin transfer involves the placement of dry fiber reinforcement into a closed mold and then injecting a catalyzed resin into the mold to encapsulate the reinforcement and form a composite (Chapter 20). The impetus for the use of this process comes from the large cost reductions that can be realized in raw materials and lay-up. The process can utilize low injection pressures i.e. 55 MPa (80 psi), therefore, the tooling can be lower cost plastic or a vacuum bag rather than metal.

a wind eye at speeds synchronized with the mandrel rotation, control winding angle of the reinforcement and the fiber lay-down rate. The reinforcement may be wrapped in adjacent bands or in repeating bands that are stepped the width of the band and that eventually cover the mandrel surface. Local reinforcement can be added to the structure using circumferential windings, local helical bands, or by the use of woven or unidirectional cloth. The wrap angle can be varied from low angle helical to high angle circumferential or 'hoop', which allows winding from about 4"-90" relative to the mandrel axis; newer machines can 'place' fiber at 0".1.7.4 FIBER PLACEMENT

I /11

1.7.5 PULTRUSION

Pultrusion is an automated process for the I manufacture of constant volume/shape profiles from composite materials (Chapter 23). The composite reinforcements are continuously pulled through a heated die and shaped and cured simultaneously. If the cross-sectional shape is conducive to the process, it is p f r 4 the fastest and most economical method of Fig. 1.7 The helical filament wound ply. (Courtesy composite production. Straight and cured conof Westinghouse Electric Co., Marine Division.) figurations can be fabricated with square,

c F

I

18 Introduction, composite basics and road map round, hat-shaped, angled 'I' or 'T'-shaped cross-sections from vinylester, polyester, or epoxy matrices with E and S-glass, Kevlar and graphite reinforcements.. The curing is effected by combinations of dielectric preheating and microwave or induction (with conductive reinforcements like carbon graphite) while the shape traverses the die.1.7.6 BRAIDING, WEAVING AND OTHER PREFORM TECHNIQUES3

I

Fig. 1.8 The unidirectional ply.

Braiding, weaving, knitting and stitching represent methods of forming a shape, generally be the same in any transverse direction. This is referred to as preforming, with the composite the transverse isotropy assumption; it is fibers before impregnation (Chapter 18). The approximately satisfied for most unidirecshape may be the final product or some inter- tional composite plies. These properties are typically modified by mediate form such as a woven fabric. The braiding process is continuous and is transformation relative to the laminate axis amenable to round or rectangular shapes or where these may not be the same as the ply smooth curved surfaces and can transition axes. In a multidirectional laminate there can be easily from one shape to another. The other fabric preforming techniques are as many as 21 stiffness constants. Strength preweaving, knitting and the non-structural dictions are equally as complicated because of stitching of unidirectional tapes. Stitching sim- directional differences, i.e. compression is not ply uses a non-structural thread, such as nylon always equal to tension, and because the sevor Dacron, to hold dry tapes at selected fiber eral failure theories are complex. As the angles. Preforming in this manner results in a complexity of the matrix calculations increase, higher-cost raw material but saves labor costs it becomes evident that errorless mathematical for orientation of individual lamina. The manipulations are impossible without the aid stitched preform has known, stable fiber ori- of computers. Chapters 30 and 32 elaborate on entations similar to woven fabric, without the the techniques of laminate analysis and the crossovers which could reduce compressive applications of laminates to structures strength.1.9 DESIGN OF COMPOSITES 1.8 MECHANICS OF COMPOSITE MATERIALS

The 1,2,3 axes in Fig. 1.8 are special and are called the ply axes, or material axes. The 1 axis is in the direction of the fibers, and is called the longitudinal axis or the fiber axis. The longitudinal axis is typically the highest stiffness and strength direction. Any direction perpendicular to the fibers (in the 2,3 plane) is called a transverse direction. Sometimes, to simplify analysis and test requirements, ply properties are assumed to

The design process for composites involves both laminate design and component design and must also include considerations of manufacturing process and eventual environmental exposure. These steps are all interdependent with composites and the most efficient design must involve true concurrent engineering. Figure 1.9 shows the various concerns that should be a part of the composite design process at the initiation of the design process, and continuously from there on.

