- Author / Uploaded
- Bryan G. Crowther
Handbook of Rubber Bonding
Bryan Crowther (Editor) The (Revised Edition) Editor: Bryan Crowther rapra TECHNOLOGY Rapra Technology Limited Sh
3,298 1,159 3MB
Pages 400 Page size 524.409 x 708.661 pts Year 2008
Recommend Papers
File loading please wait...
Citation preview
Handbook of Rubber Bonding
Bryan Crowther (Editor)
The Handbook of Rubber Bonding (Revised Edition)
Editor: Bryan Crowther
rapra TECHNOLOGY
Rapra Technology Limited Shawbury, Shrewsbury, Shropshire, SY4 4NR, United Kingdom Telephone: +44 (0)1939 250383 Fax: +44 (0)1939 251118 http://www.rapra.net
First Published 2001 by
Rapra Technology Limited Shawbury, Shrewsbury, Shropshire, SY4 4NR, UK
©2001, Rapra Technology Limited Revised and Reprinted 2003
All rights reserved. Except as permitted under current legislation no part of this publication may be photocopied, reproduced or distributed in any form or by any means or stored in a database or retrieval system, without the prior permission from the copyright holder. A catalogue record for this book is available from the British Library. Cover photograph reproduced with permission from Rubber Chemistry and Technology, 1994, 67, 4582. Copyright 1994, Rubber Division, American Chemical Society, Inc.
ISBN: 1-85957-394-0
Typeset by Rapra Technology Limited Printed and bound by Rapra Technology Limited
Contents
Introduction .......................................................................................................... 1 1
Substrate Preparation Methods ....................................................................... 3 1.1
1.2
1.3
1.4
Metal Preparation - General Techniques ................................................ 3 1.1.1
Structure of Metal Substrates - Metallography .......................... 3
1.1.2
Bonding ..................................................................................... 5
1.1.3
Rubber Component with Metal Support ................................... 5
1.1.4
Metal Pre-treatments ................................................................. 6
Pre-treatments of Plastics and Rubbers ................................................ 12 1.2.1
Introduction ............................................................................. 12
1.2.2
Studies of Pre-treatments for Plastics ....................................... 13
1.2.3
Hydrocarbon Rubbers with Little or No Unsaturation ............ 19
1.2.4
Unsaturated Hydrocarbon Rubbers ......................................... 20
1.2.5
Halogenated Rubbers .............................................................. 25
1.2.6
Miscellaneous Rubbers ............................................................ 26
1.2.7
Discussion ................................................................................ 27
1.2.8
Summary ................................................................................. 29
Bonding Rubbers to Plastic Substrates ................................................. 29 1.3.1
Introduction ............................................................................. 29
1.3.2
Plastics Substrate Preparation .................................................. 31
1.3.3
Degreasing and Solvent Cleaning ............................................. 35
1.3.4
Adhesive/Bonding Agent Choice .............................................. 36
Substrate Preparation for Bonding Using the Wet Blast Process ........... 42 1.4.1
Summary ................................................................................. 42
1.4.2
The Wet Blast Phosphating Plant ............................................. 42
i
The Handbook of Rubber Bonding
2
3
1.4.3
Comparison Between Conventional and Wet Blast Phosphating .. 45
1.4.4
The Wet Blast Phosphating Plant ............................................. 46
1.4.5
Advantages of the Wet Blast Phosphating Plant ....................... 47
Rubber to Metal Bonding ............................................................................. 57 2.1
History................................................................................................. 57
2.2
Bond System Characteristics ................................................................ 62 2.2.1
Adhesive Characteristics .......................................................... 62
2.2.2
Compound Characteristics....................................................... 64
2.3
Adhesion .............................................................................................. 66
2.4
Effective Bond Formation .................................................................... 71
2.5
Post Vulcanisation Bonding ................................................................. 73
2.6
Factors Affecting Bond Integrity .......................................................... 73
2.7
Bond Failure Types .............................................................................. 74
2.8
Bond Test Procedures ........................................................................... 76
2.9
Summary .............................................................................................. 77
Rubber to Metal and Other Substrate Bonding ............................................. 81 3.1
3.2
Introduction ......................................................................................... 81 3.1.1
Foreword ................................................................................. 81
3.1.2
History .................................................................................... 81
3.1.3
Types of Bonding ..................................................................... 82
3.1.4
The Bonding Process - An Overview ........................................ 83
3.1.5
Development of Bonding ......................................................... 84
3.1.6
Bonding Agent Reliability ........................................................ 84
3.1.7
The Environment and Solvent Use ........................................... 86
3.1.8
Methods of Reduction in Solvent Emissions ............................ 87
Substrates and their Preparation .......................................................... 87 3.2.1
ii
Mechanical Treatment of Metals ............................................. 88
Contents
3.3
3.2.2
The Abrasion Process ............................................................... 90
3.2.3
Levels of metal cleanliness ....................................................... 92
3.2.4
Time Window .......................................................................... 93
3.2.5
Chemical Preparation of Surfaces ............................................ 94
3.2.6
Future Developments ............................................................... 96
Bonding Agent Preparation .................................................................. 97 3.3.1
3.4
3.5
3.6
3.7
3.8
Solvent-borne Bonding Systems ............................................... 97
Bonding Agent Application and Use .................................................... 98 3.4.1
Application Methods ............................................................... 98
3.4.2
Waterborne Bonding Systems ................................................... 98
3.4.3
Bonding Agent Thickness......................................................... 99
Post Vulcanisation Bonding ............................................................... 100 3.5.1
Post Vulcanisation Bonding Applications............................... 100
3.5.2
Choice of Bonding Agent for Post Vulcanisation Bonding ..... 100
3.5.3
Rubber Substrate Preparation for PV Bonding....................... 101
3.5.4
Metal Substrate Preparation .................................................. 101
3.5.5
Methods of Application ......................................................... 101
Waterborne Bonding Systems ............................................................. 103 3.6.1
History .................................................................................. 103
3.6.2
Differences Between Solvent and Waterborne Bonding Agents .. 103
3.6.3
Suggested Spraying Equipment and Conditions ..................... 105
3.6.4
Application and Substrate Temperatures ............................... 105
3.6.5
Film Thickness ....................................................................... 106
3.6.6
Layover .................................................................................. 106
3.6.7
Progress in Performance......................................................... 106
Health and Safety in the Workplace ................................................... 109 3.7.1
The Safety Data Sheet ............................................................ 109
3.7.2
Perspective ............................................................................. 110
Bonding Agent Testing ....................................................................... 110
iii
The Handbook of Rubber Bonding
3.9
Shelf Life Considerations ................................................................... 112 3.9.1
Shelf Life Categories .............................................................. 113
3.9.2
Procedures for Re-certification of Bonding Agents ................ 113
3.10 Troubleshooting ................................................................................. 115 3.11 Summary ............................................................................................ 120 4
Bonding Rubber to Metals with Waterborne Adhesive Systems .................. 125 4.1
4.2
4.3
5
4.1.1
Solvent Elimination by the Rubber Industry .......................... 126
4.1.2
Techniques Necessary in Bonding of Rubber to Meet Local Environmental Pollution Limits ................................... 127
Waterborne Bonding Systems ............................................................. 127 4.2.1
Structure of Organic Solvent-based Bonding Systems ............ 127
4.2.2
Structure of Waterborne Bonding Systems ............................. 127
4.2.3
Fundamentals of Waterborne Bonding Agent Application ..... 128
4.2.4
Waterborne Bonding Systems in Factory Usage ..................... 128
4.2.5
Metal Preparation - For Waterborne Bonding Systems .......... 129
4.2.6
Waterborne Bonding Agent Application ................................ 129
4.2.7
Waterborne Bonding Agent Storage Stability ......................... 130
4.2.8
Non Bond Advantages of Waterborne Bonding Systems ........ 130
4.2.9
General Comments - Waterborne Bonding Agents ................. 130
Waterborne Bonding Agents - A Factory Experience ......................... 131 4.3.1
Thickness Effects ................................................................... 131
4.3.2
Pre-bake Resistance ............................................................... 133
4.3.3
Primers .................................................................................. 134
4.3.4
Polymer Range ....................................................................... 134
4.3.5
Product Range ....................................................................... 134
4.3.6
Current Disadvantages of Waterborne Bonding Agents ......... 134
Rubber to Rubber Bonding ......................................................................... 137 5.1
iv
Introduction ....................................................................................... 125
Bonding of Unvulcanised Rubbers ..................................................... 137
Contents
5.1.1
Tack/Autohesion .................................................................... 137
5.1.2
Influence of Vulcanisation System .......................................... 139
5.1.3
Influence of Filler Type .......................................................... 140
5.1.4
Effects of Plasticisers/Process Oils .......................................... 141
5.1.5
Effects of Tackifiers ............................................................... 141
5.1.6
Effects of Other Ingredients ................................................... 142
5.1.7
Effects of Surface Modification .............................................. 142
5.1.8
Effects of Surface Roughness ................................................. 144
5.1.9
Influence of Contact Time/Pressure/Temperature ................... 144
5.1.10 Effects of Blooming ................................................................ 145 5.1.11 Effects of Ageing .................................................................... 146 5.1.12 Testing of Tack/Autohesion Levels ......................................... 147 5.1.13 Adhesion Theories ................................................................. 148 5.2
Bonding of Vulcanised Rubbers to Unvulcanised Rubbers ................. 150
5.3
Bonding of Vulcanised Rubbers ......................................................... 152 5.3.1
Strip Bonding of Tyre Retreading Components ...................... 152
5.3.2
Effects of Strip Thickness ....................................................... 155
5.3.3
Effects of Surface Roughness ................................................. 156
5.3.4
Effects of Temperature on Bonding ........................................ 156
5.3.5
Effects of the Chemical Nature of Polymers/ Polymeric Additives/Surface Roughness ................................. 156
5.3.6
Urethane Adhesive Systems .................................................... 158
5.3.7
Surface Treatments to Improve Bonding ................................ 158
5.3.8
Effects of Contact Time/Surface Bloom .................................. 159
5.4. The Mechanism of Adhesion of Fully Cured Rubbers........................ 159 6
Rubber-Brass Bonding ................................................................................. 163 6.1
Introduction ....................................................................................... 163
6.2
Mechanism of Rubber-Brass Bonding ................................................ 165 6.2.1
Reviews ................................................................................. 165
6.2.2
Recent Mechanistic Studies .................................................... 165
v
The Handbook of Rubber Bonding
6.2.3
Updated Rubber-Brass Adhesion Model ................................ 170
6.2.4
New Evidence for Ageing of the Interfacial Sulphide Film ..... 177
6.2.5
Compounding for Brass Adhesion ......................................... 180
6.2.6
Additives to Compounds for Brass Adhesion ......................... 181
6.2.7
Developments in Metal Pre-treatments .................................. 184
6.2.8
Developments of Novel Alloys for Bonding to Rubber .......... 189
6.2.9
Miscellaneous ........................................................................ 190
6.2.10 Summary ............................................................................... 190 7
8
Review of Tyre Cord Adhesion ................................................................... 197 7.1
Introduction ....................................................................................... 197
7.2
Accepted Mechanisms of Rubber-Brass Bonding ............................... 198
7.3
Ageing of the Rubber-Brass Bond ...................................................... 200
7.4
Metal Organic Cobalt Salts ................................................................ 201
7.5
The Role of Resins and Silica/Resin Systems ...................................... 205
7.6
Summary ............................................................................................ 208
Rubber to Metal Bonding Using Metallic Coagents .................................... 213 8.1
Introduction ....................................................................................... 214
8.2
Metallic Coagents .............................................................................. 215
8.3
8.2.1
Scorch Safety ......................................................................... 217
8.2.2
Tensile Properties ................................................................... 219
8.2.3
Tear Strength ......................................................................... 220
Experimental ..................................................................................... 221 8.3.1
8.4
8.5
vi
Materials ............................................................................... 221
Results and Discussion ....................................................................... 229 8.4.1
Adhesion to Metals ................................................................ 229
8.4.2
Adhesion to Fibres and Fabrics .............................................. 235
Summary ............................................................................................ 238
Contents
9
Rubber to Fabric Bonding ........................................................................... 241 9.1
Introduction ....................................................................................... 241
9.2
Adhesive Systems ............................................................................... 241
9.3
9.4
9.5
9.2.1
Aqueous Fabric Treatments ................................................... 241
9.2.2
Solvent-Based Adhesive Systems ............................................ 248
9.2.3
In Situ Bonding Systems......................................................... 249
Mechanisms of Adhesion ................................................................... 250 9.3.1
Dip/rubber Interface .............................................................. 250
9.3.2
Dip/textile Interface ............................................................... 252
Other Factors Affecting Adhesion ...................................................... 253 9.4.1
Storage of Treated Textiles ..................................................... 253
9.4.2
Adhesion in Service ................................................................ 254
Environmental Aspects ...................................................................... 254 9.5.1
Storage and Handling ............................................................ 254
9.5.2
In Process ............................................................................... 255
9.5.3
Wastes and Disposal .............................................................. 255
10 Bonding Rubber with Cyanoacrylates ......................................................... 259 10.1 Introduction ....................................................................................... 259 10.2 Liquid Cyanoacrylates ....................................................................... 259 10.3 Curing of Cyanoacrylates .................................................................. 260 10.3.1 Factors Affecting Cure ........................................................... 261 10.3.2 Cure Speed ............................................................................. 263 10.4 Types of Cyanoacrylate ...................................................................... 263 10.4.1 Bonding to Acidic and Porous Substrates............................... 264 10.4.2 Toughened Cyanoacrylates .................................................... 265 10.4.3 Flexible Cyanoacrylates ......................................................... 266 10.4.4 UV Curing Systems ................................................................ 266 10.5 Design Considerations ....................................................................... 266
vii
The Handbook of Rubber Bonding
10.5.1
Minimise Peel and Deavage Loads .................................... 267
10.5.2
Bond Line Thickness ......................................................... 268
10.5.3
Special Requirements for Bonding with Cyanoacrylates .... 269
10.5.4
Internal and External Mould Release Agents .................... 269
10.5.5
Successful Joint Design ...................................................... 269
10.6
Bonding to Silicone Rubber ............................................................. 270
10.7
Environmental Resistance ............................................................... 270 10.7.1
Glass Bonding ................................................................... 272
10.7.2
Hot Strength ..................................................................... 272
10.8
Activators ........................................................................................ 274
10.9
Application Methods for Cyanoacrylates ........................................ 275 10.9.1
Pressure/Time Systems ....................................................... 275
10.9.2
Syringe Systems ................................................................. 276
10.10 Health and Safety and Handling Precautions .................................. 276 10.11 Typical Applications........................................................................ 277 10.11.1 Bonding Nitrile, Polychloroprene and Natural Rubbers .... 277 10.11.2 Bonding EPDM ................................................................. 277 10.11.3 Bonding Santoprene and Silicone Rubbers ........................ 279 10.11.4 Bonding Medical Devices .................................................. 279 10.12 Troubleshooting .............................................................................. 280 10.12.1 Blooming of Cyanoacrylates ............................................. 280 11 Bonding Silicone Rubber to Various Substrate ............................................ 285
viii
11.1
Introduction .................................................................................... 285
11.2
Why Bond Silicone Rubber? ............................................................ 286
11.3
Material Combinations of Interest - Examples ................................ 287 11.3.1
Silicone to Silicone Bonding (Soft and Soft) ...................... 287
11.3.2
Silicone to Plastic Bonding (Soft and Hard) ...................... 288
11.3.3
Silicone to Metal Bonding (Soft and Hard) ....................... 288
11.3.4
Why Use Silicone Rubber for Such Composites? ............... 288
Contents
11.4
Some Applications of Silicone Rubber Composites ......................... 290
11.5
Bonding Concepts ........................................................................... 291
11.6
11.7
11.8
11.9
11.5.1
Undercuts .......................................................................... 291
11.5.2
Primers .............................................................................. 292
11.5.3
Self-adhesive Silicone Rubbers .......................................... 292
11.5.4
The Build-up of Adhesion ................................................. 292
Bonding of Liquid Rubber (LR) ...................................................... 293 11.6.1
Properties of Self-adhesive LR ........................................... 297
11.6.2
Limitations of Self-adhesive LR ......................................... 298
Bonding of Solid Rubber (HTV) ..................................................... 299 11.7.1
Self-adhesive HTV Silicone Rubber Applications .............. 299
11.7.2
Applications for Self-adhesive HTV .................................. 301
11.7.3
HTV Used in Other Bonding Applications ........................ 303
Processing Techniques ..................................................................... 303 11.8.1
Liquid Rubbers in Inserted Parts Technology .................... 303
11.8.2
LR in Two-component Injection Moulding Technology (Two Colour Mould) ......................................................... 306
Silicone to Silicone Bonding (Soft and Soft) .................................... 308
11.10 Cable Industry ................................................................................ 309 11.11 Duration of Bonding Properties ...................................................... 309 11.11.1 Duration of Bonding - Chemically Bonded Composites .... 311 11.12 Alternatives to Injection Moulding ................................................. 313 11.12.1 Adhesives .......................................................................... 313 11.12.2 Welding ............................................................................. 313 11.12.3 Mechanical Bonding Techniques After Moulding .............. 314 11.13 Summary ......................................................................................... 314 12 Failures in Rubber Bonding to Substrates ................................................... 319 12.1.1
Introduction ...................................................................... 319
12.1.2
Incorrect Moulding Procedures ......................................... 328
ix
The Handbook of Rubber Bonding
12.1.3
Incorrect Production Quality Testing Procedures .............. 329
12.1.4
Corrosion in Service .......................................................... 330
12.1.5
Product Abuse ................................................................... 333
12.1.6
Other Failure Modes ......................................................... 333
12.1.7
Factors Affecting Adhesion of Rubbers ............................. 334
12.1.8
Topography of Substrate ................................................... 335
12.1.9
Surface Conditions of Adherend ....................................... 335
12.1.10 Classification of Rubber According to their Wettabilities .. 336 12.1.11 Bonding - Interphase or Interface Considerations ............. 337 12.1.12 Problems in Adhesion........................................................ 339 12.2
12.3
Rubber Bonding in Power Transmission Belting ............................. 339 12.2.1
Introduction ...................................................................... 339
12.2.2
Power Transmission Belt Failure Modes ............................ 340
12.2.3
Adhesion Systems in Power Transmission Belts ................. 346
12.2.4
Adhesion Testing in Power Transmission Belts .................. 347
Undesirable Adhesion Occuring Under Service Conditions (Fixing) .. 349 12.3.1
Factors Affecting ‘Fixing’ .................................................. 349
12.3.2
Prevention of ‘Fixing’ ........................................................ 351
12.3.3
Other Methods of Preventing ‘Fixing’ Examined Experimentally ................................................. 351
Abbreviations and Acronyms............................................................................. 357 Author Index ..................................................................................................... 363 Company Index ................................................................................................. 371 Main Index ........................................................................................................ 373
x
Contributors Derek Brewis Loughborough University, Institute of Surface Science and Technology, Department of Physics, Loughborough, Leicestershire, LE11 3TU, UK. Richard Costin The Sartomer Company, 502 Thomas Jones Way, Exton, PA 19341, USA. Bryan Crowther 49 The Avenue, Bengeo, Hertford, Hertfordshire, SG14 3DS, UK. Kenneth Dalgarno School of Mechanical Engineering, University of Leeds, Leeds, LS2 9JT, UK. Steve Fulton OMG Limited, Ashton New Road, Clayton, Manchester, M11 4AT, UK. Robert Goss Henkel Loctite Adhesives Limited, Watchmead, Welwyn Garden City, Hertfordshire, AL7 1JB, UK. Jim Halladay Lord Corporation, Chemical Products Division, 2000 West Grandview Boulevard, PO Box 10038, Erie, PA 16514-0038, USA. Richard Holcroft 5 Brooklands Drive, Birmingham, West Midlands, B14 6EJ, UK. Peter Jerschow Wacker-Chemie GmbH, Johannes-Hess Strasse 24, D-84489 Burghausen, Germany. Rani Joseph Department of Polymer Science and Rubber Technology, Cochin University of Science and Technology, Cochin 682022, Kerala, India. Mike Rooke Henkel Loctite Adhesives Limited, Watchmead, Welwyn Garden City, Hertfordshire, AL7 1JB, UK.
Commercial rubbers
The Handbook of Rubber Bonding
Berndt Stadelmann Wacker-Chemie GmbH, Johannes-Hess Strasse 24, D-84489 Burghausen, Germany. Walter Strassberger Wacker-Chemie GmbH, Johannes-Hess Strasse 24, D-84489 Burghausen, Germany. Wim van Ooij Department of Chemical and Materials Engineering, University of Cincinnati, Cincinnati, OH 45221-0012, USA. Patrick Warren Lord Corporation, Chemical Products Division, 2000 West Grandview Boulevard, PO Box 10038, Erie, PA 16514-0038, USA. Ron Woodcock 5 Lower Leicester Road, Lutterworth, Leicester, LE17 4NF, UK. David Wootton 95 Greenhill Road, Bury, Lancashire, BL8 2LL, UK Keith Worthington Chemical Innovations Limited (CIL), 217 Walton Summit Centre, Bamber Bridge, Preston, PR5 8AL, UK.
Introduction
Although many volumes of information have been published about the subject of adhesion of materials in general, it is some forty years since a publication has been devoted solely to the subject of the bonding of rubbers to various substrates. Three very successful Rapra Technology, conferences on the subject of the bonding of rubber have shown that there is clearly a need for such a publication to be devoted to this topic of wide industrial significance. Although from time to time manufacturers of bonding agent systems publish papers in trade journals there is generally a dearth of available information for the factory practitioner to consult. The subject matter for this present volume has been selected to cover a wide range of interests, both in terms of products and applications. Rubbers in many applications need the support of, or reinforcement by, a variety of materials ranging from fibres to metals. To ensure optimisation of the properties from these composites it is necessary to ensure that the optimum adhesion levels are achieved, both initially and to be maintained throughout the service life of the products. Rubbers are bonded to a variety of substrates in many products, in numerous applications, to meet the needs of the modern world. The Rubber Bonding Handbook draws together the expertise of a number of world authorities engaged in improving the bonded product to meet the ever increasing demands placed on composites and components manufactured from rubbers bonded to metals, fabrics, fibres and plastic substrates. The papers included in this volume have been written by experts in their fields, many of whom have world-renowned reputations. Thus the information they include in their chapters can be considered to be the most up-to-date, state-of-the-art discussions of their respective areas of research and knowledge. The topics range from in depth discussions of such fundamental topics as the mechanisms of bonding of rubbers to brass, bonding techniques for adhesion to fabrics through to methods of preparation of substrates and the development of bonding agent systems for adhesion to metals and plastic substrates. Bonding with silicone rubbers and cyanoacrylate adhesives for post vulcanisation bonding are also included. A section dealing with information related to adhesion, failure and other adhesion related topics such as ‘fixing’ and practical reasons for a variety of bond failures, either during production or service are also covered.
1
The Handbook of Rubber Bonding Although there is some discussion of relevant theory in various sections of text, the emphasis in this volume has been to concentrate on the practicalities of bonding of rubbers, to themselves and substrates. It is considered that this type of information is of immediate interest to the practising technologist dealing with shop floor problems on a daily basis. It is hoped that the publication of definitive papers on the subject of adhesion of rubbers will be of considerable value to the practitioner in factories engaged in the previously seldom discussed variety of bonding applications being carried out by the rubber industry. Because of the legislation now in progress of being implemented by the rubber industry to eliminate sources of environmentally hazardous chemicals, there is information on the development and applications of waterborne bonding systems.
Acknowledgements I would like to express my appreciation of the help and assistance given to me in the editing of this publication. To Claire Griffiths (Editorial Assistant), Sandra Hall for typesetting, to Steve Barnfield for the cover design, Rebecca Dolbey for editorial advice and particularly to Frances Powers (Commissioning Editor), for her support, patience and guidance on general editorial matters. Bryan Crowther November 2000
2
1
Substrate Preparation Methods B. Crowther (Section 1.1) D. Brewis (Section 1.2) K. Worthington (Section 1.3) R. Holcroft (Section 1.4)
1.1 Metal Preparation – General Techniques
1.1.1 Structure of Metal Substrates – Metallography There is little written about the subject of metallography with respect to the bonding characteristics of the various metals used within the hot bonding process carried out by the general goods, rubber to metal bonding profession. Some work has been carried out in the field of adhesives for aeronautical applications [1]. In general only a few of the metals or adhesives described for this type of bonding have much application in the rubber to metal bonding factory, except perhaps if one is post vulcanisation bonding. The lack of fundamental metallography studies in the hot bonding of rubbers to metals is mainly due, no doubt, to the lack of influence which the bonding technologist has in these matters. He is usually told the grade of metal to be used and proceeds to find the best way, according to current factory processes, equipment, practices and experience, to deal with the problem. He can of course discuss the nature of his problem with his bonding agent supplier, who can in turn consult his research department if the problem is really abstruse. Perhaps a better understanding of metallography would enable the factory technologist to choose the best way to pre-treat his customer-dictated metal for his factory processes, or to discuss his customer’s ‘real’ metal requirements. To understand some of the problems associated with the achievement of good rubber to metal bonds it is worth considering some of the scenarios involving the atomic structure of metals at their surfaces. A metal, or an alloy of metals, naturally assumes a crystalline structure and it is likely that it will have a regular shape and lattice structure, with some voids in the interstices. As with rubber compounds, metals are formed by mixing a number of components together which disperse relative to each other, but never, except maybe in the case of pure metals, become one totally uniform uninterrupted phase. Most metals are used as some type of alloy, i.e.,
3
Commercial rubbers
The Handbook of Rubber Bonding steel consists of iron mixed with carbon in varying proportions to produce the different grades of commercial product. Also minor proportions of other metals are added to give different processing and end use characteristics to the steel, e.g., chromium, manganese, molybdenum, nickel, tungsten. The finishing processes of steels can also seriously alter the ability of adhesive to bond to them, due to the altered surface microstructure. With a pure metal its strength will depend on the size of the crystals making up its structure. In general small crystals make strong metals, whilst metals with large crystals, such as zinc, are weaker. The strength of a metal is also affected by the amount of impurity which may be present, as the impurities tend to arrange themselves at the interfaces between the crystals, thus preventing perfect crystal contact. In most metal alloys, as with rubber alloys or blends, the individual metals remain in discrete, but dispersed domains within the metal alloy structure. In an alloy the metal crystals involved, during cooling, have different shrinkage values and thus tend to move apart, allowing either voids to occur or when chemically hardened, other metals to infiltrate into these voids or interstices, at or near the surface. The individual crystals of the metals during cooling and shrinkage can join together to form chain structures, giving interlocking of the various metallic crystals. In some metal mixtures there is a mutual solubility and in these cases all crystals of the metal are the same. Although as a rubber to metal bonder one is not very interested in the metals structure within the mass of the metal, one must consider what is happening in and on its surface layers. Most metals form oxide layers on their surfaces, some of which, like iron are porous and thus continual oxygen ingress enables the oxide layer continually to increase whilst in aerobic conditions. Other metals such as aluminium form a dense oxide film which does not permit oxygen ingress and thus protects the metal underneath from further oxidation. Both metals types are being oxidised, albeit at different rates and this oxidation can be termed as a form of corrosion. Although the rubber to metal bonder must take the precautions necessary to prevent this type of corrosion continuing under his processing conditions, once the bonding agent has been applied, the condition at the metal interface becomes anaerobic and thus further oxidative corrosion is prevented (see Sections 12.1.2.2 and 12.1.5). There are a great variety and complexity of steel microstructures available to the component specifier, which complicate any cleaning procedure carried out prior to bonding. Incorrect chemical cleaning of low carbon and stainless steels, for example, can result in iron oxide ‘smutting’ of the surface leaving a deposit difficult to remove entirely [1] during metal cleaning. These deposits may subsequently give an extremely weak bonding surface and, as a result a bonded product which fails easily under low working stresses in service.
4
Substrate Preparation Methods However, as far the rubber to metal bonder is concerned he must avoid situations which can cause galvanic corrosion, a far more serious condition, which can propagate under the bonding system to cause eventual degradation of the bond and inevitable failure. Galvanic corrosion [2] is caused by the formation of an electrolytic cell between the different metal crystals within a structure in the presence of such agents as acids and salt water. Acids can be generated from degeneration of compounding materials or cleaning and degreasing fluids (see also Section 12.1.4). Certain metals are manufactured for their ability to prevent corrosion, e.g., stainless steel, and they contain chromium to enhance their corrosion resistance. If the level of chromium used in the metal’s makeup is high, then the very tough layer of chromium oxide, which forms on the metal’s surface as the anti-corrosion layer, is also exceedingly difficult to bond to rubbers. Wherever possible therefore stainless steels to be used in conjunction with bonding systems for rubbers should contain as low a chromium level as possible.
1.1.2 Bonding The bonding mechanisms of the multiphase systems involved in making a rubber to metal component are complex and the chemistry of the reactions involved not totally disclosed or understood. In the region of the metal contact the interactions are deemed to be a combination of mechanical and chemisorption processes. From the patent literature and some of the more recent reviews of rubber to metal bonding [3, 4], it can be seen that the primers contain a variety of halogenated rubbers and resins, which are known to have a high ability to wet out metal surfaces, thus ensuring the greatest degree of interface contact. In addition these rubbers and resins act as barriers to the migration of external corrosion catalysts of the metal surface. The resins and rubbers probably form an interpenetrating network of polymer chains within the adhesive system, thus giving strength and structure to the primer and rubber bonding coats. Bond quality depends to a large extent on the ability of all interfaces to freely exchange chemical entities. Any contamination of surfaces will upset the surface chemistry at that point and will reduce the bond strength.
1.1.3 Rubber Component with Metal Support Engineering products for a wide range of applications are made by the use of rubbers bonded to metals during the vulcanisation of the rubber. The quality of bond achieved during the manufacture of this type of component must be of sufficient integrity, not only to be stronger than the rubber itself, but also to outlast the active life of the rubber constituent of the components. To this end, the design of the component and
5
The Handbook of Rubber Bonding metal part must be carefully considered to ensure that no undue stress concentrations are created in the area of the bond between the rubber and the metal. Components consisting of moulded rubber bonded to metal, carried out during high temperature vulcanisation, can have inherent stresses simply due to shrinkage of the rubber when cooling from the vulcanisation temperature and the coefficient of thermal expansion relationship of the rubber/metal combination. The ‘shrinkage’ of the rubber in the system will be different for each type of rubber being used and is dependent also upon the compound hardness, or degree of filler present. Allowances for the rubber shrinkage must be made in determining the shape of the mould cavity and hence the component’s final shape. The environment in which the component is to work will also affect the stresses to which the rubber-metal bond will be subjected. Some oil and solvent environments will penetrate a bond at the interface and thus may weaken or destroy the integrity of the bond until the stress becomes relieved by failure. Corrosion of the metal component of the bonded unit by salt environments can also be a major problem and thus due concern and allowance must be made for the service conditions in which the rubber to metal component will be resident. Corrosion of the bonded metal under the bonding system can also occur if the metal pre-preparation is carried out with acidic degreasing fluids. Care must be taken that degreasing fluids are and remain, neutral in pH throughout their use in the application. Recovery of used solvents and redistillation can significantly change the pH of a solvent. This can be a particular problem with chlorinated solvents, where after redistillation the distillate can be acidic in nature. To effect good long-lasting bonds between rubber and metals it is essential that both materials presented to the interface be clean and free from detritus. The presence of oils and the possibility that compounding ingredients can exude or bloom from the rubber surface, before or after moulding, or during the service life of the component must also be taken into consideration and remedied.