Design of composites 191.9.1 LAMINATE DESIGN RECOMMENDATIONS

1. Take advantage of the orthotropic nature of the fiber composite ply. 0 To carry in-plane tensile or compressive loads align the fibers in the directions of these loads. 0 For in-plane shear loads, align most fibers at -c 45" to these shear loads. 0 For combined normal and shear in-plane loading provide multiple or intermediate ply angles for a combined load capability. 2. Intersperse the ply orientations. 0 If a design requires a laminate with 16 plies at *45", 16 plies at 0", and 16 plies at 90, use the interspersed design (90,/ -c 45,/0,),s rather than (90,/ .+ 45,/10,)s. Concentrating plies at nearly the same angle (0" and 90" in the above example) provides the opportunity for large matrix cracks to form. These produce lower laminate allowables, probably because large cracks are more injurious to the fibers, and more readily form delaminations than the finer cracks occurring in interspersed laminates. 0 If a design requires all 0" plies, some 90" plies (and perhaps some off-angle plies ) should be interspersed in the laminate to provide some biaxial strength and stability and to accommodate unplannedComposite Material Environmental Considerations

0

loads. This improves handling characteristics, and serves to prevent large matrix cracks from forming. Locally reinforce with fabric or mat in areas of concentrated loading. (This technique is used to locally reinforce pressure vessel domes). Use fabric, particularly fiberglass or Kevlar, as a surface ply to restrict surface (handling) damage. Ensure that the laminate has sufficient fiber orientations to avoid dependence on the matrix for stability. A minimum coverage of 6 to 10% of total thickness in 0, ?45", 90" directions is recommended.

3. Select the lay-up to avoid mismatch of properties of the laminate with those of the adjoining structures - or provide a shear/separator ply. Poisson's ratio: if the transverse strain of a laminate greatly differs from that of adjoining structure, large interlaminar stresses are produced under load. Coefficient of thermal expansion: temperature change can produce large interlaminar stresses if coefficient of thermal expansion of the laminate differs greatly from that of adjoining structure. 0 The ply layer adjacent to most bonded joints should not be perpendicular to the direction of loading. Thicken the composite in the joint area, soften the composite by adding fiberglass or angle plies and select the highest strain-capability adhesive.4. Use multiple ply angles. Typical composite laminates are constructed from multiple unidirectional or fabric layers which are positioned at angular orientations in a specified stacking sequence. From many choices, experience suggests a rather narrow range of practical construction from which the final laminate configuration is usually selected. The multiple layers are usually oriented in at least two different angles, and possibly three or four; (go, O0/&", or

Component

Fig. 1.9 Design considerations for composites.

20 Introduction, composite basics and road map

attempt to standardize the raw materials and their test methods by publication of specifications (Appendix A). However, these standards have not reached the level of use to allow complete dependence upon them without supplier-user interaction and user testing. The fabricators of composites will rely on specifications for control of fiber, resin and/or the prepreg. Many prepreg resin and fiber Further suggestions can be seen in Chapter 31. vendors will certify only to their own specifications which may differ from those shown; users should consult the vendors to determine 1.10 COMPOSITE TESTING what certification limits exist before commitTo ensure consistent, reproducible compo- ting to specification control. nents, three levels of testing are employed: As part of raw materials verification, comincoming materials testing, in-process testing posite design effort and final product and control and final structure verification. verification mechanical testing of composite test specimens will be performed. The testing of composite materials offers unique chal1.10.1 INCOMING MATERIALS TESTING lenges because of the special characteristics of Incoming materials testing seeks to verify the composites. Factors not considered important conformance of the raw materials to specifica- in metals testing are very important in testing tions and to insure processibility. The levels of composites (Chapters 34,39). knowledge of composite raw materials do not approach those for metals, which can be bought to several consensus specifications and REFERENCES will appear generally identical although pur- 1. Foral, R.F. and Peters, S.T., Composite chased from many manufacturers. Although Structures and Technology Seminar Notes, 1989 there are fewer suppliers for composite raw 2. Hercules Data Sheet for AS-4/3901-6 prepreg H050-377/GF Prod Hdbk (4)/jc/2 materials, the numbers of permutations of 3. Agarwal, B.D. and Broutman, L.J., Analysis and resins, fibers and manufacturers prevents the Performance of Fiber Composites 2nd edn, John kind of standardization necessary to be able to Wiley and Sons, New York, 1990 p. 103 buy composite raw materials as if they were 4. Mayorga, G.D. in International Encyclopedia of alloys. ASTM (American Society for Testing Composites, (ed. S.M. Lee) Vol 4, VCH and Materials), SAE/AMS/NOMETCOM Publishers, N.Y., N.Y., 1991 (Society of Automotive Engineers, Aeronautical 5 . Tsai, S.W. and Pagano, N.J. in Composite Materials Workshop, (eds. S.W. Tsai, J.C. Halpin Materials Standards/ Nonmetallic Materials and N.J. Pagano), Technomic Publishing Co., Committee) and SACMA (Suppliers of Lancaster, PA, 1978, p. 249 Advanced Composite Materials Association) 0 / ~ 0 / 9 0 cover most applications, with 0 0 between 30 and 60 degrees). Unidirectional laminates are rarely used except when the basic composite material is only mildly orthotropic (e.g. certain metal matrix applications) or when the load path is absolutely known or carefully oriented parallel to the reinforcement (e.g. stiffener caps).