1.1.4 Metal Pre-treatments Metals must be suitably pre-treated for satisfactory bonds to be achieved with rubbers. Two basic methods of preparation are used: • Mechanical, • Chemical.
6
Substrate Preparation Methods
1.1.4.1 Mechanical Methods Metals, especially the more common iron and steel types, come from the foundry and metal plate stamping shop, coated with oil, grease and most often with a generous layer of oxide and rolling mill scale formed on the exposed surfaces. Oxide films can also develop further during storage prior to use by the bonding shop. All these materials must be removed from the surfaces and from the voids in the metal, to ensure that the oils and greases which otherwise may be trapped unseen cannot exude under the increased temperature of vulcanisation, when they become more mobile or volatile. Surface oxides must be removed for they are often only loosely structured in their attachment at the metal substrate and will rupture and detach themselves under duress, causing the metal/ adhesive bond to fail. Once the original oxide layer has been removed, the freshly exposed metal will immediately start to build a new oxide film which must be minimised by rapid degreasing and application of primer/adhesive coat.
• Initial degreasing Metals must be degreased as the first step in any metal preparation process, otherwise oil and grease contamination of blasting media, chemical treatments and machinery can result in severe factory quality problems and unreliable and variable bonding. Traditionally the most usual method of grease and oil removal from the metal surface has been by degreasing in the vapour of a chlorinated solvent such as trichloroethylene or 1,1,1-trichloroethane or perchloroethylene. The chlorinated solvent used must have a neutral pH, otherwise the acidic condition can cause the initiation of underbond corrosion. Re-distilled chlorinated solvents, especially if recovery is carried out in-house, must be adequately checked for neutrality. The metal parts must dwell in the solvent vapour until such time as the metal reaches the temperature of the vapour and condensation has ceased. The solvent will have had the best opportunity to work at its most efficient in grease removal under these conditions. Direct contact with the degreasing solvent is not an efficient way of removing greases from metal surfaces, always leaving a molecular layer at least, still lying on the ‘cleaned’ surface. This cleaning method should not be used for metals to be used in bonding. All air lines in the bonding shop must have oil/water filters connected to them to remove the possibility of oil/water emulsion being sprayed onto the metal surfaces before, after or during bonding agent application. Air compressors are notorious for allowing oil seepage into the pressure vessel, together with an amount of water, which then usually causes an oil/water emulsion to be formed. This emulsion in contact with cleaned metal surfaces will give corrosion or reduce bond formation to a minimum level through the deposit of a film of oil.
7
The Handbook of Rubber Bonding The current legislation trend and environmental pressure for the industry is to move towards the use of alternative means of removal of contaminants from the metal surfaces (see Section 1.4). Equipment is available which uses water and detergents to remove these oils and thus present a more environmentally favourable working atmosphere. The action of the detergent can be supplemented by the use of ultrasonic agitation to remove oxide flakes. These systems being water-based require efficient drying of the metals, especially in the areas between contacting metals, otherwise further oxidation of the cleaned metal will rapidly take place. Careful choice of the detergent is also necessary otherwise its residues can detract from the bond strength achievable. The water quality being used in the degreasing system final wash process will have to be determined to prevent deposit of any salts or metallic ions. The ideal final wash is with de-ionised water. Alternative solvents, if used in a vapour degreasing system must have a similar evaporation rate to that of the presently used chlorinated solvents. Otherwise too rapid evaporation of the condensed solvent on withdrawal of the metals from the solvent vapour will result in rapid surface cooling of the metal, with resultant condensation of water, especially in conditions of high humidity.
• Alkaline removal of oils and greases An alternative method of removal of the metal preparation oils and greases is to use an alkaline cleaning method. The alkaline solution is used either in dip tanks or tumbler spray units (see Section 4.1). The strength of alkaline, the temperature used and the necessary dwell time in the solution to remove the amount of grease encountered will be determined in individual factories. The length of time required for oil and grease removal can be anything up to two hours. The alkaline tanks have to be followed by water rinse tanks to remove the alkaline dip from the metals, followed by drying.
• Solvent dip methods for large scale removal of greases Solvent dip methods are generally expensive to run and do not usually, unless a number of dip tanks are used, completely remove oils and greases from the metal surfaces. Contaminants are easily carried from tank to tank and it is difficult to ascertain whether the metal surface is completely cleaned after its passage through the tank series. This method would not normally be used for anything other than small scale operations. Fast drying solvents such as methylene chloride and acetone evaporate so quickly that they lower the temperature of the metal surface and water condenses.
8
Substrate Preparation Methods
• Removal of surface oxides Metals, after degreasing, have to be blasted with a sufficiently abrasive material to remove the surface oxidation layer. The usual medium used for ferrous substrates is steel or chilled-iron grit to BS EN ISO 11124-4 grades G12 to G24 [5] (see also Section 4.2.2). Alumina or other non-ferrous grits such as quartz sand and carborundum may be used on ferrous metals, but their use on non-ferrous metals is essential to prevent the possible formation of galvanic cells. Initially impingement of the metal surface with abrasive grit has the effect of gouging the surface of the steel to give a larger surface area for bonding, but with use the grit wears and its efficiency decreases. The type of grit used must be coupled to the type of metal being treated. Incorrect grit/ metal combinations can lead to formation of galvanic cells remaining on the surface of the blasted metals and the commencement of underbond corrosion. Grits larger than about 30-50 mesh diameter soon lose their irregularities and grittiness, effectively turning into shot at which stage they must be discarded. The hardness of the steel grits should be a Rockwell C hardness of 60 – 65. Iron or steel shot should not be used as these tend to give cavitation of the blasted metal surfaces, followed by peaning over of the sharp metal pinnacles, often trapping loose shot, blasted material, etc., in the peaned over cavities. These cavitations and their contents cause weaknesses and possible underbond corrosion sites, resulting in ultimate failure in service. The service life of the blasting media should be established for efficiency and quality of surface finish. Grit in use should be cleaned of dust resulting from removed oxide scale and its own degradation products and be downgraded or discarded if it becomes too worn. Revolving drum blast machines give the best production efficiency for metals which are stout enough to resist damage from the tumbling action involved. The metal parts are tumbled on a rubber belt inside a revolving drum whilst being bombarded with the abrasive medium. Once the metal surface has been adequately cleaned of oxide contamination, dusted off and once more degreased, it is vital that the application of a bonding agent primer coat be carried out as quickly as possible to ensure that the re-oxidation of the metal surface is kept to a minimum. Ambient temperature, humidity and dust must all be controlled if the optimum bond strength is to be achieved. To consistently ensure optimum bond quality, metal components, whether unprimed or primed, should be kept in enclosed cabinets. At no time should cleaned and degreased metals be handled with bare hands. Human skin, however clean it may appear, always carries a surface layer of oils and fats, which are bond killers. Neither should metals, whether in the ‘just cleaned’ state, or
9
The Handbook of Rubber Bonding treated with bonding agent, be handled with ‘press gloves’. Press gloves are usually heavily contaminated with a variety of materials, from oil, to mould release agents and sweat. Clean, frequently discarded cotton gloves are the best protection for handling metals. They should not be allowed to become dirty and sweat ridden.
1.1.4.2 Chemical Methods The alternative metal pre-treatment processes to grit blasting use a variety of different chemical routes. It is sufficient to say here that these can be very efficient, but do occupy rather large factory floor areas and can, if not controlled correctly give variable quality of prepared surface. The usual chemical pre-treatment systems consist of acid etching of the surface followed by several water dips and subsequent phosphate or in some circumstances cadmium plating and passivating (render inert). Many of these treatments will have been carried out by the metal processor and are not the rubber bonder’s processes.
• Treatments for stainless steels (see also Section 3.3) There are various systems for the pre-treatment of stainless steels which consist of treating the metal surface with strong acids to attack crystal grain boundaries in the alloys and chromium poor regions around chromium carbide particles. All the methods give surface roughness to the stainless steel which enhances the bond to the adhesive. Mixtures of nitric, hydrofluoric, sulphuric or chromic acid are suggested as most suitable. However, the nature of the substrate alloy and the heat treatment experienced all have a bearing on the bondability of the metal.
• Phosphate coating (see also Sections 1.2, 3.3 and 4.2.5) Steel is often phosphate coated for use within the engineering and decorative laminate industries to reduce corrosion. Iron or zinc phosphate can be used. However, although used for some years as a corrosion protection technique for rubber to steel bonding, it can be difficult to control the process, with a resultant variable thickness of phosphate deposit of varying crystalline structure. If too thick a phosphate layer is obtained it becomes too friable and lacking in the cohesive integrity required to maintain a rubber to metal bond under load during service. If only a moderate phosphate coat is produced it is often necessary to ‘passivate’ the areas of steel, only minimally covered or lacking in a coating of phosphate, by treating with chromic acid to form chromium oxide to prevent corrosion. However, chromium oxide does not readily react with a bonding agent (see Section 3.1). Chromic acid is a restricted material and alternative materials can be recommended by bonding agent suppliers for the passivation or sealing of the phosphate coating.
10
Substrate Preparation Methods The nature of the phosphate deposited on the surface of the steel depends to a large extent upon the nature of the microstructure of the steels and the orientation of its underlying crystal lattice. Hardened steels having a martensite structured surface (consisting of interlacing rectilinear fibrous elements arranged in a triangular shape) support a fine flake phosphate structure. Cold-rolled steel can, having acquired a different surface orientation structure during the rolling process, acquire a lumpy large flake phosphate structure, which is easily broken apart under service stress. Any water going to drain from these processes is a potential pollution hazard and must be tested for zinc content, as this is a hazardous material. Any zinc present must be removed or limited to 1 – 2 parts per million.
• Zinc coating or galvanising To be effective the zinc coating must be hot dipped onto the freshly cleaned metal, to give a ‘galvanised’ finish. Bonding to this finish is not easy, but sometimes demanded by the component specifier. The crystalline structure of the galvanised zinc and its dipped coating thickness, can result in the flaking off, under stress, of some of the coating, resulting in bond failure (see also Section 1.1.1). The recommended treatment [6] for cleaning a galvanised finish is a) degrease metal part b) abrade the galvanised surface with grit c) degrease then apply adhesive as soon as possible or a) immerse in a solution of 20 parts by weight concentrated hydrochloric acid with 80 parts by weight de-ionised water, for 2 – 4 minutes at 25 °C b) rinse thoroughly in cold, running de-ionised water c) dry for 20 – 30 minutes in 70 °C oven d) apply adhesive as soon as possible The second method of zinc coating is more widely used.
• Zinc sheradising A method used to give what is in effect a fused zinc surface to a steel component can be specified which gives very good environmental protection for the steel component.
11
The Handbook of Rubber Bonding The steel part to be bonded is baked whilst being tumbled in zinc dust. The process is not generally suitable for delicate metal parts and causes problems with zinc build-up in screw threaded components (the latter must be protected by a sleeve or require a die running down the thread to clear it). After treatment exposed zinc surfaces do of course oxidise if stored incorrectly, but this is not usually a problem. The oxide forms after both methods of zinc coating.
• Aluminium - anodising Aluminium is usually electrolytically anodised, in the presence of an acid, either sulphuric, chromic or phosphoric, to give a tough resistant oxide film, which usually forms good bonds with the usual bonding systems. The anodising must be carried out with care and with a mind to the type of crystalline structure being formed on the aluminium surface. A uniform reticulated structure is desired, not a microscopically fragmented rippled surface, sometimes called ‘ice flows’ [7], which are unstable, easily fractured, and therefore too unstable to maintain good adhesive quality. If anodising is to be carried out by a custom plater he will need to be informed of the type of anodised structure desired. N.B. The final stages of any ‘wet’ metal preparation process for metals to be bonded to rubber is to ensure that all chemicals used in the processes have been removed in the final water rinse tank, and then to ensure that all faces of the metal parts are fully dried prior to bonding agent application. All warehouse metal storage areas must be held at least 5 – 10 °C above the dew-point and ideally as near to ambient temperature in the bonding agent application shop which should be in the region of 18 – 22 °C minimum.
• Metal preparation - for waterborne bonding systems Although the general principles used for solvent-based adhesives apply, the cleaning of metals for the application of waterborne bonding systems becomes much more critical. Scrupulously clean metals are vital, to ensure maximum wettability of the prepared metal bonding surface. Lord Corporation [8] suggest that calcium modified phosphating of metals is preferable to conventional grit blasting with its potential for ‘re-infecting’ the metal surface after initial degreasing by using contaminated grit. Proper housekeeping should eliminate such problems.
1.2 Pre-treatments of Plastics and Rubbers 1.2.1 Introduction In many cases, rubbers are joined to other materials during the process of vulcanisation. However, in other cases, rubbers are joined to other materials after vulcanisation. With this
12
Substrate Preparation Methods second group, it is often necessary to pre-treat the rubbers before bonding. Pre-treatments range from physical methods such as a solvent wipe or abrasion to chemical methods such as treatment with trichloroisocyanuric acid (TCICA). Physical methods may remove cohesively weak layers from the polymer. This is essential to good bonding unless these layers can be absorbed by the adhesive. However, physical methods will only be effective if the underlying rubber possesses suitable groups which can interact strongly with the adhesive. Chemical methods may also remove weak layers or chemically modify them so that they are more compatible with the adhesive; in addition chemical methods may roughen a surface. However, an effective chemical method will also modify the chemistry of the rubber so that the interaction with the adhesive is increased. In general, rubbers contain a greater variety and quantity of additives than plastics; fifteen components in a particular formulation is quite common. These additives or compounding ingredients as they are often called, may well create a cohesively weak layer on the rubber surface. On the other hand, plastics usually contain a small number of additives and usually in relatively low concentration. Over the last 50 years many methods have been developed to pre-treat plastics and rubbers. Partly because of the much simpler formulations, pre-treatments for plastics have been the subject of much greater scientific interest. Our understanding of pretreatments for plastics is therefore much greater than that for rubbers. Some of the key studies on pre-treatments of plastics will therefore be outlined in Section 1.2.2. Pre-treatments for rubbers have been developed on an empirical basis but some scientific studies of successful pre-treatments have been undertaken. Methods for different rubbers will be reviewed in Section 1.2.3. Rubbers will be considered in groups, namely hydrocarbons that possess little unsaturation, unsaturated hydrocarbons, halogenated rubbers and miscellaneous materials.
1.2.2 Studies of Pre-treatments for Plastics These studies may seem out of context in a book concerned with bonding of rubber but the great deal of work carried out with plastics can be used to understand the problems of rubbers. Some of the most important pre-treatments for plastics were developed in the 1950s. These include the corona and flame treatments for polyolefins [9, 10, 11, 12, 13] and the use of sodium complexes for fluorinated polymers [14 – 17]. The plasma treatment was developed somewhat later [18, 19], as was halogenation [20, 21]. It was suspected that these treatments were chemically modifying the surfaces of the plastics but there was little direct evidence as the analytical methods available at the time were not 13
The Handbook of Rubber Bonding sufficiently surface-sensitive. However, in the 1970s a new method for studying surface chemistry became available, namely X-ray photoelectron spectroscopy (XPS) which is also known as electron spectroscopy for chemical analysis (ESCA). This method is able to characterise and quantify the chemical changes caused by pre-treatments. XPS analyses the first few atomic layers of a material. This is important as some pre-treatments only modify a few nanometers of a polymer. Reflection infrared techniques in the 1970s were often unable to detect changes to the surface chemistry of polymers caused by the pre-treatments. Three of the earliest pre-treatment studies were by Dwight [15], Collins [22] and Briggs [23]. Dwight treated polytetrafluoroethylene (PTFE) and fluorinated ethylene-propylene copolymer (FEP) with sodium in liquid ammonia and sodium naphthalenide in tetrahydrofuran (THF). X-ray photoelectron spectroscopy showed extensive defluorination of the polymers together with formation of carbon–carbon double bonds and various oxygen-containing groups. Collins treated PTFE with ammonia and air plasmas. Again, XPS showed extensive defluorination and in the case of the ammonia plasma, nitrogen containing groups were introduced. Briggs [23] was the first to quantify the chemical modification caused by a pre-treatment. Briggs studied the changes caused by chromic acid etching of low density polyethylene and polypropylene. Some of the results are given in Table 1.1. Angular variation studies, i.e., the angle of incidence of the X-ray beam was varied, showed that in the case of polypropylene, the depth of the chemically modified layer was only a few nanometers.
Table 1.1 XPS data for polyolefins treated with chromic acid [23] Polymer
Treatment
LDPE
PP
Surface composition (atom %) C
O
S
None 1 min/20 °C 6 h/70 °C
99.8 94.4 85.8
0.2 5.2 13.1
0.4 1.1
None 1 min/20 °C 6 h/70 °C
99.8 93.4 94.0
0.2 6.3 5.7
0.3 0.3
NB: The sulphur originates from the attack of the polyolefin by the sulphuric acid present in the chromic acid LDPE: low density polyethylene PP: polypropylene C: carbon O: oxygen S: sulphur Reproduced with permission from D. Briggs, D. M. Brewis and M. B. Konieczko, Journal of Materials Science, 1976, 11, 7, 1270. ©1976, Kluwer Academic Publishers
14
Substrate Preparation Methods A given pre-treatment may result in the introduction of several different chemical groups. There are two methods by which these groups may be quantified and both involve XPS. The first method involves derivatisation reactions and the second method the use of high resolution spectra. The basic idea behind the derivatisation method is to use several reagents each of which will react with only one of the groups introduced by the pre-treatment. There are two other requirements. Each reagent should introduce an atom, e.g., fluorine, that is not already present in the surface and each reaction should proceed to 100% conversion. The method is illustrated by the work of Gerenser [24] where some corona treated polyethylene was derivatised. The reagents and derivatisation reactions are shown in Figure 1.1 and the results of the experiments are shown in Table 1.2.
Figure 1.1 Derivatisation reactions to identify functional groups introduced by pretreatments; a) peroxide groups reacting with sulphur dioxide, b) alcohol group reacting with hexafluoroacetic anhydride, c) carbonyl group reacting with hydrazine, d) epoxide group reacting with hydrogen chloride, e) carboxylic acid group reacting with tertiary amine. (Reprinted from L. J. Gerenser, J. F. Elman, M. G. Mason and J. M. Pochan, Polymer, 1985, 26, 8, 1162. ©1985, with permission from Elsevier Science)
15
The Handbook of Rubber Bonding
Table 1.2 Quantification of surface functionalities after corona treatment using derivatisation Functional group
Group conc. X 102 * Initial
Washed
C
OOH
1.2
0.9
C
OH
1.7
1.1
C
O
1.8
0.9
C
2.3
1.1
1.6
0.8
NO3
0.8
0.4
Total [O] tagged
13.8
7.7
Actual [O] incorporated
~18
~10
O C
O C
OH
Footnote: Allowing for the fact that some of the groups contain more than one oxygen atom, it can readily be calculated that the concentration of oxygen atoms involved in the derivatisation reactions was 13.8%; this is the amount of oxygen tagged. The actual amount of oxygen incorporated the corona treated surface was found by XPS to be 18%. This means that other oxygen-containing groups were present and/or the reactions with the above groups did not go to completion. *Moles of functional species per unreacted initial carbon atom Reprinted from L. J. Gerenser, J. F. Elman, M. G. Mason and J. M. Pochan, Polymer, 1985, 26, 8, 1162. ©1985, with permission from Elsevier Science
The second method to quantify the chemical groups introduced by a pre-treatment involves obtaining a high resolution spectrum of the photoelectrons from the C1s core level and resolving this into the various contributions. This approach is illustrated by Beamson [25] who examined a rubber-modified polypropylene which had been subjected to a corona discharge treatment. The high resolution C1s spectrum is given in Figure 1.2 and
16
Substrate Preparation Methods
Corona treatment after derivatisation
the information on the groups introduced is given in Table 1.3. This method is much quicker than the derivatisation approach but requires an instrument with very good energy resolution and great care in attribution of the various peaks.
Figure 1.2 High resolution C1s spectrum of corona treated polypropylene [25]
Table 1.3 Assignment of peaks for corona treated polypropylene [25] Peak no.
Position (eV)
Area (%)
Assignment
1
285.0
91.7
C-C, C-H
2
286.5
1.2
C-O
3
287.1
2. 3
C-O-O
4
288.1
2.3
C=O
5
288.9
1.2
COOH
6
289.5
1.3
O=C-O-C=O *
* The assignment at 289.5 eV is tentative
17
The Handbook of Rubber Bonding Strobel [12] compared the effectiveness of various gas-phase reactions for polypropylene, by determining how much oxygen was introduced into the polymer surface (the O:C atomic ratio) in a given time. These results are summarised in Table 1.4. It can be seen that to achieve a given level of chemical modification, flame, corona and plasma require much shorter treatment times than ozone or UV or a combination of UV plus ozone. The pre-treatments described above represent just a few of the many studies relating to the mechanisms of pre-treatments for plastics. However, it is clear that much is known about pre-treatments of plastics relating to: • Quantification of the chemical changes caused by pre-treatments, • The depth of the chemical modification, • Identification and quantification of chemical groups, • The rate of chemical modification. In contrast, much less work has been done relating to the mechanisms of pre-treatments for rubbers.
Table 1.4 Surface analysis of treated polypropylene films [12] Treatment
Exposure time (s)
None
—
XPS O:C atomic ratio 0.0
Corona (1.7 J/cm2)
0.5
0.1 2
Corona (0.17 J/cm2)
0.05
0.0 7
Flame
0.04
0.1 2
Remote air plasma*
0.1
0.1 2
Ozone
1800
0.1 3
UV/air
600
0.0 8
UV/air plus ozone
600
0.1 4
*The plasma was produced by a microwave generator and passed 100 mm down a tube onto the polymer surface Reproduced with permission from M. Strobel, M. J. Walzak, J. M. Hill, A. Lin, E. Karbashewski and C. S. Lyons, in Polymer Surface Modification, Ed., K. L. Mittal, VSP, Utrecht, 1996, 233. ©1996, VSP BV
18
Surface analysis
Substrate Preparation Methods
1.2.3 Hydrocarbon Rubbers with Little or No Unsaturation 1.2.3.1 Ethylene-Propylene Rubbers Ethylene-propylene rubbers (EP) have low total surface energies with small polar components. As would be expected, the adhesion of paints and adhesives to untreated EP is poor. To achieve good adhesion to EP, the introduction of suitable functional groups is necessary unless a diffusion mechanism can operate. Bragole [26] found that UV treatment of EPDM coated with a thin layer of benzophenone resulted in large increases in the adhesion of acrylic, epoxy and urethane paints to the polymer. Ellul [27] subjected EPDM/polypropylene and natural rubber/polypropylene blends to various halogenation treatments, namely fluorine/carbon dioxide, sodium hypochlorite/ acetic acid and bromine water. With the natural rubber blend, there was a substantial uptake of fluorine, chlorine and bromine in the surface regions as indicated by energy dispersive X-ray analysis and with all three pre-treatments the adhesion to an acrylic tape was greatly enhanced. In contrast, with the EPDM blend, fluorine was the only reagent which reacted with the rubbers and only this treatment resulted in a significant increase in adhesion to the acrylic tape. The above results can be explained in terms of the different concentrations of carbon–carbon double bonds in the two blends. Substantial incorporation of chlorine and bromine could occur with the natural rubber-polypropylene blend but not with the EPDM blend. However, fluorine gas will react readily with saturated hydrocarbons [28, 29] and therefore the incorporation of fluorine into the EPDM blend is not surprising. Lawson [30] using X-ray photoelectron spectroscopy (XPS) found that trichloroisocyanuric acid (TCICA) in ethyl acetate did not chemically modify EPDM. Lawson [31] also found that a corona treatment improved the wettability of EPDM as indicated by glycerol contact angles and the use of a series of formamide/2-ethoxyethanol mixtures (ASTM D2578 [32]). However, the contact angles increased significantly over a period of one hour, indicating molecular rearrangement with the polar groups introduced by the pre-treatment tending to move to the bulk of the rubber. No improvement in a peel test involving a polyurethane coating was observed. Minagawa [33] treated an EP rubber with UV and sputter etching. Large increases in adhesion were reported. However, the treatment times were long, being 10 minutes for ion etching and one hour for the UV treatment. Scanning electron microscopy (SEM) indicated the two methods caused considerable roughening of the surface. XPS and Fourier transform infrared analysis (FTIR) indicated the introduction of substantial quantities of oxygen-containing functional groups. Kondyurin [34] noted only modest improvements, at best, after treating EPDM with UV, despite clear infrared evidence for the formation of hydroxyl and carbonyl groups after treatment.
19
The Handbook of Rubber Bonding
1.2.3.2 Butyl Rubber Butyl rubber consists of ≥95% of isobutylene units with a small quantity of isoprene which permits crosslinking via sulphur vulcanisation. Butyl rubber has a low surface energy and in addition organic components with a low cohesive strength may exist on the surface. In one study [35] butyl rubber was subjected to several treatments which normally cause substantial chemical modification to polymer surfaces. The treatments included chromic acid etching, corona discharges, flames, bromination, UV radiation and potassium permanganate. Most of the treatments had little effect on the adhesion to an epoxide. It was concluded that much chain scission occurred with the result that suitable functional groups were not introduced in sufficient quantity into long polymer chains. Such chemical modification is necessary for good adhesion unless a diffusion mechanism is operating.
1.2.4 Unsaturated Hydrocarbon Rubbers 1.2.4.1 Natural Rubber Natural rubber (NR), being essentially a hydrocarbon, has a low surface energy. Some of the components in a formulated rubber, such as zinc oxide and carbon black, may substantially increase the surface energy, whereas organic additives such as extender oil and antioxidants may migrate to the surface and create a potentially weak boundary layer. Pettit and Carter [36] found that chlorine gas, acidic sodium hypochlorite and an organic chlorine donor in a organic solvent all much improved the peel strengths of joints involving NR and a polyurethane adhesive. Oldfield and Symes [37, 38] found that aqueous or organic-based chlorination gave much higher joint strengths than a solvent-wipe, abrasion or cyclisation (see Table 1.5). Oldfield and Symes used X-ray fluorescence, infrared analysis and contact angle measurement to study the TCICA treatment. X-ray fluorescence showed the amount of chlorine introduced into the polymer increased with the TCICA concentration; with a 3% TCICA solution, they estimated the chlorine content in the treated NR was 16.7% w/w. Reflection infrared analysis indicated that chlorine substituted at the allylic position in the polymer backbone. Substantial improvements in wettability were achieved especially if the concentration of TCICA was at least 0.8%. Lawson [30] pre-treated various rubbers, including NR, with a 3% w/v solution of TCICA in ethyl acetate and used XPS to study the chemical changes caused by the pre-treatment. In agreement with Oldfield, they concluded that the chemical modification was mainly substitution rather than addition at the carbon–carbon double bond.
20
Substrate Preparation Methods Table 1.5 Effect of pre-treatment on the peel strengths (N mm-1) of NR-epoxide-NR [37] Pre-treatment
Peel strength
Locus of failure
Toluene wipe
0.1
I
Abrasion on grinding wheel
1
I
Acidified hypochlorite
10
R
Cyclisation
1
I
TCICA in ethyl acetate
18
R
I - apparent interfacial ; R - cohesive in rubber Reproduced with permission from D. Oldfield and T. E. F. Symes, Journal of Adhesion, 1983, 16, 2, 77. ©1983, Gordon and Breach Publishers
Extrand [39] treated NR surfaces in an acidified sodium hypochlorite solution and used contact angle measurements and reflection FTIR to study the changes caused by the chlorination. They studied ‘pure’ NR, a peroxide cured formulation and a conventionally cured formulation. Contact angles of glycerol on the rubber surfaces reduced after chlorination as shown in Table 1.6.
Table 1.6 Effect of chlorination on the contact angles between glycerol and various rubber surfaces [39] Contact angle (°)
Substrate
Before treatment
After treatment
‘Pure’ rubber
64
11
Peroxide cured
46
30
Conventionally cured
82
30
Reproduced with permission from C. W. Extrand and A. N. Gent, Rubber Chemistry and Technology, 1988, 61, 4, 688. ©1988, Rubber Division, American Chemical Society
Peel strengths 21
The Handbook of Rubber Bonding With regard to the infrared study, bands at 660, 750 and 1260 cm-1 were assigned to the effects of chlorination. In addition, bands at 780, 916 and 1410 cm-1 were almost certainly due to chlorination. Kusano [40] found that neither corona nor plasma treatments improved peel strength with a polyurethane adhesive despite improved wettability as indicated by water contact angles. FTIR indicated substantial oxidation after the corona treatment but only minor oxidation after the plasma treatment.
1.2.4.2 Styrene-Butadiene Copolymers Styrene-butadiene rubber has a low surface energy, but this may be considerably increased by the incorporation of various components. Organic additives such as antioxidants will tend to migrate to the surface thus creating a potential weak boundary layer. Pettit [36] found that treatment of SBR with chlorine gas, acidified sodium hypochlorite or an organic chlorine donor in an organic solvent resulted in large increases in peel strength for SBR-polyurethane-SBR joints. Oldfield [37] found that physical treatments were inferior to three chemical pre-treatments (see Table 1.7).
Table 1.7 Effect of pre-treatments on the peel strengths (N mm-1) of SBR-epoxide-SBR joints [37] Pre-treatment
Peel strength
Locus of failure
Toluene wipe
0.2
I
Abrasion on grinding wheel
1
I
Acidified hypochlorite
12
R
Cyclisation
12
R
TCICA in ethyl acetate
11
R
I - apparent interfacial; R - cohesive within rubber Reproduced with permission from D. Oldfield and T. E. F. Symes, Journal of Adhesion, 1983, 16, 2, 77. ©1983, Gordon and Breach Publishers.
22
Substrate Preparation Methods Using X-ray fluorescence, they estimated the chlorine concentration in the first few microns of the SBR after treatment with TCICA at various concentrations. With a 3% solution, the resulting chlorine concentration was 16.1% w/w. Pastor-Blas [41] found that physical treatments such as abrasion did not result in significant increases in the peel strengths obtained with a polyurethane adhesive. On the other hand, treatment with TCICA in ethyl acetate resulted in large increases in peel strength. On the basis of the relative amounts of chlorine and nitrogen introduced into SBR, Lawson [30] concluded that both substitution and addition reactions were significant when this rubber was treated with TCICA in ethyl acetate. Similar results were obtained with polybutadiene. Pastor-Blas [42] studied the effect of TCICA concentration in ethyl acetate. For solutions up to 2% w/w mainly chlorinated hydrocarbon and C–O species were reported. At between 2 and 5% w/w an excess of unreacted TCICA was indicated while above 5% w/w there was a detrimental effect on adhesion due to a weak boundary layer consisting of isocyanuric acid. Pastor-Blas [43] treated an SBR formulation with TCICA solutions in ethyl acetate having concentrations ranging from 0.5 – 7% by weight. The chemical changes caused by the pre-treatments are shown in Table 1.8. Rubber strips were bonded with a solvent-based polyurethane (PU) and the joint strengths determined in a T-peel test. After peeling, the test pieces were examined using a variety of techniques; XPS and FTIR confirmed that the treatment introduced various chemical groups. The peel strengths were obtained after treatments with 0.5, 2 and 7% w/w. The highest peel strength was obtained with the 2% solution.
Table 1.8 XPS studies of SBR treated with solutions of TCICA in ethyl acetate [43] Surface analysis (atom %)
Wt% concentration of TCICA
C
O
Si
N
Cl
S
0
92.27
2.8
1.5
-
-
-
2
92.7
4.3
1. 0
1.0
0.8
0.2
7
91.5
4.6
0.7
1.9
0.9
0.4
Reproduced with permission from M.M. Pastor-Blas, J.M. Martín-Martínez and J.G. Dillard, Journal of Adhesion, 1997, 62, 1/4, 23. ©1997, Gordon and Breach Publishers.
Peel strengths
23
The Handbook of Rubber Bonding In a related publication, treatments with fumaric acid in a butan-2-ol/ethanol mixture and TCICA in butan-2-ol were compared [44]. In general, the TCICA was more effective at enhancing the peel strength achieved with a solvent-based PU adhesive. Infrared analysis indicated the treatments were probably effective by removing zinc stearate (reduction in peak at 1540 cm-1) and the introduction of carbon-oxide functionalities (1704 cm-1 and 1670 cm-1 for the TCICA and fumaric acid, respectively). With TCICA, C–Cl bonds were also observed. Pastor-Sempere [45] treated two styrene-butadiene rubbers with fumaric acid in a butan2-ol/ethanol mixture. This resulted in improved adhesion in both cases, but the improvement with one formulation was significantly greater than the other. The lower peel strength was attributed to the presence of paraffin wax and zinc stearate. Roughening prior to treatment with fumaric acid resulted in additional improvements with both rubbers. Infrared analysis indicated that the fumaric acid was effective by introducing C=O bonds and by reducing the concentration of zinc stearate. In addition, the fumaric acid caused a roughening of both rubbers. Later Pastor-Blas [46] demonstrated that high concentrations of TCICA could lead to the formation of weak boundary layers. Treatment of two SBR materials with a 7 wt% solution of TCICA in ethyl acetate resulted in poor peel strengths unless the treated surfaces were vacuum dried for one hour at 1.34 Pa. Other methods have been shown to considerably improve the bondability of SBR materials. Aqueous solutions of an organic chlorine donor or the use of an electrochemical method resulted in large increases in peel strength with a water-based PU adhesive [47]. Kusano [40] found that corona and plasma treatments resulted in large increases in peel strength with a PU adhesive. Lawson [31] reported that a 10 second corona treatment improved the water wettability of an SBR. He also reported cracking of the rubber which he ascribed to the ozone generated in the discharge. Styrene-butadiene block copolymers SBS thermoplastic rubbers have a low surface energy. Therefore, to achieve good adhesion to SBS a chemical pre-treatment may be necessary. A complicating factor is that migratory organic additives may lead to a weak layer. Pettit [36] found that treatment of SBS with chlorine gas, acidified sodium hypochlorite or an organic donor in an organic solvent resulted in large increases in peel strength with a polyurethane adhesive. As with SBR, aqueous solutions of an organic chlorine donor and an electrochemical method were also effective with SBS [47]. Pastor-Blas [48] treated SBS with TCICA solutions (0.5, 2 or 7 wt%) in ethyl acetate. The SBS was bonded with a PU and the joint strengths determined in a T-peel test. The failed surfaces, after peeling, were examined by a variety of techniques including XPS and FTIR.
24
Substrate Preparation Methods It was concluded that the highest strength (3.3 N mm-1) was obtained with the 0.5% solution. It was concluded that the stronger solutions weakened the surface regions. FTIR and XPS showed that the treatment introduced chlorine and oxygen functionalities.
1.2.5 Halogenated Rubbers Introduction of bromine and chlorine atoms in hydrocarbon polymers will enhance adhesion. In the case of PE, introduction of bromine to a Br:C ratio of 0.05:1 resulted in high adhesion to an epoxide adhesive [49]. However, the quantity of halogen in bromoand chloro-butyl rubbers is low and poor adhesion to these polymers is not unexpected especially if organic additives are present on the surfaces. Oldfield [37] only obtained modest adhesion to untreated bromobutyl rubber (see Table 1.9). Of the treatments they investigated only TCICA in ethyl acetate resulted in very high peel strengths, although aqueous chlorination gave a substantial improvement. Using X-ray fluorescence, Oldfield and Symes found that the uptake of chlorine into bromobutyl rubber was very much less than that observed with NR, SBR and nitrile rubber, as would be expected from the relative number of carbon–carbon double bonds using XPS. Lawson [30] found chlorobutyl rubber did not take up any measurable amount of chlorine in treatment with TCICA in ethyl acetate. The reason for the large improvement in bondability with bromobutyl observed by Oldfield is unclear. It may be that the TCICA was acting as an oxidising agent rather than a chlorinating agent. However, Lawson did not observe any introduction of oxygen-containing groups with chlorobutyl rubber.
Table 1.9 Effect of pre-treatments on the peel strengths (N mm-1) of bromobutyl rubber–epoxide–bromobutyl rubber joints [37] Pre-treatment
Peel strength
Locus of failure
Toluene wipe
1
I
Abrasion on grinding wheel
1
I
Acidified hypochlorite
3
I
Cyclisation
0.1
I
TCICA in ethyl acetate
20
R
I - apparent interfacial; R - cohesive in rubber Reproduced with permission from D. Oldfield and T. E. F. Symes, Journal of Adhesion, 1983, 16, 2, 77. ©1983, Gordon and Breach Publishers.
25
The Handbook of Rubber Bonding Polychloroprene (CR) has much more chlorine than the chlorobutyl rubber examined by Lawson and good adhesion to untreated CR would be expected provided there was no weak layer was on the surface. If such layers exist a suitable solvent treatment or abrasion should result in good adhesion. Cyclisation has been recommended as a pre-treatment [50, 51]. Lawson noted a large uptake of chlorine, nitrogen and oxygen on treatment of polychloroprene with TCICA, indicating addition across the carbon–carbon double bond. Lawson [31] reported that a corona discharge treatment of CR increased its surface energy, but did not improve the peel strength with a polyurethane coating. Minagawa [32] reported large increases in adhesion with CR after UV irradiation or sputter ion etching. However, the treatment times were long, being 10 minutes with ion sputtering and one hour with the UV treatment. SEM indicates that the two methods caused considerable roughening of the surface. XPS and FTIR indicated the introduction of substantial quantities of oxygen-containing groups.
1.2.6 Miscellaneous Rubbers 1.2.6.1 Silicone Rubber (see also Chapter 12) Adhesion to untreated silicone rubber is difficult. The poor adhesion may be due to a low surface energy (approximately 24 mJ m-2) or a layer of low cohesive strength or a combination of these two factors. Plasma treatment has been shown to substantially improve the wettability of silicone rubber [50-57]. Peel strengths were measured in one study and found to be much increased by plasma treatment [53]. Swanson [58] found that coating a silicone rubber with photoactive reagents and then exposing the surface to UV resulted in a large increase in joint strengths obtained with a cyanoacrylate adhesive. Combette [59] reported that microwave or radio frequency plasma treatment of silicone rubber with a gas rich in oxygen gave high peel strengths with an epoxide adhesive.
1.2.6.2 Nitrile Rubber Nitrile rubber is moderately polar and good adhesion would be expected between a polar adhesive like an epoxide and the untreated polymer provided no weak boundary layers were present. This was found to be the case by Oldfield [37] as can be seen in Table 1.10. High adhesion values were obtained with a solvent wipe. Cyclisation and TCICA treatments resulted in large increases in adhesion. X-ray fluorescence indicated substantial uptake of chlorine in the latter case [37]. Peel strengths 26
Substrate Preparation Methods Table 1.10 Effect of pre-treatments on the peel strengths (N mm-1) of nitrile rubber-epoxide-nitrile rubber joints [37] Pre-treatment
Initial strength
Locus of failure
Toluene wipe
8
R
Abrasion on grinding wheel
5
I
Acidified hypochlorite
8
R
Cyclisation
18
R
TCICA in ethyl acetate
21
R
I - apparent interfacial; R - cohesive in rubber Reproduced with permission from D. Oldfield and T. E. F. Symes, Journal of Adhesion, 1983, 16, 2, 77. ©1983, Gordon and Breach Publishers.
1.2.6.3 Polyurethanes Polyurethanes (PU) have relatively high surface energies. Adhesion problems with PU substrates are, therefore, likely to be due to cohesively weak material, such as mould release agents on the surface. Abrasion is one of the main methods recommended as a pre-treatment [50]; such a pre-treatment can remove cohesively weak material and expose strong material of relatively high surface energy. Cryoblasting, in which carbon dioxide particles are fired at a substrate, has been shown to be capable of removing silicone release agents from PUs and thus giving large improvements in adhesion [47].
1.2.7 Discussion As noted in Section 1.2.1, there have been many detailed studies relating to the pretreatment of plastics. Much is now known about these pre-treatments including the chemical groups introduced, their concentrations and the depth of chemical modification. In contrast, the number of studies involving rubbers is much lower and in general the studies have been much less informative. One of the reasons for this is that rubbers usually contain several additives, often in relatively high concentrations. These additives make an understanding of the pre-treatments much more difficult. Because of the wide range of formulations for a particular rubber, it is also more difficult to generalise about pre-treatments than it is with plastics. For example, it is known that some formulations of SBR are considerably easier to pre-treat than others.
27
Peel strengths The Handbook of Rubber Bonding The four groups of rubbers considered above will now be discussed. Conclusions about pre-treatments for rubbers will then be presented. Hydrocarbon materials with few carbon–carbon double bonds will be considered first. The most important examples in this group are ethylene-propylene rubbers which may be crosslinked with peroxides or sulphur systems in which case small quantities of dimers are polymerised with ethylene and propylene (EPDM). As EP rubbers contain no polar groups it will normally be necessary to chemically modify the polymers to enable them to interact strongly with polar adhesives such as epoxides and polyurethanes. In the case of plastics such as polyethylene and polypropylene, large increases in adhesion can be achieved by treating with a flame [9, 10], corona [11, 13], plasma [18, 19], or etching solution [23]. It would be expected that EP rubbers would respond in the same way to these pre-treatments. However, this is not always the case. Thus, Lawson [31], found that a corona treatment of an EPDM did not improve the peel strength to a polyurethane coating. It is probable that the reason for the poor adhesion is a layer of low molecular weight material on the EPDM. During corona treatment this layer, rather than the underlying polymer, would be oxidised. Hence, the polyurethane coating would not be able to interact strongly with the EPDM. Even if the EPDM was oxidised by the corona treatment, there would still be a cohesively weak layer on its surface. Many rubbers possess carbon–carbon double bonds. In such cases there is the possibility that pre-treatment may be effective by addition or substitution reactions. Thus some reagents may be effective with unsaturated hydrocarbons such as SBR and SBS but not with EP rubbers. This is demonstrated by the work of Lawson [30] who found that treatment with TCICA in ethyl acetate resulted in the introduction of substantial quantities of chlorine into SBR, polybutadiene and NR, but not into EPDM. Several methods have been shown to be effective at pre-treating unsaturated hydrocarbon rubbers. These include treatment with concentrated sulphuric acid, acidified sodium hypochlorite and TCICA in ethyl acetate. The last method is the most commonly used commercially but in many countries legislation is being introduced to reduce the use of organic solvents. Promising results have been obtained with new solvent-free methods, namely an electrochemical method involving a highly reactive complex ion, and a method involving a water-soluble organic chlorine donor [47]. Like hydrocarbon rubbers, silicones have low surface energies and interactions with polar adhesives will be low unless the surface chemistry is modified. Plasma treatments improve the wettability [52, 53, 54, 55, 56, 57] or bondability [58, 59] of silicones. It is generally accepted that the introduction of a wide range of functional groups makes a polymer much more bondable. The effect of introducing individual chemical groups into polyethylene was demonstrated by Chew [60]. Thus, bromine, carbonyl, hydroxyl
28
Substrate Preparation Methods and carboxylic acid groups were all shown to greatly increase the bondability of polyethylene to an epoxide adhesive. This is in line with the general experience that polymers possessing halogens or oxygen-containing groups are much easier to bond than polyolefins. Whether rubbers containing such groups are easy to bond depends very much on whether the bonding surface is covered by low molecular weight (MW) additives or contaminants. On the one hand, Oldfield [37] achieved high peel strengths with chemically unmodified nitrile rubber whereas Brewis [47] obtained low peel strengths with an as-received polyurethane. However, after the removal of a silicone release agent by cryoblasting, much higher peel strengths were obtained [47].
1.2.8 Summary • Methods are available to pre-treat all rubbers but additives or processing aids may make successful pre-treatment much more difficult. • TCICA in various organic solvents is very effective with those rubbers possessing carbon-carbon double bonds. However, legislation restricting the use of organic solvents is being introduced in many countries. Promising new pre-treatments include the use of water-soluble organic chlorine donors and an electrochemical method in which a highly active complex ion is generated. • With some polymers containing suitable chemical groups, e.g., PU, simply removing cohesively weak material from the surface may be all that is necessary to achieve good adhesion.
1.3 Bonding Rubbers to Plastic Substrates 1.3.1 Introduction This section is based mainly on first hand personal experience and is not intended to be an overview of bonding. It covers the basic practical principles of bonding rubbers to a variety of plastics materials. It is typical to find that those who are skilled in the art of moulding and bonding rubbers have little affinity to plastics materials and vice versa. As for polyurethanes; these are something else altogether. This chapter will concentrate on those plastics and rubbers which are likely to have uses in the manufacture of composite materials (see Appendix 1.1).
29
The Handbook of Rubber Bonding
1.3.1.1 Why Use Plastics? • Cost, • Weight saving, • Technically superior, • Environmentally more acceptable, • Fashion/style.
1.3.1.2 What Form Does the Plastics Material Come In? • Moulded components, • Cast components, • Sheet or film, • Tube/pipe or rod. Fabric, fibres and filaments are obviously important forms and uses of plastics materials. Although the basic principles of bonding plastics apply to fibres and fabrics, the other factors involved in bonding them are a subject in themselves and will not be discussed further. In the bonding of rubbers it is assumed that the plastics component is an item which has been preformed and it is this which will be treated with a bonding agent. In most cases the rubber will be moulded onto the primed surface, by techniques including the following: • Injection moulding, • Reaction injection moulding (RIM), • Compression moulding, • Transfer moulding, • Extrusion blow moulding, • Lamination, which could involve post vulcanisation bonding, • Autoclave vulcanisation - rollers, pipes, hoses, stators, • Casting at zero or low pressure - casting of PU. The basic principles should apply to any form of plastics material and to any method of moulding.
30
Substrate Preparation Methods Of course there is always the potential to mould the plastics material onto the vulcanised rubber, but this is rare. In practice, this type of moulding is an example of post vulcanisation bonding.
1.3.2 Plastics Substrate Preparation In preparing metals for bonding, steel in particular, the idea is to produce a surface which is free of contamination, is easy to wet, has a ‘sharp’ irregular surface to promote a mechanical key and controlled oxidation (see Figure 1.3).
Figure 1.3 Metal surface sites for bond
Fortunately for the commonly used metals this ‘controlled’ oxidation occurs naturally after grit blasting or acid etching. In the case of plastics, no such convenient oxidation process takes place. However, each material will have a unique surface layer containing potential sites for bonding:
• Polyamides
The polar group NH-C=O is capable of hydrogen bonding through the activated C=O group and via the N-H group. The N-H leaves a reactive site for chemical reaction with silanes, epoxies, isocyanates and any chemical adducts, which can release such species or any other species, which can react with an active hydrogen. Of course the amide group needs to be on the surface to be able to undergo hydrogen bonding or chemical reaction and steric hindrance will reduce the capability of such groups to partake in bonding, which is especially so in the case of aromatic polyamides.
31
The Handbook of Rubber Bonding
• Polyesters
The C(=O)O, ester group will partake in hydrogen bonding through both oxygen atoms, especially the activated carbonyl group. Some polyesters will be less easy to bond if steric hindrance is likely. Even PBT proves difficult to bond and often requires further treatment.
• Polyurethanes
In theory PU should be very active towards bonding, with an activated N-H and a carbonyl group, as described for polyamides. However PUs are never that easy to bond and could be due to surface oxidation and/or surface hydrolysis, it is normal to remove the surface, degrease and prime before the surface is too old.
• Polyureas
The sites for hydrogen bonding and chemical reaction are significant and polyureas are generally easy to bond. Being more oxidation resistant and hydrolysis resistant than the urethane group is significant.
• Polycarbonates
A regular repeating stable carbonyl group is available for polar attraction and hydrogen bonding.
• PPS (and PPO)
32
Substrate Preparation Methods As for polycarbonates, a regular repeating stable polar sulphur (oxygen) atom allows for polar attraction and hydrogen bonding. However, in the case of the polyolefines, there are no obvious adhesion sites:
• Polyethylene
• Polypropylene
For the bonding of these an oxidation process is essential. When one looks at the surface of metals and plastics under an electron microscope the disruption in that surface explains why bonding is never straightforward. The surface is often described as a weak surface layer and in the case of plastics one can include the surface stresses, general contamination, the presence of abhesive ingredients, i.e., process aids which have migrated to the surface. Some high temperature moulding processes may lead to variable and unwanted oxidation and/or reversion (crosslink degradation) at the surface. Therefore, one can accept the general opinion that the surfaces of plastics do need some form of abrasive or chemical treatment to remove the weak surface layer, or at least reduce it to an adequate level, as shown by the number of publications on the subject [61-69]. Putting it in simple terms the level of surface preparation depends on the performance requirements of the bond. To apply a pressure sensitive decal, no surface treatment is a feasible option, but to make a suspension mount then the plastics surface will require controlled treatments. Most engineering plastics can be treated with alumina or steel grit as for metals. However, in the real world it is quite normal to find that grit blasting is impractical for many reasons, including: • Loss of shape, especially in thin sections, • The reduction in dimensions is not reproducible,
33
The Handbook of Rubber Bonding • Surface damage, such as fibrillation and plastic flow, • Trapped (embedded) grinding media and other contaminants. The harder and the thicker the surface to be bonded the better it is for grit blasting. Similarly, the more highly filled plastics respond much better to blasting than unfilled plastics, and thermosets, especially glass-filled thermosets, are usually very successfully prepared by blasting. If a standard grit blasting process gives problems then the use of a finer grit in any standard grit blasting machine should be thoroughly tested to determine if there is an effective optimum grit size. Abrasive and chemical techniques include the following: • Treatment with abrasive belts, • Hydrosonic/ultrasonic cleaning, • High pressure water/detergent cleaning, • Acid etching, but effluent control means that this is not feasible for anything other than high priced specialities and for long running applications, • The satinisation process for POM is an example of acid etching and involves a slurry containing p-toluene sulphonic acid, • Phenol treatments of polyamides. This includes RFL treatments, • Alkali etching. As for acid etching, the action is mild surface hydrolysis and loosening of ‘debris’ on the surface, • Oxidation with relatively mild oxidising agents. Hydrogen peroxide and sodium hypochlorite are often cited, but a low hazard system worthy of testing out is ammonium persulphate, • Powerful oxidising agents, such as sulphuric dichromate etching, • Abrasion in an aqueous abrasive slurry. Since this involves effluent waste, it is seldom used on a large scale, but is an effective laboratory method, especially when combined with a mild acid, alkali or oxidising agent, • Direct oxidation by flame, or hot air. Normally only applicable to simple shapes, like extruded film, tube and rod,
34
Substrate Preparation Methods • UV treatments. Again this has restricted use, mainly films, • Plasma treatments. Yet to become a mainstream treatment for rubber to plastics bonding, • Corona discharge, • Chlorination.
1.3.3 Degreasing and Solvent Cleaning Degreasing has always been considered an integral part of ideal surface preparation, but under current environmental pressures, it is quite normal to find it has been partly eliminated or even totally eliminated. The need for thorough degreasing becomes more relevant where the environmental resistance of the bond is important and especially where an abrasive technique has left a contaminated surface. Degreasing of plastics with solvents can cause problems: • Stress cracking of the surface, where the effect can remain undetected, • Absorption and even adsorption of a solvent of a similar solubility parameter to the plastics material. This can be a very serious problem, since retained solvent within the bond line could well act as a release agent. If solvent degreasing/cleaning is going to be employed, then a fast drying solvent which has a relatively low solvating power towards the plastics being degreased needs to be used. Aqueous degreasing can be effective, especially when fully automated. However, any aqueous process can leave a surface which requires desorption of water, which adds another process. Unfortunately, for low pressure moulding and casting in particular, the ultimate bonds are often only achieved if desorption of the adsorbed water and gases is specified. This is most evident with polyamides, some polyesters, PU, melamine and urea resins and some epoxy resins. However, in the majority of high pressure moulding processes adsorbed water and gases do not appear to affect bonding, but long term environmental tests may show up a problem. A general guide to reduce the effects of water adsorption is to dry the plastic’s surface, prime with the bonding agent, dry the primed surface and give the component a prebake (the coated dried surface is heated, prior to the moulding process). Pre-bakes can
35
The Handbook of Rubber Bonding
Table 1.11 A brief summary of the preferred treatments Plastics group
Chemical treatments
Degrease
Grit blasting
Other treatments
A1, A2
Yes Take care with acrylics if in doubt use alcohol
Yes Check for optimum grit size
POM-satinisation polyesters difficult surfaces respond to alkali or ammonium persulphate treatments
Nylon 6 and 66 desorb at >100 °C, especially for low pressure moulding and casting of PU other forms of abrasion work generally for these materials
B
Yes
No
Yes Strong oxidising agents
TPOs flame treat, UV, corona discharge treat. may be followed by chlorination
C
Yes Yes Avoid ketones Unless 50° PVC and PVDC. Shore D Avoid aqueous degreasing of PU
PTFE, PVF treat with sodium naphthalene
PTFE, PVF prime with a thin coat of Cilbond 30/31, dry and fuse at >200 °C
be as little as 10 minutes at 70 – 90 °C, which could be part of the drying process, up to 30 minutes at 150 °C, which would be an additional process (see Table 1.11). For plastics group definitions see Appendix 1.1
1.3.4 Adhesive/Bonding Agent Choice 1.3.4.1 Post Vulcanisation Bonding This includes adhesive bonding and bonding with vulcanising bonding agents under the influence of heat and pressure, in those cases where the plastics component needs to be adhered to the preformed vulcanised rubber.
36
Degrease
Grit blasting
Chemical treatments Substrate Preparation Methods
This may be the only method of manufacture for some products and there is a host of adhesives available for plastics, some of which are described in the literature [63 – 69] for example. The main adhesives for bonding plastics to rubbers include cyanoacrylates, two-part urethanes, two-part epoxies, hot melt reactive urethane prepolymers, heat reactive contact cements and silane treatments. Many adhesive bonding applications require a unique answer and it is difficult to make generalised recommendations, as you can within limits, with vulcanisation bonding.
1.3.4.2 Vulcanisation Bonding This is bonding the rubber during the vulcanisation process. The ideal situation is where no bonding agent is required, but in the real world it is rare to find situations where no bonding agent, whether an internal bonding agent (added to the rubber) or a conventional (external) agent is necessary.
• Primers for the Plastics Substrate for Vulcanisation Bonding In theory the primer should match the polarity of the plastics substrate, but this could infer the need for a range of primers depending on the polarity of the plastics to be bonded. In practice, bonding agent primers contain curable polar resins and less polar rubbery polymers, which may or may not be crosslinkable. This gives some versatility in the bonding of a range of polar plastics. An ideal primer would contain highly polar curable resins and speciality polymers. The speciality polymers would vary in their polarity along the polymer chains, giving it variable polarity, a positive attribute in the bonding of a range of plastics. The polymer could be produced by grafting a polar monomer (or monomers) onto an unsaturated polymer such as NR, IR, BR or even NBR, which leads to a polymer which has certain properties: 1. It still contains unsaturation and segments of the original main chain polymer. This means it can crosslink and intermix with the rubber being moulded. 2. Any polar groups on the ungrafted polymer (for example C–N groups in NBR) take part in polar bonding to the plastics substrate. 3. The grafted monomer(s), being polar, can also partake in polar bonding.
37
The Handbook of Rubber Bonding 4. If the grafted monomer retains reactivity it can take part in chemical bonding. Such reactivity could include isocyanates, silanes, epoxides, or even heat reactive adducts, such as blocked isocyanates. 5. If the grafted monomer results in a large and highly polar site, it is possible for this moiety to behave in a way which appears similar to solvent welding (surface softening), but in this case the ‘solvent’ is the polar moiety. This phenomenon is a particular feature of one type of speciality one coat technology, because this ‘welding’ not only applies to the plastics surface, but also to the rubber surface, whether the rubber is in an uncured state or cured state. Though it has been compared to solvent welding the phenomenon described above shows no thermoplasticity, in fact heat and solvent resistance are the big features of this type of technology, along with the capability of post vulcanisation bonding. 6. The ability of the polymer and resin in the primer to react with each other generally improves the environmental resistance of both the bond and the bonding agent. 7. For improved heat resistance, aliphatic chlorine should be avoided in the polymers. For general purpose vulcanisation bonding, conventional primers are available from the established suppliers of bonding agents and all such suppliers can cite many examples of rubber to plastics bonding (see Table 1.12). For improved adhesion and improved environmental resistance the more reactive primers can exhibit advantages, such that in some tough applications, they are the only choice.
• Cover coat/top coat If one is required it should be chosen only with regard to the rubber/rubber being moulded, just as for rubber to metal bonding. (See Table 1.12.)
Summary With attention to detail, most plastics can be bonded to rubbers, provided one accepts the limitations of the rubbers, the plastics and the adhesive system chosen to bond them. It is the aim of those who recommend the adhesives/bonding agents to ensure the bonds are fit for purpose, but it is normal to find that the component manufacturer wants to see no failure attributable to the adhesive.
38
Substrate Preparation Methods
Table 1.12 Rubbers, vulcanisation bonded to plastics - systems and techniques Plastics Materials
Rubber
Environmental Resistance of the Bonds
Bonding System
Special Treatments
PPO
VAMAC (Dupont)
Heat to ≥180 °C
Cilbond 22 Cilbond 60W
Grit blast PPO (200 – 400 μm grit) and degrease with acetone or use alcohol for large or awkward shapes
PPS
NR SBR
Glycol resistant to ≥160 °C
Cilbond 21T Cilbond 22
As above
POM
VMQ
Heat to >>160 °C
Cilbond 65W
Satinse POM with p-TSA
Cilbond 89 Cilbond 22
Abrade or grit blast with fine grit. Degrease with MEK Prebake first thin coat of primer at ≥100 °C use Cilbond 22 for PV bonding
ARAMID XNBR HNBR Heat and fluids to
>170 °C
PTFE
FKM
Hot oils to >>160 °C Cilbond 65W
Sodium treat PTFE and prime as soon as possible
PP FILLED
EPDM
Water to 100 °C
Cilbond 89
Oxidise/flame treat PP. Prebake a first thin coat of primer
GRP
PU rotation cast
Bonds outperforms the PU for roller and pipe coatings
Cilbond 41+B Cilbond 49SF+B
Belt abrade and water or hydrocarbon degrease. Allow first coat to dry >2 h and second coat for >4 h, but 200 °C
Pigmentable, any colour
TPE
++
++
+/–
–
–
+
EPDM
+
–
+/–
–
–
–
Natural Rubber
–
–
–
–
–
–
Silicone LR
++
++
+
++
++
++
TPE: thermoplastic elastomer
++ = very good
+ = good
– = negative
Self-adhesive LR is one of the only rubbers which can be processed fully automatically, producing no waste and at a maximum productivity. Parts with a complex geometry can be realised using quite reasonable injection moulding machines with a low clamping force. Table 11.1 summarises the properties of silicone rubbers and the advantages and disadvantages of some groups of other rubbers that potentially could be used in rubber bonding applications. Please note that this table is intended as a rough guide and is not intended to be a complete and a final statement on the rubbers mentioned therein. As for silicone rubbers other than LR, most of the above mentioned points apply, too. Silicone HTV rubbers can be delivered in a wide variety of preforms which allows the processor to optimise the production process. Such preforms range from blocks and strips to coils and even pellets, use of the latter allowing a more or less continuous operation. Another, useful property of silicone rubbers is their unique electrical behaviour. Silicone rubbers, if processed correctly exhibit extremely high volume resistivity, high dielectric strength and excellent tracking and arcing resistance. This is also the reason why substantial quantities of silicone rubbers are used by the cable, cable accessory and insulator and electronics industries.
289
The Handbook of Rubber Bonding
11.4 Some Applications of Silicone Rubber Composites Silicone rubbers, both, peroxide and addition curing, are used in the manufacture of composite parts. Known applications focus on automotive, medical, sanitary, household, transmission and distribution (T&D), electronics and food appliances. Typical applications for silicone HTV and LR rubber composite parts are: Automotive
Connector seals Crank shaft seals Exhaust pipe hangers Ignition cables Membranes Multi-functional switches Radiator seals Rain sensors Spark plug boots
Medical
Anaesthetic tubings (composite) Body contact electrodes Catheters Respiration masks Various pads Various valves, e.g., dialysis apparatus X-ray opaque shunts
Sanitary and household
Gaskets for WCs Gaskets in tap water equipment O-rings (composite) Various valves
Transmission and distribution (T&D)
Medium and high voltage cable accessories Medium and high voltage insulators
Electronics
Anode caps and cables Key pads (composite) Various gaskets
Food appliances
Baby care articles Food dispensing valves Various gaskets
This list of applications gives just a brief idea of items currently produced using twocomponent injection moulding techniques or insert parts.