POLYESTER AND VINYL ESTER RESINSFrank A. Cassis and Robert C. Talbot

2

2.1 INTRODUCTION AND HISTORY

Organic polymers are divided into two types, reinforced-thermoplastic and thermoset. With thermoset polymers such as unsaturated polyesters and vinyl esters, a chemical reaction cross links the material so that it cannot be returned to liquid form. Other common thermosetting polymers include epoxy and phenolic resins. Thermoset plastics made with polyester and vinyl ester resins represent the major portion of the reinforced plastic composites industry today. Early workers on unsaturated polyesters soon learned that despite the possession of reactive double bonds, these resins were sluggish in reacting with themselves. Even with effective catalysts, they still required high temperatures and lengthy cure times to complete the cross linking reaction. The key to modern day application of unsaturated polyesters was the discovery by Carlton Ellis in 1937l that the addition of reactive monomers, such as styrene, gave mixtures that would copolymerize many times faster than homopolymerization. The styrene addition produced the added benefit of an easily handled liquid material that could be pumped, transported and fabricated into a finished plastic by a myriad of molding processes. Developments during the 1940s accelerated the commercial applicability of unsaturated polyesters to the position they hold today. Styrene became readily available and lower in cost as a result of the US Government's sponHandbook of Composites. Edited by S.T. Peters. Published in 1998 by Chapman & Hall, London. ISBN 0 412 54020 7

sored production of styrene-butadiene rubber. At the same time, scientists found that styrenated polyesters could yield high strength, light weight structures when reinforced with glass fibers. They also learned that fiberglassreinforced polyesters had excellent electrical properties and that large structures could be molded at low pressures with low cost tooling. As a result, commercial development proceeded rapidly after World War I1 with materials and molding research moving in many directions. In the 20 years that followed, polyester and vinyl ester resins matured rapidly and by the mid-l970s, the composites fabricator and end user had numerous options with these matrix systems to achieve the desired properties in the finished part.2.2 POLYESTER RESINS

The reaction of an organic acid with an alcohol results in the formation of an ester. By using a difunctional acid and a difunctional alcohol (glycol)a linear polyester is produced (Fig. 2.1).0 0 II II H-(-0 - C - R - C - 0

- R -)" -OH

Fig. 2.1

Properties of polyesters can be varied by using different combinations of diacids and glycols. These products are thermoplastic polyesters and they are made with various acids and

Polyester resins 35glycols such as the following: -

Acids Phthalic anhydride Isophthalic acid Terephthalic acid Adipic acid

GZycols Ethylene glycol Propylene glycol Neopentyl glycol Diethylene glycol

The reaction product of terephthalic acid and ethylene glycol is the well known polyethylene terephthalate (PET) which is used to make polyester fibers and polyester plastics such as clear plastic bottles for soft drinks. Unsaturated polyesters are produced by replacing part of the saturated diacid with an 2.2.1 UNSATURATED POLYESTER CLASSES unsaturated diacid such as maleic anhydride or fumaric acid (Fig. 2.2). The former is vastly Unsaturated polyesters are divided into types preferred since it is lower in cost, easily han- or classes depending on the structure of the dled and produces only half the water that basic building block. These are orthophthalic, would be generated in the reaction when isophthalic, terephthalic, bisphenol-fumarate, fumaric acid is used. chlorendic and dicyclopentadiene.