290
primers self bonding silicone LR
Bonding Silicone Rubber to Various Substrates
Figure 11.1 Alternatives for bonding silicone rubber
11.5 Bonding Concepts Figure 11.1 shows schematically the construction of the three most important concepts of bonding.
11.5.1 Undercuts This is the most traditional bonding concept. It allows the thermoplastic substrate to be injection moulded in the same machine as the silicone rubber. Undercuts in the substrate can be applied by special mould design, holes or other shapes, e.g., in steel sheet inserts. The possibilities for using undercuts are numerous. However, the use of undercuts has the disadvantage that the mould design for injection moulding of both the thermoplastic substrate and silicone rubber becomes more complicated. In many cases this is due to the fact that the design should allow proper venting during the filling stage of the silicone rubber part of the mould. Undercuts always imply a high proportion of ‘dead corners’ where air is entrapped, leading to incomplete filling of the mould and finally to insufficient anchorage.
291
The Handbook of Rubber Bonding Another disadvantage of this concept is that the bonding contact between silicone rubber and substrate is achieved only by mechanical ties. In other words, the interface between substrate and silicone is a source for leakage, e.g., in a gasket application. In some cases more silicone rubber is used to fill undercuts than would be required for the rubbery function of the part.
11.5.2 Primers The application of primers does not allow for the injection moulding of the substrate on the same machine as the silicone rubber part. A primer is usually applied onto degreased substrates by dipping or spraying. The primer solution can be used as delivered or diluted in hydrocarbons. After application and drying it is necessary to pre-cure the primer at an elevated temperature. In most cases 10 minutes at 130 °C is sufficient.
11.5.3 Self-adhesive Silicone Rubbers Self-adhesive silicone rubbers do not require a primer and, therefore, the plastic substrate can be injection moulded on the same machine as the silicone rubber. Self-adhesive silicone rubbers for injection moulding are restricted to LR mainly. The most peculiar property of these rubbers is the fact that the mould does not have to be treated with a special release agent (which is expensive and complicated to use) when working with these silicone rubbers. However, at the same time it is possible to achieve sufficient bonding to steel without having problems with the self-adhesive silicone sticking to the mould, as is the case with conventional bonding agents for rubber to metal components. Most recently self-adhesive HTVs have been developed. However, they require special release agents or a specially structured mould suface. They are mainly intended for bonding silicone rubber to steel.
11.5.4 The Build-up of Adhesion In general, maximum adhesion is not achieved as soon as the part is moulded, rather it increases after storage of the material for some time. At room temperature this time
292
Bonding Silicone Rubber to Various Substrates could be as long as a few weeks. After that, adhesion can become stronger than the mechanical properties of the rubber. In other words, when trying to separate the rubber from the substrate, the rubber will break. The storage period can be shortened significantly, if it is carried out at an elevated temperature. Approximately 0.5 to 1 hour at 80 °C to 130 °C will provide sufficient adhesion development. The storage or ‘post cure’ temperature is determined by the properties of the substrate, as, for example in thermoplastics, warping might be the result of too high a temperature. It is also essential, that demoulding should be stress free, so that a disruption of the contact surface between silicone rubber and its substrate cannot take place. Once this interface is broken, it is unlikely that proper adhesion will be possible later. Adhesion depends strongly on the surface properties. It is built up faster on smooth surfaces than on rough, sand blasted or electrically eroded substrates. Among other reasons this can be explained by the fact that smooth surfaces are intimately covered to 100% before the cure of the rubber. Uneven surfaces can be interpreted as surfaces with very tiny undercuts, and hence much less than 100% will be covered by uncured rubber, as curing usually happens before these micro-undercuts are filled. As mentioned in Section 11.5.1, undercuts that are not entirely filled might lead to loss of anchorage. When working with standard silicone rubbers stronger adhesion to smooth surfaces is observed. However, cohesive failure is never found when separating the rubber from a substrate which has not previously been primered.
11.6 Bonding of Liquid Rubber (LR) The curing characteristics of LR will be described first as they differ from those for HTV. LR consists of two parts, each of which contains reactive components. Once the components are mixed, they start to cure. This curing speed is virtually zero at room temperature. It rises dramatically when the mixture is heated in the mould. This allows the processor to produce moulded parts by injection moulding. LR has viscosities between 300 and 8000 Pa-s and exhibits extreme shear thinning properties. In other words, once sheared, its viscosity drops and its consistency is very much like honey. HTV rubbers have viscosities far above those of LR. The appearance of an HTV is more solid than liquid (see Figure 11.2). The curable composition of LR (component A mixed with component B) consists of two sorts of polydimethylsiloxane (PDMS) molecules. One contains vinyl groups on the ends
293
The Handbook of Rubber Bonding
Figure 11.2 Chemical structures and mechanism for addition curing LR and HTV
of the chain and the other contains at least three hydrogen atoms across the chain. In order to provide sufficient mechanical strength and rheological properties to allow proper processing, this mixture contains a filler, usually fumed silica. A platinum catalyst and a set of inhibitors control the curing characteristics. This mixture is called addition curing, because silicon-vinyl groups and silicon-hydrogen groups add to each other in the presence of platinum and heat. An advantage of this curing mechanism is the absence of peroxides and their by-products and, moreover, the fast speed of curing. HTV rubber is available as peroxide or addition curing. In some applications LR is not capable of ‘out performing’ HTV rubber. LR is perfect for a vast number of applications, but if very high mechanical properties are required, HTV will be the rubber of choice. The recently developed self-adhesive HTV silicone rubbers are addition curing, one component materials which remain processable for several weeks at room temperature. Self-adhesive liquid silicone rubbers are always delivered as two components. This is due to the fact that once components A and B are mixed the processing time is approximately 3 – 10 days at room temperature. Table 11.2 shows a rough comparison between the curing behaviour of a peroxide curing HTV, an addition curing HTV and an addition curing LR.
294
Curing time Wacker Chemie GmbH Bonding Silicone Rubber to Various Substrates
Table 11.2 Curing time for HT and self adhesive LR Type of rubber/cure
Curing time (s)
Addition Cure LR
25
Addition Cure HTV
50
Peroxide Cure HTV
120
Injection moulding of a test slab of 6 mm thickness Moulding temperature 180 °C Cold runner mould Materials used: Addition Cure LR: ELASTOSIL LR 3070/40 Peroxide Cure HTV: ELASTOSIL R 401/60 (standard HTV) Addition Cure HTV: ELASTOSIL R 4070/60, (self-adhesive HTV, one component) all produced by Wacker Chemie GmbH
Table 11.2 shows clearly that LR is the material that by having the fastest cure rate, will achieve the highest productivity among the group of silicone rubbers and most of all the other rubbers. LR can be bought ready to use and then processed fully automatically without any waste, providing parts with the highest precision and quality which do not require any mechanical secondary treatment such as for example cryogenic deflashing. Self-adhesive LR can be processed to form a composite part using: • Two machines and a robot, or • One machine and two-component (often called two-colour) equipment. In the first case the substrate is produced in the first machine. It is either injection moulded, punched from a metal sheet, etc. A robot then transfers and inserts it into the LR-mould on the second machine. Self-adhesive LR is then injection moulded onto the part. After the curing time the composite is carefully removed. The second case, using only one machine, involves injection moulding the plastic part in a thermoplastic cavity of the two-component mould. On solidification the moulding is extracted by rotating the indexing platen or using a robot and then inserted into the cavity for the LR which is then injection moulded onto it. Demoulding takes place as in the first example.
295
The Handbook of Rubber Bonding It is quite clear that it is essential to take into account the limits of the part to be coated and the requirement of a perfect ‘thermal household’ in the mould during the production of composite moulded parts. The mould temperature of the LR cavity should be as high as possible and the inserted part should be at the highest possible temperature when placed into the silicone cavity, to allow for utmost productivity, which depends very much on the speed of curing. Some proposed conditions are shown in Table 11.3. It is again clear that in injection moulding, LR will be the most viable option for such applications as it allows for a wide processing ‘window’ for adhesives with low curing temperatures.
Table 11.3 Curing times and minimum moulding temperatures for LR and HTV Minimum moulding temperature
Curing time (s)
Addition Cure LR
140 °C
25
Peroxide Cure HTV
170 °C
120
Addition Cure HTV
150 °C
50
Low moulding temperatures also favour the formation of flash. This can be avoided by using state-of-the-art mould technology. The result of a composite moulding was checked with test specimens of a well known ‘dog bone’ shape, as shown in Figure 11.3. These test pieces were produced by injection moulding. They break at a maximum load of 4 kg or even above, if good adhesion has been built up. In many cases the silicone breaks. In other words, its mechanical properties are weaker than the force of adhesion between plastic and rubber. Another similar test is the peel test [11]. A strip of self-adhesive material is peeled off the substrate and the force needed for peeling is recorded. In order not to tear the rubber during the peeling test it is reinforced with a glass fabric. Duration of adhesion (in particular under service conditions) is possibly one of the trickiest questions in connection with self-adhesive LR. In fact, it is unpredictable at present.
296
Bonding Silicone Rubber to Various Substrates PA
Self bonding LR
Figure 11.3 Test specimen for testing self-adhesive LR as a composite with polyamide (PA)
Surprisingly this is not only because of the lack of extended history but also because of the infinite number of possible different service conditions. The number of factors that could influence the duration of adhesion is vast. Some examples are: • • • • • • • • •
Constant mechanical stress, Dynamic mechanical stress, High temperatures, Low temperatures, Varying temperatures, Migration of additives in the substrate, Surface of the substrate, Ageing of the substrate, Immersion in oil, solvents, chemicals in general.
11.6.1 Properties of Self-adhesive LR Practically, all properties known for LR can be incorporated into a self-adhesive material. Such modifications include LR with enhanced oil, tear and/or heat resistance. In the case of the widely used self-lubricating LR, self-adhesive behaviour can be achieved as well. Even conductive self-adhesive LR is theoretically possible.
297
peel test The Handbook of Rubber Bonding
Table 11.4 Self-adhesive silicone rubber ELASTOSIL LR 3070/40 Test results of peeling test according to DIN 53531 [11] Chemical composition
Manufacturer
Tear force [N/mm]
PA 6
Leuna-Miramid GmbH
15,0
R
Ultramid B3EG6
PA 6.6
BASF
16.0
R
Ultramid A3EG6
PA 6.6
BASF
19.9
R
Miramid SE 30 CW
PA 6.6
Leuna-Miramid GmbH
18.3
R
Grilamid TR 55
PA 12
EMS-Chemie
11.9
RM
Grilamid LV 3H
PA 1 2
EMS-Chemie
17.3
R
Crastin SK 605
PBT
Du Pont
15.5
RM
Thermoplast Miramid VE 30 CW
M = adhesion tear (adhesion failure) R = rubber tear (cohesion failure) RM = combines R&M
Table 11.4 gives a summary of results of peel strength achieved with different substrates. The number of substrates is small in order to allow a very quick overview. Slabs 60 x 25 x 2 mm were injection moulded and in a 2K injection moulding process overmoulded by 6 mm of LR. The peeling force was determined after storage for 48 hours at room temperature. No post-cure or heat treatment was applied. None of the substrates mentioned in the table was primered. The table shows various types of plastics (including PA 6, PA 66 and others).
11.6.2 Limitations of Self-adhesive LR The use of self-adhesive LR is limited. Firstly, such materials cannot necessarily be used in applications with food contact, as post curing is required and the formulations often do not comply with respective legislation such as recommendations by FDA / 177.2600 [12] and BgVV XV ‘Silicones’ [13]. Post curing takes place over several hours at 200 °C. Most thermoplastics will not withstand this thermal treatment. But, even much lower post curing temperatures are possible resulting in longer post curing cycles.
298
Bonding Silicone Rubber to Various Substrates Secondly, if an application implying self-adhesive LR requires reliable bonding over the whole service period, extensive testing will be necessary prior to its use. At the moment self-adhesive LR cannot be used in medical device technology as no testing according to protocols like ISO 10993-1 [14] or USP Class VI [15] has been carried out. Medical devices should be using post-cured silicone rubbers only. One of the latest developments is a self-adhesive liquid silicone rubber which is compliant to BgVV and FDA, as post cured. It adheres to PBT and PA only.
11.7 Bonding of Solid Rubber (HTV) The bonding of HTV to various substrates works analogously to the bonding of LR. HTV has much lower curing speeds at any given temperature and this enables it to be used in many different applications, which for the most part are described in Section11.4. In addition, processing of HTV differs in most cases from that of LR. In moulding, compression moulding techniques are used rather than fully automatic waste free injection moulding. Consequently a deflashing procedure is required after the moulding step. In many cases this is done by cooling the parts with liquid nitrogen. The rubber becomes brittle and the flash can then be removed (this is also called cryogenic deflashing). In the case of two-component composites such deflashing steps may become critical, if the components have different coefficients of thermal expansion and because of a strong increase of brittleness in case of thermoplastic substrates. In most cases HTV rubbers are bonded to solid substrates using undercuts or primers. Self-adhesive HTV rubbers work in the same way as their LR counterparts. However, moulding is more critical because of substantially longer curing times which can lead to demoulding problems, as mould stickiness increases dramatically with curing time. For such purposes the processor might need to prime his moulds with a special PTFE primer. It should be noted that self-adhesive HTV sticks to steel when moulded!
11.7.1 Self-adhesive HTV Silicone Rubber Applications Self-adhesive HTV silicone rubber is formulated in a similar way to self-adhesive LR. The ready-to-use formulation contains all adhesion promoters. However, other than LR, which is two component and addition curing only, self-adhesive HTV can be purchased both, as peroxide and addition curing. Because of fundamental differences in curing speed between self-adhesive peroxide cured HTV and addition cured HTV the applications differ quite substantially.
299
The Handbook of Rubber Bonding Furthermore, HTV silicone rubbers exhibit better mechanical properties than LR. As a consequence applications might differ for the self-adhesive versions of these two family of products, apart from much better adhesion to steel in case of self-adhesive HCR.
• Properties Table 11.5 gives a general overview of the properties of HTV silicones and LR. It is clear that HTV will never be 100% replaced by LR. So far the mechanical properties and heat resistance of HTV rubbers are more favourable than those of LR. However, this summary should emphasise the characteristics of each family of materials and give an idea of how to select the most suitable material for an application.
Table 11.5 Summary of properties of LR versus HTV silicone rubber Liquid rubber
HTV silicone rubber
+ self-adhesive LR available
+ self-adhesive HTV available
+ production highly automated
+ more reasonable moulds
+ very complex geometry of parts possible
+ lower cost for small scale production
+ lower cost for large scale production
+ heat stability higher than LR
+ no post treatment of parts
+ chemical resistance higher than LR
+ no yellowing such as with peroxide HTV
+ better mechanical properties than LR
+ weight reduction by lower density
+ available in pellets – easy handling
+ lower curing temperature possible
+ calendering possible
– relatively high investment costs
+ extrudable
+ addition curing + low toxicity + no peroxides therefore often no post cure + much shorter cycle times + less flashes than HTV
– lower transparency than LR, yellowing
+ available as addition curing + low toxicity + no peroxides therefore often no post cure + shorter cycle times than peroxide HTV + no yellowing for addition cured HTV + positive
300
– negative
Bonding Silicone Rubber to Various Substrates Table 11.5 quite clearly shows why some applications are exclusively restricted to either HTV or LR.
11.7.2 Applications for Self-adhesive HTV 11.7.2.1 Peroxide Curing Self-adhesive HTV The application of self-adhesive peroxide curing HTV is restricted to rollers or similar applications. They work much more effectively than primers. Table 11.6 shows a comparison between a roller using a traditional primer and one using self-adhesive peroxide cured HTV (see Figures 11.4 and 11.5).
Table 11.6 Comparison of rollers produced with primers and with self-adhesive HTV ELASTOSIL R Adhesive Base 90, hardness 90 Shore A Silicone roller with primer Test results at roller function tester Line pressure N/mm
rpm
°C
Diameter mm
Abrasion
Structural break
Notice (test time, surface)
30
200
200
179
No
No
240 min, OK
40
200
200
179
No
Yes
96 min, whole rubber layer was peeled off
Result
5% adhesion to the metal core
Figure 11.4 Silicone roller with primer – rubber peeled off entirely
301
The Handbook of Rubber Bonding
Table 11.7 Silicone roller with ELASTOSIL R Adhesive Base 90 (red colour) Test results at roller function tester Line pressure N/mm
rpm
°C
Diameter mm
Abrasion
Structural break
Notice (test time, surface)
30
200
200
181
No
No
240 min, OK
40
200
200
181
No
No
240 min, OK
50
200
200
181
No
Yes
188 min, silicone layer broken
Result
100% adhesion to the metal core
Figure 11.5 Silicone roller with primer – cohesive failure in adhesive base
In this case self-adhesive HTV is used as a substitute to the primer. As it is chemically different from primers, which are not silicone rubbers, it is considered in this section.
11.7.2.2 Self-adhesive Addition Curing HTV This material cures much faster than the peroxide curing offset, which results in much shorter curing times. This makes it suitable for moulding applications. It needs not to be processed in a PTFE treated mould, if the surface of the mould has the right electroeroded structure. The reader may understand that this development is very recent and therefore no adhesion results can be provided in this review. However, one can predict that it adheres stronger to any substrate than self-adhesive LR.
302
Bonding Silicone Rubber to Various Substrates
11.7.3 HTV Used in Other Bonding Applications Using primers or undercuts the field of applications for HTV is increased dramatically. The primers can be applied to ‘normal’ solid substrates as well as to fabrics and single fibres. Depending on the scale of production, HTV can be used for all applications as described in Section 11.4. The only limitation is the speed of curing and the respective moulding temperatures. In all cases HTV cures slower and at higher temperatures than LR. In most cases this moulding temperature and curing time is responsible for the fact that it is hard to combine plastics with HTV. Most typically, metals, steel, iron, etc., will be the substrates.
• Other applications A very popular technique using HTV is the combination of moulded silicone parts together with extruded HTV profiles, tubing or mouldings. Examples of such composites are catheters used as medical devices, electrical cables, long rod insulators and composite insulators.
11.8 Processing Techniques This section will summarise the processing techniques applied in the manufacture of composite parts from LR and HTV silicone rubber.
11.8.1 Liquid Rubbers in Inserted Parts Technology The two alternative concepts for the production of composite products are the thermoplast machine and the LR machine. The prerequisite for a successful product is the choice of suitable materials. The advantages of using the LR machine are: • A higher mould temperature can be used for the LR cavity, resulting in shorter cycles, • A shorter cycle can be used, • Parts of complex geometry can be produced. The main disadvantage is investment cost and that the machine occupies a large space. As mentioned earlier, the insert parts can be produced by any apparatus. This can be extrusion with subsequent cutting, injection moulding, two or multiple colour (also two or multiple component) injection moulding, press moulding, punched and bent metal sheets, cast iron, sintered metal, ceramics in any shape, cut glass, etc., or moulded or extruded silicone rubber (as in catheters and insulators).
303
The Handbook of Rubber Bonding In order to achieve short production cycles it is advantageous to use pre-heated inserts. In some cases, primers are applied to the degreased insert in addition to undercuts even when using a self-adhesive LR. The application of such primers takes place by spraying, dipping or printing with subsequent drying and (only after drying) curing for 10 minutes at approximately 130 °C (lower for more sensitive insert materials) or higher. Hence, primered insert parts always are pre-heated. Using an unprimed LR, pre-heated inserts are advisable. This allows substantial heat transfer from the insert into the silicone, which allows faster curing. Another very important property is the thermal stability of the insert. The processor never faces problems using steel or other metals, as their thermal stability is determined by their melting point. However, phase transitions in the plastic or metallic part, e.g., from one crystal structure to another might be dangerous for the subsequent performance of the composite part. For glass or ceramics, brittleness and eventual consequences of thermal shock treatment have to be considered. The trickiest inserts are thermoplastics. They exhibit a relatively low melting or softening temperature that ranges from approximately 60 °C up to 200 °C or higher. In many cases, injection moulded, extruded or press moulded plastic inserts have incorporated ‘residual stresses’ from their method of manufacture. Such structures, often caused by process induced molecular and/or crystal orientation can relax at elevated temperatures which often leads to warping of the inserts. Table 11.8 shows some examples of melting temperatures of thermoplastic materials. These temperatures can be taken as a guide only, as in most cases softening takes place approximately 30 °C to 50 °C below the actual melting point which is quoted in the literature. It is also clear that a melt temperature will be reached only after sufficient contact time. For plastic materials, an insert should enter the hot mould at the optimum temperature that assures no or negligible change in quality when it leaves the mould after being coated with a silicone moulding. One of the most important prerequisites for successful production of composites is a very short heating time in the LR injection stage. Short heating times will be insufficient to thermally harm the plastic but long enough to allow curing of the silicone rubber. For such purposes, so-called faster curing LR have proven to be advantageous. At a given temperature and geometry, savings in cycle times (not heating time) have reached between 25 to 70% in comparison to ‘standard LR’, as indicated in Table 11.9 which shows a compilation of t90 values, the time needed to reach 90% vulcanisation in a Goettfert rheometer.
304
Temperatures Curing times Bonding Silicone Rubber to Various Substrates
Table 11.8 Melting transition temperatures of various thermoplastic materials Start of melting (°C)
Polymer LDPE
105
HDPE
132
PB
124
PP
160
PA
170
PBT
230
Table 11.9 Curing times as average t90-values for standard and fast curing LR Liquid Rubber type
ELASTOSIL LR
t90-values at 130 °C, (s)
t90-values at 150 °C, (s)
t90-values at 170 °C, (s)
Standard, general purpose (GP)
3003
50
38
26
Fast curing, GP
3004
38
27
24
Fast curing, NPC
3005
39
27
24
Standard, oil bleeding, NPC
3089
40
38
26
Fast curing, oil bleeding, NPC
3080
39
31
28
Fast curing, oil resistant, NPC
3013
33
25
23
Fast curing, low inflammability
3001
30
25
22
italic: fast curing bold: corresponding figures between two temperatures underlined: corresponding figures bold and underlined: show the efficiency of fast curing grades NPC = no post cure
The cycle time essentially consists of heating time and time for demoulding, closing, opening of the mould and metering of fresh material. Therefore it strongly depends on the shape of the individual product which is manufactured.
305
The Handbook of Rubber Bonding Table 11.9 demonstrates quite well that new fast curing grades achieve the same or a similar curing result at a certain time but at much lower temperature as the standard curing LR. Such grades are highly suitable and their use advantageous for two-component injection moulding applications. Bold and/or underlined figures in this table indicate the time-temperature relationships. It goes without saying that such fast curing technology also exhibits highly advantageous curing characteristics in ‘simple’ one-component injection moulding. In order to obtain more reliable predictions of curing times for the injection moulding process itself, the processor must conduct preliminary trials in a mould of related geometry or even a real prototype mould. Consequently, relatively short cycles at lower temperatures allow the use of polymer substrates with lower melting temperatures and less sensitivity to heat treatment as the contact times (curing times) decrease much more than the cycle times.
11.8.2 LR in Two-component Injection Moulding Technology (Two Colour Mould) This technology uses moulds with special indexing plates or handling robots that allow the transfer of the thermoplastic substrate part from the plastic cavity into the cavity for LR moulding. A specific problem in such mould design is the thermal separation – ‘cold’ in the thermoplastic cavity and ‘hot’ in the silicone rubber cavity. This is quite important and thus needs to be optimised. A peculiar fact about this moulding technique is that the plastic cavity requires a ‘hot runner’ and that the LR mould requires a ‘cold runner’. Needless to say, here we look at further two challenges for the mould designer with respect to thermal household.
11.8.2.1 Injection Moulding of Plastic Substrates In the two-component injection moulding processes the slower of the two steps is cycle determining. The thermal properties of both silicone and plastic materials require ‘compromise’ with respect to temperature in both cavities which deviates from the ideal conditions for simple injection moulding. In ordinary thermoplastic moulding, to achieve optimum cycle times for the thermoplastic part a temperature is needed that is as low as possible, or that is optimum for the injection moulding of that specific material. In two-component injection moulding for the production of thermoplastic/silicone composites, the moulding temperature for the thermoplastic should be as high as possible for two reasons. Firstly, when the thermoplastic solidifies, residual stresses are frozen in. This can lead to
306
Bonding Silicone Rubber to Various Substrates relaxation processes once the polymer is in contact with hot silicone rubber and consequently warping of the plastic substrate later. If the moulding temperature is high enough, residual stresses will generally be lower. Secondly, a higher moulding temperature for the plastic will allow its residual heat to contribute to the curing of the silicone rubber. For the injection moulding of the plastic parts, hot runners allow a waste free production. The plastic is injection moulded as in a conventional process. When sufficient solidification is reached, the mould opens, and an index plate (this is a rotating platen which takes the plastic parts from the first cavity and puts them into the liquid rubber cavity by rotating them through 180 degrees), robot, etc., extracts the plastic moulding and transfers it into the silicone cavity.
11.8.2.2 Finishing of the Composite - The Moulding of LR onto the Plastic Substrate The plastic is inserted into the hot silicone cavity. The contact surface between plastic and the walls of the cavity should be kept as small as possible, in order to minimise an excess flow of heat into the plastic. Liquid rubber is injected via a cold runner into the mould. The rubber is cured at as low a temperature as possible and the composite is extracted. The advantage of this set up is that less space needed but the disadvantage is that the cycle maybe slightly longer. It is of utmost importance to operate the liquid silicone cavities at the lowest possible temperature that allows sufficient curing times for the rubber but which will not harm the plastic substrate. A qualitative description is shown in Figure 11.6. Figure 11.6 shows the strong dependence of solidification time for the thermoplast and/ or curing time for the liquid silicone rubber from moulding temperature. It is not intended to be symmetric as it may differ for various pairs of materials. The term ‘optimum’ indicates that at a given temperature the time for solidification equals the time needed for curing the silicone. Shorter curing times in the silicone cavity lead to shorter contact times for the thermoplastic. The shorter the contact time, the higher the temperature of the silicone cavity that can be used without it being critical for the plastic substrate. In other words, the temperature of the silicone rubber mould strongly depends on the speed of curing of the liquid rubber used in the processing.
307
The Handbook of Rubber Bonding
Figure 11.6 Simplified choice of processing conditions
State-of-the-art technology allows the manufacture of such composites in 32 cavity moulds with modern cold runner technology using needle shut off valves. Materials suitable for two-component injection moulding are both normal and self-adhesive LR. At the time of the release of the component from the mould, adhesion is at its weakest level. Therefore, in the case of self-adhesive or primer technology, it is essential to demould cautiously as, once silicone and substrate are separated, no adhesion will form again. In the case of self-adhesive LR, upon cooling and after sufficient storage and/or further heat treatment, adhesion will develop to its utmost extent. This will be explained further in Section 11.11.1. It is essential that the plastic material used for the moulding of the hard substrate solidifies quickly at a high mould temperature.
11.9 Silicone to Silicone Bonding (Soft and Soft) Such applications are widely used across all fields which use silicone rubbers. The technique is quite simple. In the first step a tubing or a moulding is cured to an incomplete degree of curing. This is achieved by curing at a lower temperature or too short a curing time for complete vulcanisation. These parts are then inserted into another silicone cavity (HTV or LR) and coated with the silicone rubber. It is advisable to make silicone-silicone composites from silicone rubbers with only one curing system. In other words, a peroxide
308
Bonding Silicone Rubber to Various Substrates curing HTV silicone rubber should be combined with other peroxide curing HTV silicone rubbers. Addition curing HTV can be combined with addition curing HTV or with LR, the latter invariably being addition curing. This technique is used to produce medical catheters, components which require various combinations of two hardnesses, and technical parts, e.g., electronic keypads. As an example consider silicone keypads. The first silicone components are conductive pills. They can be incorporated using inserted part technology as well as in two-component injection moulding, the latter preferably using LR. In inserted part technology, these pills can be produced after moulding or extrusion by punching. The second component, a LR or HTV, is injection moulded onto the conductive pellets. Again, it is highly advisable to use the same curing system for both components conductive pills and electrically insulating keypad. It is quite clear that this moulding technique can be utilised to produce simple two or multiple coloured parts.
11.10 Cable Industry In some cable applications a bond between insulating silicone and conductive wire or conductive silicone is necessary for various reasons. For cables used in measuring devices and applications used under similar circumstances, adhesion of the insulator to the wire is necessary when the cable is pulled out of a plug. If there is no adhesion the insulating silicone will just come off the wire, if it is pulled strongly. A typical application is ignition cables where some automobile manufacturers specify an adhesion force of 70 N per 5 cm of ignition cable. This type of specification has been setup to assure proper adhesion when the ignition cable is disconnected from the spark plug. If a conductive silicone rubber is used instead of a copper wire, adhesion is built up during the curing process, when the outer insulating layer cures onto the inner conductive silicone lead.
11.11 Duration of Bonding Properties It is technically possible to produce very strong bonding between silicone rubber and another material. However, the silicone-plastic, silicone-metal or silicone composite obtained usually
309
The Handbook of Rubber Bonding undergoes ageing in a severe environment typical for the proposed application. Examples are oil or coolant resistant gaskets, spark plug boots, electronic housings, anode caps, etc. After receiving a set of different stresses a chemical and moreover a physical bond can weaken or even be disrupted. Earlier in Section 11.6 a number of examples of influencing factors are given to indicate what could affect the bonding process. Under normal conditions, chemical bonding will not change over time provided the article is not deformed, in particular not deformed dynamically, and provided thermal, mechanical and chemical stresses cannot influence it negatively. In such a case this bonding will persist for years, maybe decades. For mechanical bonding using undercuts or even clamping (clamping in fact is related to undercuts but it is a result of a secondary operation), the situation changes, as the material faces a stress relaxation process (also called ‘set’ of a certain mechanical parameter). This means that each composite has an initial stress distribution as moulded. This stress decays over time. As this stress is also responsible for the strength of the mechanical bond or anchorage the decay should be as slow as possible. A closely related and prominent example for that is the compression set which is analogous to the ‘tension’ set and related to stress relaxation. If an article is compressed over a certain period of time, the elastic force acting against the compression usually decays. In other words, once a material cures at a high temperature around undercuts it will be under tension upon cooling down to room temperature - either slight compression or elongation. The resultant set starts to grow from this very first moment. If a moulded part is under a certain load (stress) during its use and moreover it has some residual stresses built in during the production, such a set can become critical for the performance of the composite. In order to achieve low mechanical setting properties, the silicone rubber has to be post cured. A post cure of composites is not possible in many cases as hardly any thermoplastics will withstand a heat treatment of several hours or more at 200 °C. As an example Table 11.10 shows the compression set of various rubbers as post cured and NPC. It is quite remarkable that special NPC materials exhibit a low compression set (and hence a lower tendency to the decay of internal stresses) without the need for post-curing. This group of materials contains certain additives which result in a low compression set as moulded and after post cure. However, as NPC grades they are suitable for technical applications only. Table 11.10 contains some ranges for the compression set. Such variations originate with batch to batch fluctuation and the accuracy of determination of the compression set.
310
Bonding Silicone Rubber to Various Substrates
Table 11.10 Comparison between compression sets (CS) (Compression set; 25% deformation for 22 h at 175 °C) post cure and NPC. Duration of post cure is 6 h at 200 °C ELASTOSIL type
CS NPC [%]
CS post cure [%]
LR standard
LR 3003/60
60 – 70
10 – 20; max. 35
LR fast curing
LR 3004/60
60 – 70
10 – 20; max. 35
LR self-adhesive
LR 3070/40
60 – 70
10 – 20; max. 35
LR fast curing NPC
LR 3005/60
15 – 25; max. 30
10 – 20; max. 35
LR oil resistant NPC
LR 3013/60
20 – 30; max. 30
10 – 20; max. 35
HTV standard
R 401/60
15 – 25; max. 30
10 – 20; max. 35
HTV oil resistant NPC
R 701/60
20 – 30; max. 30
10 – 20; max. 35
All materials from Wacker Chemie GmbH
The compression set is an interesting parameter and not only relevant for mechanical bonding but also for the overall performance of a gasket or any other related application – irrespective of whether it is a simple moulding or a composite structure.
11.11.1 Duration of Bonding - Chemically Bonded Composites From investigations carried out by Wacker-Chemie GmbH it was found that bonding quality increases over time. Nonetheless, a bond cannot be guaranteed to last over a certain time under all possible circumstances. It is imperative for processors to apply sufficient testing to the bonding. Once a composite part is used under certain conditions where the function of the part essentially depends on the bonding force, sufficient testing of a prototype or a trial series is inevitable prior to use in the field. In many cases an enhanced ageing test under laboratory conditions will provide the solution. A viable solution could be the combination of primer and self-adhesive material, or even a combination between self-adhesive and mechanical anchorage. Mechanical anchorage will ensure a residual stability of the composite, and the self-adhesive material will guarantee hermetic sealing.
311
The Handbook of Rubber Bonding In many other cases adhesion is only required until the assembly of a part. In this case a growth or loss of adhesion over time becomes irrelevant. Having mentioned the factors influencing the bond between the silicone rubber and the substrate and bearing in mind the vast number of possible material combinations the necessity of preliminary extensive tests for adhesion is obvious. Figure 11.7 shows the development of adhesion of self-adhesive LR over time. During injection moulding, i.e., on a very short time scale, adhesion is quite weak. Once the part is carefully extracted from the mould and stored for a certain period of time (weeks or months) the adhesive force increases significantly - sometimes to an extent where silicone and substrate cannot be separated and the silicone rubber will break before the hard and soft components can separate. This quite long storage time leading to the desired adhesion can be shortened substantially by using a heat treatment. Figure 11.7 shows that adhesion increases with the course of time. This property is used as evidence to assume that composites made of self-adhesive materials or materials and primers will not fall apart after any time under normal circumstances.
Figure 11.7 Formation of bonding of self-adhesive ELASTOSIL LR 3070/40 as a function of time
312
Bonding Silicone Rubber to Various Substrates
11.12 Alternatives to Injection Moulding Alternatives to multi-component injection moulding are worth considering too. These techniques have not yet been referred to in the present chapter. If the available mould technology is outdated (there are only a few good mould makers capable of providing the appropriate mould design for LR and/or HTV), it is also possible to choose adhesives or even welding (the latter may be considered a special form of twocomponent press curing) for bonding. A further alternative is to use clamping technology which is related to undercuts. Some brief remarks on applications and their required properties will follow.
11.12.1 Adhesives RTV1 (room temperature vulcanising) adhesives are available for joining silicone parts to one another, or to join silicone rubber mouldings or extrusions to a number of different substrates. Adhesive joining also achieves a chemical and/or physical bond between materials which cannot be combined in one of the processes mentioned previously. As RTV1 systems cure at room temperature, the processor faces less temperature restrictions. However, curing takes much longer at room temperature than at high temperatures. It can last as long as 24 hours, depending on the geometry of the adhesive layer. In some cases, it is advantageous to change to an HTV adhesive. This will be the case if the contact surface between the joined parts is too large to provide an unhindered path for the moisture in the air - a reactant in the crosslinking process of RTV1 materials. HTV adhesives require heated parts and will only work with materials suitable in terms of thermal stability and bonding properties. PTFE and polyvinylidene fluoride (PVDF) cause problems because no suitable bonding agents exist. A number of RTV1, RTV2 and HTV self-adhesive materials are available which will not only work as an adhesive at first sight but also in ‘cured in place gasket’, a technology of growing importance in the automotive industry and many other areas. A silicone adhesive can also be used for many types of general purpose bonding of materials other than silicone rubbers.
11.12.2 Welding Welding is widely used for bonding silicone rubbers, and is closely related to injection or press moulding. However, usually one does not apply a complicated mould technology
313
The Handbook of Rubber Bonding or a process similar to HTV and LR processing. It is also related to the adhesive bonding technique previously described. The method is quite simple. Extrusions and/or mouldings are mated in a heated ‘cavity’ and an HTV material is pressed into the gap between the parts. The HTV material cures and a chemical bond remains, as the HTV crosslinks internally to the same extent as with the surfaces of the parts to be joined. Welding is a technique widely used in plastics processing, however, welding of silicone rubbers is different. In this case a cold liquid is heated to solidify by crosslinking. In other words, no melting process with subsequent solidification upon cooling is involved at all, which is a contradiction to common experiences with thermoplastics.
11.12.3 Mechanical Bonding Techniques After Moulding A silicone rubber can be bonded to a solid substrate if it is clamped into or onto it. A very good example is the baby soother. The silicone nipple is injection moulded by a common LR process and then post cured. After that the nipples are clamped into a plastic construction or even overmoulded by thermoplastic. In clamping it is essential to protect the silicone part from any mechanical damage such as cuts or scratches. These ruptures of the silicone moulding will lead to potential sources for defects during its use. Mechanical damage after clamping can also be a risk. Another example is insulators for cable connectors which have been previously injection moulded from LR or HTV. Then, they are expanded and held in that state by a plastic spiral which is removed when it is applied onto the cable (thus retaining the original shape of the insulator). A large number of similar applications can be found in the automotive, medical fields, etc. To minimise risks of mechanical rupture a range of silicone rubbers, both LR and HTV grades are available in different levels of mechanical strength and Shore hardness. Table 11.11 gives a brief survey and comparison with ‘standard’ properties. This table is simplified as it refers to tear resistance only.
11.13 Summary For additional information on general aspects of silicone rubber bonding and latest developments in silicone rubber technology, the reader is encouraged to refer to literature such as [17, 18].
314
Tear resistance Bonding Silicone Rubber to Various Substrates
Table 11.11 Comparison of tear resistance between standard silicone rubbers and high tear grades Tear resistance Hardness (Shore A) ASTM D624-00 Remark [16] [N/mm]
Material
ELASTOSIL Grade
LR standard
LR 3003/40
40
25
Addition cure standard grade
LR high tear
LR 3043/40
40
40
Addition cure high tear grade
LR standard
LR 3003/50
50
28
Addition cure standard grade
LR high tear
LR 3043/50
50
47
Addition cure high tear grade
HTV standard
R 401/40
40; Elongation at break ca. 600%
20
Peroxide cure standard grade
HTV high tear
R 420/40
40
50
Peroxide cure high tear grade
HTV high tear
R 4105/40
40; Elongation at break up to 1000%
50
Addition cure high tear grade low modulus
HTV standard
R 401/70
70; Elongation at break ca. 600%
20
Peroxide cure standard grade
HTV high tear
R 420/70
70
50
Peroxide cure high tear grade
HTV high tear
R 4105/70
70; Elongation at break up to 1000%
50
Addition cure high tear grade low modulus
There is much more that could be written on the subject of bonding silicone rubber. The main intention has been to provide a brief overview of what it is currently possible to achieve using silicone rubbers in the rubber industry. Fabric reinforced tubing and related applications have been left out, as there are too many processing techniques to be covered in a chapter of this size. This latter field is closely related to coextrusion. The remarks in Section 11.10 on cables, give some idea of what can be done with ‘general purpose’ coextrusions. This processing technique opens up a much wider set of possible material combinations. This is due to the large differences in the process of extrusion compared to injection or press moulding.
315
The Handbook of Rubber Bonding A further topic not mentioned here is the bonding of silicone on plastic inserts or moulded preforms, the surface of which has been plasma treated. Another less important technology is the necessity of priming the mould with a release coating, e.g., Teflon. These two techniques imply additional production steps and are therefore less relevant for the future of silicone bonding to plastics. As mentioned, existing technology allows the use of primerless moulds (not to mix up: primering mould = improvement of mould release properties and primer to achieve bonding). Summing up, composite technology is capable of providing simple but very elegant technical solutions. Composites have been known for decades or maybe even centuries. Composites comprising soft and hard materials are quite new and the extraordinary speed of development of their scope of applications is not only incredible but also unforeseeable. We would like to predict that such technology and in particular technology focussing on silicone composites is going to be underestimated at any time - too many substrates, an indefinite number of ideas and the creativity of material scientists account for that.
References 1.
K. Pohmer, Kunststoffe, 2000, 90, 2, 94.
2.
K. Pohmer, Kunststoffe Plast Europe, 2000, 90, 2, 34.
3.
P. Jerschow, Presented at the Processing of Liquid Silicone Rubber, Seminar, IKV Aachen, Germany, December 1998.
4.
P. Jerschow, Presented at the Injection Moulding of Composites of Hard and Soft Materials Seminar, SKZ Stuttgart, Germany, December 1998.
5.
K. Pohmer, Presented at the Injection Moulding of Composites of Hard and Soft Materials Seminar, SKZ Stuttgart, Germany, June 1999.
6.
P. Jerschow, Kautschuk Gummi Kunststoffe, 1998, 51, 6, 410.
7.
P. Jerschow, Presented at the IRE conference, Manchester, UK, June 1999.
8.
VCI, Umwelt und Chemie von A - Z, Herder Verlag, Freiburg, 1990, Germany, 8th Ed., p.136.
9.
A. Tomanek, Silicone und Technik, Hanser Publishing, Munich, 1990, p.42.
316
Bonding Silicone Rubber to Various Substrates 10. J. Comyn, Presented at the Rapra, Rubber Bonding Conference, Frankfurt, Germany, 1998, Paper No.1. 11. DIN 53531-1 Determination of the Adhesives of Rubber to Rigid Materials by the One-plate Method, 1990. 12. FDA, Code of Federal Regulations, 177.2600 13. BgVV, German Health and Veterinary Authorities, Ed., Franck Kunststoffe, 1995, Chapter 15. 14. ISO 10993-1 Biological Evaluation of Medical Devices - Part 1: Evaluation and Testing, 2002. 15. United States Pharmacopeia, current XXIII USP Class VI protocol, Washington DC. 16. ASTM D624-00e1 Standard Test Method for Tear Strength of Conventional Vulcanized Rubber and Thermoplastic Elastomers, 2000. 17. P. Jerschow, Presented at the Rapra, Rubber Bonding Conference, Amsterdam, The Netherlands, 2000, Paper No.14. 18. K. Wieczorek, Presented at the Rapra, High Performance Rubbers Conference, Berlin, Germany, 2000, Paper No.10.
317
The Handbook of Rubber Bonding
318
12
Failures in Rubber Bonding to Substrates B. Crowther (Sections 12.1 and 12.3) and K. Dalgarno (Section 12.2)
Bonds between rubber and substrates can fail for a number of reasons. Section 12.1 deals with some of the causes of rubber to metal bond failures. Section 12.2 examines the type of failures which are adhesion related, in fabric or cord reinforced power transmission belts. Section 12.3 discusses a phenomenon which causes service failures of rubber components, mainly in sealing applications. This phenomenon arises through a ‘bond’ which is formed between the rubber (nitrile) and the metal mating surface of a valve or similar, which is of sufficient strength to rupture the rubber surface when the valve is opened.
12.1.1 Introduction Bond failures in rubber to metal products are fortunately of relatively rare occurrence. When failures do occur they can stem from a number of fundamental areas and the faults are generally very characteristic of those problem areas. The main areas of bond inconsistencies and failures are: • faulty product design, • faulty metal preparation, • incorrect moulding procedures, • incorrect production quality testing procedures, • corrosion in service, • product abuse, • other failure modes.
12.1.1.1 Rubber to Metal Bonded Components Rubber to metal bonded components have been designed and manufactured since the early days of the rubber industry and their technology and manufacture has been discussed elsewhere [1]. However, the design of the rubber to metal component in the modern car engine mount has provided other problems for the rubber manufacturer not related to
319
Commercial rubbers
The Handbook of Rubber Bonding its engineering properties. Simple sandwich mountings cause no problem to the rubber to metal bonder, neither do the multi-plate products such as bridge bearings, but modern engine mounts of complex fail-safe nature can often be totally impossible to mould in one operation. This is due to the simple fact that complete bonding of rubber to the complex interlocking metals would create a product which, after bonding, would be totally locked into place in the mould, with no means of removal. This type of product can of course be made by a different bonding procedure known as partial post vulcanisation bonding. The technique is to partially bond within the mould and then to add the final metal plate externally to the mould. These bondings are more expensive because of the additional operations to complete the metal application necessary for the satisfactory operation of the unit.
12.1.2.1 Product Design and its Effects on Bond Failure All rubber to metal product designers must consider the effects of stresses exerted between the metal and rubber at the interface. This is particularly important at the edges of a component. Correct shaping of this part of the moulding will remove the stresses away from vulnerable conjunction points of metal and rubber. It is very important that rubber section corners should be radiused at the point of contact with metal components which exceed the area of the bonded rubber portion. This small modification of the rubber portion design can increase fatigue life by a factor of 7. In components which have the metal and rubber of the same dimension, then it is vital that the metal component be chamfered back to allow rubber to be moulded over a larger area than the end of the metal presents. The ideal shape for this removal of metal should be a wedge shaped chamfer which will give maximum bonding area and stress relief. This type of chamfer, which is simple to machine and does not detract from the component design or function can increase fatigue life by a factor of 6. It is particularly important in the finishing of rubber components that feather edges of rubber are not mutilated by brushes, grinding wheels or other excess rubber removing devices. Too enthusiastic an application of a buffing or grinding wheel against a metal to remove excess rubber will very quickly generate sufficient heat to successfully debond most rubber products. Excessive heating from any external source to the metal to which the rubber is adhered will result in debonding taking place. Likewise ‘nicking’ of the rubber surface can result in the tearing of the rubber back into the mass during flexing in service. If the body of the flange of rubber becomes pierced then this will allow atmospheric pollutants and corrosion creating materials to penetrate to the region of the bond and to commence the process of corrosion, usually electrochemical in nature, which will gradually undermine the bonding agent primer and destroy the bond.
320
Failures in Rubber Bonding to Substrates
12.1.2.2 Faulty Metal Preparation Many bonding failures can be traced back to faulty metal preparation, poor application of bonding agents or careless handling techniques on the factory floor.
• Metal Cleaning Poor metal cleaning, or the use of contaminated cleaning medium, can be spasmodic in appearance unless good housekeeping is practised. Metals must be thoroughly cleaned of any grease or oil protective coating applied to prevent corrosion, before grit blasting to prevent oil/grease contamination of metal cleaning grits or chemical solutions used for the removal of corrosion and foundry products. Abrasives used for the cleaning of the metals, after the removal of greases, must be monitored to ensure that the correct type and grade is being used for the application (see Section 3.2.2, Tables 3.2 and 3.3). Incorrect metals used as abrasives can leave metal particles on the cleaned surface, which will set up galvanic cells under the applied primer. Grease contaminated abrasive will apply grease to already cleaned surfaces. Round ‘shot’ must not be used as a cleaning media only grits with the correct properties (see Sections 1.1.4.1 and 3.2.2), otherwise unwanted detritus can be caught into cavities formed by the shot and then the lips of the craters ‘peaned’ over and the contaminant trapped. Finally, worn abrasive will not efficiently clean the surface of the metal and will also create considerable dust which may be left on the surface of the metal. Improperly or insufficiently treated metal being prepared for bonding, may mean that scale and corrosion are not completely removed. Most of this type of surface lying material is only loosely attached to the metal surface and can, under load in service detach itself. The result, of course, is a partial delamination of the contact area for the rubber with the metal. Once this detachment has taken place there will a gradual widening of the detached area usually accompanied by some corrosion. Application of the primer coat to freshly prepared metal must be as quick as possible to prevent atmospheric agents causing corrosion. Oxide films (corrosion) are not usually securely adhered to the surface of the parent metal and thus can be easily pulled away. If the bonding primer has adhered to the corrosion layer it too will be pulled away from the desired contact between the primer and metal. This corrosion may not be visible to the naked eye, but can result in underbond corrosion continuing after vulcanisation. Obviously ambient conditions in the metal preparation area dictate the timing and speed of primer application. Dust from the metal cleaning operation must be removed from the surface of the metal before the application of the primer. Observation of the application of the primer should also show that there has been a complete wetting out of the metal surface by the primer.
321
The Handbook of Rubber Bonding This is particularly necessary with water-based primers and bonding agents. Solvent borne bonding agents ‘wet out’ a clean metal surface with far greater ease than do the water-based products. The water-based bonding systems are much more sensitive to the presence of greases on the metal surface, and even though the metal degreasing has been carried out efficiently it will be necessary to ensure that no finger marks are found on the areas to be bonded. Solvent-based systems are capable of absorbing and dispersing small amounts of fats without causing problem to the bond. Water-based systems cannot absorb and disperse fats and therefore greater care has to be taken in the handling of metals.
• Metal cleaning using non-solvent systems The trend for metal cleaning is to move away from using solvents in general and chlorinated systems in particular. A number of alternative cleaning systems exist, but those which use water must be used in such a way that the metals are totally dry and corrosion free when being presented for bonding agent application. N.B. Beware of cross contamination of abrasive and metal treatment plant by personnel other than correct operatives. It has been known for engineering departments and individuals to not clean extremely greasy components and car engine decarbonisation parts prior to using an abrasive plant and thus to completely foul the abrasive supply. The result has been observed to cause extensive bonding problems until the contaminated abrasive was discarded and the plant cleaned of traces of grease and flakes of carbon.
• Preparation of metal surface by chemical modification (anodising, plating, sheradising) The alternative metal pre-treatment process to grit blasting uses a variety of different chemical routes. It is sufficient to say here that these can be very efficient, but do occupy rather large factory floor areas and can, if not controlled correctly, give prepared surfaces of variable quality. The usual chemical pre-treatment systems consist of acid etching of the surface followed by several water dips and subsequent phosphate or in some circumstances cadmium plating and passivating.
• Treatments for stainless steels There are various suggested systems for the pre-treatment of stainless steels which consist of treating the metal surface with strong acids to attack crystal grain boundaries in the
322
Failures in Rubber Bonding to Substrates alloys and chromium poor regions around chromium carbide particles. All the methods give surface roughness to the stainless steel, which enhances the bond to the adhesive. Mixtures of nitric and hydrofluoric acid, sulphuric or chromic acid are suggested as most suitable for this. However, the nature of the substrate alloy and the steel’s heat treatment all have a bearing on the bondability of the metal.
• Phosphate coating (see also Section 1.4) Steel is often phosphate coated for use within the engineering and decorative laminate industries to reduce corrosion. Iron or zinc phosphate can also be used. However, although for some years used as a corrosion protection technique for rubber to steel bonding, it can be difficult to control the process, resulting in a variable thickness of phosphate deposit of varying crystalline structures. If too thick a phosphate layer is obtained it becomes too friable and lacking in cohesive integrity to produce the adhesive strength required to sustain a rubber to metal bond under load during service. If only a moderate phosphate coat is produced it is often necessary to ‘passivate’ the areas of steel, only minimally covered or lacking in a coating of phosphate, with chromium oxide to prevent corrosion of the areas of minimal phosphate cover. Chromium oxide, does not readily react with a bonding agent. Chromic acid is an hazardous chemical and alternative materials can be recommended by bonding agent suppliers for the passivation or ‘sealing’ of the phosphate coating. The nature of the phosphate deposited on the surface of the steel depends to a large extent upon the nature of the steel’s microstructure and the orientation of its underlying crystal lattice. Hardened steels having a martensite structured surface (consisting of interlacing rectilinear fibrous elements arranged in a triangular shape) support a fine flake phosphate structure, whereas cold-rolled steel, having acquired a different surface orientation structure, can acquire a lumpy large flake phosphate structure, which is easily broken apart under stress. Any waste water draining from these processes is a potential pollution hazard and must be tested for zinc content, as this is a hazardous material. Any zinc present must be removed or limited to about 1 – 2 parts per million.
• Zinc coating or ‘galvanising’ Metals treated in this way are supplied to the rubber bonder already in a treated form. To be effective the zinc coating must be hot dipped to the cleaned metal, to give a ‘galvanised’ finish. Bonding to this finish is not easy. The crystalline structure of the galvanised zinc and its dipped coating thickness, can result in the flaking off, under
323
The Handbook of Rubber Bonding stress, of some of the coating, resulting in bond failure. The recommended treatment [2] for cleaning a galvanised finish is: i)
degrease metal part,
ii)
abrade the galvanised surface with grit,
iii)
degrease then apply adhesive as soon as possible.
or i)
immerse in a solution of 20 pbw concentrated HCl with 80 pbw deionised water, for 2 – 4 minutes at 25 °C,
ii)
rinse thoroughly in cold, running deionised water,
iii)
dry for 20 – 30 minutes in a 70 °C oven,
iv)
apply adhesive as soon as possible.
The second method is more widely used.
• Zinc sheradising A method can be specified which has to be carried out by specialist processors to give what is in effect a fused zinc surface to a steel component, which gives very good environmental protection for the steel component. The steel part to be bonded is baked whilst being tumbled in zinc dust. The process is not generally suitable for delicate metal parts and does cause problems with zinc build-up in screw threaded components (the latter would have to be protected by a sleeve or would require a die running down the thread to clear it). After treatment exposed zinc surfaces do of course oxidise if stored incorrectly, but this is not usually a problem.
• Aluminium - anodising Aluminium is usually anodised electrolytically, in the presence of an acid, either sulphuric, chromic or phosphoric, to give a tough resistant oxide film, which generally forms good bonds with the usual bonding systems. The anodising must be carried out with care and the type of crystalline structure being formed on the aluminium surface must be considered. A uniform reticulated structure is desired, not a
324
Failures in Rubber Bonding to Substrates microscopically fragmented rippled surface, sometimes called ‘ice flows’ [3], which is unstable, easily fractured, and therefore unable to maintain good adhesive quality. If anodising is carried out by a custom plater he will need to be informed of the type of anodised structure desired. N.B. The final stage of any ‘wet’ metal preparation process for metals to be bonded to rubber is to ensure that all chemicals used in the processes have been removed in the final water rinse tank and then to ensure that all faces of the metal parts are fully dried prior to bonding agent application. All metal storage areas must be held at least 5 – 10 °C above the dewpoint and ideally be as near to the ambient temperature in the bonding agent application shop, which itself should be in the region of 18 – 22 °C minimum.
12.1.2.3 Application of bonding agent to metals Metals must be completely grease and dust free when the primer is applied. The primer must be properly prepared, especially if new supplies are being prepared from bulk containers, to ensure that all the constituents of the bonding primer system is in suspension when being applied to the metal surface. The application of the primer is the most critical part of the metal pre-treatment process, for if not correct, bonding will either be patchy or non-existent. If spray application is being used, especially hand spraying, then care must be taken to ensure that the primer spray hitting the metal surface is capable of wetting out the complete surface and not ‘dry’ (loss of all solvent) at the time and point of contact with the metal. If ‘dry’ then ‘cobwebbing’ (the condition of the bonding agent resulting from drying out before reaching the substrate surface and is in the form of fine filaments) will occur and although the metal may appear to be covered, poor and patchy bonding will result. Application of the primer to the metals must be carried out in such a way that there is no possibility of entrapping air between the primer and the metal. Any such trapping will act as a buffer between the primer and the metal and no bond will be achieved. The bond between the primer and the metal is essentially a mechanical one taking place around the asperities of the cleaned metal surface; although some degree of chemisorption also takes place. Intimate contact with the metal is therefore essential for good bonding to take place.
• Addition of primer and rubber bond coat Bonding agents, especially the primer systems, are very prone to settlement during storage. The result of the settlement of the suspended materials to the bottom of the drum is the formation of a layer of quite hard material which is very reluctant to
325
The Handbook of Rubber Bonding return to the suspended state. It is therefore important that as soon as possible after delivery, and certainly some 24 hours prior to use, the drums contents are agitated continuously to achieve re-suspension of the deposited materials. Drum rolling on ball mill rollers for 24 hours prior to 24 hours stirring with a motor with suitable paddles to lift the suspended material is the only way to ensure that satisfactory reproducible bondings are continuously achieved. At the end of the stirring period the critical visual quality control must be carried out together with a cone bonding test to ensure quality of the bonding agent. Viscosity and flow cup checks do not fully indicate the state of the dispersion of the vital ingredients of the bonding system and cannot readily be used in any case with the water-based systems. Hydrometer measurements are suggested by bonding manufacturers for measurement of the degree of dispersed materials with these emulsions. If multiphase waterborne systems are being used then special considerations concerned with storage conditions and stirring must be observed. Having a surfactant system incorporated to facilitate the preparation and stability of the product, it is necessary to strictly observe the manufacturers instructions, as otherwise flocculation will occur. Once an adequate suspension has been achieved it is then essential to keep that suspension operational whilst treating the metal components. The ideal method is to use a continuously agitated system which does not allow settlement but is not so agitated as to cause bubbling, especially with water-based materials. Before the application of the rubber bonding adhesive layer to the primer it is necessary to ensure that contamination has not occurred to the primer surface in the drying and storage period. Dust should be removed and any parts with embedded contamination in the primer layer must be recycled, otherwise failure will result. Good housekeeping is essential to achieve a well run bonding shop. The rubber bonding coat, which is usually the second layer of bonding agent to be applied to the metal, must only be applied when the solvent from the application of the primer has all evaporated. Failure to allow adequate drying time for the primer coat will result in either bubbling of the surface of the primer or a separation of the two bonding layers at the interface. This is usually seen as polished bonding agent surfaces being evident on both rubber and metal. Application of tie cements to the surface of the rubber bonding coat to assist with the bonding of difficult compounds also need care, to ensure that the solvent is totally removed from the various bonding agent layers before passing the prepared metals to the moulding presses.
326
Failures in Rubber Bonding to Substrates The addition of the primer and bonding coat to the metal should be strictly within the specified thickness parameters laid down by the supplier of the bonding agent system. Too little bonding agent or over-diluted systems being used for spraying application will often be insufficient to give an even coating of the bonding agent to cover the hills and valleys of the asperities created during the cleaning process. It is vital that all ‘peaks’ are covered adequately and evenly by the bonding agent otherwise bond failure by tearing can be initiated from such peaks. Conversely the addition of too much bonding agent is unwise not only economically but also from a bonding integrity point of view. The primer is a rigid coat when vulcanised and the bonding coat only semi-flexible and thus too much of either system coating will tend towards a weakness line being formed at the juncture of the bonding agent systems. This fault is usually characterised by bonding agent attached to the surface of the rubber in the failure area being visible. Incomplete drying of the layers of bonding agent prior to the application of the next coat is visible in a similar manner to that described for too much bonding agent being used. Often the detached bond will show that there have been bubbles formed in the bonding agent layer with the result that the outer bubble layers adhered to the rubber are brittle and easily fracture, with resultant bond failure. This problem is much more prevalent with the water-based systems, although heated metals help. N.B. Use of waterborne bonding systems has some major requirements which differ from the way solvent-based systems are handled. Failure to follow the following rules may well result in significant bond failure problems: • Initial storage requires precautions against low temperatures which will result in possible freezing or flocculation. Similarly too high a storage temperature will have similar effects on flocculation. • Too violent agitation of adhesive will result in flocculation. • ‘Dried’ waterborne bonding agent solidified on the edges of a drum, and on equipment, cannot be redissolved back into solution (differs from solvent-based system). Consequently dried agent MUST NOT BE ADDED BACK INTO A SOLUTION, as otherwise failures will result from ‘bits’ being deposited with the bonding agent onto the metal. These ‘bits’ will not bond and will inevitably result in localised bond failures. • Cleaned metals need to be heated to help water to evaporate. It may also be necessary to provide some degree of forced drying to ensure economic drying times.
327
The Handbook of Rubber Bonding • Dilution of the waterborne systems requires deionised or distilled water, and definitely not tap water. • Cleanliness of metals is more critical when using waterborne systems. Solvent-based systems can absorb very small quantities of oily contamination without affecting bonding quality. Waterborne systems cannot absorb oils but are repelled by them, and if small traces of oils are present they will prevent the bonding agent ‘wetting out’ the surface and will result in ‘patchy’ cover of the metal by the bonding system.
12.1.2 Incorrect Moulding Procedures It is unlikely that incorrect vulcanisation time will result in poor bond strength, because of normal factory control systems. An exception can occur when an incorrect batch of compound has reached the presses. If correct control limit gates for compound viscosity and Mooney scorch are applied for the compound then the cure-rate should have been designed to allow for adequate compound flow, and consolidation in the mould cavity prior to onset of any cure, to ensure that totally unvulcanised compound comes into contact with the bonding agent surface. If this condition is met then the ability to generate the maximum bond between the bonding agent and the rubber will be maximised. Any compound condition other than this will result in variable bonding values and could lead to field failures in service. Quality control procedures should be such that this cannot happen. Incorrect metal handling techniques or improper storage of prepared metals can be a serious problem at the moulding press and can be the cause of subsequent failures. This type of fault is the responsibility of the manufacturing and quality control departments and should be quickly corrected. If control is lax, dust layers on prepared metals, or contamination by sweaty and greasy moulders gloves, can give bond failures. Greasy thumbprints on metals or rubber blanks can have similar effects. Some organisations prefer to heat treat (bake) metals which have been treated with bonding agents, to ensure that the degree of plasticity still present in the bonding agent does not enable it to be swept off the surface of the metal whilst the rubber compound is being injected or transferred into the mould cavity. Careful mould design can usually eliminate most bonding agent sweeping problems by careful direction of the flow of the injected rubber compound. If it is necessary to prebake the bonding agent to the metal surface, care must be taken to ensure that the prebake does not fully crosslink the bonding agent and thus leave no crosslinking possibilities between the bonding agent and the rubber. Overbaked bonding agent systems usually give bond failures which leave the rubber/bonding agent system faces quite shiny and polished (see also Section 4.3.2).
328
Failures in Rubber Bonding to Substrates It is far more likely that incorrect component removal techniques from the mould cavities, causing overstressing of a very vulnerable bond, will affect the integrity of the component. One problem is the over-enthusiastic application of bonding agent causing overspill onto metal edges which come into contact with flat areas of the mould. Bonding agent will bond to mould surfaces just as well as to component metals and can cause severe product removal problems which will result in stressing of the components. Levering out of mouldings with metal tools, instead of hydraulic ejectors, will also give uneven stresses to the bond. The other very significant danger in moulding of rubber to metal components is the prescribed or accidental application of release agent to the mould cavities to ease release of the finished product. Release agents can be trapped between the rubber and the bonding agent faces and prevent an adequate bond being achieved. Ideally no release agent should be used in a mould producing bonded components, neither should it be used in the vicinity of such presses. Moulding shop air contamination by minute droplets of release agents is very easily achieved and these will condense on any exposed metal surface. Ideally the press shop producing bonded components should be at a positive pressure to neighbouring shops to prevent cross contamination taking place.