In the esterification reaction with maleic anhvdride, the unsaturated acid isomerizes to the fumarate structure which copolymerizes with styrene much faster than the maleate form. A high degree of isomerization to the fumarate structure is essential to produce an unsaturated polyester with high reactivity. Although the isomerization of maleic anhydride is usually from 65-95% in the esterification reaction, some commercial resins are deliberately formulated with the more expensive fumaric acid to obtain maximum reactivity with the monomer employed.

CH = CH

Orthophthalic resinsThese are commonly referred to as ortho or general purpose resins and are usually based on phthalic anhydride, maleic anhydride and propylene glycol. Since the acid groups in phthalic anhydride are on adjacent carbons of the benzene ring, it is very difficult to produce resin molecular weights as high as those achievable with isophthalic and terephthalic acid. Accordingly, resins made from phthalic anhydride have poorer thermal stability and chemical resistance than their iso/tere counterparts.

0 = c-0-c = 0Maleic anhydride HI 1

t

\

HII

HOOC- C = C

- COOH

Fumaric acid

Fig. 2.2

The resultant polyester contains reactive double bonds (unsaturation) along the entire polyester chain, which becomes the site for the eventual cross linking to produce the cured plastic (Fig. 2.3).0 0 0I

Isophthalic resinsThese resins are produced from isophthalic acid and are characterized by greater strength, heat resistance, toughness and flexibility than their ortho cousins. In isophthalic acid, the acid groups are separated by one carbon of the benzene ring which increases the opportunity to produce polymers with greater linearity and higher molecular weight in the esterification reaction (Fig. 2.4).

HO

It II II I II HO ( C-R-C-O-R'-O-C-C=C-C-O-R'-O H )nH

Fig. 2.3

36 Polyester and viny ester resins ester can be three times longer than for an is0 resin. As a result of this, researchers have turned to polyethylene terephthalate scrap from the previously mentioned fiber and plastic operations to develop an economical source of terephthalic polyesters. This scrap can be effectively depolymerized by using different amounts of propylene glycol at elevated temperatures. The glycolyzed product is then reacted with maleic anhydride and diluted with styrene monomer to produce a cost effective terephthalic polyester. Bisphenol A fumarate resins These resins are unsaturated rigid polyesters made by reacting bisphenol A with propylene oxide to produce the glycol shown in Fig. 2.5. This propoxylated bisphenol A is then reacted with fumaric acid to form an unsaturated polyester. The bisphenol structure illustrated above imparts a high degree of hardness, rigidity and thermal stability to this particular resin. Chlorendic resins These unique polyester resins are based on HET acid (hexachlorocyclopentadiene) or the anhydride shown in Fig. 2.6. When reacted with an unsaturated acid and a stable glycol such as neopentyl, an extremely rigid unsaturated polyester results with outstanding thermal stability and resistance to oxidizing environments. The inherent chlorine in the resin chain imparts some fire retardancy as well.

Phthalic anhydride

lsophthalic acid

0c=oH O

Terephthalic acid

Fig. 2.4

Therephthalic resins Unsaturated polyesters can be produced from terephthalic acid with the expectation that the resin property improvement obtained in going from phthalic anhydride to isophthalic acid will be matched in going from isophthalic to terephthalic acid. This, however, is not the case and terephthalic resins appear to offer only a slight advantage in heat distortion temperature over their isophthalic counterparts. Other important resin properties such as modulus, hardness and overall chemical resistance favor the is0 resins. Because of its lower solubility and poorer reactivity, therephthalic acid requires the use of esterification catalysts or pressure processing to produce a resin economically. Without these, processing time for a terephthalic poly-

H HO - C - CH2 - O I

o

l

@

H .. 0 - CH2 - C - OH I

CI C Q ;? !