12.1.3 Incorrect Production Quality Testing Procedures The forming of the required shape and the vulcanisation of the rubber in a rubber to metal bonding process or the production of any rubber component, does not complete the crosslinking and full attainment of the ultimate network structure. Cure times for products are calculated using curemeter information to usually 90 or 95% cure. The 5 to 10% additional cure comes from retained heat in the product after demoulding has taken place. Rubber to metal bonded components are required to mature after vulcanisation, to enable the full crosslinking, structural networking and chemisorption linkage processes to take place between the various layers of the metal/primer/bonding agent/rubber complex. Some structural strength within the bond is achieved immediately after full vulcanisation time has been reached, sufficient to allow removal from the mould. However, it is important to ensure that full stressing and testing of the bond be carried out only after a period of 24 hours has elapsed. Rubber has a poor thermal conductivity and thus rubber to metal bondings which often have a low surface to volume ratio, i.e., large rubber volume, take a considerable time to cool to ambient temperature. This heat retention is often compounded by the close location to the vulcanising press of the skip into which the product is placed at the end of the cure cycle, the moulding shop ambient temperature, and the piling up of hot product as the work shift continues.
329
The Handbook of Rubber Bonding Vulcanisation of the rubber does not cease completely at the end of the press/cure cycle. As the temperature of the moulded product reduces, the rate of vulcanisation decreases, but it does still continue, even into service life. Many maturation reactions take place as the product returns to ambient temperatures and even in service some crosslinking reactions will still take place over an extended period of time. The filler, usually carbon black for most components, will form its own reinforcement structure over a period of time, thus contributing to the required product properties. Testing of the product too early can result in overstressing of the rubber in critical stress zones and this in itself can lead to premature failure of the product in service. Particularly prone areas of bond stress in the early part of the components existence are concentrated by protrusions of metal components, such as bolt heads, into the rubber mass. An extensive load created in the area of such protrusions concentrates in their vicinity and localised bond failure can occur. This may not be evident to the observer but cavitation may have been created which can expand under stressing of the product in service, with resultant complete failure through rubber tear. Proof testing of product after manufacture using a calculated limited stress can, if incorrectly carried out, be a source of subsequent product failure. It is necessary to calculate a suitable set of stress parameters to use for testing the product, i.e., a load and extension, such that the product does not become overstressed and does not retain permanent evidence of damage as a result of proof testing. This can appear as distortion or incorrect dimensions.
12.1.4 Corrosion in Service Corrosion in service can occur from a number of different sources: • electrochemical attack - salt or similar, • galvanic sources - underbond, • overheating, • chemical attack.
12.1.4.1 Electrochemical Electrochemical corrosion of a metal beneath a rubber coating necessitates the presence of an aqueous phase. There will also need to be anions and cations to provide conductivity in the aqueous phase and the presence of oxygen to support the cathodic
330
Failures in Rubber Bonding to Substrates reaction. To be present at the site of potential corrosion the materials have to find passage through the rubber layer, usually by migration. If an electrical potential is present the migration of water through the rubber is facilitated [3]. A similar effect occurs influencing the migration rate for cations by chemical type, i.e., Ca+ migrates faster than Na+. Thus the rate of cationic delamination occurs at a faster rate in calcium chloride solution than in sodium chloride.
• Anodic Residual stress in the metal component gives rise to an anodic condition which is largely eliminated when the metal is fully annealed. Cold working of a metal by plastic deformation will result in the electrical potential of the metal being moved in the direction of the areas of greatest activity and will be anodic in character. Incompletely cleaned metal surfaces can also be the source of potential differences. Presence of dissolved gases and the formation of protective layers during the corrosion processes are all factors affecting the generation and degree of the corrosion and its spread over the metal surface. Anodic undermining can occur beneath a rubber coating or between the components of a rubber to metal bond. Anodic undermining and delamination occur very slowly in comparison with cathodic delamination.
• Cathodic Alkali can be generated by the cathodic half of a corrosion reaction or the cathodic reaction may be driven by means of an electrical potential. When the cathodic reaction occurs between the rubber and metal surface the pH of the solution under the rubber may be as high as 14. Many factors (summarised by Leidheiser [3]) concerned with cathodic delamination are detailed. No definitive mechanism for this type of delamination has been determined although a number of suggestions have been put forward [3]. These include alkaline attack on the polymer, surface energy considerations and attack of the oxide at the interface. Moisture absorption into textile and fibre reinforcements in specialised seal compounds is often possible where exposed fibre ends are exposed to the surface of the rubber compound, either during moulding or subsequently during trimming operations. As the amount of moisture absorbed alters with atmospheric conditions of humidity and temperature there will be a variation in the activity of the electrolyte formed in the textile/rubber and thus the corrosivity of the compound to contacted metals.
331
The Handbook of Rubber Bonding
12.1.4.2 Galvanic Attack Galvanic attack usually occurs as the result of an incorrect choice of abrasive metal cleaning media. Incompatible metals used in assembly of components in service can also result in galvanic cells being generated and the resulting corrosion will, with time, cause bond decay.
12.1.4.3 Overheating The bond between rubber and the metal surface is essentially a mechanical one achieved by the use of various resins and chlorinated rubbers, which when heated and crosslinked within their mass form a rigid network. Subsequent application of a high temperature directly to the area of the bond will result in the loosening of the grip of the primer material, and under the application of a load the bond will fail. This method is sometimes used to reprocess metals from components which have been rejected by quality tests.
12.1.4.4 Corrosion by Chemicals - Not Electrochemical This type of attack can be from a variety of sources, some of which can be deliberately applied to the component. Application of unsuitable metal finishing paints or metal preservatives to the moulded component can soften the primer and/or bonding agent layer with the subsequent loss of product integrity and failure of the component. This type of corrosion can also occur, fortunately infrequently, in rubber to metal bonded components which have been designed to meet specific, often onerous service conditions. High levels of plasticisers necessary to meet very low temperature operating conditions have been known, after long periods of service, to migrate through the layers of bonding agent and effectively debond the component. The mechanism of this type of debonding is not fully understood. However a possible explanation which can be given is as follows. Primer to metal adhesion is fundamentally a combination of a mechanical keying, with a degree of chemisorption taking place. Bonding primers can be based on a complex system of materials with high resin, high chlorinated rubber combinations which probably give an interpenetrating type of structure. Obviously the penetration into and through such a primer layer by a plasticiser from the rubber layer will have the effect of softening and loosening the grip of the primers contact with the metal surface irregularities. If the bond adhesion failure occurs at the primer/bonding agent interface, a very similar explanation can be put forward. At this interface the reactions taking place are a combination of interdiffusion, adsorption and chemical crossbridging and migration of the plasticiser
332
Failures in Rubber Bonding to Substrates can cause problems of debonding by either softening of the crossbridging or by chemical interference. In addition of course there may be a chemical attack at the metal surface of the bonded component by the migrating plasticiser which will cause pitting and erosion.
12.1.5 Product Abuse It is difficult to be precise about details of product abuse, as the form this takes can be wide-ranging. Some of the abuse to which rubber to metal bonded products are submitted is totally preventable whilst others are purely accidental. The most frequent cause of failure are: • exposure to high temperatures, • exposure to oils and solvents, • painting to match machinery to which bonding is fitted, • serious misalignment of flexible couplings, • service loading too high compared to original specification, • corrosive ambient service conditions not originally specified, • mechanical interference, e.g., grinding of outer metal surface generating high heat. These are a selection of the problems that can occur to cause either failure or unsatisfactory operation, through ignorance of the properties and susceptibilities of rubber to metal bonded components.
12.1.6 Other Failure Modes Evidence of metallic particles clinging to the rubber surface can be an indication of a number of factors: • poor cleaning and removal of foundry materials from the surface of the metals prior to the coating with bonding agent, • failure to remove all ‘dust’ from metal cleaning from metal surface prior to addition of first bonding agent coat, • poor adhesion of plating applied to metals as decorative or anticorrosion finishes, • inadequate removal of zinc dust if zinc ‘sheradising’ has been used to give metal corrosion resistance in service without painting,
333
The Handbook of Rubber Bonding • incorrect coating being laid down during phosphating process of metal treatment, resulting in flaky, porous layer being deposited on the metal. The strength of this flaky coating is insufficient to resist workload stressing of rubber to metal bond, • incorrect anodised coating on aluminium metals which resists bonding to the primer. Some decorative anodised finishes have this problem.
12.1.7 Factors Affecting Adhesion of Rubbers The adhesion of rubbers is further complicated by the rheology of rubber and its added compounding constituents. Blending two or more rubbers will give more complexity to the flow and adhesion behaviour of the system. The ability of the rubber compound to flow and establish intimate contact with the substrate is of paramount importance, to enable the forces discussed previously to establish the conditions required for optimal adhesion. Any material which interferes with the establishment of the interface must be avoided. The establishment of a stable, strong interface between the two materials is the foremost requirement for successful adhesion. The processes which are likely to determine the interface formation are: • the ability of the adhesive/rubber mass to flow uniformly, • complete wetting of the adherend by the flowing adhesive/rubber mass, • the stabilisation of the rubber/adhesive interface.
12.1.7.1 Rubber Flow The ability of the rubber to flow must be a priority in the compound design. Without the ability to flow readily the rubber compound will not be capable of wetting out the surface of the adhesive film and thus the necessary interfacial contact will not be completely achieved. Reduction of the molecular weight of the rubber is beneficial, but other means of promoting flow may be problematical, i.e., addition of oils and plasticisers. Oils, plasticisers and waxy materials will almost certainly exude to the rubber surface either immediately after compound preparation or during the establishment of the interface. Similarly some plasticisers are known to travel to and into the interface layers with time, breaking down the adhesion and resulting in complete failure of the bond.
12.1.7.2 Complete Wetting of the Adherend Complete wetting of the adherend surface is essential if the best interfacial bond strength is to be achieved. However, because of incomplete wetting not all of the surface transforms
334
Failures in Rubber Bonding to Substrates into an interface. The surface energies of the adhesive mass and the adherend rarely coincide and thus ready and complete wetting is rarely achieved. The areas of incomplete interfacial bonding impairs bond strength which in turn leads to bonding defects or failure.
12.1.7.3 Stabilisation of the Rubber/Adhesive Interface It is necessary to establish the greatest area of interface between the adherend and adhesive mass. A high contact pressure between the adherend and adhesive mass will ensure the formation of an interface despite the adherend being resistant to wetting purely on grounds of interfacial chemistry. The yield of the interface is proportional to the contact pressure, though naturally the interfacial area will decrease and the surface area increase when the contact pressure is removed. For a stable interface to be retained an area of interface must exist in a stable condition after removal of the contact pressure. Stabilisation of the interface requires that the excess energy in interface formation is minimised by binding between the dissimilar surfaces (for example rubber and metal). Thus the bond in rubber adhesion results from the formation and stabilisation of the interface via the three processes outlined previously.
12.1.8 Topography of Substrate The topography of a substrate is considered to be a direct influence on the occurrence and degree of adhesion. The adhesion is created by the adhesive being locked around protrusions and into cavities of the adherend surface. The rougher the surface, the greater the adhesion strength. The cleanliness of the surface, the ability of the adhesive to wet out the total surface area in contact and the viscosity of the adhesive being critical factors in the achievement of good bond strength. The viscosity of the adhesive must be sufficiently low to enable it to flow into the ‘valleys’ and to ensure that a minimum of air, or solvent vapour, is trapped in the depths of a cavity, but be sufficiently high to ensure that asperities in the surface to be bonded are adequately covered. Failure to achieve the correct adhesive viscosity and a maximum adhesive/substrate coverage will diminish the overall bond strength.
12.1.9 Surface Conditions of Adherend The effecting of intimate contact of the surface of the adherend to be bonded and the adhesive depends on two major factors: • the surface roughness of the adherend, • the flow characteristics of the adhesive.
335
The Handbook of Rubber Bonding
12.1.9.1 Surface Roughness of the Adherend Rowe [4] argued that many substances are polycrystalline in nature, as, for example, the oxide layers on metals and in roughening metal surfaces the crystalline form of structure is likely to persist, but with a different size of crystal. Wake [5] points out that the empirical fact remains that for metals, the apparently rougher surface given, as the particle size of grit used in grit blasting is increased, the poorer the adhesion obtained. This is probably because although visually the surface of the adherend appears rougher, in fact the irregularities themselves have smoother walls when viewed under an electron microscope. Even polished, lapped, metal surfaces have irregularities of about 10-7 m and such pits can trap air. If no air is trapped in the pits of the metal then one can see that the surface wetting and adhesive coverage would be complete. Any entrapment of air will lead to incomplete adhesive/metal coverage and much will depend upon the contact angle of the adhesive and capillary attraction. The critical angle for perfect contact between adhesive and substrate approaches zero.
12.1.9.2 Flow Characteristics of the Adhesive The viscosity of the adhesive is obviously a significant factor in the correct application of the correct and complete contact layer on the adherend. A number of adhesive factors will need to be addressed: • The optimum viscosity of the adhesive will need to be determined by experimentation to achieve optimum bond strength. • The application temperature of the adhesive should be consistent and not vary widely due to ambient fluctuations. • Similar controls are desirable for the substrate to ensure consistent adhesive flow. Too low a temperature for the substrate would reduce adhesive flow rate, while too high a temperature would result in too much flow, causing adhesive to drain away from asperities (peaks). A further complication would be encountered when using solvent-based adhesives in possible solvent evaporation causing increased viscosity. Solvent loss can occur during application through temperature fluctuations, or simply by evaporation from open adhesive containers in the case of the more volatile solvents.
12.1.10 Classification of Rubber According to their Wettabilities Lee [6] classifies rubbers by wettability, ranking them according to Zisman’s critical surface tension γc (mJ/m2), see Table 12.1. 336
Failures in Rubber Bonding to Substrates
Table 12.1 Classification of rubbers by wettability Rubber
Critical surface tension γc
Isobutylene - isoprene
27
cis-Polyisoprene
31
cis-Polybutadiene
32
Styrene butadiene
33
Epichlorohydrin
35
Chlorosulphonated polyethylene
37
Acrylonitrile butadiene
37
Polychloroprene
38
12.1.11 Bonding – Interphase or Interface Considerations The interphase is a thin region which exists between the bulk adherend and the bulk adhesive [7]. A surface oxide, either native or one produced by pre-treatment, is present on most metal adherends. These oxide layers will frequently be contaminated, even after cleaning. The net effect of absorbed layers is: • that the adsorbed layer dominates over the bulk material for separation less than the thickness of the adsorbed layer, • that the bulk dominates where the separation is large compared with the adsorbed thickness, • that the adsorbed layer not only acts as a spacer but causes additional screening of the bonding reaction [8]. A primer is often applied in a production process after pre-treatment and before the application of the adhesive. Typical thickness for the oxide are 0.003 – 0.4 μm, for the primer 5 – 10 μm and for the adhesive 10 – 15 μm. The interphase region is expected to have mechanical properties different from either the adherend or the adhesive. Filbey and Wightman [9] comment that measurement of these properties is important in understanding adhesion, for example, poorly durable bonds are often a consequence of
337
The Handbook of Rubber Bonding poor interphase properties. Thus, one of the frontier areas in adhesion science today is determining interphase properties. Interfacial imperfections, such as trapped air bubbles [7, 10] can be the site for high localised stresses. The air responsible for the bubbles may have been trapped into the adhesive during mixing or occur when bridging of high viscosity adhesive occurs over cavities in the substrate surface. Some degree of interfacial imperfection may arise from even normal curing conditions, because volume changes can occur within adhesives en route to full cure. Gaseous by-products and solvent residues can also be the source of problems. Plueddemann [11] has detailed the stresses which can occur due to adhesive shrinkage and coefficient of shrinkage difference between adherend and rubber in the interfacial area. Minford [12] discusses at some length cohesive versus adhesive failure. On the topic of failure due to water desorption of the adhesive Laird [13] has shown that water can progress by diffusion along the interface as much as 450 times faster than by permeation. Adhesive systems can also be sensitive to certain of the strong polar solvents as they contain polar elements themselves. Dilution of an adhesive can be achieved by the use of this principle but after the adhesive has been cured in the bondline the same solvent can attack the adhesive and destroy the adhesion at the interface. A further interfacial factor can be the presence of non adsorbable or non desorbable contaminating films (as previously mentioned above) at the interface. Such materials can be oils, fatty acids, plasticisers from the rubber and metal processing oils from inadequately cleaned metal components. Some of these lubricants can be absorbed by the adhesive if it is solvent-based but in the case of the new waterborne rubber to metal systems this absorption cannot take place, for the systems are neither miscible or compatible with oils. These new waterborne systems have a critical tolerance level for surface contamination of the metal and if this is exceeded then wetting out of the metal by the adhesive will not, at the worst be possible, or at the best complete. When examining the surface of a failed bonding it is extremely difficult to determine whether the failure has taken place at the original interface or whether a new interface has been opened, either in the adhesive or in the rubber, for the distance from the old to the new interface can be extremely small. Thus it may not be possible to determine whether contamination of the interface was the cause of the failure or not. Examination of the surface of the failed bond using very specialised equipment such as Secondary Ion Mass Spectroscopy (SIMS), Ion Scattering Spectroscopy (ISS) and Auger Electron Spectroscopy (AES) can greatly assist with determination of some of the causes of this type of failure.
338
Failures in Rubber Bonding to Substrates
12.1.12 Problems in Adhesion Most polymeric adhesive systems undergo some degree of shrinkage during the process of crosslinking and this will create stresses within the adhesive/substrate interface. Adhesives with filler content have a lower shrinkage and thus lower stresses. The presence of water at the interface will cause interference with the establishment of a good bond. Some materials occurring as the substrate are hydrophilic in nature and will always attempt to achieve a monomolecular water film on their surfaces. This can occur, after the substrate has been cleaned and before the adhesive has been applied, from the atmosphere direct to the substrate surface, or later after the bond has been established, by diffusion. The presence of a water film at the interface can result in leaching of materials from the substrate which then can cause corrosion of the substrate, with resultant progressive loss of adhesion as corrosion spreads under the adhesive. Removal of solvent from the adhesive layer can present problems especially if evaporation from the adhesive surface is prevented in some way. If the bond is made before most of the solvent has been removed it will result in significant internal stresses in the adhesive and also lower adhesion strength. If the layer of adhesive is thick and solvent loss from the outer surfaces is rapid, viscosity changes in the outer adhesive layers will also hinder solvent loss. If the adhesive also is of the crosslinking type there can be some solvent left indefinitely internally. Crosslinking of the adhesive restricts molecular chain movement and thus prevents the movements necessary to allow the solvent molecules to travel through the mass. Solvent remaining trapped in the adhesive, usually close to the interface, results in a softer than expected adhesive/substrate interface and thus allows easier breakdown of the adhesion by stripping, etc. Choice of the solvent for the adhesive and its molecular configuration and thus size plays a considerable part in the establishment of a satisfactory bond.
12.2 Rubber Bonding in Power Transmission Belting 12.2.1 Introduction Power transmission belts are largely rubber composites which transmit mechanical power between shafts through either friction or a combination of friction and the engagement of formed teeth in pulley grooves. All power transmission belts have fibre reinforcement of some description, and so they all rely to a great extent on the quality of the bonding between fibres and rubber compounds in order to function correctly. However, there is relatively little literature concerning the role of the adhesion system in ensuring that belts retain structural integrity throughout their working life. This chapter reviews the role of
339
The Handbook of Rubber Bonding the adhesion system in belting applications, considering synchronous belts, V and Vribbed belts and conveyor belts. Whilst conveyor belts are not strictly power transmission belts some consideration of the failure of these belts has been taken into account in characterising power transmission belt failure modes. The section begins with an assessment of which belt failure modes are adhesion related, before examining the adhesion systems used in belting applications, and the tests which can be used to examine adhesion strength in belts. As the chemistry of rubber bonding is covered elsewhere in this volume the emphasis within this chapter is on mechanical and applications-based issues rather than on the chemistry of the adhesion systems.
12.2.2 Power Transmission Belt Failure Modes Four different types of belt are considered: synchronous belts, V-belts, V-ribbed belts and conveyor belts. Each type of belt has a distinct set of failure modes and so each is considered in turn below. Only failures which can be considered to be belt failures have been considered, rather than belt/pulley system failures, so that failures as a result of pulley misalignment, for instance, have not been included. For adhesion related failure modes, methods of predicting belt failure where they have been developed, are outlined.
12.2.2.1 Synchronous Belts Most of the literature pertaining to synchronous belts considers the automotive application of these belts and so this section inevitably has an automotive bias. Figure 12.1 shows the structure of a synchronous belt. As Figure 12.1 shows the belt is made up from an rubber compound with two reinforcements, the first being a tension bearing cord which runs through the middle of the belt, and the second a facing fabric which covers the belt
Figure 12.1 Synchronous belt construction
340
Failures in Rubber Bonding to Substrates teeth, strengthening them and protecting them from wear. The tension bearing cord is normally glassfibre or aramid, with the facing fabric normally polyamide. The belts transmit power through the engagement of the moulded teeth in pulley grooves. The belts are manufactured by first putting a fabric sleeve around a mandrel which has the belt teeth formed in it. The cord is then helically wound along the mandrel (this means that in a finished belt the cords do not lie along the axis of the belt but at a slight angle to it). The rubber compound (mixed but uncured) is then placed around the cord before an outer casing applies pressure to force the rubber compound into position and heat is applied to cure the rubber compound once it is in position. Once the curing process is complete individual belts are cut to width from the stock. Similar methods are used to manufacture most types of power transmission belt. Test results from experimental studies on belt life suggest the following as the major belt failure modes: tooth root cracking, wear, cord failure and fabric separation [14, 15, 16, 17, 18], and this classification has support from field data [19, 20]. Figure 12.2 shows examples of tooth root cracking, cord delamination and fabric separation failures. Tooth root cracking is the prevalent failure mode for belts and the literature suggests two mechanisms for its generation. The most commonly reported mechanism is fatigue and eventual failure of the facing fabric in the tooth root, followed by rapid crack propagation through the tooth rubber compound, usually across the rubber compound/
Figure 12.2 Synchronous belt failure modes. a) tooth root cracking, b) cord delamination, c) fabric separation
341
The Handbook of Rubber Bonding cord interface leaving the tooth (if it remains on the belt) unable to support load and the belt unable to transmit power. Kido [21] point out that this mechanism is often accelerated by wear of the fabric, leaving less material to support the loads. Iizuka [18] reports on both the mechanism outlined above and on a second mechanism to explain tooth root cracking, in which the failure initiation site is at yarn interfaces in the cord; cracks originating in this region propagate first into the tooth rubber compound and then out to the facing fabric in the tooth root. Iizuka observed this second mechanism in belt life tests at torques of around 5 Nm, with facing fabric fatigue observed in tests carried out at a higher torque, and relates the initiation of cracks at the yarn interfaces to local maxima in the belt curvature whilst engaged on the pulley [22]. There are also a number of reported mechanisms for cord failure in the belts. The most obvious of these is the belt tension being so high that the cords simply cannot support the load, as identified by Koyama [14], while Murakami and Watanabe [23] suggest that localised bending in the cord causes debonding within the cord, resulting in interfilament abrasion and a reduction in the tensile strength of the cords leading to cord failure. Dalgarno [17] observed belt failure through debonding of the cord at the sides of the belt, leading to failure when the cord at the side of the belt becomes trapped between belt and pulley. One further source of belt tensile failure is back cracking [21, 24]. In this case failure is initiated through cracking in the back cover of the belt, generally associated with significant ageing of the rubber compound. The cracks have been observed to run across the back of the belt, and once the cracks have propagated to such an extent that the belt cords are exposed, belt tensile failure follows. Fabric separation failure occurs when the belt teeth and fabric land become detached from the belt cords [17] and is essentially seen as purely an adhesion failure, although there may be links between this failure mode and the tooth root cracking failures observed by Iizuka [18], originating from cracks developed in the cord itself through internal delamination. Wear causes belt failure through changing the tooth profile to such an extent that the belt teeth can no longer support the required load [25]. Overall it is interesting that adhesion within the belt can be the root cause of most types of failure, with fabric fatigue (often accelerated by wear), wear itself, and rubber compound cracking, the root causes of those failures not related to adhesion. The observations made by various researchers suggest that both the cord/rubber interface and the fabric/rubber interface are potential failure initiation sites, with interyarn and interfilament adhesion within the cord providing further possible failure initiation sites. In attempting to identify parameters which allow the belt life to be predicted within the adhesion related failure modes identified above, the most common approach has been to use measures of belt distortion. Dalgarno [17] examined belt life data from belt failures within the tooth root cracking, fabric delamination and cord separation failure modes,
342
Failures in Rubber Bonding to Substrates and showed that the belt tooth deflection correlated well with belt life regardless of the failure mode. Childs [26] developed a failure analysis based on a combined tooth deflection and bending distortion measure, and Iizuka [27], in examining the same data as Childs [26], used the curvature of the belt as the parameter to which belt life was related. Both Childs and Iizuka show that there is good correlation between belt life and their distortion measures, and Iizuka concludes that the two measures may in themselves be related.
12.2.2.2 V and V-ribbed Belts • V-belts Recent literature on the failure of V-belts has concentrated on raw edge V-belts of the type shown in cross section in Figure 12.3, this type of V-belt being the most commonly implemented belt in automotive applications. As Figure 12.3 shows, the V-belt consists of three distinct zones, a rubber compound which forms the bulk of the V, the tension cord (normally polyester) embedded in a softer cushion rubber, and a rubber impregnated fabric. In addition the rubber compound may be axially reinforced by chopped fibre to give the rib a greater resistance to deformation in the axial direction. To improve the flexibility of the belts the belt may be ‘cogged’ as shown in Figure 12.4.
Figure 12.3 Typical raw edge V-belt cross section
Figure 12.4 Side view of cogged V-belt
343
The Handbook of Rubber Bonding The failure of these belts has been studied in detail by Gerbert and Fritzson [28, 29]. With support from field data and laboratory tests they identified four common failure modes for the belts: tensile failure, cord separation, radial cracking, and wear. Belt tensile failures were associated with short belt lives and may be assumed to occur where the belt has been underspecified, or the drive is ill-conditioned. Cord separation was defined as crack propagation along or across the belt, close to the cord layer. Debonding between cord and cushion rubber, and between cushion layer and the V rubber compound were both observed, together with crack propagation in the V rubber compound and in the cushion rubber. Edge cord separation was also observed, and this is thought to be a similar failure mode to cord delamination in synchronous belts. Radial cracking involves the growth of cracks from the bottom of the V towards the belt cord, with the cracks eventually leading to the disintegration of the belt. Belt wear is obviously an ongoing process for the belts, and becomes a failure mode when the shape of the belt has changed so much as a result of wear that the belt is no longer able to transmit power at the required level. The cord separation failure mode is obviously adhesion related, and Gerbert and Fritzson [30] present a systematic method for recording the development of delamination failures. A 100% cord separation failure is defined as when the delamination occurs either across the entire belt width or along the belt for a distance equal to the belt width. Delamination which has not reached this level is then classified through a percentage value of the length of the crack to the width of the belt, with 50%, 10% and 1% the classifications used in this case. The 1% damage level represents the first recorded observation of damage through the failure mechanism of interest. One attribute of this classification system is that it allows the development of damage through more than one failure mechanism to be recorded through the life of the belt, so that failure modes which are developing simultaneously can be monitored, and situations where one failure mode is initiated by another clearly understood. Gerbert and Fritzson also developed a procedure for predicting belt life within the cord separation failure mode, based on the shear stress in the cushion region. The overall shear stress arises from four individual components of shear stress: i) the shear stress due to friction between belt and pulley, ii) a shear stress arising from the fact that the outer cords in the belt carry a higher tension than central belt cords (as a result of the belt section bending axially), iii) the cord layer and the rubber compound having a different resistance to longitudinal bending, iv)
344
shear stresses around the cords as a result of radial pressure on the cord layer.
Failures in Rubber Bonding to Substrates Of these shear stress components i), ii) and iii) all act along the length of the belt, with iv) acting across the belt width. The combined effect of these four shear stresses is shown to have a good correlation with belt life where cord separation is the failure mode, and is used as part of a belt life prediction procedure (based on finite element analysis) which encompasses all the failure modes identified by Gerbert and Fritzson.
• V-ribbed belts Figure 12.5 shows a schematic of the structure of a V-ribbed belt. The most commonly reported mechanical failure mode for this type of belt is wear, with delamination not generally perceived as a problem. This is of interest as most flexible composite elements will have a delamination failure mode of some description, and so the most obvious question to ask is why the V-ribbed belt does not. The probable answer is that the belt cord in a V-ribbed belt is isolated from the major distortions of the belt through its position above the belt/pulley interface. The large majority of the distortion of the belt takes place in the belt ribs, away from the cord. If the four shear stresses identified previously by Gerbert and Fritzson for a V-belt are considered, it can be seen that of the four i) and ii) do not apply to the cord layer in a V-ribbed belt. The cord layer in a Vribbed belt therefore does not incur shear stress to the same extent as that in a V-belt. Thus the V-ribbed belt may be considered a better design in composite terms, with the individual elements of the composite more effectively employed and protected.
Figure 12.5 V-ribbed belt construction
12.2.2.3 Conveyor Belts Kozhushko and Kopnov [30] identify a number of possible failure modes for fabric conveyor belts. Abrasive wear, fabric breakage or joint failure are all possible failure modes but the failure mode of most interest here is that of fatigue delamination, and it is
345
The Handbook of Rubber Bonding this mode that Kozhushko and Kopnov investigate. Their approach is to consider the delamination as a shear fatigue mechanism with delamination occurring when the shear strain magnitude reaches the permanent shear fracture strain, and with the permanent shear fracture strain seen as a variable which reduces with the number and size of loading cycles a belt has undergone. Kozhushko and Kopnov test this theory through fatigue testing and show reasonable correlation between theory and experiment, and conclude that the approach has value in belting applications but that further work is required to fully validate the approach.
12.2.2.4 Prediction of Belt Life in Delamination Failure Modes If we take an overview of how the belt delamination failure modes have been analysed by researchers to develop belt life prediction routines it is clear that the different methodologies have a lot in common, and it may be of value to highlight how these belt life prediction routines have been developed. The first step is to identify the belt life determining parameter, and for both V-belts and conveyor belts this has been identified as the shear deformation, as would be expected for delamination failures. For synchronous belts the belt life determining parameter has generally been identified as a measure of belt distortion, the success of these belt distortion measures in correlating with belt life would suggest that the belt distortion measures and the shear deformation are related. With this key parameter identified the precise nature of the relationship between the parameter and belt life must be identified. The approaches for synchronous belts and conveyor belts outlined above both sought this relationship through correlation with experimental data, whilst the approach outlined for V-belts was rather more fundamental, using crack growth analysis to follow the progression of the failure (but still using experimental data to validate the approach). All that then remains is for a model of belt behaviour to be developed which can relate the belt operational parameters to the belt life determining parameter, and then an analytical structure exists which allows the belt life to be predicted for a given set of operating parameters.