CI-c-CI

-c=o

ICH3

ICH3

Fig. 2.5

Fig. 2.6

Hydrocarbon SolventsTable 2.72: Hexene-1 (4)cn2 = cn-cn2-cn2-cn2-cn,RESEARCH GRADE PURE GRADE TECHNICAL GRADE

37

FORMULA PROPERTIES

'Literature values.

Table 2.73: cis-Hexene-2 (4)

Table 2.74: Mixed 2-Hexenes (4)

FORMULA

cn3-c

- c-cn2-cn2-cn,RESEARCH GRADE

_ _ _ _ _ _ ~

FORMULA PROPERTIES

1trxe-

cH3-cH cn-cH2-cn2-cn3PURE GRADE TECHNICAL GRADE-

-

PROPERTIES Composition. weight percent~ ~

Hexene-1 ___-_-.___~_-tranr-Hexene-2 cis-Hexene.2 Hexe!Er-?p.Normal Hexane lsoolefins _ _ _ . _ Heptene-l __ tranr-Heptene-3 cis.Heptene.3 tranr-Heptene.2 cis.Heptene2~~ ~

_

0.1 0.2 99.6

~-

~

~

~

.

0.1

Composition, weight percent .. Hexenel tiant-Hexene-2 . cis Hexene-2 Hexener-3 Normal Hexane lsoolefins -_____ Heptene-l

.-.

03

__

i:it99Omin

0.8_

~- - - _

2 ~1 -

--

-

--__-

-

-~

tranr-Heptene-3 cis-Heptene3 tranr-Heptene 2 ca-Heptene2 Purity by freezing point. mol % Freezing point. F Boiling point, F Distillation range. F Initial boiling point Dry point Specific gravity of liquid at 60160 F at2014 c API gravity at 60 F Density of liquid a t 60 F. Ibdgal Vapor pressure a t 70 F. psis 100 F, psia 130 F, psia Refractive index, 2010 Color, Saybolt Acidity. distillation residue Nonvolatile matter, gramdl00 ml Flash point, approximate, F~

.

. .

..

-

-

Purity by freezing point, mol % Freezing point, F Boiling _ _ _ -~ -~ point, F Distillation range, F __ Initial boiling point Dry point 0m 7 Specific gravity of liquid at 66 at 2014 C API gravity at 60 F Density of liquid at 60 F. m a l Vapor pressure at 70 F. psia 100 F, psia -___ 130 F, psia _ _ Refractive index. 20/0 _ ._ Color. Saybolt Acidity, distillation residue Nonvolatile -__- matter. gramdl00 ml Flash point. appoximate. F *Literature values.

99.28 -222.04' 156.00'

.~~~

~~~

~-

.

~

____

__- .

0.6 _________ 920* 0.68720'5.760' 2.4' 4.9' 9.1' 1.39761. +30

155.0 __.___.__ 155.1 -. 0.684~

155.0 155.1 0.686~-

~~.~

- . . ~75.4 5.69 2.4 ____ 5.0 -~ .. . -~ ..~~~~

~ _ _ _ - . ~ -

74.8 5.71 2.4

-.. - ~-.

__

.~

5.09.2 1.396 +30 neutral -. 0.0005 -5~

.

1.396 +30 _ neutral -~ _. 0.0005 -5

!.2._.

-

- -~.

__~

. . ..

38 Polyester and viny ester resins In addition to tailoring the resin for specific applications by varying the building blocks, the properties of unsaturated polyesters can often be altered significantly by selection of the esterification process. This is particularly true with isophthalic/terephthalic polyesters which are slower reacting than phthalic anhydride. By using a two stage or modified two stage reaction with these aromatic diacids, the molecular structure of the resultant polyester can be changed to markedly improve heat distortion temperature, hydrolytic stability and chemical resistance2.In the two stage process the aromatic acid and glycol are fully or partially reacted before the faster reacting unsaturated acid is added to the cook. This processing method, compared to charging all ingredients at once (one stage method), also leads to a more random distribution of the unsaturation in the polymer chain which changes the character of the final cross linked network in the cured resin. Cure plays one of the most important roles in the chemical resistance developed by unsaturated polyester resins. Theoretically, the curing reaction should go to completion at room temperature with all the double bonds converted to single bonds in the three-dimensional network. However, complete cross linking is rarely achieved at ambient temperatures. This then will result in reduced chemical resistance and, quite often, poorer than expected mechanical properties. In addition, unreacted diluent (styrene ) can remain in the not-so-well cured polymer leading to major problems when the polyester is used for food grade applications. Accordingly, maximum chemical resistance and certain other property improvements can most often be achieved by utilizing elevated temperatures for post cure of the polyester resin finished product. Unsaturated polyester resins are used in the manufacture of a broad range of plastic products. A high percentage of these products utilize reinforcing materials, particularly fiberglass. It is estimated that less than 20% of the polyester resins produced are utilized in applications which do not involve reinforcing materials. These so-called casting applications include buttons, bowling balls, putties, cultured marble, gel coats and decorative products. The marble industry and the more recently developed polymer concrete industry represent outstanding applications for highly filled unsaturated polyesters which offer very economical materials to the building and construction industry. Fiberglass reinforced polyesters (FRP) are used in the manufacturing of boats, automobile and truck parts, building panels, corrosion resistant equipment such as pipes, tanks, ducts, scrubbers, etc., appliances and business equipment, electrical equipment, construction products such as grating and railing, sporting equipment and consumer products that are almost endless. According to the Composites Institute of the Society of Plastics Industry (SPI), automotive, construction, marine and corrosion resistant equipment are the four largest FRP markets, in that order, in the United States which produces 2.5 billion pounds of FRP annually. Mechanical properties are most often the critical factor in the selection of a polyester resin for a specific application. Testing of mechanical properties for both resin castings and fiberglass remforced composites is carried out using standardized ASTM (American Society for Testing and Materials) tests for all plastics. ASTM D-638 Standard Test Method for Tensile Properties of Plastics ASTM D-790 Standard Test Method for Flexural Properties of Plastics ASTM D-695 Standard Test Method for Compressive Properties of Rigid Plastics ASTM D-256 Standard Test Method for Impact Strength (IZOD) of Plastics ASTM D-648 Standard Test Method for Heat Distortion Temperature of Plastics ASTM D-2583 Standard Test Method for Barcol Hardness of Plastics

Polyester resins 39As mentioned earlier, glycol selection has a produce a rigid polyester which tends to be significant effect on the properties of poly- hard, brittle and lower in tensile elongation. esters. Ether glycols are of great value in Higher unsaturation also leads to higher heat increasing tensile elongation and impact distortion temperature resins. The latter is also strength which is of great importance in auto- achieved by formulating higher molecular motive, casting and liner applications. A weight resins with the chlorendic, bisphenol A principal deficiency of polyester resins is lack and dicyclopentadiene building blocks. As of alkali resistance because the ester linkages expected, all of these resin classes are more are subject to hydrolysis in the presence of brittle and have low tensile elongation. The caustics. Accordingly, increasing the size of the major exception in this scenario are the glycol has the same effect as reducing the con- iso/terepolyesters. Using the multi-stage procentration of attackable ester linkages. Thus, a cessing methods described earlier, these resins resin containing neopentyl glycol, propxylated can be formulated with reasonably high molebisphenol A, or trimethyl pentanediol will cular weights (more linearity) to give very exhibit improved water and chemical resis- tough resins having a good balance of tentance which is highly important in gel coats, sile/flexural