12.2.3 Adhesion Systems in Power Transmission Belts As with most rubber composites the adhesion system used in power transmission belts is based on an resorcinol/formaldehyde/latex (RFL) type system. General overviews of RFL adhesion systems from a tyre cord perspective have been previously published by Takeyama and Matsui [31], and more recently by Solomon [32]. Bonding with RFL systems is achieved through applying an RFL coating to the fibre structure prior to the
346
Failures in Rubber Bonding to Substrates belt being manufactured. The cords and rubber compound are then positioned and cured, with the RFL also curing at the same time. RFL systems work through using the resin part of the RFL to adhere to the fibre, with the latex part used to provide adhesion to the rubber compound through crosslinking. For some fibrous materials a two-stage RFL system is required. This involves precoating the fibre structure with an adhesion promoter of some description prior to the RFL coating being added. Polyester, glass and aramid fibres normally require a two-stage adhesion system. Broadly the strength of an RFL adhesion system will depend on the relative reactivity of all of the components of the composite, on how completely the cord is treated (wetting and penetration), and on the effectiveness of any diffusion mechanism. As the basic RFL system and its application is well documented elsewhere, the RFL system will be considered here from a power transmission belting perspective (see also Chapter 9). The introduction of new materials to belting applications is the most common reason for re-examining the RFL system, and Kubo [33] states that a re-examination of the adhesion system is essential if the perceived benefits of switching to a new material are to be obtained in practice. Kubo specifically examined adaptations to the RFL system required to ensure good adhesion to hydrogenated nitrile rubber (HNBR) for belting applications. Several authors have examined which specific RFL system is of most value where aramid fibres are being utilised in rubber composite applications [34, 35, 36]. Aramid fibres have excellent mechanical properties but are difficult to adhere to other materials, and as such much effort has been expended to develop adhesion systems which allow their excellent mechanical properties to be exploited in belting applications. Often linked to new materials introduction is the development of belts for high temperature applications, and the implications of this on conveyor belt adhesion systems has been examined by Sarkar [37].
12.2.4 Adhesion Testing in Power Transmission Belts Tests to assess the level of adhesion of fibre structures to rubber compounds can be divided into three classifications: • pull out tests to test the adhesion of individual cords to rubber compounds, • peel tests to test the adhesion of fabrics or a row of cord to rubber compounds, • belt tests. Cord pull out tests are commonly used to assess the adhesion between a single cord and a block of rubber compound. As the name suggests the tests concentrate on measuring the load required to pull a cord from a block of rubber compound. There are various
347
The Handbook of Rubber Bonding configurations which the pull out tests can take [31] with the T configuration [38], and in particular the H configuration [38, 39, 40, 41] most common where belting applications are considered. For the H configuration ASTM D4776 [42] is the most commonly quoted standard. However, Brodsky [43], in presenting a three cord adhesion test, questions the value of the H pull out test as it does not allow the stress state at the point of failure to be known. The three cord adhesion test presented by Brodsky deliberately generates a known stress state in order to overcome this limitation. Both Kubo [33] and Klingender and Bradford [40] observe that temperature and ageing are important parameters in defining the pull out strength in cord/rubber adhesion systems, with the pull out strength generally declining with an increase in either temperature or amount of ageing. Kubo also presents a methodology for assessing whether or not a cord material has been affected chemically by the adhesion system or the rubber compound. Cords are bonded into rubber compound blocks using the adhesion system of interest, before being dissolved out of the rubber compound blocks using toluene. The strength of cords treated in this way can then be compared to the strength of cords as new. Peel tests are generally used to assess the adhesion between rubber compounds and rows of reinforcing cords or fabrics. The most common format for these tests is a T-peel, where separation between one-half of the rubber compound and the reinforcing layer will be induced or manufactured in. The separated halves of the specimen are then attached to the jaws of a tensile tester and the force required to continue to separate rubber compound from reinforcement layer recorded. Such tests have been used by a number of authors with respect to belting applications [33, 34, 36, 37, 38], with the most commonly quoted standard being ASTM D4393 [44]. The importance of temperature and ageing are again both highlighted in the literature to ensure that the tests are as representative of application as possible. Both the pull out tests and the peel tests are rather idealised methods of assessing adhesion strength and may be considered as most appropriate for comparative testing of different adhesion systems rather than an indicator of potential belt performance. The relationship between the results of the standard pull out and peel tests and more representative tests resulting in belt failures through delamination has not been investigated in the literature. It is possible however to carry out a peel test on a synchronous belt which has been shown to be an indicator of the potential for a belt to fail through fabric separation [45]. The test procedure was to initially cut and peel back the facing fabric and belt teeth from the belt until three belt teeth were separated from the belt. The cut was then carefully extended under the fourth tooth into the tooth root area, taking care not to damage the fabric in any way. The cut section of belt was then T-peel tested as described above, with the fabric being peeled from the rest of the belt as the test progressed. The results of this test together with the results from belt life tests showed that adhesion systems with low peel strength values were more likely to fail through fabric separation.
348
Failures in Rubber Bonding to Substrates
12.3 Undesirable Adhesion Occuring Under Service Conditions (Fixing) This topic does not strictly comply with the title of the chapter, but represents a problem that can be very serious in nature and cause total failure of a seal system under service conditions. When nitrile rubber vulcanisates and metals are kept in contact, under pressure, for long periods of time an adhesion force develops between their surfaces; this force has become known as ‘fixing’. The strength of the bond between the two materials can be termed the fixing strength. ‘Fixing’ when it occurs in a component in the field can be very dangerous as it often occurs within the sensor mechanisms for such applications as gas control units and motor fuel systems for vehicles. Less critical occurrences can be seen when attempting to remove a hose from a metal coupling after a period of service; apart from requiring a cleanup of the metal surface before reassembly with a new hose this case causes little problem. Mori [46] examined a number of factors concerned with the phenomenon and developed tests to determine the problem and suggested four indicators of preventative measures to eliminate or control ‘fixing’ (see Section 12.3.2). Their compounds were moulded using a mould surface with an average surface roughness of 0.41 μm. Samples were compressed, in a similar manner to a conventional compression set test between sheets of the desired test metal. The adhesion between the metal and rubber were tested after periods of time and the load required to separate the metal and rubber sheets were plotted. The strength of the ‘fixing’ was found to increase with the time of compression contact between the two materials. Initially the increase was rapid and then followed by a slow steady increase with time. The two rates were considered to be quite different and were attributed to two different mechanisms, i.e., physical and chemical (see Section 12.3.1). The initial physical bond is defined as FSo and the final fixing strength including the chemical bond is defined as FS.
12.3.1 Factors Affecting ‘Fixing’ • Environmental factors It is known that the effects of fixing are more prevalent under conditions of high humidity and high ambient temperatures. The rate of the chemical force generation is strongly influenced by the increase in ambient temperature which causes generation of chemical bonds from the nitrile surface of the rubber to the metal surface. The effects of humidity and the presence of bloomed materials on the vulcanisate surface were also investigated [46]. These blooms were created artificially and wiped from solution
349
The Handbook of Rubber Bonding onto the surface of the rubber before the fixing test was carried out. Dependent upon the type of metal being tested there is also a corrosion factor to be considered. This usually proceeds from the surface of the metal plate, travelling towards the centre as corrosion proceeds. The corrosion forms an amorphous oxide layer on the metal surface which acts as a weak bonding layer. This type of corrosion occurs rather easily and is widely known as crevice corrosion [47, 48]. Crevice corrosion results from a galvanic cell forming in areas of stagnant water formed in small recesses in the metal surface. The interface of both the materials in contact must have a significant effect on the ‘fixing’ strength. Their topography caused by moulding of the vulcanisate and the metal finishing and/or plating has a significant influence on the fixing strength. The rougher the surfaces of both materials, the less will be the points of surface contact and this will reduce the strength of a bond developing. The hardness of the compound will also become a significant factor; the harder the compound the lower will be its ability to deform and follow the contours of the metal surface; this will affect the level of FS0. If however the increase in hardness of the rubber compound is produced by the addition of sulphur then there will be an increase in the chemically active units on the nitrile rubber surface and thus bond strength because of copper/sulphur bonds being formed gives an increase in FS.
• Material factors The ‘softer’ the metal the higher will be the ‘fixing’ force, FS0. This relationship is also reflected in the properties and thickness of the oxide films which are generated on the surface of the metals. Metal surface tensions and their ability to attract and hold water affect the level of FS. The surface of metals with low adsorbed moisture will generally give a high FS. The lower FS found in nickel and iron could well be attributable to their lower surface tension. Crystal size within the metal surface in contact with the rubber could well also be a significant factor in determining areas of contact. Metals which develop only a thin oxide film and having a chemically active surface will obviously generate a higher bond strength. However, copper will have a thick oxide film but this and the metal itself are very chemically active to sulphur and polysulphide chains. Aluminium and zinc plates give a low FS, since their oxide films are generally thick and chemically inactive to sulphur and polysulphide chains. The surface treatment of copper therefore has a significant effect on the ‘fixing’ properties. Organic metal treatments [47] give low FS when used in contact with NBR because the plates have a low surface tension below 0.3 mN/cm. The polarity of the NBR has an influence on the level of FS0 which increases with the acrylonitrile (ACN) level. FS also increases with the polarity of the NBR, but to a far less extent than does FS0. 350
Failures in Rubber Bonding to Substrates Carbon black fillers generally do not have much effect on ‘fixing’ except in the differences associated with general hardness increase or decrease. However the high ‘fixing’ strength of NBR may have an association with the included carbon black as it is widely known that carbon black provides active radicals in NBR during mixing and blending. These radicals can easily change to peroxides and give carboxyl groups at the surface of the NBR. The groups will be physically absorbed into the metal surface or react chemically with them. Silica however is generally inert and in fact with increasing dosage the value of FS decreases as the hardness increases. The generation of strong adhesive forces can be considered to result from chemical interaction between the metal and NBR. This generally results from the reaction of the metal with sulphur compounds and carboxyl compounds to give their metal salts, which have a high bond energy. Polysulphide reactivity is high for copper and brass plates.
12.3.2 Prevention of ‘Fixing’ Mori [46] summarises the fixing phenomenon by suggesting that first the segmented molecules of the vulcanisate diffuse into the metal surface and then polar groups such as a nitrile groups are adsorbed onto the surface. At this time, secondary order bond forces such as van der Waals forces and hydrogen bond forces are generated between both materials. Following this chemical reactions occur at the contact points between the two materials, and then both materials are combined by the first order bonds formed. As a result, strong adhesional forces are generated between the two materials. They suggest that that four preventative measure may be used: • control of contact area, • suppression of molecular motion of the NBR segments, • inhibition of interfacial reactions, • introduction of inactive crosslinks and side chains.
12.3.3 Other Methods of Preventing ‘Fixing’ - Examined Experimentally • Surface irradiation with UV light [46] Vulcanisates pre-treated by exposure to UV irradiation are influenced by the atmosphere in which the treatment is carried out. NBR treated in this way in an oxygen atmosphere did not significantly reduce in FS0, but there was remarkable increase in the FS value. This was deemed to show that the active groups such as -
351
The Handbook of Rubber Bonding OH, -CO, and -COOH are present near the surface of the metal and are also formed on the surface of the NBR vulcanisates during the UV treatment in air. If the UV treatment is carried out in an argon atmosphere then the NBR vulcanisates had a greatly decreased value of FS. This however would not be considered an industrial feasibility.
• Prevention of ‘Fixing’ by Bloom Generation Wax films, especially paraffin wax were not found to be effective as a preventative for ‘fixing’, but in fact accelerated it. If the wax is close to melting or melted then a ‘sucking’ effect will be encountered between the two materials. Mori found that amide type materials in blooms with long alkyl or alkenyl groups, provided excellent bloom films with low surface tension and high melting points on the surface of the NBR vulcanisates. Provided that the temperatures did not exceed 60 °C the amide bloom inhibited the chemical reactions occurring during ‘fixing’. The use of blooming agents is very effective for preventing the fixing between metals and NBR vulcanisates. The materials used were stearamide and methylene bis-erucamide.
References 1.
B. G. Crowther, Rubber to Metal Bonding, Rapra Review Report, 1996, Vol. 8, No 3, Rapra Technology Ltd., Shrewsbury 1996.
2.
C. L. Mahoney in Handbook of Adhesives, 3rd Edn., Ed., I. Skeist, Van Nostrand Reinhold, 1990, p.74-93.
3.
H. Leidheiser, Presented at the ACS, Division of Polymeric Materials: Science and Engineering, Fall 1986, Paper No.12
4.
D. Rowe, Private communication to W. C. Wake, 1969.
5.
W. C. Wake, Adhesion and the Formulation of Adhesives, 2nd Edn., Applied Science Publishers, London, 1982.
6.
L-H. Lee, Journal of Polymer Science, 1967, A2, 5, 1103.
7.
A. de Bruyne, The Extent of Contact Between Glue and Adherend, Aero Research Technical Notes, Bulletin No. 168, Duxford, UK, 1956.
8.
A. N. Gent and R. P. Petrich, Proceedings of the Royal Society, 1969, A310, 433.
352
Failures in Rubber Bonding to Substrates 9.
A. Filbey and J. P. Wightman in Adhesive Bonding, Ed., L. H. Lee, Plenum Press, New York 1991, Chapter 7.
10. D. Bascom, The Origin and Removal of Microvoids in Filament Wound Composites, NRL Report 6268, May 24, 1965. 11. E. P. Pleuddemann, Presented at the 25th Annual Technical Conference of the SPI Reinforced Plastics/Composites Institute, Washington, DC, 1970, Section 13-D, 1. 12. J. D. Minford in Adhesive Bonding, Ed., L. H. Lee, Plenum Press, New York, 1991, Chapter 9. 13. A. Laird, Glass Surface Chemistry for Glass Reinforced Plastics, Final Report, Navy Contract W-0679-C (FBM) 1963. 14. T. Koyama, M. Kagatoni, T. Shibata, T. Sato and T. Hoshiro, Journal of the Japanese Society of Mechanical Engineers, 1979, 22, 169, 1988. 15. T. Koyama, M. Kagatoni and T. Hoshiro, Presented at the Design Engineering Technology Conference, Cambridge, MA, USA, 7-12 Oct. 1984, ASME Paper No.84-DET-217. 16. T. H. C. Childs, A. Coutsoucos, K. W. Dalgarno, A. J. Day and I. K. Parker, Presented at the International Conference on Belt Transmissions, Hiroshima, Japan, 1991. 17. K. W. Dalgarno, A. J. Day and T. H. C. Childs, Proceedings of the Institution of Mechanical Engineers, Part D, 1994, 208, 1, 37. 18. H. Iizuka, K. Watanabe and S. Mashimo, Fatigue and Fracture of Engineering Materials and Structures, 1994, 17, 7, 783. 19. K. W. Dalgarno, Presented at the 152nd Meeting of the ACS Rubber Division, Cleveland, Ohio, Fall 1997, Paper No.47. 20. J. A. Stubbs, Industrial Lubrication and Technology, 1994, 46, 6, 7. 21. R. Kido, T. Kusano and T. Fujii, SAE Paper No.960712, 1996. 22. H. Iizuka, S. Tsutsumi, K. Watanabe, S. Mashimo and N. Ohsako, Presented at the International Rubber Conference, Kobe, Japan, 1995, Paper No.25B-17. 23. Y. Murakami and M. Watanabe, SAE Paper No.880415, 1988.
353
The Handbook of Rubber Bonding 24. K. Hashimoto, M. Oyama, N. Watanabe and Y. Todani, Presented at the 128th Meeting of the ACS Rubber Division, Cleveland, Ohio, Fall 1985, Paper No.5. 25. K. W. Dalgarno, T. H. C. Childs, A. J. Day, M. H. Hojjati, D. Q. Yu and R. B. Moore, Kautschuk Gummi Kunststoffe, 1997, 50, 4, 299. 26. T. H. C. Childs, K. W. Dalgarno, M. H. Hojjati, M. J. Tutt and A. J. Day, Proceedings of the Institution of Mechanical Engineers, Part D, 1997, 211, 3, 205. 27. H. Iizuka, G. Gerbert and T. H. C. Childs, Journal of Mechanical Design, 1999, 121, 2, 180. 28. G. Gerbert and D. Fritzson, Presented at the Power Transmission and Gearing Conference, Chicago, USA, 1989, 59. 29. D. Fritzson, Journal of Mechanisms, Transmissions, and Automation in Design, 1989, 111, 3, 424. 30. G. G. Kozhushko and V. A. Kopnov, International Journal of Fatigue, 1995, 17, 8, 539. 31. T. Takeyama and J. Matsui, Rubber Chemistry and Technology, 1969, 42, 1, 159. 32. T. S. Solomon, Rubber Chemistry and Technology, 1985, 58, 3, 561. 33. Y. Kubo, O. Mori, K. Ohura and H. Hisaki, Rubber Chemistry and Technology, 1991, 64, 1, 8. 34. Y. Iyengar, Journal of Applied Polymer Science, 1978, 22, 3, 801. 35. W. E. Weening, Presented at the 124th Meeting of the ACS Rubber Division, Houston, Texas, Fall 1983, Paper No.100. 36. H. Janssen, Journal of Coated Fabrics, 1996, 25, April, 276. 37. P. P. Sarkar, S. K. Ghosh, B. R. Gupta, A. K. Bhowmick, S. Chakraborty and A. Sen, Presented at the 132nd Meeting of the ACS Rubber Division, Cleveland, Ohio, Fall 1987, Paper No.92. 38. H. A. Daan, Kautschuk und Gummi Kunststoffe, 1985, 38, 10, 904. 39. N. A. Darwish, G. Samay and A. Boros, Polymer Plastics Technology and Engineering, 1994, 33, 4, 465.
354
Failures in Rubber Bonding to Substrates 40. R. C. Klingender and W. G. Bradford, Elastomerics, 1991, 123, 8, 10. 41. G. I. Brodsky, Presented at the 128th Meeting of the ACS Rubber Division, Cleveland, Ohio, Fall 1985, Paper No.40. 42. ASTM D4776-02 Standard Test Method for Adhesion of Tire Cords and other Reinforcing Cords to Rubber Compounds by H-Test Procedure, 2002. 43. G. I. Brodsky, Rubber World, 1984, 190, 5, 29. 44. ASTM D4393-02 Standard Test Methods for Strap Peel Adhesion of Reinforcing Cords or Fabrics to Rubber Compounds, 2002. 45. K. W. Dalgarno, Synchronous Belt Materials: Durability and Performance, University of Bradford, UK, 1991, Ph.D. Thesis. 46. K. Mori, A. Watanabe and M. Saito, Rubber Chemistry and Technology, 1988, 62, 2, 195. 47. K. Mori and Y. Nakamura, Nippon Kagaku Kaishi, 1987, 725. 48. A. Kondo, Polymer Digest (Tokyo), 1980, 32, 86.
355
The Handbook of Rubber Bonding
356
Abbreviations and Acronyms
ABS
Acrylonitrile-butadiene-styrene
ACM
Ethyl acrylate copolymer
ACN
Acrylonitrile
AEM
Analytical electron microscopy
AES
Auger electron spectroscopy
ASTM
American Society for Testing and Materials
ATR
Attenuated total reflectance
BATNEEC
Best available technology not entailing excessive cost
BDTPTS
Bis (di-ethyl thiophosphoryl) trisulphide
BGDMA
1,3-Butyleneglycol dimethacrylate
BIIR
Bromo butyl rubber
BR
Polybutadiene rubber
BTSE
Bis-[triethoxysilyl]ethane
CBS
N-cyclohexyl-2-benzothiazole sulphenamide
CIIR
Chloro butyl rubber
CM
Cement/metal failure
COR
Coefficient of restitution
CPD
Controlled product design
CP
Cement/primer failure
CPE
Chlorinated polyethylene
CR
Polychloroprene
CS
Compression set
CSM
Chlorosulphonated polyethylene
DCBS
N,N-dicyclohexyl benzothiazole sulphenamide
DCPD
Dicyclopentadiene
DIOP
Diisooctyl phthalate
DOA
Dioctyl adipate
357
Commercial rubbers
The Handbook of Rubber Bonding DOS
Dioctyl sebacate
DPG
N, N´-diphenyl guanidine
DPTT
Dipentamethylene thiuram tetrasulphide
ECO
Epichlorohydrin
EDX
Energy-dispersive X-ray (analysis)
EELS
Electron energy loss spectroscopy
ENR
Epoxidised NR
EP
Ethylene-propylene rubber
EPC
Easy processing channel black
EPDM
Ethylene propylene diene monomer
EPR
Ethylene propylene rubber
EPTPE
EPDM/PP-based thermoplastic rubbers
ESCA
Electron spectroscopy for chemical analysis
ETO
Ethylene oxide
EV
Efficient vulcanisation
EVA
Ethylene vinyl acetate copolymer
EVM
Ethylene-vinyl-acetate rubber
FDA
Food and Drug Administration (USA)
FEP
Fluorinated ethylene propylene copolymer
FKM
Fluorocarbon rubbers
FMEA
Failure mode and effect analysis
FMQ
Silicone rubbers
FPM
Fluoropolymer
FRP
Fibre-reinforced plastic
FTIR
Fourier transform infrared analysis
GP
General purpose
GPF
General purpose furnace black
GRP
Glass-reinforced plastic
HAF
High abrasion furnace black
HB
Brinell hardness
HDPE
High density PE
358
Abbreviations and Acronyms HMMM
Hexamethoxymethylmelamine
HMT
Hexamethylene tetramine
HNBR
Hydrogenated acrylonitrile butadiene rubber
HR
Hexamethylenemelamine-resorcinol
HRC
Rockwell hardness C
HRH
Hexamethylenemelamine-resorcinol-hydrated silica
HRL
Heat resistant litharge
HTV
High temperature vulcanising
HV
Vickers hardness
HVLP
High velocity low pressure
IAD
Isopropyl azodicarboxylate
IASF
Intermediate superior abrasion furnace black
IIR
Butyl rubber
IPA
Isopropyl alcohol
IR
Isoprene rubber
IRHD
International rubber hardness degree
ISO
International Organisation for Standardisation
ISS
Ion scattering spectroscopy
LDPE
Low density polyethylene
LR
Liquid rubber
MBT
Mercaptobenzothiazole
MBTS
Mercaptobenzothiazole disulphide
MEK
Methyl ether ketone
MF
Melamine-formaldehyde
MHF
Maximum torque
MIBK
Methyl isobutyl ketone
MOR
2-(4-Morpholinyl mercapto)-benzthiazole
MW
Molecular weight
NBR
Acrylonitrile butadiene rubber
NOBS
N-oxy di-ethylene benzthiazyl sulphenamide
359
The Handbook of Rubber Bonding NPC
No post cure
NR
Natural rubber
NRTPE
NR-based thermoplastic rubbers
OBTS
N-oxydiethylene benzothiazole sulphenamide
ODC
Ozone depleting chemicals
ODR
Oscillating disk rheometer
PA
Polyamide
PB
Polybutadiene
PBT
Polybutylene terephthalate
PC
Polycarbonate
PCB
Printed circuit board
PDMS
Polydimethylsiloxane
PE
Polyethylene
PEEK
Polyetherether ketone
PES
Polyethersulphone
PET
Polyethylene terephthalate
PG
Propylene glycol
PHR
Parts per hundred rubber
PIXIE
Particle-induced X-ray emission
PNR
Polynorbornene
POM
Acetal (polyoxymethylene)
PP
Polypropylene
PPO
Polyphenylene oxide
PPS
Polyphenylene sulphide
Pt
Platinum
PTFE
Polytetrafluoroethylene
PU
Polyurethane
PV
Post vulcanisation
PVC
Polyvinyl chloride
PVDC
Polyvinylidine chloride
PVDF
Polyvinylidene fluoride
PVF
Polyvinyl fluoride
360
Abbreviations and Acronyms R&D
Research and Development
RC
Rubber/cement failure
RF
Rubber failure
RFL
Resorcinol/formaldehyde/latex
RFR
Resorcinol/formaldehyde/resin
RH
Relative humidity
RIM
Reaction injection moulding
RPN
Risk priority number
RT
Room temperature
RTV1
Room temperature vulcanising, one component
RTV2
Room temperature vulcanising, two component
SBR
Styrene butadiene rubber
SBS
Styrene-butadiene-styrene
SDS
Safety data sheet
SEBS
Hydrogenated SBS
SEM
Scanning electron microscopy
SEV
Semi-efficient vulcanisation
SG
Specific gravity
SIMS
Secondary ion mass spectroscopy
SMR
Standard Malaysian rubber
SNMS
Secondary neutral mass spectrometry
TAC
Triallyl cyanurate
T&D
Transmission and distribution
TCAT
Tyre cord adhesion test
TCBQ
Tetrachlorobenzoquinone
TCE
Trichloroethane
TCICA
Trichloroisocyanuric acid
TEM
Transmission electron microscopy
TETD
Tetraethyl thiuram disulphide
Tg
Glass transition temperature
THF
Tetrahydrofuran
361
The Handbook of Rubber Bonding TIC
Trichloroisocyanuric acid
TLV
Threshold limit value
TMTD
Tetramethyl thiuram disulphide
TMTM
Tetramethyl thiuram monosulphide
TOFSIMS
Time-of-flight secondary ion mass spectrometry
TOTM
Trioctyl trimellitate
TPE
Thermoplastic elastomer
TPO
Thermoplastic olefinic elastomer
TPU
Thermoplastic PU
TR
Thin rubber
TRIM
Trimethylolpropane trimethacrylate
TSA
Toluene sulphonic acid
UF
Urea-formaldehyde
USP
United States Pharmacopoeia
UV
Ultra violet
VAMAC
Ethylene acrylic terpolymer
VMQ
Silicone rubber with methyl and vinyl substituents
VOC
Volatile organic compounds
VP
Vinyl pyridine
VS
Vinyl silane
WC
Water closet
WDS
Wavelength dispersive X-ray (analysis)
XPS
X-ray photoelectron spectroscopy
XNBR
Carboxylated NBR
ZDA
Zinc diacrylate
ZDMA
Zinc dimethacrylate
362
Author Index
A Ahmad S. 211 Ahn J. H. 179, 191, 193, 209 Alarcon-Lorca F. 212 Aldred D. H. 256 Allen K. W. 54 Alliger G. 78 Alstadt D. M. 78 Andries J. C. 160 Ayres R. L. 51
B Ball J. J. 168, 192, 211 Bantjes A. 194 Barker L. R. 211 Barr T. L. 210 Basaran M. 209 Bascom D. 353 Beamson G. 16, 52 Beatty J. R. 160 Beecher J. F. 192 Beers R. N. 212 Benarey H. A. 51, 78 Benderly A. A. 52 Benko D. A. 212 Bhowmick A. K. 51, 78, 160, 192, 193, 209, 210, 354 Biemond M. E. F. 167, 192, 193, 201, 209 Binder H. 51 Biswas A. 192, 210 Blow C. M. 78 Blythe A. R. 51