properties plus higher tensile corrosion resistant equipment, construction elongation and heat distortion temperatures. products and many consumer products. Obviously then, when the end use criteria The major effect on polyester physical prop- requires the 'something more' than is offered erties is, however, provided by the by general purpose polyesters (orthophthalics unsaturation content in the polyester polymer. and dicyclopentadienes),the formulator turns Higher unsaturation makes for more cross to iso/terepolyesters which have no disadvanlinking and a stiffer cured composite. tages compared to general purpose resins Accordingly, the formulators' selection of other than slightly higher cost. unsaturated acid to saturated acid ratio which Table 2.1 summarizes the property and determines cross linking density can move the application status for the various classes of resin flexural modulus from rigid to resilient unsaturated polyesters. to very flexible. In most cases, a 1/ 1 ratio willTable 2.1 Properties and applications of unsaturated polyesters

ClassOrthophthalics, dicyclopentadiene

CharacteristicsRigid, resistant to crazing, light in color Tough, good impact and overall mechanical properties, resistant to environmental elements and moderate chemical attack. Highly resistant to aromatics Rigid, high heat distortion, highly resistant to oxidizing chemical environments Rigid, high heat distortion, highly resistant to most chemical environments particularly caustics

UsesBoats, tub/shower, spas, marble, consumer products, buttons, corrugated sheet, building panels, seating, decorative products Automotive parts, gel coats, electrical, bowling balls, trays, gasoline, tanks, septic tanks, swimming pools, tooling, aerospace products, corrosion, construction products Corrosion resistant tanks, ducting, stacks, industrial vessels Corrosion resistant tanks, piping, stacks, industrial vessels

Isophthalics/terephthalics

Chlorendic Bisphenol A fumarates

40 Polyester and uiny ester resins2.3 VINYL ESTER RESINS

Vinyl ester resins are the most recent addition to the family of thermosetting polymers. Although several types of these resins were synthesized in small quantities during the late 1950s, it was not until the mid-1960s that commercialization, principally by Shell and Dow 2.3.1 VINYL ESTER RESIN TYPES Chemical led the push to establish an extremely important segment of todays com- Aside from the fire retardant versions of vinyl posite industry. Vinyl esters are unsaturated ester resins which are discussed in the next resins made from the reaction of unsaturated section, there are two basic types of vinyl carboxylic acids (principally methacrylic acid) esters having commercial significance. These with an epoxy such as a bisphenol A epoxy are the general purpose lower molecular resin. The typical structure of a vinyl ester weight vinyl esters and the higher heat resistant vinyl esters with greater cross link resin is shown in Fig. 2.8. The structure of vinyl ester resins shows density. several important features which account for the resultant exceptional properties of vinyl General purpose vinyl esters ester resins. There is an epoxy resin backbone with a high molecular weight that provides These are principally methacrylated epoxies excellent mechanical properties combined made by the reaction of methacrylic acid with with toughness and resilience. Secondly, vinyl a bisphenol A epoxy resin. When dissolved in esters display terminal unsaturation which styrene monomer they provide a thermosetting makes them very reactive. They can be dis- resin with good heat resistance, excellent solved in styrene and cured like a mechanical properties (particularly high tenconventional unsaturated polyester to give sile elongation) and outstanding chemical rapid green strength. Obviously, the vinyl resistance to acids, bases, hypochlorites and ester structure also enables convenient many solvents. homopolymerization which could lead to high heat distortion products. Finally, vinyl esters Heat resistant vinyl esters have much fewer ester linkages per molecular weight which combined with the acid resistant These vinyl esters have higher density cross epoxy backbone, give outstanding chemical linking sites available which leads to a more resistance (acids, caustics and solvents) to this heat resistant polymer network. They are proclass of resins. duced from novolac modified epoxy resins

Although vinyl esters have often been classified as polyesters, they should be designated separately because they are typically diesters with a recurring ether linkage provided by the epoxy resin backbone.

OH H - C- CH2- 0 G I

T

O

OH O - CH2-CI - H

0 7H2I C-CH3II

CH3

7%0I

c=o

I

c = oI

C-CH3II

CH2

CH2

Fig. 2.8

Vinyl ester resins 41and methacrylic acid which provides more cal properties can be 'tailored' to meet the unsaturation sites and higher molecular requirements of specific applications. Another weight due to the epoxy backbone. These unique property of vinyl ester is the bondabilvinyl esters increase the heat resistance by ity of these resins to other surfaces. They are 17-27C (30-50F) over the general purpose not as good as epoxy resins in this charactertypes. This often translates to higher useful istic, but obviously the epoxy resin operating temperatures for vinyl ester based component gives them a boost over other reinforced plastics even in corrosive environ- unsaturated polyesters in this area. A case can ments. The higher-density cross linked vinyl also be made for vinyl esters providing better esters are less resilient (lower tensile elonga- fiberglass wet out in FRP composites due to tion) but still have excellent mechanic