Bobrov A. P. 182, 193, 194 Bomal Y. 193, 210, 211 Bond K. M. 51, 136 Boros A. 354 Borroff E. M. 161, 256 Bourrain M. P. 210 Bowden P. J. 212 Bradford W. G. 348, 355 Bragole R. A. 19, 52 Brandon D. G. 55 Brass I. 52 Brecht H. 51 Brewis D. M. 29, 51, 52, 53, 54, 283 Briggs D. 14, 51, 52, 54, 209, 283 Brindoepke G. 211 Brodsky G. I. 348, 355 Brunat W. 54 Buchan S. 78, 191 Burkhart T. 211 Burrows F. H. 212 Busscher H. J. 54 Butcher D. 193, 211 Butler D. P. 239
C Carlson G. L. 54 Carne R. J. P. 53 Carter A. R. 20, 53, 161 Cartier J. F. 51 Cayless R. A. 52 Chakraborty S. 354 Chakravarty S. N. 160 Chaler N. A. 170, 193
363
The Handbook of Rubber Bonding Chandra A. K. 169, 181, 192, 193, 202, 209, 210 Chen J. K. 54 Chew A. 28, 52, 54 Child T. F. 194, 212 Childs T. H. C. 343, 353, 354 Clearfield H. M. 51 Cochet P. 193, 206, 211 Coleman Jr., E. W. 78 Collins G. C. S. 14, 52 Colson J. C. 210 Combette C. 26, 54 Combette P. 212 Comyn J. 53, 317 Coover H. W. 283 Cope B. C. 53 Corish P. J. 160 Cornell J. A. 239 Costin R. 79, 239, 240 Coutsoucos A. 353 Craft J. 88 Crawford D. 239 Creasey J. R. 211 Crowther B. G. 160, 352
D Daan H. A. 354 Daft L. 59, 78 Dahm R. H. 52, 54 Dalgarno K. W. 342, 353, 354, 355 Darwish N. A. 354 Davis G. D. 51 Day A. J. 353, 354 De Bruyne A. 352 De Gryse R. 210 De P. P. 161 De S. K. 161 De Volder P. 210 De Vries J. 54 Debruyne E. 210
364
Del Vecchio R. J. 79 De Lollis N. J. 54 Denq Y. L. 54 Deryaguin B. V. 150, 161 Deuri A. S. 193, 210 Dietrick M. I. 211 Dillard J. G. 23, 53, 54 Dogadkin B. A. 160 Downes J. 212 Drake R. E. 160 Dreifus D. W. 283 Duc T. M. 54 Duffour T. 193, 210 Dwight D. W. 14, 52
E Edington R. A. 256 Eirich R. 239 Ekwall R. 239 Eley D. D. 256 Elliot D. J. 78 Ellul M. D. 52, 143, 160 Elman J. F. 15, 52 Elschner A. 192, 211 Engelhardt M. L. 212 Erickson D. E. 212 Ershov E. A. 193 Evans L. R. 193, 206, 211, 212 Everaert E. P. 54 Extrand C. W. 21, 53
F Fahrig M. 257 Feast W. J. 209 Feng Y. 240 Fernández-García J. C. 53, 161 Feuer H. O. 239 Filbey A. 337, 353 Flanagan P. 239
Author Index Flemming R. A. 210 Friedel C. 88 Fritz J. L. 54 Fritz T. L. 52 Fritzson D. 344, 345, 354 Fujii T. 353 Fujikura Y. 53 Fujino A. 239 Fuller K. N. G. 160
G Garbassi F. 55 Gatza P. E. 239 Gent A. N. 21, 53, 160, 161, 352 Gerbert G. 344, 345, 354 Gerenser L. J. 15, 52 Gesang T. 52 Ghosh S. K. 354 Ghosh T. B. 209 Gibbs H. W. 192, 193, 210, 211 Giridhar J. 169, 189191, 193, 195, 209 Goerl U. 257 Goralski E. G. 193, 212 Goryaev V. 184, 194 Grabov’ I. F. 195 Gregory H. J. 54 Grigorev M. F. 194 Grogger W. 192 Grubbauer G. 192, 209 Gupta B. R. 354
H Haemers G. 191, 210 Hall J. R. 52 Hall M. M. 51, 78 Halladay J. R. 58, 79 Hamed G. R. 159, 161, 168, 182, 192, 193, 203, 208, 210, 212 Hammer G. E. 209, 210
Han M. H. 194, 212 Hannell W. J. 210 Hansen R. H. 52 Harris S. J. 195, 212 Hartz R. E. 257 Hashimoto K. 354 Hawkins I. M. 193, 210 Hazelton D. R. 52, 143, 160 Hennemann O. D. 52 Hess R. H. 211 Hewitt N. L. 193, 211 Hill J. M. 18, 51 Hirahara H. 195 Hiratsuka S. 195 Hisaki H. 354 Hivert D. 54 Hoekje H. H. 211 Hofer F. 192, 209 Hoff C. M. 182 193, 211, 212 Hofmann W. 79 Hojjati M. H. 354 Hollahan J. R. 54 Holtkamp D. 192, 211 Hope J. C. 193, 211 Horie H. 195 Hörnström S-E. 194 Hoshiro T. 353 Howath T. 239 Huang J. 193, 208, 212 Hudis M. 54 Hummel K. 192, 209 Hupje W. H. 256
I Ichikawa M. 160 Iizuka H. 342, 343, 353, 354 Ikeda A. 239 Ikeda K. 195 Ikeda Y. 195
365
The Handbook of Rubber Bonding Ishikawa Y. 209, 210 Iwabuchi H. 53 Iyengar Y. 257, 354
J Janssen H. 354 Jayaseelan S. K. 192, 194, 212 Jazenski P. J. 78 Jeon G. S. 194, 212 Jerschow P. 316 Job L. 161 Joseph R. 161
K Kagatoni M. 353 Kang K-K. 212 Kaplan W. D. 55 Karbashewski E. 18, 51 Kato N. 53 Kawakami S. 210 Keller C. 240 Kendall C. R. 51 Khavina E. Y. 192 Kido R. 342, 353 Kilduff T. J. 52 Kim H. J. 160 Kim K. J. 52 Kimu S. 195 Klingender R. C. 239, 348, 355 Klyachtin Y. 53 Konar J. 209 Kondo A. 355 Kondyurin A. 19, 53 Konieczko M. B. 51, 52 Kopnov V. A. 345, 354 Koshal D. 283 Koyama T. 342, 353 Kozhushko G. G. 345, 354
366
Kranz G. 52 Kretzschmar T. 192, 209 Krone R. 184, 194 Krüger R. 51 Kubo Y. 347, 348, 354 Kurian J. 161 Kurosaki T. 160 Kusano T. 353 Kusano Y. 22, 24, 53
L Labarre D. 181, 193, 210 Labriola J. M. 160 Lai J. Y. 54 Lai S. M. 160 Laird A. 338, 353 Lake G. J. 160 Landrock A. H. 55 Lawson D. F. 19, 20, 23, 24, 25, 52, 53 Le Prince P. 54 Lee K. S. 78 Lee L. H. 352 Legeay G. 54 Lehrle R. S. 210 Leidheiser H. 331, 352 Levi D. W. 52 Lin A. 51 Lin Y. Y. 54 Lingjun C. 193 Loha P. 160 Lowe A. C. 52 Luginsland H-D. 194 Lüschen R. 52 Lyons C. S. 51
M Macintosh C. 256 Maeseele A. 210
Author Index Mahoney C. L. 51, 352 Manino L. G. 78 Manoj N. R. 161 Martin S. 239 Martín-Martínez J. M. 23, 53, 54, 161 Mashimo S. 353 Mason M. G. 15, 16, 52 Mathieson I. 52, 54 Matsui J. 346, 354 Matyukhin S.A. 190, 195 Maucourt J. 54 Mayer F. 51 McNamarra D. K. 51 Meier K. 257 Melley R. E. 160 Melvin T. 239 Michel G. 54 Middletown O. H. 209 Minagawa M. 19, 26, 53 Minford J. D. 51, 338, 353 Mitchell B. 256 Mittal K. 18 Mizumoto S. 195 Molitor P. 239 Montgomery R. E. 51 Montoya O. 54 Moore M. J. 136 Moore R. B. 354 Morawski J. C. 210 Morel-Fourrier C. 211 Mori K. 190, 195, 349, 351, 355 Mori O. 354 Morra M. 55 Mowrey D. H. 51, 136 Mukhopadhyay R. 192, 193, 209, 210 Munro H. S. 209 Murakami Y. 353 Murray P. F. 191, 209, 210
N Nagel W. 79, 239, 240 Naito K. 53 Nakamura M. 195 Nakamura Y. 355 Nando G. B. 161 Nangreave K. R. 53 Nawafuna H. 195 Nelson E. R. 52 Nicholas D. 52 Niderost K. J. 210 Noguchi T. 53
O Occhiello E. 55 O’Connor J. T. 283 Ohsako N. 353 Ohura K. 354 Oishi Y. 195 Okel T. A. 193, 211 Okumoto T. 160 Oldfield D. 20, 21, 22, 25, 26, 29, 51, 53 Olson L. R. 239 Oplochenko N. A. 195 Opperman G. W. 54 Orgilés-Barceló A. C. 53, 161 Orjela G. 195, 212 Othman A. B. 160 Owen M. J. 54 Oyama M. 239, 354
P Panchuk F. O. 190, 195 Parker I. K. 353 Pastor-Blas M. M. 23, 24, 53, 54 Pastor-Sempere N. 24, 53 Patrick R. L. 55 Paul R. 192, 203, 210
367
The Handbook of Rubber Bonding Pawlowski L. B. J. 256 Pelletier J. B. 210 Persoone P. 210 Peterson A. 211 Petrich R. P. 352 Pettit D. 20, 22, 53, 161 Pieroni J. K. 239 Pieroth M. 168, 192, 211 Plasczynski T. 136 Pleuddemann E. P. 194, 338, 353 Pochan J. M. 15, 52 Pohmer K. 316 Possart W. 161 Potapov E. E. 169, 192, 193 Potente H. 51 Prescott L. E. 54 Prokof’ev Y. A. 168, 183, 192, 193
R Rance D. G. 52, 54 Rangarajan V. 191 Rauline R. 209, 211 Reed T. F. 212 Rhee C. K. 160 Ridha R. A. 212 Riggs W. M. 52 Rijpkema B. 243, 256 Roach J. F. 212 Roberts A. D. 146, 160 Rodriquez G. 239 Roland C. M. 161, 239 Roseboom F. 194 Rowe D. 352
S Saito M. 355 Saito T. 53 Saito Y. 239 Sakharova E. V. 192, 193
368
Salych G. G. 192, 193 Samay G. 354 Sánchez-Adsuar M. S. 53 Sanderson C. 59, 120 Sarkar P. P. 354 Sartre A. 193, 210 Sasaki T. 53 Sato T. 353 Schaefer R. J. 211 Schlett V. 52 Schoenherr D. 51 Schonhorn H. 52 Schoon Th. G. F. 257 Schramm E. C. 52 Schürmann K. 121 Seibert R. F. 209, 212 Semak B. D. 195 Sen A. 354 Seo G. 194, 212 Setiawan L. 51 Sexsmith F. H. 51, 78 Sharma S. C. 186, 194 Shekhter V. E. 193 Shibata T. 353 Shieh C. H. 160, 161 Shields J. 54, 55 Shijian L. 212 Shoaf C. J. 256 Shofner D. L. 51 Shvarts A. G. 193 Shyu S. S. 54 Siverling C. E. 78 Sjothun I. J. 78 Skeist I. 51, 55, 283, 352 Skewis J. D. 161 Smilga V. P. 161 Snodgrass L. J. 193, 212 Snogren R. C. 55 Soldatos A. C. 160 Solomon T. S. 346, 354
Author Index Sommer F. 192 Song J. 194 Souchet J. C. 211 Sowell R. R. 54 Starinshak T. W. 212 Stohrer W. D. 52 Strain F. 211 Strobel M. 18, 51 Stubbs J. A. 353 Stuck B. 211 Subramanian V. 194 Sugimoto M. 160 Sutherland I. 52 Swanson M. J. 26, 54 Symes T. E. F. 20, 21, 22, 25, 27, 51, 53
T Takeyama T. 346, 354 Tarney R. E. 160 Tate P. E. R. 192, 206, 210, 211 Teets A. R. 239 Terashima K. 160 Tessier L. 193, 210 Thomas J. 51 Thornton J. S. 51 Todani Y. 354 Toesca S. 210 Tomanek A. 316 Tommasini F. 195, 212 Torregrosa-Maciá R. 53, 54 Torró-Palau A. 53 Touchet P. 239 True W. R. 212 Tsutsumi S. 353 Tutt M. J. 354
V Vakula V. L. 161 van der Aar C. P. J. 194
van der Mei H. C. 54 van Ooij W. J. 165, 166, 167, 169, 177, 180, 181, 185, 186, 191, 192, 193, 194, 195, 199, 200, 201, 204, 205, 209, 210, 212 Verbanc J. J. 160 Vincent M. 195, 212 Voyutskii S. S. 148, 161
W Waddell W. H. 183, 193, 211, 212 Wagner M. P. 206, 211 Wake W. C. 159, 161, 256, 352 Wallerswein S. 211 Walz G. 211 Walzak M. J. 18, 51 Watanabe 342 Watanabe A. 355 Watanabe K. 353 Watanabe M. 353 Watanabe N. 354 Watanabe T. 53 Watts J. F. 52 Weening W. E. 191, 209, 210, 243, 256, 354 Wegman R. F. 54 Weih M. A. 78 Weihe J. 51 Wenghoeffer H. M. 257 Westerdahl C. A. L. 52 Wheeler E. L. 212 Whelan A. 78 Wieczorek K. 317 Wightman J. P. 337, 353 Wiktorowicz R. 52 Wilson J. C. 193, 210 Wolff S. 211, 257 Wootton A. B. 51
369
The Handbook of Rubber Bonding
X Xie H. 240
Y Yamaguchi K. 195 Yan H. 212 Yoshi F. 53 Yoshikawa M. 53 Yu D. Q. 354
Z Zaporozhchenko V. I. 194 Zhang J. Q. 194 Zierler L. 257
370
Company Index
A Abrasive Developments 42 Advanced Elastomer Systems 259
B Bayer 60
C
Lord Corporation 12, 57, 58, 61, 63, 77, 81
M Metallgesellscahft 61 Metalok 61 Michelin & Cie 209 Morton International 128
Chemical Inovations Limited (CIL) 61
N
D
Nagase Co. 246
Degussa AG 60, 257
E E. I. du Pont de Nemours and Company 160, 256, 259 Electro-Chemical Rubber & Manufacturing Company 78 EMS American Grilon 246
G
P Par Chemie 61 Pirelli 189 Proquitec 61
U Union Carbide Corporation 160 Uniroyal Chemical Company Inc. 212
Goodyear Tire and Rubbers (US) 212
V
H
Vulnax International Ltd. 246
Henkel KGaA 61, 81, 88 Honda Motor Company 42 Hughson Chemicals 81
W Wacker-Chemie GmbH 295, 311
L
Y
Loctite 283
Yamashita Rubber Company 42
371
The Handbook of Rubber Bonding
372
Main Index
A Abrasive wear 345 Accelerator 175, 176, 177 Accelerators 2-mercaptobenzothiazole 65 mercaptobenzothiazole disulphide 65 sulphenamide 183 tetramethyl thiuram disulphide 65 Acrylic acid 215 butyleneglycol dimethacrylate 214 metallic salts 215 Addition curing 294 Additive 164 Adherend surface conditions 335 surface roughness 336 Adhesion 76, 148, 312 braided wire 235 build-up 292 corrosion 350 curing system 251 duration of 296 environmental factors 349 fabric 235, 236 factors affecting 253, 352, 349 hardness 350 humidity 349 in service 254 interfacial 205 mechanisms 250, 252 monofilament 235 prevention 351 problems 339
retention 189 rubber-to-metal 224 surface irradiation 351 T-pull 237 temperature 349 test methods 76 theories 148 tyre cord 197 Adhesion of rubbers 334 rubber flow 334 stabilisation 335 wetting 334 Adhesion peel tests 224 Adhesion promoter 226, 230 internal 226, 230, 231 Adhesion test T-pull 236 three cord 348 Adhesive 25, 62 characteristics 62 cyanoacrylate 37 dispersions 234 epoxide 25 flow characteristics 336 heat reactive contact cements 37 hot melt reactive urethane prepolymers 37 polyurethane 156 primers 71 rubber to metal 62, 63 selection 66 silane treatments 37 tie layer 232
373
The Handbook of Rubber Bonding two-part epoxies 37 two-part urethanes 37 wetting 128 Adhesive application 69 brushing 69, 71 dipping 69, 70 spraying 69 tumbling 69 Adhesive bond strength 231 Adhesive systems 241 solvent-based 248 urethane 158 Alloys metal 59 ternary 189 Aluminium 326 anodising 12 Analysis 19, 20, 24 Fourier transform infrared 19, 23 infrared 20, 24 X-ray fluorescence 20, 25 Antidegradants antioxidants 66 antiozonants 66 prevulcanisation inhibitors 66 waxes 66 Application methods 98, 101, 117 brush 98 dip 98 electrostatic 98 flowcoat 98 reverse roller coat 98 roller 98 sponge 98 spray 98 Aramid 247 ASTM test methods 122 Autohesion 137, 139, 141, 144, 148, 150
374
B Bearing pads 102 Belt failure 341 abrasive wear 345 back cracking 342 cord failure 341 cord separation 344 fabric breakage 345 fabric separation 341, 342 joint failure 345 radial cracking 344 tensile failure 344 tooth root cracking 341 wear 341, 344 Belt life prediction 344 shear stress 344 Blooming 145, 146, 147, 151, 280, 282, 283, 349 Boiling water tests 107 Bond failure 74, 319 corrosion 319 metal preparation 319 moulding procedures 319 product abuse 319 product design 319 Bond formation 71, 144 Bond integrity 73 Bond strength 190 Bondability 64 index 64 Bonded parts 72 knit lines 72 splits 72 Bonding 29, 73, 152, 337 concepts 291 duration 311 interface 337 interphase 337
Main Index mechanical ties 292 post vulcanisation 36, 73, 100, 101 properties 309 rubber to brass 163 rubber to fabric 241 rubber to metal 319 rubber to plastic 29 silicone to metal 288 silicone to plastic 288 silicone to silicone 287, 308 solid rubber 299 strip 152 test procedures 76, 77 test specimen geometry 76 vulcanisation 37 Bonding agent 93, 98, 99 application methods 98 application to metals 325 applications 93 moisture-sensitive 114 photosensitive 115 prebake 118 preparation 97 re-certification 113, 114 solvent 103 testing 110 thickness 99 waterborne 103 Bonding concepts 291 primers 291, 292 self bonding silicone LR 291 self-bonding silicone rubbers 292 undercuts 291 Bonding mechanism rubber to metal 62 Bonding systems 63, 97, 98 application 129 in situ 249 metal preparation 129 one coat 63
organic solvent-based 127 primer/cover-coat 63 waterborne 98, 103, 127, 128, 129, 130, 131 disadvantages 134 factory usage 128 polymer 134 pre-bake resistance 133 primers 134 product range 134 thickness effects 131 Bonding techniques mechanical 314 Braid Nylon 236 Brass 59, 163 composition 164 electrodeposited 59 Brass adhesion 181 additives 181 boroacylate 181 cobalt-naphthenate 181 compounding 180 neodecanoate 181 silica 180 stearate 181 Brass plating 163 Bridge bearing pads 100 Bush components 102
C Cable connectors 314 Cable industry 309 Calcium hydroxide 142 Car emissions 197 reduction 197 Carbon black 141, 351 easy processing channel 140 high abrasion furnace black 147 Chain scission 141
375
The Handbook of Rubber Bonding Chemical treatments 37, 68 phosphatising 68 Chemisorption 63 Chloropyrimidines 208 Chlorotriazines 208 Cleaning abrasives 321 metal 321, 322 Coagents metallic 60 methacrylate ester 214 triallyl cyanurate 215 trimethylolpropane trimethacrylate 214, 215 Coated components assembly 102 Coatings primer 325, 326 rubber bond 325, 326 Cobalt salts 60, 174, 187, 189, 202, 207 boroacylate 202 metal organic 197, 201 naphthenate 164, 168, 202, 203 stearate 202 Coefficient of restitution 215 Cohesive failure 232 Compounding 59, 60, 223 characteristics 64 isopropyl azodicarboxylate 138 self-bonding 60 soft rubber 59 Compression moulding 299 Compression set 224, 310 Contact pressure 144 Contact temperature 144 Contact time 144 Contamination 118 Copper sulphide 178 Cords brass 202 heavy cabled 249
376
Corona discharge 26 Corona treatment 17, 19 after derivatisation 17 Corrosion 5, 6, 62, 68, 164, 177, 186, 321, 330 anodic 331 by overheating 332 cathodic 331 chemical 332 electrochemical 330 galvanic 5, 74, 332 inhibitors 62, 205 resistance 68 underbond 69 Cotton 244 Coumarone-indene 141 Coupling agents 186, 206, 213 titanate 213 zirconate 213 Cover coat 38 Covulcanisation 140 Crosslinking 62, 141, 329 agents 61 peroxide 213, 216 polysulphide 216 Cryogenic techniques 178 Cryoblasting 27 deflashing 118, 299 Curative 234 peroxide 234 sulphur 234 Cure characteristics 224 Cure rate 224 Curing 223 peroxide 235 Curing times 295, 305 Cyanoacrylate 37, 260 acidic substrates 264 activators 274 adhesives 263
Main Index application methods 275, 277 applications blooming 280, 282, 284 bond line thickness 268 bonding EPDM 277 bonding medical devices 279 bonding natural rubbers 277 bonding nitrile 277 bonding polychloroprene 277 bonding Santoprene 279 bonding silicone rubbers 279 bonding to silicone rubber 270 cleavage loads 267 design considerations 266 environmental resistance 270 excess adhesive 281 external release agents 269 glass bonding 272 glazed appearance 284 health and safety 276 heavy molecular weight 282 hot strength 272 internal release agents 269 joint design 269 liquid 259 low strength 284 no cure 284 peel loads 267 poor adhesion 284 porous substrates 264 pressure systems 275 relative humidity 262, 281 slow cure 281 special requirements 269 syringe systems 276 time systems 275 toughened 265 Cyanoacrylate, curing 260, 261 acidic deposits 263 adhesive 262
cure speed 263 relative humidity 262 temperature 263 volume 262 Cyclohexyl benzothiazole sulphenamide 170
D Debonding 200 Deflashing 73, 118 cryogenic 118, 299 non-cryogenic 73 Degreasing 35, 37 aqueous 35 Delamination 344 Demoulding 119 Dezincification 164, 167, 184, 201, 204 Drying 102, 128, 129
E Ebonite 58, 59 Electrochemical corrosion 200 Electroplating 73 Engine mounts 57, 100, 102 car 319 fluid 57 Environmental aspects 254 of processing 255 pollution limits 127 storage and handling 254 wastes and disposal 255 Ethylene propylene diene rubber 64
F Fabric breakage 345 Fabric separation 348 Fabric treatments 241 aqueous 241
377
The Handbook of Rubber Bonding Fabrics cotton-based 250 Failure 61, 74, 145, 319 bond 235 cement/metal 74 cement/primer 75 cohesive 61, 158, 236 cord separation 344 interfacial 145, 158 rubber 74 rubber/cement 74 Fatigue life 320 Fibre reinforcement 339 Fillers 65, 140 carbon black 65 channel blacks 65 silica 65 Film interfacial 178, 179, 182 Film thickness 101, 106 Fixing. See Adhesion Flocculation 127, 128, 327 Formaldehyde 346 Formulation preparation 222
G Galvanic cell 350 galvanising. See also Zinc coating 11 Gibbs free energy 149 Green strength 137, 138, 148 Green tyres 197, 206 Grit blasting 9 34, 37, 321
H Hardness 224 Shore A 224 Health and safety 109 inorganic lead salts 110 solvents 110
378
Hexamethoxymethylmelamine 206, 207 Hexamethylene tetramine 205, 207 High abrasion furnace black 140 HTV silicones 300 properties 300 Humidity 147 Hydrogenated nitrile rubber 214
I Ignition cables 309 Infra red analysis attenuated total reflection 182 Injection moulding 313 hot runners 307 plastic substrates 306 two colour mould 306 Instruments surface tension pens 128 Instruments, measuring 71 beta backscatter 71 dry film 71 magnetic induction current 71 Interdiffusion 148, 149 Interface 62, 205 dip/rubber 250 dip/textile 252 primer to adhesive 63 primer to metal 62 rubber–metal 205 Internal mixer 223 Banbury 223 Ionic bonds 216 Ionomers 216 Isobutylene-isoprene (butyl) rubber 64 Isocyanate 248, 252
J Joint 147 butt 147
Main Index peeling 147 Joint failure 345
K Keypads silicone 309
L Latex 346 Layover 106, 119 Leaking moulds 118 Liquid rubber 227 bonding 293 silicones 300 Liquid squalene 168
M Maintenance spray equipment 70 Martensite 11 surface 323 Mass spectrometry 165 secondary ion 168 secondary neutral 168 time-of-flight secondary ion 165 Mechanical bonding clamping 310 undercuts 310 Metal 4, 6, 8, 10, 67, 129 activation 67 adhesion 229 alloys 3 carbon black 351 chemical pre-treatment 6, 10 cleaning 129 complexes 208 crystal size 350 degreasing 7
heat treatment 328 mechanical pre-treatment 6, 7 oxide layers 4 pre-treatments 6 preparation 3 removal of grease 8 removal of oil 8 silica 351 smutting 4 substrates 3 surface tension 350 Metal preparation 117 anodising 322 chemical modification 322 faulty 321 plating 322 sheradising 324 Metallic coagents 213, 215, 216 Saret 216 Metallography 3 Methacrylic acid 215 metallic salts 215 Microscopy 19, 198 analytical electron 198 electron 33 scanning electron 19, 168, 202, 208 transmission electron 168, 169 Mill 223 two-roll 223 Modulus 224 Montreal Protocol 126 Mould design 328 Mould release agents 269 external 269 internal 269 Moulding 30, 33, 35, 72 compression 30 extrusion blow 30 high pressure 35 high temperature 33
379
The Handbook of Rubber Bonding incorrect procedures 328 injection 30 operation 118 reaction injection 30 transfer 30 Moulding operation 118 Moulds 72 bonding process 72 designing 72
N Natural rubber 64, 138 Natural rubber or polyisoprene Nitrile rubber 64 Novolak resin 242, 243 Nylons 288
O Organic resins 62 Oscillating-disk rheometer 224 Oxide films 7 Oxides surface 9 Ozone exposure 142
P Paints high solvent 73 Peaning 9 Peel energy 145 Peel strengths 21, 23, 24, 26, 28, 157, 237 Peel test 110, 148, 237, 296, 298, 300, 348 T-peel 348 Phenol-formaldehyde 141 Phosphate treatment 68, 94, 95 coating 10, 323
380
Physical tests 224 Plasticisers 141 reactive super 213 Plastics 12, 13, 31, 40, 41 pre-treatment 12, 13 primers 37 substrate preparation 31 Platen temperatures 119 Polyamide 244 Polybutadiene rubber 138 maleated liquid 141 Polybutylene terephthalate 288 Polycarbonates 32 Polychloroprene 64 Polydimethylsiloxane 293 Polyester 245, 246, 248 textiles 248 Polyethylene 33 Polyethylene terephthalate 245, 288 Polyisocyanate 60, 154 Polyisocyanates Polymer micelles 104 Polymers 62 halogenated 62 polyphenylene oxide 32 solvation 214 Polyphenylene sulphide 32, 288 Polypropylene 33 Polysulphides oligomeric 168 Polyureas 32 Polyurethanes 27, 32 Post vulcanisation bonding 102 cure cycle 102 partial 320 Power transmission belts 339 adhesion systems 346 adhesion testing 347 belt life 341, 346 bonding 339 conveyor belts 345
Main Index failure 340 synchronous 340 V-belts 343, 345 Pre-treatments abrasion 13 chemical 68 chlorination 73 corona 13, 28 etching 28 flame treatment 13, 28 metal 69 phosphatising 68 plasma 28 sodium complexes 13 solvent wipe 13 trichloroisocyanuric acid 13 Prebake resistance 99, 106 Preparation methods substrate 3 Primers 62, 321 application 116 rubber to metal 62 Processing oils 66, 141 aromatic-based 66 ester-based 66 naphthenic 66 Processing techniques 303 inserted parts 303 Product abuse 333 high temperatures 333 mechanical interference 333 oils and solvents 333 service loading 333 Product design effects on bond failure 320 Profile carbon depth 180 Properties interphase 338 Pull out tests 347, 348
Q Quality testing 329
R Rayon 244 Re-certification period 114 Reactive dispersion 230, 235 Recycling 190 Reinforcing material 197 Relative tack 141 Release agent 329 accidental application 329 Resins 205 one component 207 Resorcinol 205, 346 RFL adhesive system See also Resorcinol 60, 241, 244, 246 dip 248 formulation 242 novolak resin 243 preparation 242 Rubber 6, 12, 20, 21, 25, 26, 27, 28, 29, 138 butyl 20 conventional 41 ethylene-propylene 19, 28, 138 ethylene propylene diene 138 fully cured 159 halogenated 25 hydrocarbon 19 liquid 225 natural 20 nitrile 26 peroxide cured 65 pre-treatments 12 shrinkage 6 silicone 26 solid 221 strain crystallising 138
381
The Handbook of Rubber Bonding sulphur cured 65 thermoplastic 156 unsaturated hydrocarbon 20, 28 unvulcanised 150 vulcanized 150, 152 Rubber classification wettability 336 Rubber retention 106, 107, 109 Rubber to brass bonding 198 ageing 177, 200 degradation 200 mechanisms 198 organofunctional silanes 185 Rubber to fabric bonding 247 conveyor belting 247 hose 247 V-belt cord 247 Rubber to metal bonding 205 interface 205 Rubber-to-rubber bonding ageing 146 blooming 145 contact time 144, 159 effects of surface modification 142 filler 140 ozone exposure 142 plasticisers 141 polymeric additives 156 polymers 156 pressure 144 process oils 141 strip thickness 155 surface bloom 159 surface roughness 144, 156 surface treatments 158 tackifiers 141 temperature 144, 156 vulcanisation 139 Rust preventation 73
382
S Safety data sheet 109 Scorch safety 217, 219 Scorch time 224 Sealants 225 EPDM 227 formulation 225, 226 natural rubber 225 polybutadiene rubber 225 polybutadiene glass 229 pumpable 225, 226, 228 Self-bonding HTV applications 301 peroxide curing 301 Self-bonding LR 289 limitations 298 properties 297 Shear adhesion 227 Shelf life 112, 113 Sheradising zinc 324 Shoaf system 245 Shrinkage 339 Silanes 186, 188, 208 polysulphur 187 Silica 141, 206, 207, 351 Silica/resin systems 205 Silicone HTV applications 290 Silicone LR applications 290 Silicone roller 301 Silicone rubber composites 290 applications 290 Silicone rubbers 285 bonding 286 high temperature vulcanising 285 liquid rubber 285 primer 287 processing conditions 308
Main Index undercuts 286 Solvent 126 cleaning 35 dip 8 elimination 126 Spectroscopy 14, 19, 183, 198 auger electron 198, 200, 207, 208, 338 electron energy loss 169 electron for chemical analysis 14 ion scattering 338 proton induced X-ray emission 208 reflection infrared 14 Secondary Ion 338 secondary ion mass 198 x-ray photoelectron 14, 19, 23, 183, 198, 208 Spraying equipment 70, 105 Sputter etching 19 Squalane 171, 175 Squalene 170, 171, 175, 184 Stabilizers 62 viscosity 62 Stainless steel 10 pre-treatment 10 treatments 322 Stearic acid 176 Steel cords 182, 184, 197 brass plated 201 brass-coated 197 Storage stability 130 Strength adhesive 165 interfacial 165, 167 Stress cracking 35 Stresses at interface 320 Styrene butadiene rubber 64, 139 Substrate 3, 69 preparation 101
rigid plastics 69 topography 335 Sulphidation 175, 199 Sulphide copper 165 interfacial 165 interfacial film 177 zinc 165, 174 Sulphurating species 171 Surface analysis 19 Surface primers 270 Suspension bushes 100 Sweep tests 111
T T-peel test 23 Tack 137, 141, 144, 145, 147, 148, 150 relative 137 Tackifiers 141 Tear resistance 220, 221, 315 Temperatures melting transition 305 Tensile elongation 224 modulus 224 properties 219 strength 108, 219, 220, 224 testing 148, 224 Testing 69, 147 autohesion levels 147 dog bone shape 296 tack 147 Test specimens 296 water break 69 Textiles 248 treated 253 Thermal ageing 177, 200 Threshold limit values 110
383
The Handbook of Rubber Bonding Tie coat layer 230, 233 cements 326 Top coat 38 preparation 117 Torque values 224 Treatment 34 abrasive belts 34 acid etching 34 alkali etching 34 chlorination 35 corona discharge 35 high pressure water 34 hydrosonic/ultrasonic cleaning 34 mechanical 68 oxidation 34 phenol 34 plasma treatments 35 satinisation 34 UV treatments 35 Tyre cords 163, 178, 179, 187, 189, 198 coating 191 Tyres 150 precured retreading 150 radial 163 reinforcement 197 retreading 152, 154
U Unvulcanised rubbers bonding 137
V V-belts 249 cut-edge 249 Viscosity 97, 115, 130 reduction 213 sensitivity 114 viscometry 114 Volatile organic compounds 62
384
Vulcanisation 30, 139, 213, 329, 330 autoclave 30 bonding 30 efficient 65 peroxide 213 semi-efficient 65 sulphur 219 system 65
W Waterborne bonding 12 metal preparation 12 Welding 313, 314 Wet blast process 42, 43, 47 degreasing agent 43 phosphate treatment 45 phosphating plant 42 VAQUA pump 43, 44 Wood rosin 141
X X-ray analysis energy dispersive 19, 168, 169, 183 wavelength dispersive 169 X-ray emission particle induced 183
Z ZDA scorch retarded 218 ZDMA scorch retarded 218 Zinc coating 11, 323 Zinc diacrylate 60 Zinc dimethacrylate 60, 214 Zinc sheradising 11 Zirconates 214 neoalkoxy 214